Speaking to Domestics in Lebanon:

Power Issues or Misguided Communication?

Nabelah Haraty, Ahmad Oueini & Rima Bahous

Lebanese American University, Beirut, Lebanon


The Lebanese use a combination of Arabic and English telegraphic speech, along with gestures and other forms of speech adjustments to address their domestics. This pattern of inadequate speech is based on the misconception that domestics understand exactly the same way they speak. Using interviews, questionnaires, and participant observations, the researchers identified some of the underlying issues, power and trust, related to this form of fragmented speech. The investigators recommended that communication with domestics be in one language and in complete sentences, not only for the sake of language acquisition but to ensure a fair treatment of foreign helpers.

Keywords: Foreigner Talk, Critical Discourse Analysis, Code Switching

1. Introduction

In Lebanon, the practice of hiring domestics, natives of Sri Lanka, the Philippines, Ghana, and Ethiopia, is commonplace among middle and upper-middle class households. One pattern of communication with these individuals is typically foreigner talk (Holmes, 2001), that is, speaking in telegraphic speech, without prepositions, with simplified syntax, at a lower rated delivery, using gestures and a combination of Arabic and English phrases, depending on the level of proficiency of English of both addressor and addressee. The purpose of using such a fragmented pattern is presumably to facilitate comprehension for the non-native speaking domestic.

2. Research Questions

The investigators aimed at:

  1. Identifying the way Lebanese madams speak to their domestics; and
  2. Investigating the personal and cultural motives behind such rudimentary means of communication.

To answer the questions, the researchers opted to review the literature on critical discourse analysis, foreigner talk, and code-switching.

3. Critical Discourse Analysis

Critical Discourse Analysis (CDA) is a movement of scholars, interested in discourse, who focus on social issues. The researchers typically study different forms of the abuse of power in relation to class, ethnicity, and gender. These scholars seek to know how discourse enacts and contributes to the reproduction of inequality. 'Critical' means exposing the hidden social relations in discourse since such social relations of inequality involve the dominance of powerful groups. According to Crystal (1992), CDA aims "to reveal the hidden power relations and ideological processes at work in spoken or written texts" (p.22). Stubbs (1997) adds that CDA raises serious social issues. There is a certain connection between ideology, language, and power (Thomas, 2004). Also, there is a link between how the world is represented in different texts, and the way people in general think about the world. Fairclough (1989) defines CDA as "connections between language use and unequal relations of power" (p.1). Van Dijk (2001) adds that in CDA there is dominance, power, and inequality between social groups.

Van Dijk (1993) states that CDA is motivated by pressing social problems that would be better dealt with through discourse analysis. Therefore, one would look at social inequality, for instance, and try to analyze critically the extracts collected in order to understand the social differences better. Critical discourse analysts believe that by doing so, they are not only fighting injustice but also finding a solution for it. They hope that through critical understanding they will be able to make some changes (Van Dijk, 1997). However, one might wonder how functional these changes are. Their aim is to fight injustice, dominance, and power through the analysis of texts and conversations, but in truth the end result might not be what they aimed for. Clyne (2003) discusses the different arguments raised regarding the social significance of a language in a contact situation. His main claim is that the choice of a language reflects ‘societal patterning’ (p. 44) (see also Fishman 1972) i.e. the choice of a language depends on who you are addressing and when.

The CDA researchers may examine carefully how powerful some speakers can be in the abuse of power in some situations. However, the key point is the context, for it might transfer, interfere, or even reinforce some relations. Later, Van Dijk (2001) claims that the social power of institutions and groups is the central notion in critical discourse. He defines power in relation of control. A person who can control the mind and the acts of others has power. This power can be attributed to social status, fame, knowledge, money, etc.

4. Foreigner Talk

Foreigner Talk is the register used by native speakers when they address their non-native counterparts (Holmes, 2001). This pattern is influenced by a host of variables such as the topic of the conversation, the age of the participants, and in particular the level of proficiency of the learner. Therefore, foreigner talk is not to be thought of as a static set of features, but rather as a dynamic process that varies in accordance with various situational factors. One of its functional characteristics is that it involves simplifications leading to informal speech being influenced more by age than linguistic competence (Scarcella and Higa, 1981). Hatch (1983) suggested that foreigner talk is meant to promote communication and establish a special kind of affective bond between both: native and non-native speakers. Although foreigner talk resembles motherese, the simplified, high-pitched, infant-directed speech (Cook-Gumperz and Kryatzis, 2001), there are main differences related to both the input and the interactional features. Freed (1980) found that foreigner talk has a higher proportion of statements relative to motherese.

Many studies (see for instance Hatch, 1983; Chaika, 1994; Holmes, 2001) dealing with foreigner talk generally focused on English as a Second/ Foreign Language, students in educational settings and on ways in which native speakers of English modify their speech to accommodate their less proficient addressees. They found positive effects on second language acquisition. Some of the findings have important implications for the study at hand.

Ferhaadi’s (1988) informative analysis of the adjustments made in an individual’s speech according to the age and native language of the person being addressed, concluded that variation in speech according to the addressee’s linguistic competence and/or age probably does occur and that the results support reduced registers such as foreigner talk. Ellis (1985) identified a function of foreigner talk that is rarely mentioned in literature: to mark the role between the speakers - the ‘talking down’ function. This involves the use of ungrammatical simplifications: the omission of grammatical functions and a special lexicon.

