An account of shifting access rituals among the Ngoni of Tanzania

The current paper is an attempt at providing an ethnographic description of Tanzanian Ngoni access rituals (greetings) considering verbal and visual aspects of these communicative routines. Three methods of data collection were used, namely: role-play, semi-structured interview, and observation. The role-plays were used in order to see how the Ngoni apply their knowledge of this kind of access ritual still acquired in the course of growing up in their communities. Role-plays were recorded with a digital camcorder after the interviews had taken place. Semi-structured interviews were meant to provide insights into diachronic developments such as how the expression of politeness through greetings may have changed. Observation method was meant to capture behavioural patterns shown during greeting exchanges and notes were taken immediately thereafter. The main findings are that due to pervasive contact between Ngoni and Swahili, greetings tend to be brief today and there is a widespread use of code-switching by all age groups. The tendency to prefer Swahili is viewed by older members of the community as lack of respect whereas the younger generations regard the use of Swahili as more appropriate and as a reflection of social changes that have been taking place in the community.


Introduction
This paper provides a first ethnographic description of Tanzanian Ngoni greetings considering verbal and visual aspects of these communicative routines.On the basis of data obtained through role-play, observation and interviews three questions will be answered: How do Ngoni apply their knowledge of greeting routines acquired in the course of growing up in their communities?Which behavioural patterns are performed during greeting exchanges?Have expressions of politeness through greetings changed over time?Due to pervasive contact between Ngoni and Swahili, greetings tend to be brief and there is a widespread use of codeswitching by all age groups.The tendency to prefer Swahili today is viewed as lack of respect by elder Ngoni whereas the younger generations regard the use of Swahili as more appropriate and as a reflection of social changes that have been taking place in the community.
Linguistik online 84, 5/17 ISSN 1615-3014 64 The Ngoni ethnic group is found in the south-western part of Tanzania, and occupies about two-thirds of the inhabitants of the Songea District in Ruvuma Region (Ngonyani 2003: 1).Their language is called Kingoni (N.12).A recent estimate of the number of Ngoni speakers is 258'218 (Languages of Tanzania 2009: 3).The name Ngoni derives from the collective name Nguni, referring particularly to the Zulu cluster of languages of South African Bantu languages (Ebner 1987: 9;Ngonyani 2003: 1;Moser 1983: 9).For about a century now the language and its culture have been in contact with the Swahili (G.40) speaking community, besides contacts with other neighbouring ethnolinguistic groups such as Matengo (N.13), Ndendeule (N.101) and Manda (N.11) (Guthrie 1971).In the villages where the study was done a small number of speakers of Bena (G.63) and Hyao (P.21) was identified (see Map).
This article addresses Ngoni greeting exchanges (henceforth GEs) as constituting socially important access rituals, and functioning as a marker of politeness.Greetings have been considered by different scholars as fulfilling different functions such as, for instance, invitation to social relationships or phatic communion (Brown 1980: 114); cherishing relationships already in existence; expression of polite behaviour (Yahya-Othman 1995: 211) and acknowledgement of recognition (Schegloff 1968).
A greeting exchange in the Ngoni community is determined by factors such as the age (or seniority) and the sex of the interlocutors, social relations between participants, speech event and location of the GE.Ngoni greetings will generally be regarded as expression of polite behaviour while not delving into a discussion in the following sections on where exactly the greeting exchange occurs in an encounter between Ngoni interlocutors.

