Intergenerational interviews in Negev Arabic: Negotiating lexical, discursive and cultural gaps

Roni Henkin
Abstract

Communication strategies used for conversational repair in Negev Arabic are examined here in a 170,000-word corpus of intergenerational interviews, with university students interviewing their relatives, over age 55, in the Bedouin community in the Negev, southern Israel. Since the traditional language and narrative style of the elderly are largely unfamiliar to the young generation in terms of lexicon, discourse structure and cultural norms, progressivity was often interrupted for purpose of repair. Other-initiated self-repair sequences were particularly frequent: the student asks a metalinguistic or referent-tracking question, or inquires about past customs, and the interviewee explains; additional turns may contain candidate understanding moves and confirmation, before resuming progressivity of the narrative sequence. Gaps were sometimes mediated by a middle-generation ‘broker’ interlocutor. Conversational repair was found to be frequent in facilitating both intelligibility and comprehensibility in these intergenerational conversations.

Keywords:
Publication history
Table of contents

1.Introduction

This study investigates the communication strategy of repair typical of intergenerational interviews in Negev Arabic (NA), as spoken by the Bedouin population in southern Israel. It is intended as a contribution to research on intergenerational communication within a highly collectivist culture, where the urbanized, educated and bilingual young generation is far removed in lifestyle from elderly relatives, especially those living outside the towns.

Section 1 introduces sociolinguistic concepts pertaining to Arabic and its speakers in Israel: dialects, diglossia, bilingualism and Hebraization. NA is sub-classified into two generational varieties. The conceptual framework of the paper is also outlined in Section 1, as couched in conversational repair theory. Section 2 presents the present study – its methodology, research questions and predictions. The findings follow in Sections 35, where I analyze repair strategies in cases of gaps in shared lexical, discursive and cultural knowledge. Section 6 summarizes and suggests issues for further study.

1.1Arabic in Israel

Arabic is well-known for its many dialects. In Israel, Arabic dialects are mostly of the sedentary Syro-Lebanese-Palestinian type. Exceptions are two large groups of Bedouin dialects: one in the north of the country and the other in the south – the Negev.

While a variety of regional dialects is common in many languages, Arabic is additionally characterized by a complex sociolinguistic situation known as diglossia (Bassiouney 2009Bassiouney, Reem 2009Arabic Sociolinguistics. Edinburgh: Edinburgh University Press. DOI logoGoogle Scholar, Chapter 1): a rich diglossic scale ranges from each of the local vernaculars to Modern Standard Arabic (MSA). The range of intermediate levels incorporates elements from MSA and from prestigious areal dialects, within a colloquial base. This continuum reflects rich sociolinguistic variation by dialect, gender, age, education and profession. Diglossic (code-)switching, also known as intra-lingual code alternation, characterizes educated speakers. They maneuver with great flexibility on the diglossic scale according to discourse considerations such as prestige, importance, eloquence, complexity, sophistication and seriousness (Albirini 2011Albirini, Abdulkafi 2011 “The Sociolinguistic Functions of Codeswitching between Standard Arabic and Dialectal Arabic.” Language in Society 40: 537–562. DOI logoGoogle Scholar) whereas uneducated speakers, especially elderly women, have a narrower range at their disposal.

Another sociolinguistic factor that increases variation in some regions of the Arabic-speaking world is bilingualism. This too is a scale, ranging from minimal to maximal fluency in both languages. Some degree of bilingualism characterizes most adult Arabs in Israel, where Hebrew, as the official and hegemonic national language, is necessary for almost all purposes outside the home community. Predominating in bureaucracy, high education and, in fact, all public space discourse, Hebrew enjoys high market value. For Arabic-speakers, “at times and in some domains it is more important than Arabic” (Amara 2013Amara, Muhammad 2013 “Arab Population of Israel. Sociolinguistic Aspects.” Encyclopedia of Hebrew Language and Linguistics Vol. 1, 124–128. Leiden: Brill.Google Scholar, 125) and their level of bilingualism varies according to sociolinguistic parameters such as location of residence, age, gender, education, profession, mobility and contact with Hebrew-speakers.

Monolingual Arabic-speakers too are under the influence of Hebrew, as Hebraization is pervasive in the form of loans, calques and linguistic landscape (Henkin 2010 2010Negev Arabic: Dialectal, Sociolinguistic, and Stylistic Variation. Semitica Viva Series no. 48, Otto Harrassowitz. Wiesbaden.Google Scholar, 4.3.1; Amara 2013Amara, Muhammad 2013 “Arab Population of Israel. Sociolinguistic Aspects.” Encyclopedia of Hebrew Language and Linguistics Vol. 1, 124–128. Leiden: Brill.Google Scholar, 127). Bilingualism and Hebraization are strongest among the young, the highly educated and those working with a majority of Hebrew-speakers. Their in-group code is often their local Arabic dialect with much code-switching to Hebrew.

1.2The Bedouin Negev – Linguistic history and varieties

The Negev desert in southern Israel lies on the historical continuum of Bedouin culture and language, extending from the core, the dialects of the Arabian Peninsula, through south Jordan to Sinai (Henkin 2010 2010Negev Arabic: Dialectal, Sociolinguistic, and Stylistic Variation. Semitica Viva Series no. 48, Otto Harrassowitz. Wiesbaden.Google Scholar, 30; Shawarbah 2012Shawarbah, Musa 2012A Grammar of Negev Arabic. Wiesbaden: Harrassowitz.Google Scholar, 10–14).

The core of contemporary NA sub-classifies to sub-dialects of the three major traditional confederations: Tiyāha (Ty), ʕAzāzmih (Az), and Tarābīn (Tr), and some large tribal factions, such as Ḏ̣ullām (Ḏ̣l) and Gdērāt (Gd; Shawarbah 2012Shawarbah, Musa 2012A Grammar of Negev Arabic. Wiesbaden: Harrassowitz.Google Scholar, 4). These vary in minor but distinctive details. All the varieties are mutually intelligible in discourse concerning everyday matters. However, much of the lexicon pertaining to the traditional nomadic lifestyle varies somewhat in the sub-dialects and is by no means intelligible to all, as we shall see.

This original Bedouin stock was augmented in the course of the 19th century by many ex-peasant families known as ḥumrān (Ḥm) ‘reddish’ or ‘light skinned folk’, mostly from Egypt and the Gaza strip, who originally came as hired laborers and then settled with their families. They generally did not intermarry with the veteran Bedouin sumrān ‘dark skinned folk’, and have retained distinct ex-peasant dialects that did, however, undergo a high degree of Bedouinization through accommodation to the dialects of their respective patron families. In 2000 they numbered 19,028 (Henkin 2008Henkin, Roni 2008 “Diachronic and Synchronic Accommodation in Negev Arabic.” In Between the Atlantic and Indian Oceans: Studies on Contemporary Arabic Dialects. Proceedings of the 7th AIDA Conference, ed by. Stephan Procházka, and Veronika Ritt-Benmimoun, 237–250. Lit Verlag.; Shawarbah 2012Shawarbah, Musa 2012A Grammar of Negev Arabic. Wiesbaden: Harrassowitz.Google Scholar, 6–7).

In the pre-State era the Negev Bedouin led a semi-nomadic lifestyle, relying on the skills of ex-peasant incomers for some agriculture. The 1948 war left just about 12% of the original 95,000 Negev Bedouin in Israel. They were relocated in a 1,000 sq. km. area east and northwest of Beer-Sheva, named the Enclosure Zone (Hb. sayig). Following the end of military rule over the Negev in 1966, seven government-planned Bedouin towns were constructed in this zone, and these are populated by about half of the Negev Bedouin population (Shawarbah 2012Shawarbah, Musa 2012A Grammar of Negev Arabic. Wiesbaden: Harrassowitz.Google Scholar, 4–5) that now numbers 292,288 (Population and Immigration Authority, July 2022). The rest live in 46 tribal encampments and villages, lacking in infrastructure, water and electricity, known officially as ‘the unrecognized villages’. Since 2002 eleven of these villages gained official recognition.

In the villages and encampments, especially outside the main Northern Negev zone, the traditional cultural patterns are retained to a wide extent, as are the sub-dialects. The Bedouin towns, however, are more heterogeneous: on the one hand, communal services such as clinics, pharmacies, banks and commercial centers are joint for members of all families; on the other hand, the confederations ‘live together but separately’, avoiding intermarriage, especially between Bedouin and non-Bedouin, and maintaining a degree of socio-spatial and socio-cultural separation as well as political competition (Shawarbah 2012Shawarbah, Musa 2012A Grammar of Negev Arabic. Wiesbaden: Harrassowitz.Google Scholar, 6). So, mosques and pre-schools located inside those urban neighborhoods serve a majority of the dominant families. Family-based residence patterns also dominate the educational system, whereby both registration of pupils and employment of teachers and principals are determined by area of residence. In general, an increase in the number of tribal schools in the Negev promotes the retention of tribal dialects (Shawarbah 2012Shawarbah, Musa 2012A Grammar of Negev Arabic. Wiesbaden: Harrassowitz.Google Scholar, 7).

In the traditional patriarchal society, multi-generational families lived together, but today many families are split, with the elderly remaining in their original location, while their offspring have moved to a town. So the town-dwellers have lost daily contact with their elderly relatives outside the towns. This factor is crucial for the issue of intergenerational communication. Their different lifestyles and geographic distance make present day communication challenging, as we will see.

Palestinian Arabic dialects enter the Negev through several routes: women from Hebron, Gaza, Nazareth, Ramle-Lod and other areas who marry into Negev families pass their native dialects on to their children; also a very large (though nowadays steadily decreasing) corps of teachers is from outside the Negev (Shawarbah 2012Shawarbah, Musa 2012A Grammar of Negev Arabic. Wiesbaden: Harrassowitz.Google Scholar, 7), and as they teach in their native dialects, the young generation is familiar with these from school. Dialectal leveling in the direction of prestigious urban dialects, such as those of Jerusalem or Haifa, is promoted by exposure to mass media and internet resources. MSA and other elevated varieties also spread in this channel, especially via religious programs.

The extent of levelling, bilingualism and Hebraization are very different in the two main demographic varieties that concern us in this study, Traditional Negev Arabic (TNA) and Young Negev Arabic (YNA).

1.2.1Traditional Negev Arabic

The term Traditional Negev Arabic, introduced by Cerqueglini (2015b 2015b “Spatial Frames of Reference in Traditional Negev Arabic: Language-to-Cognition Correlation.” Cognitive Processing 16 (1): 185–188. DOI logoGoogle Scholar, 185), subsumes the various sub-dialects of NA as spoken by the elderly, primarily those that are illiterate and monolingual. This group includes effectively all elderly women and those men whose education, if any, consists of some elementary classes, and whose contact with the external Hebrew-speaking environment is limited. Their everyday in-group speech, at the colloquial ‘lower’ end of the diglossic scale, retains sub-dialectal characteristics from times when all the tribes lived separately, with predominant illiteracy, restricted intermarriage and very little contact outside the community (particularly for women).

