caton greetings from yemen.pdf

7/27/2019 caton greetings from yemen.pdf

1/20

"Salm tahyah": Greetings from the Highlands of Yemen

Author(s): Steven C. CatonSource: American Ethnologist, Vol. 13, No. 2 (May, 1986), pp. 290-308Published by: Blackwell Publishing on behalf of the American Anthropological AssociationStable URL: http://www.jstor.org/stable/644133Accessed: 04/02/2010 13:22

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at

http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless

you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you

may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/action/showPublisher?publisherCode=black.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed

page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

Blackwell Publishing andAmerican Anthropological Association are collaborating with JSTOR to digitize,

preserve and extend access toAmerican Ethnologist.

http://www.jstor.org
http://www.jstor.org/stable/644133?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=blackhttp://www.jstor.org/action/showPublisher?publisherCode=blackhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/644133?origin=JSTOR-pdf


2/20

salam tahiyah:greetings from the highlands of Yemen

STEVEN C. CATON-Hamilton College

introductionThe ethnography of communication claims that it is possible to describe speaking as rule-

governed behavior. While it was important that this point be demonstrated by a number ofexcellent articles and monographs describing ways of speaking in different cultures (Schegloff1968; Labov 1972; Gumperz and Hymes 1972; Irvine1974; Bauman and Sherzer 1974), thefield is not in my opinion related significantly enough to the concerns of contemporary, espe-cially symbolic or interpretive,anthropology. In other words, speaking entails more than aknowledge of how to use signs "appropriately"in certain social situations-a view of be-havior in the perspective of an older, more normative sociology-it also involves the commu-nication of cultural meanings that are interpretedby actors in social contexts. By linking theethnography of communication with what Geertz (1973) has called "cultural interpretation,"sociolinguistics can be made to seem more vital and interestingto the concerns of anthropol-ogists.

Attentionto the "native point of view" was alreadyevident in the earliest formulation of theethnography of communication by Dell Hymes (1974). He referredto it as the "emic" modelof speaking, defined as the "participants'own explanations and conceptualizations of theirbehavior" and "their 'homemade models' " (Hymes 1974:11-12). A particularly elling use ofthe emic model to help explain speech behavior can be found in Heath (1983), a studyof speak-ing in black and white working-class communities of the Carolina Piedmont. And certainly,this article will depend heavily on an "emic" view of speaking, especially where the sayyid'speech community is concerned, as isdescribed below. In Basso (1979) we areconfronted withthe Apache's parody of "Whiteman" ways of speaking thatdeeply offend Indian sensibilities.Itis a double-edged, metapragmaticevent: at once a scathing condemnation of "Whiteman"speech (and by implication the white man's moral character) as well as an affirmation ofApache ways of communication (which are connected with their concepts of human dignity).

Basso's approach is what one might call a "culturalinterpretation"of speaking, which I in-tend to emphasize in this study of Yemeni greetings. In my view, too little has been done towed the ethnography of communication with culturalanthropology'sconcern forsymbols andtheir meanings.There is, of course, another side to this intellectual coin. Whereas interpretiveanthropologyhas made use of linguistic data to elucidate culturalconcepts central to the society under in-vestigation, rarelyhassuch use been motivated by linguistically informedparadigms.In MiddleEasternethnography, to take an example with which I am more familiar, the interpretive ap-

The North Yemeni speech event of greeting as a rich semiotic act is interestingtoanalyze from both a linguistic and an anthropological point of view. It is arguedthatan indexical or pragmaticapproach combined with an "interpretive"or sym-bolic understandingof culture may lead to the most interesting insights into verbalforms and theirsocial meanings. Inparticular,it is argued thatdifferent constructsof the "person" are created in the speech event. [Arabic, ethnography of com-munication, interpretiveanthropology, and self]