2
Theoretical background and framework of the study Politeness research has made and is still making a lot of advances since the publication of the seminal work by Brown and Levinson (1987) who viewed politeness in connection with the notion of "face".Brown and Levinson (1987: 62) defined "face" as the basic wants which every societal member knows that every other member desires, and which also is in the interest of every member to satisfy.So initially politeness was understood as a potentially universal concept and in relation to attempts by participants in an interaction to avoid causing discomfort or embarrassment to the collocutors.While the basic idea of "face" remains unchallenged, the argument that realisation of politeness is universal has been overly criticised and has led to some conceptual modifications.Mills (2011: 24-26), for instance, argues that in different cultures, politeness can be realised in different ways based on different values and premises, and has different functions other than protecting face.As an example, she mentions discernment of individuals as an expression of politeness in specific contexts so that they are addressed and treated accordingly.The notion of "discernment" was initially provided by the pragmatician Sachiko Ide (1989) who brought up this notion as a critique to universality of the politeness model proposed by Brown and Levinson (1987).Ide's main argument has been that in Japanese culture honorifics are used strategically as a marker of politeness.
On the possibilities of different parameters to politeness, Lackoff (1975: 53) had argued earlier that there are many types of behaviour that can be called "polite".She characterises some forms of politeness as linguistic, some as non-linguistic, and some as mixed.Different forms may be regarded as polite in some settings and impolite in others, some as neutral and some as downright rude.As such Lackoff (1975: 67) proposed three "universal rules" for politeness, which are not mutually exclusive: (i) formality (the need to keep aloof), (ii) difference (the need to give options using different ways such as euphemisms, hedges etc.) and (iii) camaraderie (which is the necessity to show sympathy).As can be seen, these different postulations point to the need for specificity in studying politeness in different cultures since universality cannot cut across the board.
Another scholar who has convincingly argued towards a more balanced understanding of politeness seems to be Haugh (2013).In his study on evaluations of (im)politeness as a form of social practice, Haugh (2013: 53-54) has argued that, despite all the criticisms levelled against the study of "facework" politeness is still evaluated with regard to facework and norms as evaluated by participants.In a way this is to say that the need to protect the face of the collocutors is still prominent even in the proposed replacements.
Greeting exchanges have been viewed by Goffman (2010Goffman ( , 1971) ) as "access rituals" and he argues that they "occur between individuals at the point where they find themselves about to enjoy a period of heightened access to each other".As such these exchanges "mark a transition to a condition of increased access" (Goffman 2010: 78-79).Goffman (2010: 62) further defines a ritual as "a perfunctory, conventionalized act through which an individual portrays his respect and regard for some object of ultimate value to that object of ultimate value or its stand-in".Goffman (1971: 79) groups greetings into two categories, namely "passing greetings" and "engaging greetings".Whereas a passing greeting can be regarded as a mere "ceremonial acknowledgement of the possibility of access", engaging greeting exchanges invite further access.
When it comes to identifying linguistic elements of greeting encounters, Braun (1988: 7) has further differentiated between different categories of address forms which comprise "words and phrases used for addressing".Address forms refer to the collocutor, they contain a strong element of deixis and from a cross-linguistic perspective they are mainly derived from three word classes, namely pronouns, verbs, and nouns, the latter supplemented by words which are syntactically dependent on them.
Mostly defined by their social function (as means of phatic communion, invitation to establish a social relationship, expression of [im]polite behaviour, see above) greetings as well as leave takings will have to be defined as structured and socially meaningful parts of opening and closing sequences of verbal exchanges among participants in a particular situation and setting.
As prerequisite for the study of greetings in a given community of practice, Duranti (1997: 67)  While taking Duranti's (1997: 67) approach as a point of departure and reference we will describe verbal and non-verbal elements of greeting exchanges in Tanzanian Ngoni by analysing five concrete examples (see § 4).We will use extra-linguistic information (metadata), visual (video) and speech data (recordings) and the main focus of the linguistic analysis will be on the identification of Swahili elements in Ngoni greeting encounters among relatives (husband/wife, parent/child) in the home domain in three villages in Ruvuma region.
In this context we will also take into account that [w]hatever greetings accomplish, they do it by virtue of sociohistorical circumstances.To say that greetings are constituted by formulaic expressions only tells half of the story.The other half is how such formulaic expressions may be adapted to, and at the same time help establish, new contexts.(Duranti 1997: 88) 3 Description and analysis of Ngoni GEs In his book Mila na Desturi za Wangoni ('Culture and Traditions of the Ngoni') Nyirenda (2006: 66) observes that it is customary for the Ngoni to greet whenever they meet, "Vangoni pevikonganika, vilamukilana" ('When the Ngoni meet, they greet each other').Nyirenda is a Ngoni himself who was born and grew up in the Ngoni community.Ngoni greet each other when they meet, regardless of whether or not they know each other.Besides passing and engaging greetings, the Ngoni GE has an additional one -a working greeting which has the dual function of passing and wishing the people involved in the work success.Greetings in the past served a number of functions in the Ngoni community including wanting to know about the situation of the collocutor, a marker of recognition and respect, an expression for requesting support when needed, and also a signal of community ties.Failure to offer or return a greeting in the community attracted a reprimand by elders.
Additionally, due to contact with Swahili, the Ngoni words are being replaced by their Swahili counterparts baba ('father'), mama ('mother') etc. Pronouns are marked on the verb, and traditionally, the plural nominative form tu ('we') along with the objective form va (plural 'you'), as in tu-va-woni ('we greet you') have been used.Besides these forms of address, endearment terms are rarely used because endearing relationships are considered private, and the Ngoni avoid calling people by their names because they want to keep distance.The Ngoni ethnolinguistic group has been in contact with many other groups since their start from South Africa to date.1 This contact situation has recently been intensified due to socioeconomic forces operating at various levels: community, regional, national and global (see above).However, increased contact situation starting from around the early 2000s in the area has been seen to affect particularly the Ngoni language in terms of lexical borrowing from Swahili and the development of ambiguous attitudes among the speakers of the language towards their language (Mapunda 2013;Rosendal/Mapunda 2014).
Coming down to the household level, Ngoni households in the past lived together with their associated families in the same compound.As is typical for patrilineal communities, the father usually acquired a big piece of land, and distributed some of it to his sons (usually not daughters, on the argument that when the daughters marry they would move away and belong to their husbands' households).After having lost their cattle during the Maji Maji war (Moser 1983: 189) the Ngoni became small-holder farmers, animal keepers and hunters.Currently, they are mostly small-holder famers.
Before they started the day in the morning, the practice was for them to greet one another.
This was aimed at knowing if each household member was fine.Similarly, every member of the community was concerned about the affairs of other members of their community.The way one greeted their parents was the same way they would greet other elderly or older people.However, in the middle part of the 1970s the Government of Tanzania embarked on villagisation policy which forced all people to come and live together in villages.This disrupted the household life pattern.
Additionally, Western education brought about by missionaries and later on by the Government, brought about further alterations on the Ngoni household structure.However, it must be understood that all the changes being talked about relate to life in the rural villages, and a distinction needs to be made between life in the village and life in relatively big towns and semiurban locales today.In Ngoni villages, people have retained intimacy and life has all along been communal.As such we are here mainly talking about GEs in the villages.