The diglossic scale of these speakers is limited – formal situations may cause some elevation with koineized forms and formulaic quranic phrases. Their bilingual scale depends on past working relations with the Hebrew-speaking society. The oral narrative tradition of TNA-speakers, in contrast, is rich and highly distinctive. Much of its lexicon originates in the traditional Bedouin poetry, known as Nabaṭi poetry. This poetic tradition is largely interdialectal (Henkin 2010 2010Negev Arabic: Dialectal, Sociolinguistic, and Stylistic Variation. Semitica Viva Series no. 48, Otto Harrassowitz. Wiesbaden.Google Scholar, Chapter 5), common to the vast area of Bedouin culture. It is understood by TNA-speakers of all sub-dialects. But many items, particularly those pertaining to the traditional nomadic lifestyle, are unintelligible to the young.

1.2.2Young Negev Arabic

The language of the young generation, YNA, is more strongly influenced than TNA by diglossia, bilingualism and Hebraization. Additional sociolinguistic processes that are common across the Arabic-speaking world and affect YNA more than TNA are leveling (homogenization) and koineization (Henkin 2010 2010Negev Arabic: Dialectal, Sociolinguistic, and Stylistic Variation. Semitica Viva Series no. 48, Otto Harrassowitz. Wiesbaden.Google Scholar, 4.3.1). Leveling involves avoiding ‘broad’ or specific dialectal features in order to facilitate communication with speakers of other dialects (Bassiouney 2009Bassiouney, Reem 2009Arabic Sociolinguistics. Edinburgh: Edinburgh University Press. DOI logoGoogle Scholar, 117–122). Koineization is the process of becoming more similar to a koiné, which is composed of elements from various dialects and, once formed, spreads as a prestigious areal dialect, a kind of lingua franca (Siegel 1985Siegel, Jeff 1985 “Koines and Koineization.” Language in Society 14 (3): 357–378. DOI logoGoogle Scholar). In YNA, impact of koinés, such as Jerusalem Arabic, is felt in formal or out-group discourse, augmented by influence of non-Bedouin speakers residing or teaching in the Negev.

All these processes are on the rise in the present day setting of urbanization, compulsory education (with many of the teachers speaking urban or rural Palestinian dialects), exposure to MSA and external koiné dialects in the media. As most of the work places outside the Bedouin towns are Hebrew-speaking, as are all the universities in Israel, adult bilingualism develops rapidly in conditions of immersion. Consequently, the in-group code among YNA-speakers is characterized by Arabic-Hebrew codeswitching, the intensity of which varies in accordance with discourse situations.

Example (1) shows two-dimensional codeswitching typical of YNA conversations: a 23-year-old female undergraduate student of the Gdērāt clan, interviewed by a female undergraduate student of the Ḏ̣ullām, talks about her experience in higher education. Superscript H and E mark lexical items that are Hebrew or elevated respectively:

(1)

f-al-bidēyih kān kašēh H,11.Vowel length in Hebrew words is marked only when pronounced by Arabic speakers. This is because Arabic speakers tend to lengthen stressed Hebrew vowels considerably, due to the link in Arabic between vowel length and stress. In Hebrew, stress (not vowel length) is phonemic, so penultimate stress is marked, e.g. keʔílu ‘as if, like’, beséder ‘okay’. laʔinnih kān fīh eh… [ḥājiz as-safāh H]E. innih at-taʕlīm b-as-safāh H al-ʕivrīt H waná as-safāh H tabaʕtī ʕaṛabiy, kān šwayyih ṣaʕib yaʕniy. innih beʕētsem H jāyīn aḥna natʕallam mawḏ̣ūʕ mʕayyan fa- E bnalga ḥājizēn.

In the beginning it was difficultH, because there was eh… the [barrier of the languageH]E, that the studies were in the HebrewH languageH and me, my languageH was Arabic, it was a little difficult, yaʕniy.22.The discourse particle yaʕniy is not translated when functioning for hedging or self-repair (‘I mean, like’). It is briefly discussed under Extract (3). That actuallyH we come to study a specific subject butE find two obstacles.

Bilingual codeswitching is prominent here, with six tokens of Hebrew lexemes, none of which are necessitated by lexical gaps in NA, as they all have common NA equivalents: Hb. kašeh ‘difficult’ is in fact subsequently re-formulated in NA as ṣaʕib; the three tokens of Hb. safah ‘language’ all replace NA luġah; Hb. ʕivrit ‘Hebrew’ is NA ʕubrāniy. Hebrew discourse markers and fillers such as beʕétsem ‘actually’ in (1), and even more notoriously, Hb. keʔílu ‘like’ and beséder ‘okay’, are hallmarks of YNA.

As for diglossic switching, a ‘rise’ on the diglossic scale, typical of academic interview style, is represented here by the metaphorical abstract construct mawḏ̣ūʕ mʕayyan ‘a specific (academic) subject’ and the educated discourse marker fa- ‘but then; then however’. Some elevation may be due to the observer’s paradox, whereby “to obtain the data most important for linguistic theory, we have to observe how people speak when they are not being observed” (Labov 1972Labov, William 1972 “Some Principles of Linguistic Methodology.” Language in Society 1: 97–120. DOI logoGoogle Scholar, 113). In two studies on the peer-interview style of female YNA-speaking students (Henkin 2015 2015 “Codeswitching Patterns in Negev Bedouin Students’ Personal Interviews.” Zeitschrift für Arabische Linguistik 61: 5–34.Google Scholar, 2016 2016 “Functional Codeswitching and Register in Educated Negev Arabic Interview Style.” Bulletin of the School of Oriental and African Studies: 1–26. DOI logoGoogle Scholar), I found that the presence of recording equipment, even a small cellphone recorder, often caused a tendency to elevate speech.

1.2.3Flexibility of age group definitions and boundaries

As is conventional in dialectology, most NA research has focused on the traditional variety, now known as TNA. This is true of the studies by Blanc (1970)Blanc, Haim 1970 “The Arabic Dialect of the Negev Bedouins.” Proceedings of the Israeli Academy of sciences and Humanities. Jerusalem, 112–150.Google Scholar, Henkin (2010) 2010Negev Arabic: Dialectal, Sociolinguistic, and Stylistic Variation. Semitica Viva Series no. 48, Otto Harrassowitz. Wiesbaden.Google Scholar, Alatamin (2011)Alatamin, Mohammad 2011 “Ethnographic and Linguistic Aspects of the Negev Arabic Lexicon.” PhD diss. Ben-Gurion University of the Negev. [Hebrew] and Shawarbah (2012)Shawarbah, Musa 2012A Grammar of Negev Arabic. Wiesbaden: Harrassowitz.Google Scholar. In these, no age limit is specified for the choice of informants.

YNA is targeted only in Cerqueglini (2020) 2020 “Dialectal, Gender-Based, and Cross-Generational Variation in Negev Arabic Spatial Representations.” Coyote Papers: Working Papers in Linguistics, Linguistic Theory at the University of Arizona. http://​hdl​.handle​.net​/10150​/641485 on the spatial language of the young, and my own studies based on personal interviews, recorded by students interviewing other students (Henkin 2015 2015 “Codeswitching Patterns in Negev Bedouin Students’ Personal Interviews.” Zeitschrift für Arabische Linguistik 61: 5–34.Google Scholar, 2016 2016 “Functional Codeswitching and Register in Educated Negev Arabic Interview Style.” Bulletin of the School of Oriental and African Studies: 1–26. DOI logoGoogle Scholar, 2019 2019 “Inclusive Generic Person in Women’s Discourse in Israeli Hebrew and Negev Arabic.” Journal of Pragmatics 150: 53–74. DOI logoGoogle Scholar). Since research on age groups is so new, the age limits are not yet set by independent linguistic criteria. So, the range of the age groups varies somewhat, depending on the issue(s) targeted and on the sample of informants that were recorded. Table 1 shows the ranges according to the available publications.

Table 1.Age group ranges in current NA research
Source YNA TNA
Cerqueglini 2015aCerqueglini, Letizia 2015aObject-Based Selection of Spatial Frames of Reference in aṣ-Ṣāniˁ Arabic. Pisa University Press.Google Scholar, 46; 2015b 2015b “Spatial Frames of Reference in Traditional Negev Arabic: Language-to-Cognition Correlation.” Cognitive Processing 16 (1): 185–188. DOI logoGoogle Scholar, 185 under 40 over 60
Cerqueglini 2016 2016 “Etymology, Culture and Grammaticalisation: A Semantic Exploration of the Front/Back Axis in Traditional Negev Arabic.” In Arabic Varieties: Far and Wide. Proceedings of the 11th International Conference of AIDA – Bucharest, 2015, ed. by George Grigore, and Gabriel Bițună. Bucharest: Editura Universității din București.Google Scholar, 176 not mentioned over 60
Cerqueglini 2020 2020 “Dialectal, Gender-Based, and Cross-Generational Variation in Negev Arabic Spatial Representations.” Coyote Papers: Working Papers in Linguistics, Linguistic Theory at the University of Arizona. http://​hdl​.handle​.net​/10150​/641485, 43–44 under 40 over 70
Henkin 2019 2019 “Inclusive Generic Person in Women’s Discourse in Israeli Hebrew and Negev Arabic.” Journal of Pragmatics 150: 53–74. DOI logoGoogle Scholar, 58 18–28 over 60
Present study (see Table 2) 18–25; 26–35 over 55

Fluctuation is inevitable, at least at this stage. Any boundaries set along the scale are fuzzy and rather arbitrary due to involvement of additional factors, such as education and profession. So, a 50-year-old illiterate woman’s speech can show the characteristics of TNA, whereas the speech of an educated man of the same age would be more typical of Middle Negev Arabic (MNA), the range between TNA and YNA (see Table 2).