290 american ethnologist


3/20

proach has been brilliantly represented by C. Geertz (1979), H. Geertz (1979), Eickelman(1976), Rosen (1972, 1984), Meeker (1976, 1979), and others; all of them, to varying degrees,have noticed the importanceof "language" to understandingtheir interpretive problems. Con-sider the following passage from a recently published work by one of its leading representa-tives:

integralo the unification f the social andconceptualdomainsof Sefrou ife wasthe roleplayedbylanguage nthe formation f socialrelations.There s, of course,nothingmysteriousboutthecentralrole of speech in MiddleEastern ocieties .. As I watched he peopleof Sefroumaneuverwithin herange ftermsandmeanings vailable or hecharacterizationftheirrelationships,sawthat inguisticusageswere notsimply abels attached o an availablearray f socialpositionsand roles butwere in-tegralo theverycreationof those ties[Rosen1984:3].However, when one proceeds to examine what is meant by "language" in this importantwork,one discovers that the author rarelyventures beyond the lexicon. We are given an analysis ofnames, recapitulatedfrom an earlier work by C. Geertz (1979), key terms such as tar (shame)whose semantic range is supposed to tell us something significant about the Moroccan culturaluniverse, kinshipterms, and an occasional proverb. But if I single out Rosen, it is not becausehe is alone in his basically semantic and language-reduced-to-the-lexicon approach, for herepresents a continuation of work done on the problem of language-and-culture in Moroccosince the 1960s when interpretiveanthropologists firstbegan to do their fieldwork there, andone fears that the limitation of this kind of research will perpetuate itself. There are a few ex-ceptions to this generalization (cf. Khuri1968; Rosen 1972; C. Geertz 1983; Eickelman1985),but even these are not an extensive analysis of utterances.Speaking in all itsdiverse and com-plex forms, functioning across sociocultural contexts, ought to be the object of study, afterwhich one can determine the specific formsthat create the culturalconstructs in question. Ofthe many competing approaches to the study of speaking today, the one that probably bestrepresents(1) the amalgam of linguistics (and not just language philosophy, as in Austin 1962,Searle 1969, and Hancher 1979) and (2) a broadly conceived, holistic study is the approachknown as the "ethnography of communication."As an ethnographer of speaking in Yemen, I had become fascinated with the richness andcomplexity of the Arabic speech event of greeting. At first, I was simply concerned with un-derstandingthe "rules" of speaking in this event, given both the "traditional"theoretical ap-proach of this kind in the discipline and my own practicalneed of having to produce greetingsin context. Gradually I realized that a great many verbal and nonverbal signs in the speechevent communicated values central to a Yemeni concept of the person. Butmore importantly,it dawned on me that the speech event did not merely "reflect" or "express" the person, itcreated it in social interaction. How this is done is the subject of this article. The point is thatone cannot proceed in this venture by compiling a list of lexemes with their glosses, which areintuited to be somehow central to the cultural tradition,for this will truly leave us only with a"reflection" or "expression" of personhood, and not its creation, in language. We must studyacts of sign usage in context-what is sometimes referred to as "pragmatics" (Silverstein1976)-and then determine how, if at all, the person is constructed in them. Without such anact-oriented approach, the social construction of realitythrough language will be missed.Thedata to be analyzed in this articledemonstrate the fruitfulnessof linkingthe ethnographyof communication with interpretiveanthropology in the Middle East. Forseveral years now,the culturalconstruct of the "person" has had and continues to have a major interestfor stu-dents of Morocco (C.Geertz 1971; Eickelman1976; Rosen 1972) and elsewhere (Barth1981).

the problemWhile I was in Yemen (1979-81), I lived for approximately 1 year in a hijrah village, so-called because its inhabitants are sidah (pl. of sayyid) or reputed descendants of the Prophet