B:
Eeh! Kw-a bwina.Yes 17-be well/good 'Yes!It is fine.' Examples from the Ngoni data sample show that independent short and long pronominal forms as well as dependent subject pronouns are used in verbal greetings embedded in the dyadic GE.Although all subject concord forms may be used in GEs, the use of pronominal elements as address forms is restricted to the second person singular and plural, addressing the interlocutor either as an individual speaker (2.sg.) or as member of a family/household (2.pl.): Plural forms may also be used for a single referent to express respect/politeness like in the following example where the son addresses his mother when visiting her house while referring to himself with a verb containing a first person plural subject concord element: Tukuwoni mawu We greet you mother.('How are you mother?')Among relatives and among husband and wife a number of verbal salutations were used as greetings (see ex. 1, role play, § 4).During the first two turns the husband (B) initiates the greeting without receiving a response from his wife (A).Only in the third turn -now initiated by the wife (A) -the GE becomes symmetrical and is now exclusively carried out in Ngoni: In some cases Swahili forms are used in the Ngoni greeting.As mentioned by Ngonyani (2003: 103-104) and Moser (1983: 253-254) the Ngoni equivalent for baba is dadi whereas the kinship term kaka is used in Ngoni to refer to an elder brother.According to Ngonyani (2003: 253) the respectful Swahili address form (title) mzee 'elder' may be used in Ngoni (sg.mzee, pl.vazee) together with mgogoro, a term mentioned by Ngonyani (2003: 106) as m'gogolo 'old person'.
For the Swahili kinship term mama 'mother', Ngoni has mawu 'mother' or nyina 'mother' as equivalents.Swahili gogo for 'grandmother' is mentioned in Ngonyani (2003: 103) for Ngoni as well whereas Moser (1983: 137) also lists mbuya 'grandmother', a term which is said to be used side by side with the Swahili term in Ngoni.In the interview data some terms of address, i. e. kinship terms, were also used as plural forms such as mwavamawu 'mothers' and mwavabambo 'gentlemen', the latter form by means of a title bambo meaning 'gentleman'.
One of the (Swahili) terms of endearment used in the role play data presented below is bwana 'master, mister, sir' which in informal contexts conveys the meaning 'pal' or 'mate'.It can be used to address male as well as female interlocutors.In the role plays some other address terms were also used such as mwanangu 'my child' and wamuyangu 'my mate'.The latter term was used in the greeting exchange between son and mother but also among a middleaged couple (see examples in § 4).
Besides pronominal, verbal and nominal forms of address speakers also make use of discourse particles in GEs.Some of these have been classified as interjections by Moser (1983: 122) and Ngonyani (2003: 72).In the greeting exchanges the form yewo is used as an affirmative response to opening questions.Forms such as eeh or e(e)na 'yes' and haya 'ok' are also used.

Physical arrangement, status relationship, initiation and sequencing of GEs
As will be further illustrated in the examples below, Ngoni greetings at home are carried out while participants are seated.Differences in status and/or seniority are expressed by the sitting arrangement, i. e. a senior participant or a participant with higher status position is seated on a wooden stool (N chigoda) or on a chair during the GE.Junior participants usually initiate the GEs and their verbal salutation is accompanied by certain movements such as crouching, squatting and sitting down.The verbal exchange is further accompanied by handshakes.