Long (1983) suggested that the use of ungrammatical foreigner talk depends on the level of proficiency in the second language, and the perception of the speaker’s higher status.

A relatively large body of research dealt with this subject in non-academic settings, e.g. doctors’ clinics (e.g. Ainsworth-Vaughn, 2001) and lawyers’ offices (e.g. Shuy, 2004). Issidorides and Huletija (1992) concluded that native speakers’ ‘simplified’ or modified utterances, as in foreigner talk, actually facilitated comprehension for non-native speakers having such speech pattern. They added that using more complex language structure does not adversely affect comprehension. Finally, The British in India employed servants as valets, butlers, and other staff that let to the emergence of ‘Butler English.’ These subordinates learned a few words that they used to communicate with and especially impress their masters. Ali (2003) explains that thhe language is usually broken English characterized by "the absence of the copula and the tense usage, which is different from Standard English. The language is reduced to its bare essentials, with the emphasis on the content" (p. 1). It was considered charming and sometimes annoying when Butler’s English was ‘faultless’ (see also, Hosali, 2003).

This paper purports to demonstrate that the madams followed both approaches. On the one hand, there was a noticeable talking down because of the status issue, consistent with CDA proponents (Fairclough, 1989, 1995; VanDijk, 1993, 1997, 2001) and Ellis’ (1985) position. The latter maintained that ungrammatical foreigner talk happens in certain circumstances, mainly when the native speaker (or the addressor) has a superior status in reference to the non-native (addressee) one. Nevertheless, at times, the madams would engage in an interaction whereby they try their best to facilitate their communication with the domestics, as suggested by Issidorides and Huletija (1992). Therefore, it is rather difficult to draw a clear-cut line between the instances when the madams in the study are authoritarian and when they are more accommodating.

The lack of understanding in the first phases of language acquisition is related to the very limited linguistic resources. This reality at times could lead to tension on the part of the native and the non-native speaker. The latter is at a disadvantage because any misunderstanding or lack of comprehension will lead to a series of reprimanding comments that at best lead to humiliation. That is why the non-native tries to understand not for the sake of understanding alone but for sparing one’s feelings and pride. It takes quite a while to figure out how to resolve this tension. However, once the problem is worked out, a common basis of understanding is established. People do not communicate by words alone. Tone of voice, body language and other means carry at all times messages that either confirm or contradict the words, and sometimes are irrelevant to them (Chaika, 1994). In this study, the madams not only use gestures to emphasize a particular message but switch codes between English and Arabic.

5. Code-Switching

Code-switching is a normal practice among Lebanese people. It can be defined as a switch between languages within the same conversation (see Ferguson, 1959; Romaine, 1995; 2000; Crystal, 1995; Myers-Scotton, 1997; Auer 1998). It is the use of more than one language in a single utterance, conversation or in a course of single episode of communication. It usually occurs when an individual who is bilingual switches between two languages simultaneously and interchangeably. Thus, it is the transformation from one mode of communication to another in the same conversation. Code-switching may be used to achieve two purposes: fill linguistic gaps and/or attain multiple communicative purposes. Gumperz (1982) defines code-switching as

"… the juxtaposition within the same speech exchange of passages of speech belonging to two different grammatical systems or subsystems" (pp. 55- 99). Chaika (1994) points out that bilinguals in general code switch according to the topic of conversation within the same social situation. She also adds that code-switching may occur to emphasize a phrase. Such switch "reinforces bonds between speakers" (p. 335). Code-switching may also be resorted to in front of a third party who does not understand the target language used; thus, the speakers may switch codes to exclude him/her. Consequently, code-switching is not language interference on the grounds that it supplements speech. It provides continuity in speech rather than interference with language. It is viewed as a linguistic advantage rather than an obstruction to communication (Crystal, 1995; Poplack, 1998).

Poplack (1980) identified three types of code-switching: tag switching where it is common to hear people start their sentence with ‘you know’... or ‘I mean’; intersentenial switching characterized by shifts in clauses where one clause or sentence is in one language and the second clause or sentence is in another; and finally intrasentenial switching marked by switching among different types within the same sentence or clause. The code-switching that is the focus of this study is a mixture of all three, but it is structured on the basis of foreigner talk when used with the domestics.

6. Methodology

The researchers’ aim is to identify and question different language patterns used by the madams with their domestics and investigate both the cultural and the personal motives behind the use of such language patterns.

7. Participants

The three madams chosen for the study are educated, full-time working married women. They all have children living at home. The three madams have received their university education in American-style institutions. They are bilingual in Arabic and English. One madam is trilingual; she speaks French fluently, over and above English and Arabic. These ladies play the dual role of housewives and working women. The choice of the word ‘madam’ was inspired by the domestics themselves because this is the title they use to refer to their female employers. Usually communication takes place between madams and domestics, and rarely between husbands and domestics, given the clear gender roles that exist in their households.

The three Sri-Lankan domestics came to Lebanon to work as maids or helpers. The researchers limited their research on maids from Sri Lanka because they are the largest groups of domestic helpers in Lebanon, and more importantly, to maintain uniformity in data collection and analysis. The domestics are married with children and are in Lebanon on their first appointment. All three have minimum formal education. Their presence in Lebanon is based on the relatively attractive monthly compensation of 100 US dollars. Most have two-and-half-year contracts. These women have a limited knowledge of the English language. The study refers to them as domestics as the word domestic more accurately captures the fact that they join the household and leave two and a half years later. This setup is quite common and convenient for Lebanese families.