1.3Conceptual framework

Three basic concepts employed in this paper are negotiation of meaning, communication strategies and conversational repair. Negotiation of meaning is the joint cooperative construction of meaning that produces meaningful conversation (Varonis and Gass 1985Varonis, Evangeline Marlos, and Susan Gass 1985 “Non-native/Non-native Conversations: A Model for Negotiation of Meaning.” Applied Linguistics 6 (1): 71–90. DOI logoGoogle Scholar). Communication strategies are “mutual attempts of two interlocutors to agree on a meaning in situations where the requisite meaning structures do not seem to be shared” (Tarone 1980Tarone, Elaine 1980 “Communication Strategies, Foreigner Talk, and Repair in Interlanguage.” Language Learning 30: 417–431. DOI logoGoogle Scholar, 420). A major communication strategy for negotiation of meaning is conversational repair, which “refers to a set of practices for dealing with problems of hearing, speaking, and understanding talk” (Bolden 2012Bolden, Galina B. 2012 “Across Languages and Cultures: Brokering Problems of Understanding in Conversational Repair.” Language in Society 41 (1): 97–121. DOI logoGoogle Scholar, 100, based on Schegloff et al. 1977Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks 1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 53 (2): 361–382. DOI logoGoogle Scholar, 361). Repair occurs in ‘non-understanding routines’ which are operationally defined as “those exchanges in which there is some overt indication that understanding between participants has not been complete” (Varonis and Gass 1985Varonis, Evangeline Marlos, and Susan Gass 1985 “Non-native/Non-native Conversations: A Model for Negotiation of Meaning.” Applied Linguistics 6 (1): 71–90. DOI logoGoogle Scholar, 73). When the normally smooth course of conversation, which may be visualized as a horizontal line or path, is disrupted as a result of a loss of footing by a participant, this may trigger signal of a need for negotiation of meaning and repair, which suspends progressivity and may be visualized as a vertical drop. Progressivity continues as soon as the repair sequence has been completed (Varonis and Gass 1985Varonis, Evangeline Marlos, and Susan Gass 1985 “Non-native/Non-native Conversations: A Model for Negotiation of Meaning.” Applied Linguistics 6 (1): 71–90. DOI logoGoogle Scholar, 72ff.; Kitzinger 2013Kitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Hoboken, NJ: John Wiley and Sons.Google Scholar, 238ff.). This interaction may be seen as a balance between progressivity and intersubjectivity, or a shared perception of reality between interlocutors. Repair shows where “the principle of intersubjectivity […] has trumped the principle of progressivity” (Heritage 2007Heritage, John 2007 “Intersubjectivity and Progressivity in Person (and Place) Reference.” In Person Reference in Interaction: Linguistic, Cultural and Social Perspectives, ed. by N. J. Enfield, and Tanya Stivers, 255–280. Cambridge: Cambridge University Press. DOI logoGoogle Scholar, 260ff.).

Initial repair research, such as the pioneering study by Schegloff et al. (1977)Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks 1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 53 (2): 361–382. DOI logoGoogle Scholar, focused on settings where the interlocutors are on equal footing, communicating in their shared native language, and sharing also culture and background knowledge. In this setting, negotiation of meaning may be expected to be relatively minor: “among NS-NS speakers, shared background is maximized, and, as a result, there is little need for linguistic negotiation” (Varonis and Gass 1985Varonis, Evangeline Marlos, and Susan Gass 1985 “Non-native/Non-native Conversations: A Model for Negotiation of Meaning.” Applied Linguistics 6 (1): 71–90. DOI logoGoogle Scholar, 86). Repair is more crucial in NS-NNS discourse including foreigner talk, as studied by Long (1983)Long, Michael 1983 “Native Speaker/Non-native Speaker-Conversation and the Negotiation of Comprehensible Input.” Applied Linguistics 4 (2): 126–141. DOI logoGoogle Scholar and Vasseur et al. (1996)Vasseur, Marie-Thérèse, Peter Broeder, and Celia Roberts 1996 “Managing Understanding from a Minority Perspective.” In Achieving Understanding: Discourse in Intercultural Encounters. ed. by Katharina Bremer, Celia Roberts, Marie-Therese Vasseur, Margaret Simonot, and Peter Broeder, 65–108. London: Longman.Google Scholar, and in NNS-NNS interactions, such as EFL and ELF settings (Varonis and Gass 1985Varonis, Evangeline Marlos, and Susan Gass 1985 “Non-native/Non-native Conversations: A Model for Negotiation of Meaning.” Applied Linguistics 6 (1): 71–90. DOI logoGoogle Scholar; Pitzl 2005Pitzl, Marie-Luise 2005 “Non-understanding in English as a Lingua Franca: Examples from a Business Context.” Vienna English Working Papers 14: 50–71.Google Scholar; Watterson 2008Watterson, Matthew 2008 “Repair of Non-understanding in English in International Communication.” World Englishes 27 (3–4): 378–406. DOI logoGoogle Scholar). In such settings, interlocutors often lose conversational footing due to gaps in lexicon, grammar, pragmatics and cultural concepts. They learn to compensate by questioning particular utterances and/or requesting conversational help (Varonis and Gass 1985Varonis, Evangeline Marlos, and Susan Gass 1985 “Non-native/Non-native Conversations: A Model for Negotiation of Meaning.” Applied Linguistics 6 (1): 71–90. DOI logoGoogle Scholar, 73). So, “the more involved non-native speakers are in a dyad, the more time interlocutors will spend moving down, or in other words, in the negotiation of meaning, rather than moving forward, in other words, in the progression of the discourse” (Varonis and Gass 1985Varonis, Evangeline Marlos, and Susan Gass 1985 “Non-native/Non-native Conversations: A Model for Negotiation of Meaning.” Applied Linguistics 6 (1): 71–90. DOI logoGoogle Scholar, 83). Non-understandings, in general and particularly under these circumstances, can arise at three different planes (Watterson 2008Watterson, Matthew 2008 “Repair of Non-understanding in English in International Communication.” World Englishes 27 (3–4): 378–406. DOI logoGoogle Scholar, 386–388): intelligibility (at the level of a single word or word group), comprehensibility (at the level of a whole sentence or utterance) and interpretability (at the level of underlying intent or purpose).

Major distinctions in repair research are, first, between self-repair and other-repair and, second, between self-initiated repair and other-initiated repair (Schegloff et al. 1977Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks 1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 53 (2): 361–382. DOI logoGoogle Scholar; Kitzinger 2013Kitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Hoboken, NJ: John Wiley and Sons.Google Scholar, 230 ff.; Couper-Kuhlen and Selting 2018Couper-Kuhlen, Elizabeth, and Margret Selting 2018Interactional Linguistics: Studying Language in Social Interaction. Cambridge: Cambridge University Press.Google Scholar, 116ff.). In equal footing conversations, self-repair has been found to be most common, since other-repair, by a peer listener, could be face-threatening to the speaker being corrected. For the same reason and in the same settings, self-initiated self-repair, which suspends just the turn progressivity, is more common than other-initiated self-repair, which suspends progressivity of a sequence.

Intercultural conversations and other cases of asymmetrical knowledge status, however, are characterized by other-initiated self-repair, often in four basic steps of a ‘non-understanding routine’ (Varonis and Gass 1985Varonis, Evangeline Marlos, and Susan Gass 1985 “Non-native/Non-native Conversations: A Model for Negotiation of Meaning.” Applied Linguistics 6 (1): 71–90. DOI logoGoogle Scholar) or a ‘repair sequence’ (Kitzinger 2013Kitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Hoboken, NJ: John Wiley and Sons.Google Scholar; Dingemanse et al. 2014Dingemanse, Mark, Joe Blythe, and Tyko Dirksmeyer 2014 “Formats for Other-initiation of Repair across Languages: An Exercise in Pragmatic Typology.” Studies in Language 38.1: 5–43. DOI logoGoogle Scholar):

  1. trigger (T), i.e. the segment that causes non-understanding

  2. indicator (I), i.e. the signal showing lack of understanding

  3. response (R), i.e. acknowledgement of the non-understanding and, usually, attempts to repair it

  4. reaction to response (RR) indicating that the problem is now resolved (optional)

When relations are asymmetrical in terms of ‘the control of information’, with one interlocutor less competent or knowledgeable than the other, the trigger is typically an item in the competent speaker’s turn that baffles the less competent speaker, who then initiates the indicator in the consequent turn, since “interlocutors seeking particular information are more likely to initiate negotiation than interlocutors holding the information” (Varonis and Gass 1985Varonis, Evangeline Marlos, and Susan Gass 1985 “Non-native/Non-native Conversations: A Model for Negotiation of Meaning.” Applied Linguistics 6 (1): 71–90. DOI logoGoogle Scholar, 87). The trigger may, for example, pertain to recognitional reference (Heritage 2007Heritage, John 2007 “Intersubjectivity and Progressivity in Person (and Place) Reference.” In Person Reference in Interaction: Linguistic, Cultural and Social Perspectives, ed. by N. J. Enfield, and Tanya Stivers, 255–280. Cambridge: Cambridge University Press. DOI logoGoogle Scholar) or problems of hearing, among other possibilities. Intercultural settings, however, are characterized by additional problems pertaining to linguistic or cultural gaps.

Repair initiators or indicators are often questions, presenting the speaker as knowing less than the addressee (Dingemanse et al. 2014Dingemanse, Mark, Joe Blythe, and Tyko Dirksmeyer 2014 “Formats for Other-initiation of Repair across Languages: An Exercise in Pragmatic Typology.” Studies in Language 38.1: 5–43. DOI logoGoogle Scholar, 9), but this is not the only format. The indicator can be placed on a ‘continuum of procedures’, which is a scale of explicitness or strength ranging from non-verbal or minimal vocal ‘symptoms’, progressing to open formats, such as a question word (how?), repetition of the trigger (reprise) with or without a question word; and finally reaching maximally explicit or closed signals, including understanding checks such as ‘you mean…?’ (Vasseur et al. 1996Vasseur, Marie-Thérèse, Peter Broeder, and Celia Roberts 1996 “Managing Understanding from a Minority Perspective.” In Achieving Understanding: Discourse in Intercultural Encounters. ed. by Katharina Bremer, Celia Roberts, Marie-Therese Vasseur, Margaret Simonot, and Peter Broeder, 65–108. London: Longman.Google Scholar, 76 ff.; Pitzl 2005Pitzl, Marie-Luise 2005 “Non-understanding in English as a Lingua Franca: Examples from a Business Context.” Vienna English Working Papers 14: 50–71.Google Scholar, 54 ff.; Couper-Kuhlen and Selting 2018Couper-Kuhlen, Elizabeth, and Margret Selting 2018Interactional Linguistics: Studying Language in Social Interaction. Cambridge: Cambridge University Press.Google Scholar, 140ff.). In repair sequences, and also within one turn, progression has been shown from weak or open to strong or closed formats, reflecting advancing knowledge and responsibility for negotiation of meaning.

2.Present study

The present study investigates intergenerational conversations recorded in the year 2004.33.Two exceptions are Extract (6), recorded in 2000, and Extract (2), recorded in 2005. The interviewers, all university students, were free to choose initiating questions and topics, the one methodological restriction being that only elderly relatives were to be interviewed. This restriction prevented potential mixing between speakers from different family-groups, which could cause dialect mixing and gaps that are dialectal rather than intergenerational. It also ensured a common degree of closeness and affinity between the participants in a society where family relations are very important. In general, the elderly are very keen to pass on their heritage to young kin and are proud to see them engaged in academic studies. The students, on their part, acknowledged the elderly as repositories of historical-cultural knowledge, and were clearly motivated to access some of this heritage, making the most of their advantage as kin. The speakers were classified into four age groups, as seen in Table 2.