3.4
The use of Ngoni and Swahili forms (code-switching) in Ngoni GEs If participants do not know each other well or if the relationship between them is formal, a GE typically consists of three parts, (i) opening, (ii) chitchat and (iii) leave taking.However, among intimates the first two parts can be regarded as common.
Our data show that an opening verbal salutation exchanged between elderly and some middleaged members of the Ngoni community typically consists of dyadic exchanges structured as follows (A initiates the GE, B responds to the verbal salutation): A. In the interview data, this salutation was used by seven out of the 18 participants.Five out of six of the participants aged 51+ years used it while in the age group 31-50 two out of six employed this kind of Ngoni greeting sequence.For participants aged 15-30 years five out of six and four out of six of those aged 31-50 years the use of a Swahili verbal salutation was obviously identified as being the appropriate Ngoni choice.
In other words, nine out of all the 18 participants used Swahili forms of address in GEs: A. Shikamoo 'Hello (I greet you, who are older than myself), B: Marahaba 'Hello (I accept your greeting younger one) (also Table in  § 4.5).

Methods of data collection
Currently the Ngoni speaking area is in a situation of intense contact with Swahili, both directly and indirectly.This paper looks into how greetings and forms of address have been affected.It is initially hypothesized that the intense contact has influenced GEs and forms of address, thus necessitating in a redefinition of what is regarded to be polite behaviour as opposed to what it was regarded to be when contact was minimal.
Three methods of data collection were used in this study, namely: role-play, semi-structured interview, and observation.The use of role-play was meant to show how Ngoni speakers actually carry or carried out their GEs and how they apply their knowledge on GE norms and practices which they have acquired in the course of growing up in a community where this kind of daily routine is still performed.Role-plays were recorded with a digital camcorder after the interviews had taken place.
This set of data adds to the information gathered during the interviews.The use of semistructured interviews was meant to get insights into diachronic developments such as how the expression of politeness through greetings and address forms has evolved and may have changed over time.Interviews were semi-structured in the sense that there were guiding questions, but the order of questions and follow up questions varied from one interviewee to another.
A total of eight hours and twenty minutes (8:20) of interviews were recorded using a powerful ZOOM H4n digital voice recorder.The recording took place in the participants' houses, and consultants were already acquainted with the interview situation through their participation in previous studies.Observation method was meant to capture different behavioural patterns which participants showed when conducting GEs in daily encounters in their villages.Notes were taken immediately or immediately thereafter.In terms of other profiles of the participants, out of the 18 participants 16 had primary, one had elementary and one secondary levels of education.This composition was unavoidable because almost all the villagers in the area did not go beyond primary education, partly because secondary and higher education was not readily available in the villages.Socioeconomically, the group was homogenous.2Two of the participants were retired catechists and one ran a small retail shop, the remaining participants were all small-holder farmers.

Data analysis: role plays and interviews
As has been said before, greetings are conversational routines that serve a number of sociopragmatic functions in all human societies.They follow particular patterns which also serve to distinguish one system from another.Some patterns are unique to particular societies but some are similar across societies.
In this paper we are looking at Ngoni GEs and analyse them in their own right before providing an account of how these patterns are changing as a result of the complex interplay between contact influences (mainly from Swahili) social changes and internal dynamics within the community itself.We look at how GEs are carried out using role-play data performed by first-language Ngoni speakers in Ngoni villages along with interview data.The role-plays were performed by five pairs of participants.The first two were performed by husband and wife, the third by mother and son, whereas the last two were carried out by consultants of differing ages who were not sharing a kinship relationship.Since the participants were all practitioners of the GEs they illustrated, we believe that what is shown and described can be regarded as a true replica of Ngoni GEs.
In some cases we are using photos of the participants taken from the video recordings to provide as close an account of GEs as possible and to show the performance of non-verbal parts.
In the performance of GEs, it will be argued, non-verbal arrangements in GEs among the Ngoni are very important, possibly even more important than the verbal salutation parts.Due to this truism we are trying as much as we can to show various non-verbal behaviours which are used (or at least which are supposed to be used) when the Ngoni carry out GEs.

How Ngoni GEs are carried out
The use of role-play as a data collection method has also been used in other related studies (cf.Demeter 2007).The justification for the use of this method for this study is that the consultants used their lived knowledge of greetings which they grew up with, and which they actually practice or practiced in the past.An additional circumstantial advantage is that the role-plays were performed in the same community on which the study is based.Role-plays have also been used in the study of face work e. g. in working contexts in South Africa by de Kadt (1998).In her study it was shown that this kind of method can successfully be employed in socio-pragmatic research contexts in order to find out about how participants employ politeness strategies in "asymmetrical" Zulu conversational encounters.
In the excerpts below Swahili words are in bold.The volume of Swahili used in an otherwise Ngoni-only GE was sizeable.Participants claimed that the Swahili forms are commonly used among the Ngoni these days.Nevertheless, we regard these verbal routines as code-switches.
Code switching refers to the use of more than one language or variety within a communication transaction (Auer 1998;Myers-Scotton 1993).