greetings from the highlands of Yemen 291


4/20

Muhammad,and therefore bloodshed is forbidden within its precincts. Feudingtribesmenmaygo to trade in the marketor pray in the mosque with immunity from attack.2 This village issurroundedby sedentary tribes located in small hamlets that dot the numerous wadT-sof high-land Yemen. Iwould visit these hamlets to observe and participatein theirvarious tribalsocialevents, such as weddings, religious festivals, and dispute mediations.Inthe hijrahvillage Inoticed how extremely importantgreeting routines were to daily inter-action and how sensitively their linguistic forms markedsocially important meanings. When Ivisited tribalhamlets, Iproceeded to use the same greeting routines Ihad learned in the hijrah,butquickly stopped doing so when I observed thatthey did not greeteach other in this fashion.It was apparentthat sadah and tribesmenemployed differentgreetings in their respective com-munities, and, as Ifound out when watching them interact with each other, they could switchthese greetings depending most often on the location they happened to be in.:When Ibegan to examine greetingroutinesof these two social groups more closely, Irealizedthat linguistic forms were being used to create a public construct of the "person" in socialinteraction. I am not speaking here of the "person" in quite the same sense that Eickelman(1976), Rosen (1984), and Geertz (1971) have done in Moroccan ethnography, namely, theculturalconception of the individual (as opposed to the group) and the social organization ofpersonal networks (as opposed to groups). Rather,Iam concerned with a categorical concep-tion of the person that formsa kind of baseline forsocial interaction,afterwhich actors may tryto find out more about each other that will identify them as individuals with certain kinds ofbackgrounds, specific political affiliations,approximatewealth, and so forth,such specific in-formation then being used to "negotiate" social interaction (Geertz 1979) and reality (Rosen1984). In themselves a kind of prelude for social transactions, greetings, not surprisingly,arecrucial as a pragmatic act in which the former kind of "person" (public versus private andcategorical versus individual) is constructed.4They create the stage or set the frame for inter-action. Yet, despite being preliminaryor preparatory,greetings are important:without them,social interaction in most cases simply cannot proceed.

Specifically, I will maintain that the public categorical concept of the person involves thekey values of both honor and piety, but that their relative importance is reversed in the twocommunities. Forthe person of the sayyid, piety isdominant over honor (or, in an analysis akinto Dumont's (1966), we might say "englobes"), whereas for the person of the tribesman,honoroutweighs piety. It is true that certain social contexts may bringout subdominant sides in eachpersonal configuration-for example, a sayyid interactingwith a tribesman may wish to fore-ground his "honorableness" as a person, for whatever strategic reasons, and, conversely, atribesmanmay want to demonstrate to the sayyid that he is a "pious" individual-but this doesnot alter the fact that one value or the other is ideologically dominant for the member of onesocial group or another.Most of this article will be devoted to the pragmaticanalysis of the ways in which the person(asanalyticallydefined above) is constructed in the sayyid and tribalgreetingroutines.Theterm"pragmatic"has, unfortunately,as many differentuses in philosophical and linguistic literatureas the notion of the person has in sociology, anthropology, and psychology, so I will now tryto clarifythe sense in which I intend this analysis of greetingsto be "pragmatic."It is argued here that the study of Arabic greetingscan benefit greatlyfrom the notion of thespeech "index," as this notion has been developed by Michael Silverstein(1976) within a gen-eral framework of the ethnography of communication and within a Praguean structural-func-tional tradition which, at least in the work of Roman Jakobson and Michael Silverstein, hasdrawn on the theory of signs propounded by C. S. Peirce (1932).Of the complicated type of signs that Peirce described, three of them-the icon, index, andsymbol-have been found to be of lasting significance in semiotics. Icons are signs whose phys-ical propertiesbear some resemblance to the "object" they signal. Some examples: a road mapis an icon of some delimited territory, he word caw is an icon of the raven'scall, even the Stars



5/20

and Stripesis an icon of the 50 states. The likeness is not as close in some cases as in others,but still mustbe evident if the sign is to be classified as an icon. Indexes are signs that bear someexistential relationship(spatiotemporalcontiguity)to the "object" being signaled. Thus, an exitsign in a movie theater is an index located very near the doorway it signals. An arrow in acorridoris an index signaling spatial direction. The symbol differs from the icon and the indexinsofaras, on the one hand, the relationship between its physical sign characteristics and itsmeaning is arbitraryand, on the other, there need be no contiguity with the "object" signaled.Inotherwords, the symbol is defined in terms of what the other two categories of signs are not.Of course, no actual, specific sign is, as Peirce realized, "pure" in the sense of belonging toonly one category, for in realitythe overwhelming majorityof signs is "mixed." Forexample,the lexeme caw, though iconic of a bird's call, is also symbolic insofar as its sounds conformto the phonological conventions of English.Our analysis of speech indexes will have to takethis fact into account. Indeed, we will find iconic indexes to be prominent in the greetingevent.The sayyid, as itwere, is presentinghimself as an "icon" of the pious person and the tribesmanas an "icon" of the honorable person.I will firstpresent some key values of Yemeni society that are crucial to the social construc-tion of the person. Then I will proceed to describe the categorical notion of the person for thesayyid and the tribesman,as itdepends on these values. Preliminary o the analysis of the wayin which the person is created in the speech event of greeting is a fairly lengthy section on thenature and sequencing of the verbal and nonverbal indexes appearing in it. The final partwillbe devoted to the demonstration of the argument that these indexes do indeed construct thekinds of person I will have previously adumbrated.