Table 2.Age group ranges in the present study
Group Age Variety
1 undergraduate student interviewers up to 25 YNA
2 graduate students 26–35 YNA
3 intermediate generation, mostly parents of Group 1 36–55 MNA
4 interviewees over 55 TNA

Groups 1 and 2 were the YNA-speaking interviewers. Their turns were short, averaging 4.47 words per turn. With a total of about 11,220 words over 2,507 turns, they produced 6.6% of the total word number (approximately 170,000). Group 3, speakers of MNA, were only present in a few of the interviews, where their role was typically as intermediaries between the old and the young. I ask three research questions:

RQ1:

What kinds of gaps characterize these intergenerational conversations?

RQ2:

How are these gaps handled?

RQ3:

How is this setting special (or not) with respect to repair?

The current sociolinguistic context was expected to necessitate frequent repair. I consider it akin to NS-NNS discourse: the elderly TNA-speakers represent native speakers of the ‘authentic’ dialect, which the YNA-speaking interviewers are trying to access, as their own language has changed due to urbanization, education, koineization, mass media and communal bilingualism. Communication gaps were predicted primarily in the lexical domain, as this is considered the most dynamic level of language, highest in borrowability (Matras 2009Matras, Yaron 2009Language Contact. Cambridge: CUP. DOI logoGoogle Scholar, 6.2.2). Volatility of lexical items is manifested not only in expansion, by creation or adoption of new lexemes, but also in loss of old lexemes that have been replaced. So, it is not surprising that over time the lexicon of NA has changed more than other linguistic domains. Intergenerational conversations are thus prone to lexical gaps, jeopardizing intelligibility and necessitating repair work, especially for TNA lexicon not understood by the young.

Discursive or structural gaps were also predicted, challenging interpretability and pertaining to cohesion via linkage of discourse elements. From my personal acquaintance with the TNA oral narrative tradition I can testify to the difficulty in participant tracking for listeners unfamiliar with the events narrated. Finally, cultural gaps were predicted in view of the rapid changes in lifestyle the society is undergoing and unfamiliarity of the young with codes and norms of the past, as appearing in the speech of the elderly.

Other-initiated repair strategies expected to prevail included metalinguistic questions, candidate understanding questions and paraphrasing, as found in repair literature elsewhere (see below, 3.1–3.2). The findings presented below are classified under the three main gap-types found: lexical, discursive (or structural) and cultural.

3.Lexical repair

We will see two types of other-initiated or solicited indicator questions, both common in our corpus: metalinguistic questions and candidate understanding questions.

3.1Metalinguistic questions

Metalinguistic44.The term is used in three different ways, including ‘language about language’ (DeKeyser 2009DeKeyser, Robert M. 2009 “Cognitive-Psychological Processes in Second Language Learning.” In The Handbook of Language Teaching, ed. by Michael H. Long, and Catherine J. Doughty, 119–138. Blackwell Publishing Ltd. DOI logoGoogle Scholar, 122–123), which is the sense used here, for which some use the term ‘metalingual’. questions (MLQs) and responses explicitly target lexical items which crop up in a speaker’s text and are problematic for another, who initiates a direct, maximally explicit question and receives a response, often an explanation or re-formulation of the trigger speech turn. An example, following X as the trigger lexeme, is ‘what does X mean?’. Such direct metalinguistic queries at the extreme right end of the explicitness continuum (Vasseur et al. 1996Vasseur, Marie-Thérèse, Peter Broeder, and Celia Roberts 1996 “Managing Understanding from a Minority Perspective.” In Achieving Understanding: Discourse in Intercultural Encounters. ed. by Katharina Bremer, Celia Roberts, Marie-Therese Vasseur, Margaret Simonot, and Peter Broeder, 65–108. London: Longman.Google Scholar, 88), are the strongest other-initiated closed formats that explicitly specify the source of trouble.

Similar, but less explicit, are formats of reprise, with or without a question word (‘X?’ or ‘how X?’).55.In the repair literature, these are distinct from the metalinguistic format, which only includes explicit formulations, and are located in the middle range of the explicitness scale (Vasseur et al. 1996Vasseur, Marie-Thérèse, Peter Broeder, and Celia Roberts 1996 “Managing Understanding from a Minority Perspective.” In Achieving Understanding: Discourse in Intercultural Encounters. ed. by Katharina Bremer, Celia Roberts, Marie-Therese Vasseur, Margaret Simonot, and Peter Broeder, 65–108. London: Longman.Google Scholar, 75; Couper-Kuhlen and Selting 2018Couper-Kuhlen, Elizabeth, and Margret Selting 2018Interactional Linguistics: Studying Language in Social Interaction. Cambridge: Cambridge University Press.Google Scholar, 197). But I include all of these under the category of metalinguistic questions, defined as referring to a linguistic item that is being questioned, albeit differing in explicitness and strength. In the corpus, MLQs were always asked by the young interviewer, hence B, and answered by the elderly interviewee, hence A.66.In the numbered extracts from here on, A is of age group 4, B is of group 1 or 2 and an occasional C is a middle generation member of group 3.

Extract (2) The Bedouin and the mare

A’s life story begins with a description of her hard work as a laborer at a citrus plantation in order to sustain the family when they lived in an isolated encampment. But then they moved into the large town Rahat so that the children could go to school, and as soon as her husband saw some money coming in, he married a second wife. This painful detail is embedded, as typically in Bedouin culture, in a proverb, which implicitly criticizes the first actions the Bedouin man performs upon becoming rich:

(2)
1A al-bádawiy laṃ-ibyabṭar […] yā byatxayyal yā byajjawwaz.
The Bedouin (man), when he gets rich […], he either byatxayyal [T] or (re-)marries.
2B ēš yaʕniy byitxayyal?
What does byitxayyal mean? [I]
3A yaʕniy byišriy lih faṛás.
It means he buys himself a mare [R]
4B āh.
Yes. [RR]
5A alḥīn hū baṭáṛ, tajawwaz jīzih.
Now he got rich, he re-married.
[4F.57ʕAṢ/1F.SAṢ-Tr]77.The bracketed reference at the end of each extract specifies, in this order, age group, gender, age (when known) of the interviewee, initials, followed by a slash and the corresponding specifications for the interviewer and, hyphenated, the sub-dialect, which is always shared for all participants in the interview. So, the reference in Extract (2) designates a Group 4 female, 57-year old ʕAṢ, as interviewed by the female undergraduate relative SAṢ, both speakers of the Tarābīn sub-dialect. A third speaker, as in (5), is similarly designated in between.

The pragmatic point in 1A is indirect criticism of the androcentric equation: a mare and a wife, in this order, are two equally valued alternatives for a man to purchase and show off his new wealth. But a simple lexical gap triggers B’s MLQ, as she does not understand the first of the alternatives, buying a mare (2B). The self-repair response (3A) paraphrases the opaque verb, and lexical intelligibility is now acknowledged (4B). But from A’s point of view, deviation to the proverbial mare was totally missing her pragmatic point. So, having duly answered the ‘irrelevant’ MLQ, she regains control as the narrator, and goes back to her point. This time, in 5A, she stresses the act of re-marrying, so that it would not be missed again. She achieves this by means of an intensifying grammatical form known as the cognate complement, a deverbal noun following its source verb – in this case the deverbal noun jīzih ‘marriage’ derived from the verb tajawwaz ‘marry’: tajawwaz jīzih lit. ‘he married a marriage’.

Extract (3) Free will?

Here B raises the question of free will in choosing a life partner, an important issue for the young generation today:

(3)
1B witjawwaztiy b-wafgiyyitkiy yaʕniy, kiʔīluw H inhum saʔalōkiy gabaḷ-ma titjawwazay, walla kēf?
And you got married of your own free will, yaʕniy? LikeH they asked you before you got married, or what?
2A laʔ, wafgiyyih mā fīh, aná kunt kumān bidīlih, baddalaw biy, uwkunt ṯalāt-t-iyyām waná imʕaggdih l-al-ʕirīs.
No, there is (was?) no free will. I was also a bidīlih [T], they exchanged for me, and I walked three days on foot to the bridegroom.
3B kīf yaʕniy bidīlih?
How, yaʕniy, bidīlih? [I]
4A bidīlih… abūy ajjawwaz… jāb abūy waḥdih kimān ṯēnyih, iwṯalāt-t-iyyām waná mʕaggdih.
bidīlih… my father married… my father brought a second one (an exchange bride for himself). And I walked for three days. [R]

[4F.60FAJ/1F.JAJ-Az]

The initial question in 1B opens with an improvised abstract term wafgiyyih ‘free will’ then glossed in self-initiated self-repair with the everyday verbal phrase ‘they asked you’. Note the intensive hedging, typical of the students in this corpus: here we see both Hebrew and NA ‘like’-hedges, and the alternative question format ‘or what?’, cross-linguistically common in other-initiated repair, as in German oder? (Dingemanse et al. 2014Dingemanse, Mark, Joe Blythe, and Tyko Dirksmeyer 2014 “Formats for Other-initiation of Repair across Languages: An Exercise in Pragmatic Typology.” Studies in Language 38.1: 5–43. DOI logoGoogle Scholar, 37).

In 2A this question is answered with the absolute ‘no’, and then more explicitly ‘there is (was?) no free will’.88.This utterance is formulated in present tense, and is ambiguous with respect to temporal designation. It could be generic present, pertaining to this question in general in this culture; or, more likely, it refers to the time in question, many years ago. In this case it shows non-past reference used for habitual past in the TNA narrative style (see 4.2). A adopts the improvised term wafgiyyih ‘free will’ and adds an piece of information which in her judgement explains the irrelevance of free will in her case – the fact that she was a bidīlih lit. ‘exchange itemfm’, i.e. an exchange bride. The term bidīlih, in the feminine form with no masculine counterpart, denotes a bride given as part of an exchange deal, a common practice in the traditional society, saving dowry expenses on both sides. In this particular case her father took a bride for himself and gave his daughter in marriage to one of his new bride’s menfolk. To ensure intelligibility, A then glosses the traditional term ‘exchange bride’ by means of the corresponding verb ‘they exchanged for me’. This may be seen as self-repair for the benefit of B. Again it implies that in this four-member arrangement the chances for free will of the couples are minimal, as marriage choices are mainly a political or family matter. Having duly answered the question of free will, which for her generation is not really an issue, A resumes progressivity with her main point – the grueling three-day march to the groom’s home.