GE by an elderly couple (husband returning home from a journey)
The first GE role-play is performed by an elderly couple.The husband is 84 years old while the wife is 82.The context is that the husband is coming home from a far-away journey in another province where he had gone to visit their children who lived there.When he gets home, after having been away for a number of days, he finds his wife seated outside their home, and there is a short wooden stool (N chilimba) which is vacant.
The husband (A) immediately offers his right hand to his wife and the left one on one side of the stool, avoiding eye-contact and also facing away from each other.The wife in return also offers her right hand and the left one resting on her left knee, but not looking into her husband's eyes.The man's offer of hand goes almost simultaneously with a verbal salutation to his wife.After the first turn, the two hold their right hands and twist them left, then right and then left (three times).The husband greets again.Of days there?('What's the news after many days there?')
We have woken up.

B:
Habari za safa (She looks at her husband briefly) News of the journey?('How was the journey?') A: Tigendi bwina.(Each one puts their two hands together) We walked safely.('The journey was safe.')

B:
Mgendi salama?You (pl.) walked safely?('Was your journey safe?')Looking at the GE by this elderly couple, a number of things come to the open.The issues that emerge can be discussed in terms of (i) physical arrangements, (ii) verbal salutations and turn taking sequences, (iii) nonverbal behaviour and (iv) overall length of the GE.
Starting with the physical arrangement, the wife was already sitting down when the husband was arriving, but when they are already together the wife is sitting on the floor while the husband is sitting on a wooden short stool -chilimba.
The husband sitting on the stool but the wife on the floor may be explained by the fact that the Ngoni community is patriarchal, men being regarded as the more powerful.So the sitting arrangements reflect power relations in the family, and here it may be concluded that the wife accepts her position in the family by sitting at another level than her husband.This arrangement can be considered along with turn-taking routines which require that the subordinate (in terms of age, socio-economic status, speech event at hand etc.) has to initiate the greeting as is explained in more detail in the next section.
Regarding the turn-taking sequence, the husband initiates the verbal salutation twice but the wife does not respond.Instead she initiates it herself, and the husband responds immediately.
According to the Ngoni traditions, it is usually the subordinate who initiates a GE.But another tradition holds that the one who arrives initiates a GE (Nyirenda 2006: 66).
However, what is seen in this GE is a paradoxical situation in which both husband and wife have some reason for initiating the GE.On the one hand, it seems that the wife is aware of her social position, and so deems it unacceptable for her husband to initiate the GE.Meanwhile, the husband who is arriving is also required to initiate the GE.So in this case the wife observes the rule lest she be sanctioned by her husband.Nevertheless, when the wife initiates the GE the husband does not hesitate to respond which may also explain his readiness to bear with any sanction should there be one.This GE lasts for 21 seconds, and appears to be the longest of the greeting exchanges considered here.It takes a total of 14 turns including two silent responses.When the husband arrives from his journey and after being offered a chilimba by his wife and after sitting down, the husband initiates a verbal salutation, incidentally in Swahili.His wife does not respond to this salutation.The husband initiates a verbal salutation again, but the wife again turns it down by keeping silent.In turn, the wife initiates with a verbal salutation.The length of the GE is assumedly sustained by the length of time the husband has been away, but also by the urge by the wife to get information on not only the health of the collocutor, but also of the manner in which the journey took place and also the situation of the children and their families.
Another observation relates to the verbal salutations regarding the use of Ngoni and Swahili.
Usually code-switching and borrowing are widespread (Rosendal/Mapunda 2014), and the younger generations often code-switch between Ngoni and Swahili.However, the GE by this elderly couple suggests that even though they are both advanced in age, code-switching to Swahili is a common feature in their GE.They both use various code-switches, ranging from singly occurring Swahili words, a combination of words to a whole turn.