Further, the environment and the nature of the interaction that takes place on a daily basis in the different households will be analyzed to investigate the kind of communication that emerges as a result. The following table summarizes the characteristics of the informants:



Age of domestic



Mid 20s



Mid 40s



Early 20s

All three domestics work in houses where one of the interviewers is a frequent visitor. Thus, being interviewed by someone they already know provided them with a sense of security. The interviews took place in the houses where these domestics work so as to keep the setting as natural as possible and to help them answer questions and carry out commands easily and spontaneously without rehearsal or practice. The interviewer tried to instill a sense of trust and comfort, reassuring them that this interview was not in any way going to be used against them, nor would it jeopardize their job.

8. Data Collection

The researchers conducted semi-structured interviews with the domestics and their madams to give more freedom in the exact working and sequencing of questions and more attention to a wider variety of topics (Powney and Watt, 1987; Robson, 2000). This is in contrast to the structured interview that leads to rigidity or to the unstructured interview that is unfocused. Usually, in small-scale evaluations, semi-structured interviews tend to be favored (Robson, 2000). They provide current information and personal insights that may not result from consulting other sources. In addition, information from an interview can add a dimension of human interest to a paper that might otherwise be dry and recondite. Interviews open the door to more verbal interaction, more probing and more asking.

Another method used to collect the data was direct observation. Types of direct observation are generally used: non-participant and participant. The investigators decided to conduct a participant observation for the following reasons: Non-participant is rather difficult to apply in this situation as it needs specialized research skills and simultaneous administration, whereas the participant observation does not have to account for these constraints (Robson, 2000; Cohen, Manion, and Morrison, 2000). Also, for this research, the body language and facial expressions of the interviewee was an important point in observation.

All interviews were tape-recorded. Permission was secured from the participants before starting the data collection process. Two sets of interviews were conducted: one with the domestics and one with the madams. Two of the three Madams attended the interviews with the domestics which may have affected the validity of the data. It is in fact a limitation of the paper. Most of the questions with the domestics consisted of two to three words. The researcher had to follow the same speech pattern the madams used with their domestics. Most of the questions directed to the madams focused on the language they use when communicating with their domestics. The interviews with the madams also involved questions about the responsibilities of the domestics and issues of trust

The researchers opted for a third method for data collection to ensure triangulation. They prepared and piloted a semi-structured questionnaire. The latter was distributed at random to eighty participants of different ages who have a domestic at their homes.

9. Discussion

9.1 Interviews with the domestics

The communication with the domestics was limited to household chores, so the questions revolved around tasks that domestics do at home. The only interview that was conducted in the kitchen was with Nirma. Both the interviewer and the observer, with steaming mugs of coffee in their hands, sat on two chairs in front of the kitchen table. Nirma sat in front of them but without drinking or eating anything. She could speak a few English words and phrases. She also managed to pick up a few Arabic words to use when the need arises. Nirma came with no English education, but she showed genuine interest in learning the language. Nirma was also in charge of all the household chores. Living with a large number of people may have helped her communicate better with the interviewer. She is in fact daily in contact with three children, two other co- workers, two drivers, and a janitor. However, when the questions were not sentences or statements she hears everyday she could not answer. For instance, when she was asked about the age of the children for whom she works, she looked down because she was not sure of the answer.

Interviewer: How old? Are they big? Small …

Nirma: Don’t know.

The 'Don’t know' refers to the ages- she is not privy to the exact ages. When the question was repeated to indicate that the answer only required whether they were big or small- still the response was ‘Don’t know.’ She was probably afraid to give wrong information. She does not know how small is small or how big is big.

She seemed relaxed sitting in the kitchen and talking to the interviewer. However, when she was not sure about the answer, she would look sideways or only at her hands.

Also, Nirma preferred the English and not the Arabic language. When asked what language she speaks ‘good’ English or Arabic, she answered:

Nirma: English

Interviewer: English- why

Nirma: I love.

(Later in the conversation)

Interviewer: you want them to speak to you only in English?

Nirma: yes.

Interviewer: why?

Nirma: like go Srilanka and talk English.

Her hope is to learn English in Lebanon then go to Sri Lanka and teach the language to her son. Maybe this limited English could help him find a better job when he grows up. Nirma’s speech became rather incoherent when she referred to her son. She was using some Sri Lankan words as if she remembered the language she used with him back home.

The second domestic to be interviewed was Ruba. The interview took place in the living room in front of Renée. The interviewer, the observer, and Renée sat on the comfortable armchairs in the living room sipping coffee and eating cake. Ruba was invited to join them, but she waited for her madam’s approval before accepting to sit on a chair there next to Renée. She seemed to be quite uncomfortable sitting in the living room. She has been in Lebanon for fifteen months and came from Sri Lanka with no knowledge of either Arabic or English. In her stay, Ruba has learned Arabic because this was Renée’s wish. But still, she sprinkles her speech with English phrases here and there.

Interviewer: inti inti hibbee tih’keh Arabeh kteer mneeh?

Ruba: eh, ana hib kteer- Arabeh.

Interviewer: inth hibbeh tihkeh English?

Ruba: no- Arabeh


Interviewer: Do you like to speak Arabic well?