But the concept of exchange bride is obviously still not clear to B. It triggers an other-initiated question (3B), opening with a manner question word kīf ‘how’ followed by repetition of the trigger item bidīlih. This format, known as ‘partial repeats of the trouble-source turn plus a question word’ is the third of the five types of formats for other-initiation of repair, located in the middle-range of strength (Dingemanse et al. 2014Dingemanse, Mark, Joe Blythe, and Tyko Dirksmeyer 2014 “Formats for Other-initiation of Repair across Languages: An Exercise in Pragmatic Typology.” Studies in Language 38.1: 5–43. DOI logoGoogle Scholar, 17ff., based on Schegloff et al. 1977Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks 1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 53 (2): 361–382. DOI logoGoogle Scholar, 368).

yaʕniy in 3B is ambiguous: its original lexical meaning, as an imperfect (3rd person msc. sg.) verb, is ‘it means’, and it was accordingly translated as a full verb in cases where the meaning of a lexical item was in question (see for example, Extract [2]: 2B, 3A; Extract [4]: 2B, 3A and more). In this full lexical sense, 3B could be interpreted as ‘how does it mean an exchange item?’.99.A similar function of ‘mean’ has been noted in English. Together with rising intonation, ‘y’mean’ serves as a device for epistemic downgrading: it mitigates the assertiveness of an utterance by indicating that what follows is not claimed as a statement of fact but rather as an inference about another person’s state of mind (Dingemanse et al. 2014Dingemanse, Mark, Joe Blythe, and Tyko Dirksmeyer 2014 “Formats for Other-initiation of Repair across Languages: An Exercise in Pragmatic Typology.” Studies in Language 38.1: 5–43. DOI logoGoogle Scholar, 27). Here, however, yaʕniy seems to be functioning in its equally prevalent role as a self-repair or hedging discourse marker ‘I mean, like’, as also in 1B, in Example (1) above and in many more cases below where it was not translated (see fn. 2).1010.For an enlightening analysis of yaʕniy in Cairene women’s monologues, see Marmorstein (2016)Marmorstein, Michal 2016 “Getting to the Point: The Discourse Marker yaʕni (lit. “It Means”) in Unplanned Discourse in Cairene Arabic.” Journal of Pragmatics 96: 60–79. DOI logoGoogle Scholar.

The response in 4A opens with a repetition, as is very frequent in repair responses (Watterson 2008Watterson, Matthew 2008 “Repair of Non-understanding in English in International Communication.” World Englishes 27 (3–4): 378–406. DOI logoGoogle Scholar, 391), followed by an elaboration. Then come the details of the specific deal. But this detail too was a side issue for the interviewee. So, here again, as soon as the repair is completed for her, she gets back to her point, her three-day march.

Extracts (4)–(5) Drinking regimes

As A was describing to her grandson and his parents the layout of the major historical wells of their tribal territory, she referred to a camel drinking regime lexicalized with the verb ġabb ‘drink every other day’ (1A).1111.See Alatamin (2011Alatamin, Mohammad 2011 “Ethnographic and Linguistic Aspects of the Negev Arabic Lexicon.” PhD diss. Ben-Gurion University of the Negev. [Hebrew], 164), Shawarbah (2012Shawarbah, Musa 2012A Grammar of Negev Arabic. Wiesbaden: Harrassowitz.Google Scholar, 236, 391). As this verb is totally obsolete and also etymologically opaque it triggered a MLQ (2B) and a paraphrasing repair response (3A).

(4)
1A […] ubiyġibbuw an-nās kān.
[…] people would biyġibbuw. [T]
2B ēš yaʕniy biyġibbuw?
What does biyġibbuw mean? [I]
3A yaʕniy alyōm waṛádaw min bēkir muhum wērdīn
It means (if) today they went to the well, tomorrow they do not go to the well [R]
[4F.68JAK/2M.33SAK-Gd]

Again, yaʕniy is ambiguous. I have translated with the more plausible function for this context, as a full verb. But, at least in 3A, it could be the filler or discourse marker.

3.2Metalinguistic candidate understanding questions

Once his MLQ in Extract (4) was answered and intersubjectivity restored, B was better able to cope with new lexemes in the topic of watering regimes, and indeed the next trigger follows immediately: realizing she had underestimated the number of days the camels actually went without water, A performs a self-initiated self-repair:

(5)
1A lā bitġibb ál-bil, ṣudug, ubtirbiʕ, ubtirbiʕ kān ibtirbiʕ
they do not bitġibb the camels, actually, they btirbiʕ, btirbiʕ, they used to btirbiʕ [T]
2B ibtirbiʕ yaʕniy aṛbaʕ-t-iyyām
btirbiʕ means [they drink every] four days? [I]
3A aṛbaʕ-t-iyyām uṯalāt-t-iyyām, uhī mā ḏ̣āgat miy, ál-bil
four days and three days, and they have not tasted water, the camels [Ri]
4C min gillt-ál-miy yaʕniy min gillt-al-miy
Because of scarcity of water, yaʕniy because of scarcity of water. [Rii]

[4F.68JAK/2M.33SAK/3F.IAK-Gd]

By now the trigger verb tirbiʕ in A1, very clearly and transparently derived from the numeral arbaʕ ‘four’ and uttered in correction of the two-day drinking alternation routine, is fairly manageable. So, in 2B a hypothesis is formulated (Vasseur et al. 1996Vasseur, Marie-Thérèse, Peter Broeder, and Celia Roberts 1996 “Managing Understanding from a Minority Perspective.” In Achieving Understanding: Discourse in Intercultural Encounters. ed. by Katharina Bremer, Celia Roberts, Marie-Therese Vasseur, Margaret Simonot, and Peter Broeder, 65–108. London: Longman.Google Scholar, 82, 84), that it is four days of thirst. This format, a reprise and question ‘Does X mean Y? is classified as a form of candidate understanding, located in the middle of the explicitness continuum (this term, however, includes also cases that are not metalinguistic). Candidate understandings are relatively non-disruptive:

The resulting flatter epistemic cline can then be levelled with less effort – for instance by responding with a minimal confirmation when the candidate understanding turns out to be correct. This, in turn, permits sequences in which the repair is less obstructive to the progressivity of the conversation(Dingemanse et al. 2014Dingemanse, Mark, Joe Blythe, and Tyko Dirksmeyer 2014 “Formats for Other-initiation of Repair across Languages: An Exercise in Pragmatic Typology.” Studies in Language 38.1: 5–43. DOI logoGoogle Scholar, 28)

The answer in 3A is indeed a confirmation response, consisting of repetition and additional elaboration. At this stage B’s mother steps in as a broker or intermediary (Bolden 2012Bolden, Galina B. 2012 “Across Languages and Cultures: Brokering Problems of Understanding in Conversational Repair.” Language in Society 41 (1): 97–121. DOI logoGoogle Scholar, 99) to offer more explanatory details (4C).

Extracts (4)–(5) show the enterprise of meaning negotiation to be a joint task, balancing intersubjectivity and progessivity (Heritage 2007Heritage, John 2007 “Intersubjectivity and Progressivity in Person (and Place) Reference.” In Person Reference in Interaction: Linguistic, Cultural and Social Perspectives, ed. by N. J. Enfield, and Tanya Stivers, 255–280. Cambridge: Cambridge University Press. DOI logoGoogle Scholar, 260–261). They also show the principle of proceeding according to strength, from reprise and question word of the less knowledgeable (‘what does X mean?’, Example [4]: 2B) to the candidate understanding format showing more knowledge and responsibility (‘does X mean Y?’, Example [5]: 2B; Dingemanse et al. 2014Dingemanse, Mark, Joe Blythe, and Tyko Dirksmeyer 2014 “Formats for Other-initiation of Repair across Languages: An Exercise in Pragmatic Typology.” Studies in Language 38.1: 5–43. DOI logoGoogle Scholar).

3.3Rare breakdowns

The response in a repair sequence is usually offered willingly, as a cooperative move. But exceptions do occur, where one side loses patience.

Extract (6) Pregnant but different

(6)
1A an-nyāg mitēliy aṯ-ṯintēn
The she-camels were both mitēliy [T]
2B mitēliy? [I]
3A ʕášṛa! al-bádawiy byáʕarif al-mitliy ēš! kānak
Pregnant! A Bedouin knows what mitliy means! If you are a…[R]

[4M.75MAS/1M.25AAS-Hm]

1A contains the word mitēliy, pl. of mitliy, a passive participle from the verb ‘to follow’. It designates a she-camel followed by her young offspring, but serves also for a fecundated she-camel, as distinct from a fecundated horse mare (baṭḥa‏ or bāṭiḥ) and sheep, goat or cow (ʕášṛa; Shawarbah 2012Shawarbah, Musa 2012A Grammar of Negev Arabic. Wiesbaden: Harrassowitz.Google Scholar, 368, 374, 435). The narrator’s undergraduate son, who had grown up in a town, did not know this term, since the distinct TNA lexemes for pregnancy of each animal are all subsumed under ʕášṛaʔ in YNA. He inquired about it (2B) in a ‘trouble-presenting repeat’ of the closed, maximally explicit formats (Vasseur et al. 1996Vasseur, Marie-Thérèse, Peter Broeder, and Celia Roberts 1996 “Managing Understanding from a Minority Perspective.” In Achieving Understanding: Discourse in Intercultural Encounters. ed. by Katharina Bremer, Celia Roberts, Marie-Therese Vasseur, Margaret Simonot, and Peter Broeder, 65–108. London: Longman.Google Scholar, 86; Pitzl 2005Pitzl, Marie-Luise 2005 “Non-understanding in English as a Lingua Franca: Examples from a Business Context.” Vienna English Working Papers 14: 50–71.Google Scholar, 56) that “displays knowledge of what was heard but professes lack of knowledge as to how to interpret it” (Dingemanse et al. 2014Dingemanse, Mark, Joe Blythe, and Tyko Dirksmeyer 2014 “Formats for Other-initiation of Repair across Languages: An Exercise in Pragmatic Typology.” Studies in Language 38.1: 5–43. DOI logoGoogle Scholar, 34; Couper-Kuhlen and Selting 2018Couper-Kuhlen, Elizabeth, and Margret Selting 2018Interactional Linguistics: Studying Language in Social Interaction. Cambridge: Cambridge University Press.Google Scholar, 145; see also Kitzinger 2013Kitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Hoboken, NJ: John Wiley and Sons.Google Scholar, 249).

Clearly upset at this ignorance, the father grudgingly provided the leveled variant ʕášṛaʔ (3A), which for him is only suitable for sheep and is demeaning when used of a she-camel. His angry retort contained offensive language, implying that B was not fit to be a member of the collective, and which I omitted out of respect for both participants. Of all the examples in this paper, this is the only one that looks like a breakdown in communication. Although intelligibility was reached and progressivity was not suspended for long, the problem is topicalized, both speakers were offended and intersubjectivity was jeopardized.