GE between a middle-aged couple (meeting in the morning before starting work)
The second role-play being described here is between husband and wife.The husband is 50 years old while the wife is 45.The encounter is taking place in the morning at their home in village A. The wife had woken up earlier than her husband, and she was doing routine domestic chores outside the house.The husband has just woken up and is sitting on a short stool outside their house.The wife sees him and so she comes to greet him.
The wife (A) sits down on the floor and offers her right hand first, followed immediately by her left hand.The husband (B) in return offers his right hand and the other hand rests on his left lap, in which case both the wife's hands engulf the husband's right hand.The offer of hands by the wife goes hand in hand with a verbal salutation to the husband.Both are looking at each other very briefly and then each of the interlocutors looks in a different direction.For some time there is no direct eye contact between the two, but after a few moments the husband looks at his wife from time to time but the wife avoids eye contact, looking away.
The physical arrangement in this GE is similar to the one found in the first role play, performed by an elderly couple.The wife is again sitting down on the floor while the husband is sitting on a stool.In this situation the husband is viewed as being "higher" both in terms of the physical positioning vis-à-vis his wife in the greeting encounter and in terms of the sociopsychological configuration of the relationship.Husbands and wives in the Ngoni community have different social roles and responsibilities which distinguish them.It is usually the husband who pays dowry, who owns the farms and houses for the family.In their upbringing, boys are taught how to be heads of their families.Women are taught to be obedient and submit to their husbands.This may explain why women sit down on the floor while husbands sit on vilimba (traditional three-legged stools).In the interviews it became apparent that it is the subordinate who initiates a GE.The scope of "subordination" covers issues of age, who arrives first, socio-economic status, etc.In this regard, the wife initiates the GE and the husband responds without ado.Comparing this to the elderly couple where the husband initiates the GE, the turns in this GE were all properly responded to.Another observation that needs to be made is regarding eye-contact, proximity, and hand-shake.Eye contact among the Ngoni is brief and elongation is not acceptable.This is also observed in the very brief eye-contact in the GE by this couple.Hand-shake is long.This GE lasts 16 seconds, and immediately the interlocutors start talking about the activities of the day while still sitting.What is unique in this GE is that there are fewer turns than in the previous GE.This is explained in the interview by participants 11B and 16A.Participant 11B is of the view that if you don't meet very often, the greeting becomes longer.Similarly, participant 16A holds that "the length of a greeting is determined by how much respect one commands, and the length of time the collocutors have not met".This explains why this couple having been together in their house did not engage in a more extensive greeting exchange.
Here again the verbal salutation is characterized by code-switches between Ngoni, the matrix3 language, and Swahili, the embedded language.Both code-switch, but the husband does it more frequently than the wife.Regarding the use of the terms of address, it is usually the kind of relationship that pertains between the two which is used as the term of address.
Both the wife and the husband address each other as wamuyangu 'my mate'.The use of this kind of address term can be explained by the fact that the Ngoni do not usually call each other by name; they prefer avoidance of calling other people by their names and greet without using address terms.Another issue refers to the use of a Swahili address term used by the husband to the wife: bwana.This address term literally means 'sir, lord, master' or 'mister'.Typically this address term is used when addressing a man, and not a woman.However, in this situation is has been used generically to express intimacy and closeness, so it may be taken to mean 'mate' (see also role play 1).

GE between son and mother (at the mother's house early in the morning)
The third GE is between mother and son.The mother is an elderly woman aged 82 years, and the son is 50 years old.The GE takes place outside the mother's kitchen hut, in Village A in Songea Rural District where they both live.Their houses are about 70 meters apart.
The typical Ngoni GE between mother and son is elaborate, involving quite a number of turns before they take leave.We see that while at home, there are important physical arrangements Gastor Mapunda and Gabriele Sommer: When Shikamoo Mama/Baba replaces Tukuwoni Mawu/Dadi ISSN 1615-3014 79 which make the exchange be performed in an acceptable manner.For example, between the mother and the son, the mother is seated and the son comes closer, and takes a few moments almost simultaneously squatting, offering both hands and then uttering the verbal salutation.
After the first turn, the son is still squatting and the GE goes on.They both look at each other in the eyes while the GE is going on.The son replies (turn 2) while still squatting.He stands up and says the last turn.The mother again puts her hands together and gazes at her son who stands up and starts moving away.Ee, voha viyumuka.Yes, all are waking up.