Ruba: Yes, I like to speak Arabic well.

Interviewer: Do you like to speak English?

Ruba: no- Arabic

When asked what language she would like to teach her children when she goes back to Sri Lanka, she insisted on Arabic. At the end of Ruba’s interview she surprised the interviewer with the ‘merci,’ the French word for ‘thank you.’ She has learned this from Renée who also shifts to French every now and then. It was quite obvious that Ruba has a good relationship with her madam. All through the conversation she was happy, excited and kept giggling every time she was asked a question. She kept patting on Renée’s hand and telling her to stay, as if to show her how advanced her Arabic has become.

The third and shortest interview was with Dummy who has only been in Lebanon for two and a half months, and works for Amal. The interview again took place in the living room. The interviewer, the observer, and Amal sat on comfortable armchairs; they had juice and cookies. Amal called Dummy and asked her to sit on a chair and answer the questions. She had to repeat her request three times before Dummy slowly moved towards the chair next to the door and sat on its side. Dummy had very little command of Arabic or English. The interviewer had to repeat each question more than once as each time Dummy would hear a question, she would move her hands and her head from one side to another saying ‘no’. Then, the interviewer would repeat the question and Dummy would try to answer.

Interviewer: you speak English

Dummy: shway English- shway.

Interviewer: you speak English

Dummy: a little English- a little

The interview revealed that, with inadequate receptive and expressive English skills, Dummy was conditioned to react to certain key words in a particular way. For example, when asked how she makes Amal’s Nescafe- her reaction was the following

Dummy: yes go

She thought she was asked to go and prepare Nescafe. And then she proceeded to give details of the preparation of a cup of coffee.

Dummy: Haleeb

Interviewer: Ok

Dummy: Niss-coffee- sikka. Haleeb tree- niss coffee one- sikkar two


Dummy: Milk

Interviewer: Ok

Dummy: Nescafe- sugar. Milk three- Nescafe one- sugar two

Dummy simply repeated the ingredients Amal taught her when making her a cup of Nescafe. Her reaction was triggered by one word ‘Nescafe’ and not because she understood the question. In fact, Dummy’s answers were short phrases with a mixture of both English and Arabic with more stress on Arabic. This stresses the code switching norm which is not only common among Lebanese people but is also used by their domestics.

Interviewer: inti what ish tigil

Dummy: eh house- clean- bar’rad clean- himmam clean- killo. Clean.

Interviewer: What do you do?

Dummy: I clean the house, clean he fridge, the bathroom all of it, clean


An interesting duality could be observed: The domestics treat their madams as head of the house; they are the ones who hear them, talk to them, pay them, etc. Similarly, these domestics are also heads of their own households, or the breadwinners, whereas back home their husbands stay at home and take care of the children.

The Sri Lankan maids have a vested interest in acquiring a new language, be it Arabic or English, because they see it as a status improvement that would help them both in the short and long term. In the short term, these helpers can maintain employment beyond the three-month trial period. In the long run, they can pass this knowledge on to their children. In many ways, acquiring a new language is not entirely a choice; it is a question of employability and survival. In fact, the domestics "are in the paradoxical situation of having to communicate in order to learn and having to learn in order to communicate" (Simonot in Roberts 1996, p. 2).

Nirma’s continuous exposure to English has affected her comprehension of the questions in English more than the code switching to Arabic. When Nirma first arrived a year and a half ago, she knew almost no English. However, living with the Madam and her family for a while and having to use English with them helped Nirma communicate better in. So, this exposure to her madam and her family in English has positively influenced her command of the English language. English in this context is foreigner talk.

The comprehension of key words led her to the comprehension of the sentence. This is similar to teaching students a new word based on context clues. The questions addressed to Nirma most of the time were in foreigner talk.

Nirma’s behavior while performing her duties varies depending on the guest being served. If the guest is a regular visitor, she takes the initiative and brings the coffee to the guest. If the person is not a regular visitor, Nirma stands by the door with the tray and awaits a sign from her madam. On her own, Nirma does not dare offer the guest coffee.

This emphasizes the class distinction-the master and the subordinate, consistent with Long's (1983) observation about the feeling of superiority usually harbored by native speakers. Nirma knows that she is not allowed to interact - even if momentarily with any of the guests unless specifically allowed to do so by her madam.

Ruba was able to answer the questions in Arabic more skillfully than in the other two modes, owing it to Renée, her madam, who insists that she only use Arabic when talking to her. Ruba also responded to some questions in English because, no matter how hard Renée tries to speak exclusively in Arabic, it is inevitable that she spices up her conversation with a few English words. As for serving coffee, Renée goes to the kitchen and places the cups on a tray. If the cups and trays are fancy china, then Renée serves. If they are plain, then Ruba serves.

The most difficult part was dealing with Dummy whose command of English and Arabic language was reportedly very poor. On several occasions, the interviewer had to repeat the question, alternating between varying the pitch, using gestures, and so forth.

9.2 Interview with the Madams

The interviews with each of the Madams were conducted in the living room of their respective houses. The interviewer, the observer, and the Madam sat there discussing each of the domestic. A common theme among the three madams is the lack of total trust in their domestics.

When asked about the interaction pattern between the husband and Dummy, Amal replied as follows

Amal: …afik'ra ana my husband doesn't talk to her fina n'ool Interviewer: Why doesn’t your husband talk to her?