the values of honor and pietyFundamental to an understandingof the "person" in Yemen, as it is in most Arabsocieties,is the concept of saraf(honor),yet it is not easy to define. Both men and women possess honor

and, as Meeker (1976) has pointed out for another area of the Middle East, it is both ascribedby virtue of patrilinealdescent as well as achieved through what he calls "glorious deeds" inthe public arena. Itwould appear, however, that what is valued as a glorious deed in the sayyidcommunity is not the same as the action so recognized and prized in the tribal one. Inthe latter,as is well known, honor accrues to the individual who excels in acts of hospitality, eloquence(especially poetry), and courageous violence such as daring raids or warfare (see Bourdieu1965 and Jamous 1981 for particularlygood discussions of the way in which honor is tied tosuch acts). Inaddition, the public "control" of women (as in many cases also the possession ofland and herds) is also a "glorious deed."5 And while a sayyid's honor is dependent on someof these actions, such as the control of women and lands, hospitality,and eloquence are givenfar less emphasis in sayyid social action than in its tribal counterpart, and the more violentpublic deeds would runcounter to the ideology of peace to which they are committed, exceptin defense of Islam. In place of raidingand warfareone would have to substitute learningas a"gloriousdeed" by which a sayyid achieves honor in his community.It is not only the types of performances throughwhich honor is achieved, however, that dis-tinguishthe two communities. AsMeeker (1976) again was the first o make clear, fortribesmento engage with each other in glorious deeds, it must be presumed (at least ideologically) thatthey are equals in social statusand power; that is, one only performsa glorious deed againstanother who is as honorable as oneself, for one does not win points by challenging an inferior,nor can a superior usually be induced to take up one's challenge (forthe reason that his repu-tation will only suffer-if he wins, he will be considered a "bully"and if he fails, he riskslosingall his honor). Itis, then, important ortribesmen to establish symbolically thatthey are personsof equal strength and status, otherwise the transaction cannot be one in which honor is



6/20

achieved. Among the sadah, however, it is not as crucial that one performdeeds against another in some sort of public agon and, furthermore,hierarchy, not equality, is built into theconcept of honorable relations among men. It is part of sayyid ideology that there are clearmoral and materialdifferences between men in the community and the relative statusof thesemen should be publicly recognized if transactionsare to be carried out on an honorable footing.It cannot be emphasized enough that what is crucial for social interaction in either com-munity is (1) demonstratingone's respect (ihtiram) or the social honor of the other in somesymbolic act, which is simultaneously (2) a demand for respect of one's own honor, also dem-onstratedin a symbolic act, by the other. Thusa luncheon, as an act of hospitality, is more thanjust demonstrating respect for the other's honor; it is also a demand that the other reciprocatethe show of respect, often, though not always, in the form of another luncheon. Adults areconstantly being evaluated on whether or not they are muhtarim(respectful)6as indicated bytheir public acts.7This norm is just as stronglyfelt in hierarchical relations in the sayyid com-munityas it is among equal-status individuals in the tribalcommunity. That is, even though ahigh-rankingsayyid can expect to receive a more elaborate greetingfroman inferiorthan he isobliged to give, nevertheless he cannot neglect to demonstratea show of respect due the statusof an inferiorwithout incurringsocial criticism.In Middle Eastethnography, especially worksdevoted to Morocco (cf. Gellner 1969; Jamous1981), a conceptual opposition is often drawn between honor and baraka(blessing)where thelatteris understood to be a charismatic, often magical power inhering in the person and deedsof a religious figureor saint who is descended fromthe ProphetMuhammad.The sadah of thehijrah(and more generally in Yemen) do not, however, subscribe to this sortof mystical belief.Instead,they place greatemphasis on piety, which mightbe defined as an attitudeof reverencefor God demonstrated in the performanceof Islamicritualand strictadherence to Islamiccredoas defined by the Zaidi (SchTca)ect. As inthe case of honor and the attitudeof respect harboredby the individual, piety must be demonstrated in performingcertain religious acts.These religious acts include, of course, the famous five pillarsof Islam(prayer,fasting, alms,pilgrimage,and saying the shahadah), butwhat is not often realized is thatthe speech event ofgreeting is deeply connected with Islamic credo. It is not merely a coincidence that a verbalnoun commonly used to referto the speech event of greeting (taslim) is derived from the verbsallam (to greet) that can also mean to "bless with divine favor,"and for the reason that manyformulasof greeting in fact literallyinvoke God's blessing on the addressee:

sabah-tu (Mayyourmorning e blessed)mase-tO (Mayyoureveningbe blessed)'allahyisallim-ak (MayGodblessyou)

In other words, when greeting a person one is in a very real sense engaging in a religious act,calling on God to bestow his favoron the addressee.

Interestingly, he Qur'ancontains an explicit emic model of greetingfor devout Muslims. Forinstance, the greeting as-salam cale-kum (Peace be upon you) is the quintessentially Islamicmode of address, as revealed in this verse:

When hosecometo thee,who believein OurSigns, ay"Peacebe uponyou"(SurahVI,53)8Thus,the use of this formula indexes the addressee as a member of the community of believers(the 'Ummah) and it may be for this reason that some fanatical Muslimseschew pronouncingit when addressing kuffar(unbelievers). The greeting has even deeper resonances in Islamicfaith, for it says in the Qur'an that for those Muslims who are saved and are allowed into Par-adise, "theirgreeting will be "peace"!" (SurahXIV,23). Because the formula as-salam cale-kum is heardso often in social interaction,one is not, of course, always self-consciously awareof itsdeeper religious significance and yet there are situations-sometimes potentially desper-ate ones-where actors strategicallydraw on them forthe purpose of framingtheir transactionin Islamicterms. Here is one example:



7/20

Ata strangemeeting .. the nomadsare insuspenseof mindandmistrustachother .. After hewhis-perswithinhadsufficientlyakenknowledgeof our peaceabledemeanor,one approaching ircum-spectly,gaveusthewordofpeace,Salaam aleyk, nd t wasreadily nsweredbyus allagain,Aleykom-as-salaam.Afterhis sacrament f the lipsbetweenBeduw, here s no moredoubtamong hemof anyevilturnIDoughty 921:573].Butwith respectto the greeting,there is no more important njunction mentioned in the Qur'anthan the following:

Ifyou aregreetedcourteously, hengreetwith a betterone, or return t (at least) n kind,Godtakesaccountof allthings SurahV,86].Itis easier to illustratethe injunctionthan to explain it. Ifthe partyhailed by as-salam cale-kumreciprocates the greeting "in kind," then he or she replies with the standardwa c ale-kum as-salam; but if the addressee wants to reciprocate with a "better one," then to the above replycan be added the formulaic phrase wa rahmatu!!lh wa barakatuh-"and the Mercy of Godand His Blessings." In other words, one would be "heaping" more blessing on the addresseethan one had received. Justhow a speaker returnsa greetingwith a reciprocalor more intensiveresponse will be examined in detail shortly;for now, it should be emphasized that not only theact itself but its very structureare deeply implicated in Islamic piety. Because piety must bedemonstrated in action as enjoined by Islamiccredo, the performance of the greeting becomesa pious act.Because the sadah are closer to the literary, scripturalIslamic tradition,one would expectthem to adhere more closely to the Qur'anic model of the greeting event, and indeed we willfind this to be the case. However, piety is also a strongly felt value among tribesmen and itwould be wrong to ignore it. They too performthe "five pillars,"often as assiduously as theirsayyid counterparts, and we shall find that the notion of the greeting as a "blessing" is alsopresent in their formulaic expressions. Itshould be evident that the greeting is related to manylayersof culturalmeaning. Justas it would be a serious mistake to reduce it to a matterof civilityoretiquettewithout takingintoconsideration the concept of social honor, so would the analysisbe impoverished by omitting its religious significance.