4.Discursive gaps

In addition to problems due to lexical gaps, problems in intergenerational communication can be caused by unusual characteristics of TNA narrative style. Two such problems pertain to cohesion of the text.

4.1Participant tracking

As typical of oral Bedouin narrative (Henkin 2010 2010Negev Arabic: Dialectal, Sociolinguistic, and Stylistic Variation. Semitica Viva Series no. 48, Otto Harrassowitz. Wiesbaden.Google Scholar, 116), participant tracking in the narratives is often challenging, with change of protagonists extremely under-specified. A sequence of three consecutive verbs such as ‘(he) said…, (he) said…, (he) said.…’ could well refer to three different protagonists, with no necessary antecedents in the preceding co-text.1212.This narrative technique, used for dramatizing, has been labeled ‘ping pong style’, as it resembles an energetic ping pong game (Henkin 2010 2010Negev Arabic: Dialectal, Sociolinguistic, and Stylistic Variation. Semitica Viva Series no. 48, Otto Harrassowitz. Wiesbaden.Google Scholar). Moreover, gāḷ lit. ‘(he) said’ serves as a narrative discourse marker in the sense of ‘and then’, which adds to confusion in its interpretation. Interpretation relies heavily on cultural familiarity with these stories, traditionally told in the men’s hosting tent to an audience consisting of in-group members. All the listeners know the tribal history and appreciated stylistic conventions, such as zero-pronoun (omitting the subject pronoun of verbal sentences) for dramatic effects. In retelling these narratives to a young audience, however, repair is often needed in order to enable participant tracking.

Extract (7) Allied ambush

The Gdēriy narrator recounts an episode from the 19th century wars between the Gdērāt and Tarābīn confederations over Negev territories. Since the Tarābīn were the most powerful confederation by far, the Gdērāt allied up with the Gīsiyyih peasant group from the Hebron hills. The allies set up a double ambush, with the Gdērāt warriors at a well called Bīr Kuḥlih:

(7)
1A alḥīn al-Gdīṛāt imgaʕʕdīn gaʕd ʕa-bīr Kuḥlih, int xābir Bīr Kuḥlih, al-bīr kān an-nās ibtašṛab minnih. gaʕʕduw gaʕád l-at-Taṛābīn wat-Taṛābīn f-al-gāʕah hēḏiy min ʕind al-Ligiyyih ba-hal-gīʕēn hēdiy […]. gāḷ: walliy iymīl ʕalīku – aḏbaḥawh.
Now the Gdērāt had laid an ambush at Bīr Kuḥlih, you know Bīr Kuḥlih, the well at which the people would drink. They lay in ambush for the Tarābīn and the Tarābīn were at the plain away from al-Ligiyyih in those plains […]. He said: “And whoever comes by – kill him.” [T]
2C biygūluw l-al-Gīsiyyih?
They said to the Gīsiyyih? [broker-I]
3A biygūluw l-al-Gīsiyyih […] halḥīn ad-dinya šōbih wsāṛaw, al-gīmēn xābrih ġāṛ al-bīr, kull-ma tmīl ṣurbah, hēḏiy bitḥērib aṣ-ṣurbah hiḏīk tašṛab witjiy itḥērib wal-ʕaṭšānīn iygōṭruw yašṛabaw. bitmīl ṣurbah ibyaḏbaḥawha, bitmīl ṣurbah ibyaḏbaḥawha. gāḷaw: ēš ad-daʕáwah mālaw ṣurbtēn umā ṭallaw.
They said to the Gīsiyyih [R] […] Now it was hot and the enemy parties only knew the well, every time a small group went aside, this party would fight and this small group would drink and come back to fight, and those who were thirsty would go to drink. A small group would go aside and they (the enemies in ambush) would kill them, a(nother) small group would go aside and they would kill them. They [T] said “What’s going on? Two groups have gone aside and not reappeared.
4B at-Taṛābīn?
The Tarābīn? [I]
5A at-Taṛābīn.
The Tarābīn. [R]
[4F.68JAK/3M.BAK/2M.33SAK-Gd]

Upon naming the well, which is critical in her story, the narrator verifies that it is known to her grandson B by means of a self-initiated comprehension check, a well-known communication strategy when a potential trigger is present (Varonis and Gass 1985Varonis, Evangeline Marlos, and Susan Gass 1985 “Non-native/Non-native Conversations: A Model for Negotiation of Meaning.” Applied Linguistics 6 (1): 71–90. DOI logoGoogle Scholar, 78). But, typically, she does not specify who the speaker is in the verb ‘he said’ or, alternatively, to whom the quote was directed (1A).

Sensing that participant tracking is becoming tricky, A’s son (B’s father) steps in with a broker candidate understanding (2C), hypothesizing that was the Gdērāt instructing their allies the Gaysiyyih. This is confirmed by A via total reprise (3A) and progressivity is resumed. A little later B asks a candidate understanding question regarding identity of the referent of 4B, hypothesized to be the Tarābīn, upon realizing that their men were going to drink and not coming back. 5A is affirmation via a repair strategy of repetition.

Extract (8) But who drove you out?

The attempts to secure referent tracking are not always immediately successful, however. Repair sequence (8) lasts over 11 turns.

(8)
1A kunna al-maskin tabaʕna, lamma aṭlaʕawna mn-at-Tahámih
We used to live [at Tahámih], when they [T] made us leave the Tahámih…
2B min ṭallaʕkuw?
Who made you leave? [I]
3A ṭallaʕawna, ṛaḥḥalawna gāḷaw inna kumān
They made us leave, made us wander away, also told us…. [R]
4B bass min yaʕniy alliy ḥáka līkuw inkuw taṛḥalaw?
But who yaʕniy was it who told you to wander away? [I]
5A ġayyaṛawha innaʔ
They changed it (our location) for us [R]
6B bass yaʕniy aná basʔal: min alliy yaʕniy ḥáka līkuw inkuw ġāṛ taṛḥaluw yaʕniy min hal-manṭagah?
But yaʕniy I’m just asking: Who was it yaʕniy who told you that you have to get out yaʕniy, from this area? [I]
7A ḥakáw inna
They told us… [R]
8B yaʕniy intuw walla… min yaʕniy?
yaʕniy you, or… who, yaʕniy? [I]
9A al-ḥukūmah gāluw inna.
The government told us. [R]
10B al-ḥukūmah ṭallaʕōkuw? kēf ṭallaʕōkuw? b-al-guwwih walla yaʕniy b-ittifāg?
The government made you leave? How did they made you leave? By force or yaʕniy by agreement? [I]
11A lā, b-ittifāg, ṭallaʕawna.
No, by agreement, they made us leave. [R]
[4F.60FAJ/1F.JAJ-Az]

B’s recurring attempts to identify the referent only succeed after four identity questions of the strategy involving category-specific interrogative ‘who….?’ framed by repetition of the trigger verb (Kitzinger 2013Kitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Hoboken, NJ: John Wiley and Sons.Google Scholar, 249) as in 2B or a paraphrase thereof (4B, 6B). Her stepmother A seems not to have recognized the problem – for her the referent was probably obvious. She repeats the verb, glosses it with a near synonym and wants to resume progressivity by adding what was said to them, but still with no referent (3A). When asked again more explicitly (4B, 6B) she paraphrases with other verbs (5B, 7B). Only in 8A does she reveal the referent, the government. B then asks a candidate understanding question (10A), hypothesizing that ‘the government told us’ also answers the question of who made them leave. The repair continues to another aspect of the event, the degree of force (10B, 11A).

Note that all of B’s questions are hedged and downtoned. There are 6 tokens of the discourse marker yaʕniy ‘like, sort of’.1313.Other downtoners and hedging elements in B’s turns include bass ‘only, just’ (4B, 6B and Extract [10]: 1B). Also šwayyih ‘a little’ (Example [1]; Extract [10]: 1B), Hebrew kiʔīluw H‘like’ in Example (1).

4.2Tense use

In Labov’s (2006) 2006 “Narrative Pre-construction.” Narrative Inquiry 16 (1): 37–45. DOI logoGoogle Scholar model of narrative structure the Orientation segment presents the pre-plotline setting, preceding the chain of narrative events. This segment naturally pertains to past time. A prominent feature of Orientation segments in TNA narrative style is non-past tenses, serving to portray the whole narrative as concrete and actual. In TNA the main grammatical means for this are the present-future tense and nominal predication.1414.In Semitic languages, nominal predication enables verbless sentences, such as ana (I) walad (boy) ‘I am a boy’. In such a setting it is not always clear if the scene described is historical or contemporary – the listener is supposed to derive the time frame from pragmatic and cultural knowledge and according to the genre (Henkin 2010 2010Negev Arabic: Dialectal, Sociolinguistic, and Stylistic Variation. Semitica Viva Series no. 48, Otto Harrassowitz. Wiesbaden.Google Scholar, 7.6, 143 ff.).1515.Sometimes a single past time marker is inserted somewhere within the Orientation, not necessarily at the beginning, and that specifies the past setting for the entire segment. But this is not common in TNA. So far in the present article, when translating Orientation segments referring to a habitual past, I have used the past tense for the sake of comprehensibility, although the interviewees often used non-past forms (cf. Extract [5]: 1A; Extract [7]: 1A).

Extract (9) Today? Long ago?

This narrative segment begins with no contextual background clues to specify the scene as historic: the first word binṭušš ‘we wander’ is a present-future tense verb (1A):

(9)
1A binṭušš maʕ ál-ġanaṃ kull an-nahāṛ sārḥīn, ṣārḥīn maʕ ál-ġanaṃ, iwbingōṭir. aḥna wal-jēš ixwān, […] kull an-nahāṛ f-nuṣṣ baʕáḏ̣na.
We wander with the sheep all day long out in the pasture, out in the pasture, and we go off. We and the army are brothers […] all day long mingled together. [T]
2B alyōm, zamān?
Today? Long ago? [I]
3A zimān yaʕniy mn-aṃṃiy jābat ʕĀyših.
Long ago, yaʕniy, when my mother had Ayših. [R]
[4F.60Fʕ/1F.JAḤ-Ḥm]

In the absence of the pragmatic knowledge necessary to interpret the time frame, namely that good army-shepherd relations characterized the distant past, this triggers B’s question (2B) and A’s response, consisting of partial reprise and elaboration (3A).

Extract (10) You didn’t use to pray?!