Amal: mabuaa'teeha wij. He likes to keep distances with her.

Interviewer: Why?

Amal: to be the – to let her feel if I am loose – he's the master.

Interviewer: To show there's power at home.

Amal: Yes hatta matfak'kir ha'la el-bait bayyta mitil ma am'leh.

Amal: …you know you can say that my husband doesn’t talk to her

Interviewer: Why doesn’t your husband talk to her?

Amal: He likes to keep distances with her.

Interviewer: Why?

Amal: to be the – to let her feel if I am loose – he's the master.

Interviewer: To show there's power at home.

Amal: Yes so that she doesn’t think that the house is hers.

What transpires from this bilingual conversation is that Amal's husband does not talk to Dummy at all, nor does Dummy report back to him. All communication goes through Amal. The purpose behind the husband's abstinence from talking to Dummy is that he sees no reason to make her feel at ease around him. Amal's husband believes that his wife may be a bit too easy-going with Dummy, so in order to ensure that there is some sort of authority at home, he insists on being intimidating and unapproachable. Amal's attitude embodies what Reisigl and Wodak (2001) refer to as racism and prejudice against other social groups, which is what CDA proponents advocate against. Amal, on the other hand, welcomes such stern attitude. Both she and her husband do not want Dummy to feel or to get too comfortable with them and behave as if she were in her own home.

Renée showed that she also likes to keep a distance between herself and her domestic, Ruba.

Interviewer: Do you talk to her other than telling her what to do and what not to do?

Renée: Ch… I've kept my relationship with her even though very friendly but very professional, so I, I try not to get very personal…There should be a yani limit, there should be a distance between the boss if you want and the employee.

Renée is the best example of a madam who has a good relationship with her domestic, yet she still believes that the element of dominance and power between employer and subordinate should not be overlooked. It is a requisite rule that goes into effect as soon as a domestic is hired.

Mona also showed that the sense of feeling or getting too comfortable with her domestic bothered her.

Interviewer: Are your helpers from the same nationality?

Mona: No … three different nationalities .. Philippines, Sri Lanka, and Ghana.

Interviewer: Why not same nationality?

Mona: (joking) you want them to form a league – a force task against me? No reason is … they get too comfortable with each other … when they come from different countries they do not get too relaxed… they respect the differences and keep a distance … I don't hear them arguing … better for me.

When Mona was asked about the reason why she uses only English with her domestics, she commented as follows:

Mona: I don't want them to understand everything we say at home. Also, I do not want them to get too friendly with the drivers, the janitor, or the neighbors. I do not feel comfortable.

I don’t even let them go out on Sundays… I don’t want you to think that they are prisoners, no…


Mona does not want to worry about what goes on during her absence. She uses English to form a common code of communication with the three domestics to aid her with comprehension rather than to establish a bond with them. She does not want them to be privy to discussions that take place among family members at home. She does not want them to socialize either. One can sense this reticence of allowing the domestics to have a sense of belonging, which lends itself to security. It is quite ironic that the fear is from the domestic feeling too comfortable while in reality it should be the other way round; the domestic should feel secure and comfortable so they could improve their productivity.

The closer we get to understanding the attitudes of the madams towards their domestics, the clearer the idea becomes that the domestics and the madams do not share the same definition of being a member of the household. The madams, perhaps inadvertently, exert power over their domestics. They want them to obey and not to question, perform duties and not complain, and be alert all the time and not get too familiar. With all these factors in mind, albeit semi-consciously, the madams not only use code-switching when addressing their domestics, but change the whole nature of their communication. Downes (1998) points out that languages in general vary depending on participants’ interactions: relationship, type, social goals, etc. The imbalance of power is an ethical concern related to the hidden agenda. The language becomes more of a command in most of the cases because of the power that the employer possesses over the employees.

Dealing with the imbalance of power is the duty of the person who has the most power. The best way to deal with this imbalance of power is to address it directly as soon as there are hints that it might be thwarting open communication. However, there was hardly any attempt on the part of the madams to ease the obstacle interfering with communication, as evidenced by observation of statements uttered by the madams themselves. It is neither necessary nor accurate to deny differences in power, authority, or responsibility. Simply being around one another may create a relationship, but this doesn’t mean that the relationship is valued. The domestic will hardly feel that her presence at home goes beyond the limits of how well she serves and performs her duties. While her work is quietly appreciated by her employers, it does not transcend the realm of her household world. She is not valued as a human being nor treated as a member of a household whose opinion is sought and company desired.

Control describes a spectrum of dominance and submission. Every relationship involves some conflict and a control issue. Some relationships are created with stated differences in control. Westra (1996) discusses fear, trust and imbalance of power together because they tend to occur together. It is rare for a person to be afraid while feeling in control. It is when this control is weak or threatened that fear arises and trust becomes an issue. Because of the imbalance of power inherent in the relationship between supervisor/employee, confrontation becomes problematic. In effect, the lack of willingness to confront demonstrates a lack of trust. It is wise at times not to confront if keeping the relationship intact is more important than addressing a particular issue.

Brown and Levinson (1987) maintain that there are culturally different variations in interactional style caused by differential assessments of such factors as power, social distance and ranking of verbal imposition.