the sayyid and tribal concepts of the male personthe sayyid As descendants of the Prophet Muhammad, the sadah possess tremendous sarafthat can be individually augmented by learning, ethical acts, and religious piety, and yet itwould be wrong to exaggerate the importance of honor to their concept of the person. Ionceasked a sayyid friend of mine whether he thought the tribes had more saraf than the sadah."There'sreally no comparison," he replied. "The tribes are famous for theirsarafand we canhardly compete with them. Forus, it is more importantto be a rajuldTni, a pious (religious)man.' " Perhapswe can benefit from a Dumontian formulation of the problem by saying thatin the sayyid concept of the person piety "ideologically englobes" acts of saraf, so that, forinstance, "gloriousdeeds" of violence are only honorable in the cause of Islam or the deed ofhospitalitybecomes supplanted by prodigious feats of learningand ethical conduct. Notewor-thy also in this respect is the fact that the "control" over women is more often religiously thansocially enjoined.It is because piety dominates sarafthat we can also explain the sayyid hierarchical view ofsocial relations. Life must be lived in strict conformity with Islamic doctrine, which is bothinterpretedand taught by the religiously instructed. These scholars are in the sayyid view thenaturalleaders of society, who dictate the conduct of others' lives in order that they may leadthe "good" life. Though the tribesman adheres to Islamic credo and tries to remain pious, hejealously guards his autonomy fromearthlyauthoritythat may compromise his honor.the tribesman (qabili) As reputed descendants of Qahtan, one of the mythical founders ofthe Southern Arabs, and of theHimyaritic and Sabaean kings who controlled the ancient in-



8/20

cense trade, the tribesconsider themselves to be men who have inheritedgreatsaraf.Butas wehave been saying, a tribesmancannot rest on his laurels, he must strive to achieve more honorthrough "glorious deeds." There is always the danger thatone might lose honor by despicableacts. According to legend, the xaddam, who are the low-status servants found in many tribalvillages, aredescended fromH myaritic kings,buttheirancestors proved themselves to be cow-ards in battle and so were strippedof their tribal identity.Today, tribesmen insist that the xad-dam are not tribal,nor are they honorable.Justas it would be erroneous to deny sarafto the sayyid's concept of his person, so would itdistortrealityto arguethat the tribes have no concept of themselves as pious persons. My tribalfriendsprayed regularlyand on Fridays,the holy day of the Muslim week, they would attendthe imam's sermon in the hijrahor some other local mosque. They fasted during Ramadhanand tried to make the pilgrimage to Mekka at least once, often more than once. They readilyadmittedthat their piety could not be as deep as the sadah's, primarilybecause their illiteracyprevented them from reading many of the scripturesand commentaries and few had the incli-nation to become scholars. Inshort, they were quite willing to concede to me that in terms ofdTn religion),the sadah naturallyexcelled.If in the sadah concept of the person piety englobes saraf, the dominance of these culturalvalues is reversed in the tribalconcept of the person. Forexample, according to Islamic credo,murder is haram (forbidden),and yet honor requires revenge killing in tribal law. Temporalrulers like imams are to be resisted,even when they are the paragons of virtue, because of thefact that submission to another man's authority reduces one's own autonomy and therebythreatensone's honor. Political equality, not hierarchy, is the vision of social relations in thetribalcommunity.