When it is B who sets the temporal frame for eliciting information about past customs, it is usually explicit past, as typical of YNA conventions. So in 1B we find the past tense verb ‘was’ (and note here, too, the ample hedging):

(10)
1B al-ʕīd, yaʕniy kīf kān ʕindikuw šiy ismih ʕīd, yaʕniy iḥkiy lay išwayyih ʕinnih bass.
The feast day, yaʕniy how was this thing called a feast day with you, yaʕniy just tell me a little about it.
2A aḥkiy ilkiyyih, awwal, lā binṣalliy.
I’ll tell you, in the old days, we do not pray (=didn’t pray). [T]
3B mā kuntuw tṣalluw??
You didn’t use to pray?? [I]
4A mā kunna nṣalliy. miḥna ʕērfīn inṣalliy. ʕērfīn, yōminhum biyhill ah-hlēl, ibyiḥsbuw išhuṛṛah.[…] bingūl: hall, Ramaḏ̣ān, ṣūmuw. [R]
We didn’t use to pray. We do not (=did not) know how to pray. We know (=knew) that when the new moon rises, they calculate the months. […] We (used to) say: “New moon is here, Ramadan, start fasting.”

[4F.63ʕĠ/1F.JAJ-Az]

But the answer (2A) is formulated with a non-past tense, as typical of TNA. This triggers B’s candidate understanding question, basically a reprise but ‘repairing’ with a compound tense which is favored in YNA for non-ambiguous habitual past (3B). A’s response acknowledges by repetition, including the compound tense (4A).1616.The pre-verbal negator in 2A is also exclusive to TNA and may confuse the young. B reformulates with the general verbal negator , a change then confirmed by A (3B, 4A). But she formulates the following five verbs in the non-past, resorting to the TNA form after a single repair switch.

5.Cultural gaps

Often the generational gap is not lexical, grammatical or discursive. A cultural gap, pertaining to norms that are alien to today’s young generation and jeopardizing interpretability, may also trigger a repair sequence. If in Extract (10) the young student was shocked to hear that people in the past did not know how to pray, the immediately following Extract (11) continues A’s description of the former Ramadan customs, and B is once again shocked, this time to hear of gender mixing in celebrations of the past, when folk Islam was practised, before the advent of orthodox Islam to the Negev.

Extract (11) In the same tent??

(11)
1A biygawṭruw il-bēt aš-šēx f-ál-ʕaṛab, biygōṭruw ʕindih ibyiltammaw. ibyiltammaw al-wild, ibyiltamman al-bnittiy.
They (would) go to the tent of the Sheikh of the tribe, they (would) go to his place and get together. The boys (would) get together, the girls (would) get together [T]
2B fī nafs al-bēt?
In the same tent? [I]
3A fī nafs al-bēt. alliy bitjiy aṃṃ burgaʕ, alliy bitjiy ʕalēha širšibān.
In the same tent. A girl comes (could come) wearing a veil, another comes with colored strips. [R]
4B h ixtilāṭ bēnkin iwbēn aš-šabāb yaʕniy?
Is (was) there mixing between you and the young men yaʕniy? [I]
5A h. fīh. hāḏ̣a ʕādiy, mā fīh fargiyyih, itgūl ixwān.
There is (was). There is (was). It is (was) normal. There is (was) no difference, you could say (like) brothers. [R]
[4F.63ʕAĠ/1F.JAJ-Az]

Going back to the question of tense use, we see her too, as in Extract (10), the present tense serves for past in all of A’s turns. But here it does not cause any comprehensibility problem. Moreover, in 4B this use of the present (by means of a nominal clause) is taken up by B, although it is atypical of YNA-speakers in their own in-group discourse. This seems to be a case of up-accommodation, with the student adjusting her speech to that of the interviewee. The topic of accommodation in these interviews deserves an independent study.

If we consider Extract (11) as a clarification request, it conforms to statements that in epistemic terms, candidate understandings and clarification requests evidence cultural knowledge, graded higher than low-level lexical repair.

6.Summary and what next

TNA is a very conservative dialect, most prominently so in the domain of lexicon and stylistic features characterizing its oral narrative-poetic genres. These features are particularly challenging to outsiders. YNA, in contrast, is characterized by three sociolinguistic processes which are all absent in TNA – leveling, koineization and Hebraization – all conditioned by social factors of age, urbanization and education. As expected, these differences were found to create considerable communication gaps in intergenerational conversations.

While much of the repair research, based on equal-footing conversations, has stressed prevalence of self-initiated repairs, we focused on other-initiated repair, followed by self-repair or repair by means of an intermediary. Predominance of this sequence is clearly associated with the hierarchical status of the interlocutors: since the young acknowledge the elderly as a repository of cultural knowledge and since the side needing information typically initiates the repair, the interviewers are expected to ask, not wait for the narrators to repair on their own initiative.

In answer to RQ1, gaps were found to involve lexical, discursive and cultural knowledge. Lexical knowledge pertained to intelligibility, whereas discursive problems, such as difficulties in participant tracking and temporal setting, pertained to comprehensibility. Cultural gaps pertain to interpretability, the ability to ‘digest’ the information, integrate it in the student’s knowledge of the world.

Regarding RQ2, the typical indicator was a question by the student, who is the interlocutor in need of information, and this was answered by the elderly. In the case of lexical repair pertaining to intelligibility, the indicators were often direct metalinguistic questions (‘What does X mean?’), which could be followed by candidate understanding questions (‘you mean…?’). As for comprehensibility questions pertaining to participant tracking (‘who told you to leave?’) or to general cultural codes (‘in the same tent?’), these were characterized by downtoning and hedging, with ample interspersion of yaʕniy ‘I mean, like’. In fact, hedging and polite downtoning characterized the students’ style in general. The speech of the elderly, in direct contrast, was rich in emphasizers and intensifiers such as waḷḷāhi ‘by God’. This interactional contrast testifies to the relations: tentativeness and respect by the young speakers contrasting with authority and assertiveness on the part of the elders. Another indicator of politeness and deference is total absence of the extreme low-level open-class initiators, namely interjections of the ‘huh’-type and the generic question word ēh ‘what’, typologically very common in repair sequences. Since this format ‘puts the onus on A to figure out the trouble and solve it’ (Dingemanse et al. 2014Dingemanse, Mark, Joe Blythe, and Tyko Dirksmeyer 2014 “Formats for Other-initiation of Repair across Languages: An Exercise in Pragmatic Typology.” Studies in Language 38.1: 5–43. DOI logoGoogle Scholar, 9), which would be a sign of disrespect here, B always takes responsibility to formulate the request in a relatively closed format. It is also striking that the students’ speech throughout the large corpus was characterized by almost total absence of codeswitching and Hebraizing, which are the most prominent markers of their in-group style. This avoidance of Hebrew shows socio-cultural recognition, possibly subconscious, of the fact that Hebrew has no place in conversations with the elderly.

As for RQ3, the repair preferences which emerged seem to me to be special in several respects. Some, such as preference for other-initiated questions of both metalinguistic and cultural kinds, are typical of asymmetrical and intercultural interaction. Others, such as the discursive questions, target unique characteristics of TNA narrative style, such as the elliptic ‘ping pong’ technique and present time reference to past time.

Many gaps that surface in intergenerational NA conversations are due to changes over time in the dialect, particularly at the lexical level, but also tense use, discourse structure, narrative style. Besides these linguistic factors, socio-cultural differences in lifestyle, orthodoxy, ideology are also influential and not easy to tease apart from linguistic or dialectal factors.

Finally, I believe that the present two- or three-generational family setup was an integral factor in determining the repair style, which may differ in other settings even within the same community. I agree that “negotiation may strengthen the social or interpersonal dimension of an interaction by acting as a conversational continuant” (Varonis and Gass 1985Varonis, Evangeline Marlos, and Susan Gass 1985 “Non-native/Non-native Conversations: A Model for Negotiation of Meaning.” Applied Linguistics 6 (1): 71–90. DOI logoGoogle Scholar, 86), elaborated later by Pitzl (2005Pitzl, Marie-Luise 2005 “Non-understanding in English as a Lingua Franca: Examples from a Business Context.” Vienna English Working Papers 14: 50–71.Google Scholar, 57–58). In this holistic approach, negotiation of meaning, jointly construed as a co-operative project, promotes interpersonal rapport and bonding, and a sense of fulfilment unites all participators upon reaching understanding. An extended family setting is expected to promote feelings of mutual affection, affinity and responsibility, especially in a collectivist society that places a high value on family relations.

This preliminary study raises many issues for future investigation of intergenerational communication in the Negev Bedouin community: differences in cross-gender and same gender interviews; mutual accommodation and complementary roles; hedging by the young and intensifying by the elderly, as a function of status and gender; also mutual endearing vocatives, which are exceptionally abundant in NA (Henkin 2010 2010Negev Arabic: Dialectal, Sociolinguistic, and Stylistic Variation. Semitica Viva Series no. 48, Otto Harrassowitz. Wiesbaden.Google Scholar, 192 ff.), reflecting much appreciation and respect to the elders, and fondness to young kin. Pursuing these topics across a range of conversational genres, participant constellations, and situational contexts, would supply us with a fuller picture of intergenerational communication in this unique minority society, maneuvering between a strong tradition and rapid changes.

Funding

This study was supported by The Israel Science Foundation (personal grant 804/06).

Acknowledgements

I thank Eihab Abu-Rabiah for some linguistic and non-linguistic information on the Bedouin communities of the Negev.