GE between daughter and father (at home in front of the house)
The father is sitting on a chair outside their home.The daughter is aged 30 years and she goes to her father, crouches with her left knee up, the right one down, her body resting on her right leg, simultaneously offering her right hand to her father.The daughter's left hand holds her own right hand wrist.The father offers his right hand in return, and his left resting on his left lap.The daughter instantly initiates the verbal salutation to her father, while still crouching: As explained earlier, each age group, that is, 15-30, 31-50 and 51+, had six participants, of whom three were female and three male.In the 15-30 years group, five participants out of six, greet with the Swahili shikamoo while only one greets in Ngoni.However, the Ngoni salutation which 1A used, habali za lelu, is a calque4 from the Swahili salutation habari za leo (Ngonyani 2003).
As is shown in the Table above, all of the 18 interview participants answered the interview question "How do you greet a person who is older than yourself?".The findings show that out of the 18 consultants nine said they greeted a person older than themselves with the Swahili verbal salutation shikamoo and the response by the older person was marahaba.
Another observation comes from the verbal salutation provided by participant 16 who is an elderly man aged 84 years and who claimed to greet a person older than himself with a verbal salutation nikuwoni mzee, meaning 'hello old man'.Clearly this verbal salutation is a Swahilized version of the Ngoni form/expression tukuwoni bambo which contains the greeting nikuwoni and a term of address mzee.It is Swahilized in two ways: • nikuwoni is Swahilised in the sense that the typical Ngoni verbal greeting does not use the singular subject marker niin a greeting.This is evident in the interview recordings in general, when participants related it to translating it in accordance with the Swahili first person singular marker (ni) and plural (tu).In the interviews some participants said that the ni-kuwoni version is a rather recent development in the language.This reasoning is supported by responses provided by five participants aged 31 years and above whose verbal salutation to persons older than themselves is tu-kuwoni, and not nikuwoni.
• mzee is a Swahili word for 'old man' (for the Ngoni equivalent see Ngonyani above).
Because of co-presence of the two languages in the community, the greeter may freely use the Swahili address term or title in a GE today.Regarding the use of terms of address such as titles, kinship terms and so on, it was concluded from the interview data and also from researchers' own observations in the community, that these are less frequently used in Ngoni GEs.Usually GEs go without terms of address.However, some people still use them.For example from the question "How do you greet a person older than yourself?" 5 out of 18 participants five, i. e. 27.8 % of the consultants, used terms of address in their greetings.Terms of address were used in all the five role-plays (see § 3.4 above).
The common pattern in this case was as follows: a term of address is used in the first turnthe opening turn.The return turn by the interlocutor may or may not contain a term of address.In the role plays there are examples showing a GE in which the opening turn embodies a term of address but not the return turn.There are also examples in which both the opening and the return turns contain terms of address.There are also examples of terms of address appearing in subsequent turns.
From the examples in the data it can be seen that terms of address in Ngoni perform at least two functions: declaration and acknowledgement of the relationship between the collocutors, and secondly expression of intimacy between the collocutors.This can be seen in the following example taken from the GE between mother (B) and son (A): A: Tukuwoni mawu.We greet you mother.Another observation regarding terms of address is that the most often used terms of address are the Swahili terms baba 'father', kaka 'brother' and mzee 'old man'.These were used regardless of the age of the greeter.This observation is apparent among all the age groups in the interviews.
However, in the role plays both Ngoni and Swahili terms of address were used; mawu 'my mother (N)' and wamuyangu '(you) my mate (N)' as well as mwanangu 'my child (S)' and bwana 'sir' which was used in the data as an endearment term with the meaning 'pal' or 'mate' (see above).What is interesting to see is that even elderly participants used Swahili terms of address instead of Ngoni ones.

5
Are there changes in the way GEs are carried out?
Another question asked was whether or not there are changes in GEs as practiced by the Ngoni these days as opposed to greeting habits practiced in 'the past'.Those consultants who said that there are changes were also asked to mention what these changes are.Interestingly, all interviewees claimed that there are in fact changes in Ngoni GEs.
Members of the first age group (15-30 years) held the view that it is mostly young people who have changed their greeting behavior.Besides changes in verbal forms, changes also relate to non-verbal behavior.These days, movements and gestures which were the actual markers of politeness 'in the past' are nowadays being ignored.A5, one of the young participants, had this to say: It was not possible to greet while walking.It was necessary to stop, bow or kneel and then greet.These days they may even greet you while walking.
The youngest participant, 5B, argues that there are changes which he too has seen.However, regarding the use of Swahili in GEs, it is sometimes the parents who instruct their children to use these forms and ways of behaving.Similarly, for those who still abide by Ngoni nonverbal aspects of greeting behavior would be just one out of ten families.He argues: I don't use old greetings like tukuwoni because I don't know them.My father has taught me to greet in Swahili shikamoo.I stop and kneel when I greet, and if I am on a bicycle I alight to greet.We like the modern greeting because we do not know the old one.
Another observation is that in the society some young people may decide even not to greet at all.The most they can do is to greet you in Kiswahilishikamoo.Participant 3C, a young lady aged 28 years says that she cannot greet the Ngoni way because it is not natural for her age: Tukuwoni is for the elderly.If she/he greets me tukuwoni I respond nzuri ('fine'), shikamoo.I see her/him as outdated.If I respond in Ngoni even my fellow young people will see me as pretending to be polite.
The second group comprised the 31-50 year olds.These generally complained that the way young people greet these days is scornful.In the past kneeling was very respectful.A younger person knelt to an older person, but today they greet you while standing up.The language that they use now is Swahili, so they would tell you shikamoo or even mambo/shwari?
Another observation is that if proper Ngoni GE should be carried out, it is for people who know each other or who are related.Among the youth only a few still kneel or squat, and most of those who still do it, do so to their own parents only.Participant 9C has this to say: We use shikamoo or at least habari za lelu but not tukuwoni.The youth do not show respect when they greet: there is usually no kneeling, but only for people who are related or know each other.The youth have their own cavalier greetings Oya, mambo?('Hey, issues?').
Another observation made is that GEs used among the youth now are shorter, and they mostly greet while walking, not stopping.Among the youth themselves, they may pass even without greeting each other if they don't know each other, but old people have not changed much, they still stop and greet the Ngoni way.
As to why they don't kneel when they greet, participant 12C aged 48, is of the view that kneeling would unnecessarily glorify those people to the maximum.He says: We are now more interested in what we see on TV, video etc.So we are fond of how other people in other parts of the world do.
The last group were 51 years old or elder.One of the observations made among them is that most (but not all) old people have not changed, but the youth.They are unhappy about the use of the Swahili form shikamoo and passing by without stopping when they greet.Participant 13A laments that "They may greet you even on a bicycle.They lack respect".Participant 15C also laments regarding the way the youth carry out GEs: Linguistik online 84, 5/17 ISSN 1615-3014 88 They greet shikamoo and without stopping, they are just on the go.They only stop when they greet those whom they know.They may stop and offer you a hand and say Habari za leo mama or shikamoo mama.They don't kneel these years.They don't listen to anyone.This disrespect is rife.
Participant 17B is of the view that what the youth are doing is very disappointing.He says sometimes he is told things which he never expected: The young people may tell you things you never expect when they greet you mambo mzee?So I would usually respond: What issues?; they don't kneel; we see only scorn.They should use these greetings only among themselves but not with old people.
Participant 18C who was 67 years is the only one who does not see changes in GEs as a problem, but rather as a natural phenomenon.He says: They simply greet you shikamoo baba while still standing up.I do not see it as a problem because life changes with time.There is nothing that can be done, so we let things go.