The social distance is well observed in Amal’s case. Amal exercises her authority over her domestic and makes sure to maintain a distance. This is seen in her interaction with Dummy. Amal doesn’t use English with Dummy, when asked why:

Amal: … because it won’t be practical for me to always speak English-and if I want to say something about her to my husband-if she knows English-I won’t be able to tell my husband…

Although Amal uses Arabic with Dummy, she speaks English with her husband when she does not want Dummy to get a notion about the topic.

Renée, also contradicted herself more than once. Though she claimed she had quite a rewarding relationship with her domestic, she stated that her relationship with Ruba is friendly yet very professional. She meant that their relationship is strictly confined to household interaction.

In matters of trust, Amal was very reluctant in entrusting Dummy with Dima, her daughter. Her attitude does not stem from lack of communication, but from cultural differences. To illustrate, Amal gives her concern that Dummy who loves hot spices and chili might give some to her daughter as a treat.

Interviewer: do you trust Dima with her? …

Amal: Laa – never ever.

Interviewer: never – why?

Amal: …due to the difference in our culture- she may give her something bad to eat or drink

Amal complains that Dummy does not follow instructions. When she asked Dummy not to boil Dima’s water, Dummy boiled it. She simply did not understand what she was being asked to do.

Interviewer: Does she usually boil it? (the water)

Amal: Yes, but I did not want her to boil it yesterday.

I had to throw the water.

Interviewer: why did she do it- out of habit?

Amal: eh habit, everyday I tell her please don’t put all the clothes in the dish- in the washing machine- everyday she puts my clothes, Dima’s and my husband’s in the washing machine- the next day please Dummy don’t put it- again- I had to tell her no quite forcefully.

Amal believes that it is hard to 'untrain' domestics if they are first taught to do something. She believes they act out of habit and not out of understanding or analyzing.

All madams used Ellis’ (1985) talking down style. Ungrammatical foreigner talk occurs under special conditions, such as when the native speaker considers he holds a superior status to the non-native speaker (Ellis, 1985). Examples from the interviews abound. In Amal’s case, she did not teach Dummy any words- neither Arabic nor English. She depended on her mother's maid to train her.

Interviewer: So how do you communicate with her…

Amal: If I- yes- if I want to tell her – eh – boil water for Dima… I cannot say please boil the water for Dima- she cannot understand – I say " mayy- Dima- Nar-…" water-Dima-fire.

Amal justifies this by asserting that Dummy can not understand full sentences in Arabic or English. This also supports Issidorides and Hulsfija’s (1992) conclusions that native speakers simplify the utterances in order to facilitate the comprehension of the non-natives. From Ellis’(1985) perspective, this pattern of speech shows more power and exclusion than facilitating comprehension.

Renée speaks to Ruba only in Arabic and has more trust in her, albeit limited, reportedly because her domestic is older. Renée’s tone with Ruba is different from Amal's with Dummy. Renée explained that there is closeness in age between her and Ruba. Thus, she feels a bit self-conscious about raising her tone of voice while Mona and Amal have no qualms when it comes to addressing them in a harsh, reprimanding tone. This supports Ferhaadi’s (1988) informative analysis that adjustments are made according to age.

An interesting point to ponder is when asked to perform a task all domestics responded in English, "Yes, madam" even when the question or command was in Arabic. Neither the interviewer nor the domestic thought there was anything unusual about this exchange. The same goes for the interviews with the madams where this code-switching pattern was carried out unconsciously and the switch had gone unnoticed by both parties. There was no attempt at showing or comparing competencies in either case. It is obvious that the domestics are not fluent in either language and the madams and the interviewer code-switched as part of a common habit in Lebanon. The interviews showed that the domestics performed best in the language they hear most. What happened is that the madams assume that their domestics know some vocabulary in Arabic and some vocabulary in English. By the same token, the domestics believe that their madams know a lot of Arabic and a lot of English.

9.3 Questionnaires

Eighty questionnaires were distributed at random to different participants, including faculty, graduate students, and professionals. The common factor among all these participants was having a live-in domestic at home. Sixty-five questionnaires were returned, twenty –five of which were filled in by males and forty-five by females. All the participants answered that the duties of the domestics is to clean the house (iron, wash clothes, etc.). Many also added that the domestic has to cook, but twenty-five of the participants pointed out that their domestics were not allowed to cook. The majority of the participants do not allow the domestics to go out on their own for fear that they may steal them, get lost, do not know how to socialize, or may even get to know others who would definitely advise them wrongly. Seven of the participants said that their domestics usually go out to buy goods for the house or to walk the dog. Two added that their helpers are human beings and they have their own life. Only seven wrote that they trust their maids and thus allow them to go out every now and then.

Many of the participants claimed that the topics they mainly discuss with the domestics are food and family, and the languages they use to communicate are a mixture of broken Arabic and English. However, fifteen replied that they do not discuss topics with their helpers. Most again agreed that their domestics are easy going as they do everything they are asked to do though sometimes they do them in the wrong way. In fact, most domestics live in fear of losing their jobs and getting sent back to their home countries before their contract expired should their employers find their performance or character questionable. Ten pointed out that their maids are very stubborn as they do not do what they are asked to do, but then it seems that these participants have various problems with their helpers as they added that they wanted them out of their houses. It seems that the majority of the participants treat their domestics as second-class citizens, the same way the interviewed madams treat their domestics. The main job of these helpers is anything that has to do with housework. Very few participants trust their domestics; in fact, the majority do not allow them to leave the house on their own or have any personal contact with others. There is definitely social injustice there (see Stubbs, 1997; Van Dijk, 1993; 1997; 2001; Thomas, 2004) as the participants are using their power to order their domestics around.