Notes

1.Vowel length in Hebrew words is marked only when pronounced by Arabic speakers. This is because Arabic speakers tend to lengthen stressed Hebrew vowels considerably, due to the link in Arabic between vowel length and stress. In Hebrew, stress (not vowel length) is phonemic, so penultimate stress is marked, e.g. keʔílu ‘as if, like’, beséder ‘okay’.
2.The discourse particle yaʕniy is not translated when functioning for hedging or self-repair (‘I mean, like’). It is briefly discussed under Extract (3).
3.Two exceptions are Extract (6), recorded in 2000, and Extract (2), recorded in 2005.
4.The term is used in three different ways, including ‘language about language’ (DeKeyser 2009DeKeyser, Robert M. 2009 “Cognitive-Psychological Processes in Second Language Learning.” In The Handbook of Language Teaching, ed. by Michael H. Long, and Catherine J. Doughty, 119–138. Blackwell Publishing Ltd. DOI logoGoogle Scholar, 122–123), which is the sense used here, for which some use the term ‘metalingual’.
5.In the repair literature, these are distinct from the metalinguistic format, which only includes explicit formulations, and are located in the middle range of the explicitness scale (Vasseur et al. 1996Vasseur, Marie-Thérèse, Peter Broeder, and Celia Roberts 1996 “Managing Understanding from a Minority Perspective.” In Achieving Understanding: Discourse in Intercultural Encounters. ed. by Katharina Bremer, Celia Roberts, Marie-Therese Vasseur, Margaret Simonot, and Peter Broeder, 65–108. London: Longman.Google Scholar, 75; Couper-Kuhlen and Selting 2018Couper-Kuhlen, Elizabeth, and Margret Selting 2018Interactional Linguistics: Studying Language in Social Interaction. Cambridge: Cambridge University Press.Google Scholar, 197). But I include all of these under the category of metalinguistic questions, defined as referring to a linguistic item that is being questioned, albeit differing in explicitness and strength.
6.In the numbered extracts from here on, A is of age group 4, B is of group 1 or 2 and an occasional C is a middle generation member of group 3.
7.The bracketed reference at the end of each extract specifies, in this order, age group, gender, age (when known) of the interviewee, initials, followed by a slash and the corresponding specifications for the interviewer and, hyphenated, the sub-dialect, which is always shared for all participants in the interview. So, the reference in Extract (2) designates a Group 4 female, 57-year old ʕAṢ, as interviewed by the female undergraduate relative SAṢ, both speakers of the Tarābīn sub-dialect. A third speaker, as in (5), is similarly designated in between.
8.This utterance is formulated in present tense, and is ambiguous with respect to temporal designation. It could be generic present, pertaining to this question in general in this culture; or, more likely, it refers to the time in question, many years ago. In this case it shows non-past reference used for habitual past in the TNA narrative style (see 4.2).
9.A similar function of ‘mean’ has been noted in English. Together with rising intonation, ‘y’mean’ serves as a device for epistemic downgrading: it mitigates the assertiveness of an utterance by indicating that what follows is not claimed as a statement of fact but rather as an inference about another person’s state of mind (Dingemanse et al. 2014Dingemanse, Mark, Joe Blythe, and Tyko Dirksmeyer 2014 “Formats for Other-initiation of Repair across Languages: An Exercise in Pragmatic Typology.” Studies in Language 38.1: 5–43. DOI logoGoogle Scholar, 27).
10.For an enlightening analysis of yaʕniy in Cairene women’s monologues, see Marmorstein (2016)Marmorstein, Michal 2016 “Getting to the Point: The Discourse Marker yaʕni (lit. “It Means”) in Unplanned Discourse in Cairene Arabic.” Journal of Pragmatics 96: 60–79. DOI logoGoogle Scholar.
11.See Alatamin (2011Alatamin, Mohammad 2011 “Ethnographic and Linguistic Aspects of the Negev Arabic Lexicon.” PhD diss. Ben-Gurion University of the Negev. [Hebrew], 164), Shawarbah (2012Shawarbah, Musa 2012A Grammar of Negev Arabic. Wiesbaden: Harrassowitz.Google Scholar, 236, 391).
12.This narrative technique, used for dramatizing, has been labeled ‘ping pong style’, as it resembles an energetic ping pong game (Henkin 2010 2010Negev Arabic: Dialectal, Sociolinguistic, and Stylistic Variation. Semitica Viva Series no. 48, Otto Harrassowitz. Wiesbaden.Google Scholar). Moreover, gāḷ lit. ‘(he) said’ serves as a narrative discourse marker in the sense of ‘and then’, which adds to confusion in its interpretation.
13.Other downtoners and hedging elements in B’s turns include bass ‘only, just’ (4B, 6B and Extract [10]: 1B). Also šwayyih ‘a little’ (Example [1]; Extract [10]: 1B), Hebrew kiʔīluw H‘like’ in Example (1).
14.In Semitic languages, nominal predication enables verbless sentences, such as ana (I) walad (boy) ‘I am a boy’.
15.Sometimes a single past time marker is inserted somewhere within the Orientation, not necessarily at the beginning, and that specifies the past setting for the entire segment. But this is not common in TNA.
16.The pre-verbal negator in 2A is also exclusive to TNA and may confuse the young. B reformulates with the general verbal negator , a change then confirmed by A (3B, 4A).

References

Alatamin, Mohammad
2011 “Ethnographic and Linguistic Aspects of the Negev Arabic Lexicon.” PhD diss. Ben-Gurion University of the Negev. [Hebrew]
Albirini, Abdulkafi
2011 “The Sociolinguistic Functions of Codeswitching between Standard Arabic and Dialectal Arabic.” Language in Society 40: 537–562. DOI logoGoogle Scholar
Amara, Muhammad
2013 “Arab Population of Israel. Sociolinguistic Aspects.” Encyclopedia of Hebrew Language and Linguistics Vol. 1, 124–128. Leiden: Brill.Google Scholar
Bassiouney, Reem
2009Arabic Sociolinguistics. Edinburgh: Edinburgh University Press. DOI logoGoogle Scholar
Blanc, Haim
1970 “The Arabic Dialect of the Negev Bedouins.” Proceedings of the Israeli Academy of sciences and Humanities. Jerusalem, 112–150.Google Scholar
Bolden, Galina B.
2012 “Across Languages and Cultures: Brokering Problems of Understanding in Conversational Repair.” Language in Society 41 (1): 97–121. DOI logoGoogle Scholar
Cerqueglini, Letizia
2015aObject-Based Selection of Spatial Frames of Reference in aṣ-Ṣāniˁ Arabic. Pisa University Press.Google Scholar
2015b “Spatial Frames of Reference in Traditional Negev Arabic: Language-to-Cognition Correlation.” Cognitive Processing 16 (1): 185–188. DOI logoGoogle Scholar
2016 “Etymology, Culture and Grammaticalisation: A Semantic Exploration of the Front/Back Axis in Traditional Negev Arabic.” In Arabic Varieties: Far and Wide. Proceedings of the 11th International Conference of AIDA – Bucharest, 2015, ed. by George Grigore, and Gabriel Bițună. Bucharest: Editura Universității din București.Google Scholar
2020 “Dialectal, Gender-Based, and Cross-Generational Variation in Negev Arabic Spatial Representations.” Coyote Papers: Working Papers in Linguistics, Linguistic Theory at the University of Arizona. http://​hdl​.handle​.net​/10150​/641485
Couper-Kuhlen, Elizabeth, and Margret Selting
2018Interactional Linguistics: Studying Language in Social Interaction. Cambridge: Cambridge University Press.Google Scholar
DeKeyser, Robert M.
2009 “Cognitive-Psychological Processes in Second Language Learning.” In The Handbook of Language Teaching, ed. by Michael H. Long, and Catherine J. Doughty, 119–138. Blackwell Publishing Ltd. DOI logoGoogle Scholar
Dingemanse, Mark, Joe Blythe, and Tyko Dirksmeyer
2014 “Formats for Other-initiation of Repair across Languages: An Exercise in Pragmatic Typology.” Studies in Language 38.1: 5–43. DOI logoGoogle Scholar
Henkin, Roni
2008 “Diachronic and Synchronic Accommodation in Negev Arabic.” In Between the Atlantic and Indian Oceans: Studies on Contemporary Arabic Dialects. Proceedings of the 7th AIDA Conference, ed by. Stephan Procházka, and Veronika Ritt-Benmimoun, 237–250. Lit Verlag.
2010Negev Arabic: Dialectal, Sociolinguistic, and Stylistic Variation. Semitica Viva Series no. 48, Otto Harrassowitz. Wiesbaden.Google Scholar
2015 “Codeswitching Patterns in Negev Bedouin Students’ Personal Interviews.” Zeitschrift für Arabische Linguistik 61: 5–34.Google Scholar
2016 “Functional Codeswitching and Register in Educated Negev Arabic Interview Style.” Bulletin of the School of Oriental and African Studies: 1–26. DOI logoGoogle Scholar
2019 “Inclusive Generic Person in Women’s Discourse in Israeli Hebrew and Negev Arabic.” Journal of Pragmatics 150: 53–74. DOI logoGoogle Scholar
Heritage, John
2007 “Intersubjectivity and Progressivity in Person (and Place) Reference.” In Person Reference in Interaction: Linguistic, Cultural and Social Perspectives, ed. by N. J. Enfield, and Tanya Stivers, 255–280. Cambridge: Cambridge University Press. DOI logoGoogle Scholar
Kitzinger, Celia
2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Hoboken, NJ: John Wiley and Sons.Google Scholar
Labov, William
1972 “Some Principles of Linguistic Methodology.” Language in Society 1: 97–120. DOI logoGoogle Scholar
2006 “Narrative Pre-construction.” Narrative Inquiry 16 (1): 37–45. DOI logoGoogle Scholar
Long, Michael
1983 “Native Speaker/Non-native Speaker-Conversation and the Negotiation of Comprehensible Input.” Applied Linguistics 4 (2): 126–141. DOI logoGoogle Scholar
Marmorstein, Michal
2016 “Getting to the Point: The Discourse Marker yaʕni (lit. “It Means”) in Unplanned Discourse in Cairene Arabic.” Journal of Pragmatics 96: 60–79. DOI logoGoogle Scholar
Matras, Yaron
2009Language Contact. Cambridge: CUP. DOI logoGoogle Scholar
Pitzl, Marie-Luise
2005 “Non-understanding in English as a Lingua Franca: Examples from a Business Context.” Vienna English Working Papers 14: 50–71.Google Scholar
Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks
1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 53 (2): 361–382. DOI logoGoogle Scholar
Shawarbah, Musa
2012A Grammar of Negev Arabic. Wiesbaden: Harrassowitz.Google Scholar
Siegel, Jeff
1985 “Koines and Koineization.” Language in Society 14 (3): 357–378. DOI logoGoogle Scholar
Tarone, Elaine
1980 “Communication Strategies, Foreigner Talk, and Repair in Interlanguage.” Language Learning 30: 417–431. DOI logoGoogle Scholar
Varonis, Evangeline Marlos, and Susan Gass
1985 “Non-native/Non-native Conversations: A Model for Negotiation of Meaning.” Applied Linguistics 6 (1): 71–90. DOI logoGoogle Scholar
Vasseur, Marie-Thérèse, Peter Broeder, and Celia Roberts
1996 “Managing Understanding from a Minority Perspective.” In Achieving Understanding: Discourse in Intercultural Encounters. ed. by Katharina Bremer, Celia Roberts, Marie-Therese Vasseur, Margaret Simonot, and Peter Broeder, 65–108. London: Longman.Google Scholar
Watterson, Matthew
2008 “Repair of Non-understanding in English in International Communication.” World Englishes 27 (3–4): 378–406. DOI logoGoogle Scholar

Address for correspondence

Roni Henkin

Ben-Gurion University of the Negev

POB 653

Beer-Sheva 8410501

Israel

[email protected]

Biographical notes

Roni Henkin is Full Prof. at the Department of Hebrew Language at Ben-Gurion University of the Negev, Israel. She is author of Negev Arabic: Dialectal, Sociolinguistic, and Stylistic Variation, 2010 and also writes on Arabic and Hebrew in contact, particularly on acquisition of Hebrew as 2nd language of Arabic-speakers in Israel.