Conclusions
As is clear from the current study the issue of GEs is an important means of expressing politeness among the Ngoni of Tanzania.However, following increased contact between the Ngoni community, and internal sociocultural dynamics, greeting habits have undergone and are still undergoing serious changes.
Increased contact with Swahili and other internal dynamics such as deaths of older community members, increased modern education provided mostly through Swahili and English, increased presence of Swahili in the mass media and in the community, modernised ways of life, and opening up of the community to maize business have all pointed toward a situation that traditional Ngoni GEs will eventually be replaced.
Furthermore, preference by the youth of seemingly simple Swahili GEs makes Ngoni GEs look antiquated and uninteresting.
While prior to the 1990s the Ngoni would usually stop for a short while and exchange greetings as a sign of not only respectful behaviour but also that one cared for the interlocutor, currently the youth are of the view that such exchanges simply waste their time.In due course, what can possibly be done is describe, analyse and document Ngoni GEs with a view to preserving them for future use -maybe also among younger generations.
formulates three major points, i. e.(a) ethnographic description of greeting encounters, (b) the recording of what has been actually said, and (c) a working definition of the phenomenon that has been investigated.Studies of greeting exchanges, he claims, should not only rely on interview data, participant observation or field notes but also additionally on film recordings.For a detailed analysis of access rituals Duranti distinguishes six major criteria (1997: 71): 1. Near-boundary occurrence 2. Establishment of a shared perceptual field 3establishment of a spatio-temporal unit of interaction 6. Identification of the interlocutor as a distinct being worth recognizing

B
's the news, mate?') A: ………….. (no response) B: U-yumuk-a 2sg-wake up-FV ('Are you fine?') A: ………….(no response)A: Za magono uko?Of days there ('What's the news after many days there?')B: Ti-yumuk-a 1pl-wake up-FV 'We have woken up.' ('It's fine.')Ascan be seen from several other examples of Ngoni GEs carried out among relatives in the role plays (see § 4 for more details), intimates use nominal forms of address whereas no such form is normally used among people who are not closely related.Among the more common the role plays or mentioned during the interviews are male kinship terms or titles.Baba 'father' and kaka '(elder) brother' are kinship terms which have equivalents in Ngoni.

Figure 1 :
Figure 1: The Ngoni speaking areaA total of 18 consultants from three villages of Mgazini, Muungano Zomba and Lugagara in Songea Rural District (see Map) participated in the semi-structured interviews.Nine of them were female and the other nine were male, split into three age groups: 15-30, 31-50, and 51+.The identification of the participants was done with the help of local government leaders.The data were collected in May 2014.All the interviews lasted between 20 and 40 minutes.In terms of other profiles of the participants, out of the 18 participants 16 had primary, one had elementary and one secondary levels of education.This composition was unavoidable because almost all the villagers in the area did not go beyond primary education, partly because secondary and higher education was not readily available in the villages.Socioeconomically, the group was homogenous. 2 Two of the participants were retired catechists and one ran a small retail shop, the remaining participants were all small-holder farmers.

Table 1 : Forms of address/greeting formulae used by consultants from different villages
They do not follow the traditional Ngoni way in a number of ways.Regarding the verbal salutation, they opt for Swahilized versions and sometimes even use colloquial Swahili forms or expressions such as Mambo?/Poa ('Issues?/Cool'),or Je?/Shwari ('How?/Peace').By greeting in Swahili the youth feel globalized.An example can be taken from what participant 3C says: We young people use Swahili.Even if an old man greets me in Ngoni tukuwoni, I respond in Swahili nzuri, shikamoo.We don't use the old greetings, because to do that is to be outdated.