10. Implications

Due to the domestics’ limited knowledge of Arabic or English, most employers attempt to establish a common ground with them, using foreigner talk as well as a combination of gestures, single words, facial expressions and so forth. This common ground is highly individualized and varies from person to person living in the household and trying to communicate with these domestics. Also, this pattern of inadequate communication that goes on at home may have negative effects on children (see Freed, 1980; Ferhaadi, 1988). When there is an emergency at home and quick action is called for, it would prove more daunting to overcome the language barrier and communicate important instructions to a domestic with limited language skills. This problem is further exacerbated when the madams are phoning in instructions where they cannot use gesture or facial expressions to get their message across. From the domestic's perspective, code-switching tends to hinder her language acquisition. She will always be stuck with a fragmented second language and, when she leaves the country subsequent to her employment, she finds herself no more fluent in any language than when she first came in.

Yet, despite these disadvantages, foreigner talk continues to prevail.

Knowing that comprehension is almost always superior to expression, communication with the domestics should be in one language in complete sentences. It may be English or Arabic but not both. However, this is not likely to happen because better communication will let the domestics into their madams’ lives in ways they perhaps shouldn’t. However, this is inevitable because it is a fact that some of the madams will not go beyond the everyday basic manual chores when communicating with their domestics. This is shown in light of the interviews with the madams where one can infer that complete trust is restricted only to cleaning the house and nothing more. Paradoxically, these domestics are members of the household in terms of living under the same roof, eating the same food, and taking care of the household chores. But when a family member goes on a trip for two weeks or is admitted to the hospital, this information is not shared with domestics. Exclusion means detachment; the domestics will not feel a sense of belonging; their ties with the family are very tenuous. They are after all viewed as ‘maids, servants, subordinates’ which means they can not be included in the family so there is a clear class distinction in fully operational in this context (see Fairclough 1989; 1995; Stubbs, 1997; Van Dijk, 1993; 1997; 2001)

Hierarchy exists in every culture and society. It serves to delineate boundaries among people, assign various responsibilities, ensure accountability, and facilitate communication. However, the mere existence of a hierarchy does not give permission for employers to degrade their subordinates. When the latter occurs, it is clearly motivated by a denial of the employee's basic human right or respect and dignity. So the implications of foreigner talk, particularly in the context of communicating with domestics go far beyond the realm of inadequate communication and laziness to teach correct language. It encompasses the underlying issue of discrimination, racism and oligarchy. Perhaps this is an opportunity for Lebanese people to reexamine their motives when using foreigner talk so indiscriminately. It would be very hard to convince ourselves that our patterns of communication are innocuous, and do not have implications that touch the core issue of equality and basic human rights.

11. Conclusion

In this paper, the researchers examined critically the cross cultural communication between the madams and their domestics in Lebanon. There is an urgent need to remedy the clear social injustice that prevails in the households. It is true that in few houses, these domestics are not treated better than prisoners as they are forbidden to leave the house except if accompanied by their madams. They are not allowed to talk on the phone or even watch television. In fact, the issue of power and misguided communication should be taken a step further and more research in the area should be conducted in the country to shed light on this social injustice. These domestics should be empowered and their rights should be made clear to all. A further scope of research in this area would involve conducting experiments on whether speaking to domestics in syntactically sound sentences improve their language skills, English and/or Arabic.


Ainsworth- Vaughn, N., 2004. The discourse of medical encounters. In: Schiffrin, D. Tannen, D. & Hamilton, H. (Eds.), The Handbook of Discourse Analysis. Blackwell Publishing, Malden, Massachusetts, pp. 453 – 469.

Ali, S. I. 2003. Butler English. The Hindu Education, retrieved November 8, 2006 www.thehindu.com/thehinu/edu/2003/02/18 stories/2003021800010200.htm

Auer. P., 1998. Code-Switching in Conversation. Routledge, London.

Brown, P. & Levinson, S., 1987. Politeness: Some Universals in Language Usage. Cambridge University Press, Cambridge.

Chaika, E., 1994. Language: The Social Mirror. Heinle and Heinle Publishers, Boston, Massachusetts.

Clyne, M., 2003. Towards more language centered approach to plurilingualism. In: Dewaele, J. M., Housen, A., & Wei, L. (Eds.), Bilingualism: Beyond Basic Principles. Multilingual Matters Ltd, Clevedon, pp. 28 – 42.

Cohen, L., Manion, L., & Morrison, M., 2000. Research Methods in Education (5th. ed.). Routledge, London.

Cook-Gumperz, J. & Kyratzis, A., 2001. Child discourse encounters. In: Schiffrin, D. Tannen, D. & Hamilton, H. (Eds.), The Handbook of Discourse Analysis. Blackwell Publishing, Malden, Massachusetts, pp. 590 – 611.

Crystal, D., 1992. Introducing Linguistics. Penguin Group, London.

Crystal, D., 1995. The Cambridge Encyclopedia of the English Language. Cambridge University Press, Cambridge.

Downes, W., 1998. Language and Society (2nd ed). Cambridge University Press, Cambridge.

Ellis, R., 1985. Understanding Second Language Acquisition. Oxford University Press, Oxford.

Fairclough, N., 1989. Language and Power. Longman, London.

Fairclough, N., 1995. Critical Discourse Analysis: The Critical Study of Language. Longman, London.

Ferguson, C. A., 1959. Diglossia. Word 15, 325 – 340.

Ferhaadi, A., 1988. Impact of adults and children on the addressor. Paper presented at the Annual Meeting of the Teachers of English to Speakers of Other Languages. (22nd, Chicago, H. 1988). (ERIC Document Reproduction Service No. ED 295488).

Fishman, J. A., 1972. The relationship between micro- and macro-sociolinguistics in the study of who speaks what language to whom and when. In: Pride, J. B. & Holmes, J. (Eds), Sociolinguistics. Penguin, Harmondsworth, pp. 15 – 32.

Freed, B., 1980. Talking to foreigners versus talking to children: similarities and differences. In: Scarcella, R. & Krashen, S. (Eds.), Research in Second Language Acquisition. Newbury House, Rowley, Massachusetts, pp. 19 – 27.

Gumperz, J., 1982. Discourse Strategies. Cambridge University Press, New York.

Hatch, E., 1983. Psycholinguistics: A Second Language Perspective. Newbury House, Rowley, Massachusetts.

Holmes, J., 2001. An Introduction to Sociolinguistics (2nd. ed.). Longman, an Imprint of Pearson Education, Harlow.

Hosali, P. (2000). Butler English form and function. BR Pub. Corp, Delhi.

Issidorides, D. & Hulstijn, J. 1992. Comprehension of grammatically modified and non-modified sentences of second language learners. Applied Psycholonguistics,13 (2), 147 – 171.

Long, M., 1983. Native speaker/non-native speaker conversation in the second language classroom. In: Clarke, M. & Handscombe, J. (Eds.), TESOL’82: Pacific Perspectives on Language Learning and Teaching. TESOL, Washington D.C.

Myers-Scotton, C., 1997. Code-switching. In: Coulmas, F. (Ed.), The Handbook of Sociolinguistics. Blackwell Publishers, Madden, Massachusetts, pp. 205 – 216.

Poplack, S., 1980. Sometimes I start a sentence in English y termino in Espanol: towards a typology of code switching. Linguistics, 18, 581 – 616.

Poplack, S., 1998. Contrasting patterns of code switching in two communities. In: Trudgill, P. & Cheshire, J. (Eds.), The Sociolinguistics Reader: Multilingualism and Variation (vol. 1). Arnold, London, pp. 44 – 65.

Powney, J. & Watts, M., 1987. Interviewing in Educational Research. Routledge & Kegan Paul, London.

Reisigl, M. & Wodak, R., 2001. Discourse and Discrimination: Rhetoric of Racism and Anti- Semitism. Routledge and Kegan Paul, New York.

Roberts, C., 1996. Background to the understanding project. In: Bremer, K., Roberts, C., Vasseur, T., Simonot, M. & Broeder, P. (Eds.), Achieving Understanding: Discourse in Intercultural Encounters. Longman, London, pp. 1 – 8.

Robson, C., 2000. Real World Research: A Research for Social Scientists Practitioner Researchers (2nd. ed.). Blackwell, Oxford.

Romaine, S., 1995. Bilingualism. Blackwell Publishers, Oxford.

Romaine, S., 2000. Language in Society: an Introduction to Sociolinguistics (2nd ed.). Oxford University Press, Oxford.

Scarcella, R. C. & Higa, C. 1981. Input, negotiation, and age differences in second language acquisition. Language Learning 31, 499 – 508.

Shuy, R. W., 2004. Discourse analysis in the legal context. In: Schiffrin, D. Tannen, D. & Hamilton, H. (Eds.), The Handbook of Discourse Analysis. Blackwell Publishing, Malden, Massachusetts, pp. 437 – 452.

Stubbs, M., 1997. Language and the mediation of experience: linguistic representation and cognitive orientation. In: Coulmas, F. (Ed.), The Handbook of Sociolinguistics. Blackwell Publishers, Malden, Massachusetts, pp. 358 – 373.

Thomas, L., 2004. Language, Society, and Power: An Introduction. Routeldge, London.

Van Dijk T.A., 1993. Principles of critical discourse analysis. Discourse and Society, 4(2), 249 - 283.

Van Dijk T.A., 1997. The study of discourse analysis. In: Van Dijk, T. A. (Ed.) Discourse as Structure and Process. Sage Publications, London, pp. 1 – 34.

Van Dijk T.A. 2001. Critical discourse analysis. In: Schiffrin, D. Tannen, D. & Hamilton, H. (Eds.), The Handbook of Discourse Analysis. Blackwell Publishing, Malden, Massachusetts, pp. 352- 371.

Westra, M., 1996. Active Communication. Brooks/Cole Publishing Company, Pacific Grove, California.

About the authors:

Nabelah Haraty
Lebanese American University
P. O. Box 13 – 5053
Chouran Beirut 1102 – 2801, Lebanon
Email: nharaty@lau.edu.lb
Tel: 00 961 1 811 968 (ext. 1396)

Ahmad Oueini
Email: aoueini@lau.edu.lb

Rima Bahous
Email: rbahous@lau.edu.lb

Journal of Intercultural Communication, ISSN 1404-1634, issue 14, June 2007.
Editor: Prof. Jens Allwood
URL: http://www.immi.se/intercultural/.