Enoughhas been said about the culturalconcepts of the person. Letus now turn our attentionto the firststage in the pragmaticanalysis, an examination of the verbal and nonverbal indexesto be found in the speech event of greeting.

speech formulas and their discourse sequencingWe note, firstof all, that the "said" of discourse in greetings is characterized by the use of

(relatively)fixed unit-expressions called formulas, culturally valued patternsof speaking thatare preservedfor their own sake and in which few changes can be made, with the exceptionof such features as pitch, loudness, and so forth.9 Ferguson (1967) has noted that Arabicabounds in such formulaic expressions that are used by speakersas forms of linguistic etiquettein social interaction(see also Youssouf 1976; Rossi 1939).The first or lowest order of sequencing for these formulas is what might be called an "ex-change," where a formulautteredby the speaker(A)has coupled with it in discourse a standardreply utteredby the addressee (B).The structureof such a communicative event so neatly fitsMauss'sconcept of exchange that it is no wonder he included the greeting and other "courte-sies" within the domain of the gift (Mauss 1967 [1925]:3). On the other hand, Goffman (1981)calls this two-partstructurea "dialogic couplet," borrowinghis metaphor fromthe dramaturg-ical model of social relations,and the conversational interactionistshave captured it underthedrierphrase "adjacency pair" (Sacks et al. 1974). I preferthe notion of exchange to describethis sequence because of the moral compulsion implicit in the act of giving, receiving, andreciprocatingthe salutation.What is interestingabout the greeting exchange in Arabic, to a fargreater degree than is thecase (forexample) in English,is that in the majorityof cases one can tell fromthe linguistic formof the utterancenot only that the formulasare coupled or paired,but that one formulais clearlythe initiatorof the exchange and the other is the response. In English,this is true of the pair"How are you?"/"lam fine" butnot of "Hello"/"Hello" and many other such exchanges. Con-



9/20

trastthese with our paradigmArabicgreeting:as-salam cale-kum:: wa cale-kum as-salam. Theconjunctive wa (and) presupposes in the second "pair-part" hat some talk precedes (and fol-lows) it; therefore, we would expect the formula to come second in the exchange. Note, too,that the syntax of the sentence is of the "equational" type, the order of whose constituents isreversed in the two formulas (noun phrase + prepositional phrase,and vice versa).The syntaxof the firstformula is the more unmarkedof the two, and hence itwould seem more natural forit to appear first in discourse. The second formula, on the other hand, shifts the prepositionalphrasewith its addressee pronoun into firstposition, as if to draw attention to the addressee inthe act. Note that the exchange is a perfect icon (diagram)of the event of greeting. Speaker Ais giving a greeting and receiving one in exchange, whereas Speaker B is receiving a greetingand giving one in exchange. Inother words, they are in a reciprocal, if inverse, relationshiptoeach other, and this is beautifully mirrored n the fact that the formulas use the same words butin their inverse ordering. But the exchange is even more deeply iconic if one observes that theresponse formula may (and usually is) expanded to include the formula wa rahmatu!!ah wabarakat-uh(and the Mercy of God and His Blessings),which intensifies the illocutionaryforceof the original or firstpair-part. Recalling the Qur'anic model of speaking, we realize that therespondent is enjoined to equal or, betteryet, elaborate the greeting of the addressor, and hedoes so by "heaping" on more blessings than he has received. The point is that the verbalexpansion of the respondent's turn of talk is iconic of an intensified illocutionary force. Manyformulas,as we shall see below, take a more expanded form in the second pair-part.Besides the linkingof formulasin exchanges, there is a second or higher order of sequencingof exchanges that we may call "chains" (Goffman 1981) of discourse. Two distinct types ofchaining are discernible, with a third intermediateor mixed varietyalso commonly in use.

The simplest is an additive chaining of exchanges that we might schematize as follows:A B1. x, ::2. y, :: Y23. z, :: 2A and B are initiatorand respondent, respectively; the subscriptson the small-case letters referto the firstand second pair-part,respectively; the Arabicnumeralson the left indicate that thereis a sequencing constraintoperative in the chaining of the exchanges-that is, x, : : x2 usuallyprecedes y, : : y, and z, : : z2 in discourse, and so forth. Examplesof this type of chaining canbe found in one-on-group greetings described for sayyid ceremonial occasions below, as wellas many of the greetings to be covered in the tribalcommunity.A second, more complicated type of chaining involves the coupling in one turn of talk foreither Speaker A or B of the second pair-partof one exchange and the firstpair-partof a newexchange that we might schematize as follows:

A B1. x, :: x2;Y1 ::2. Y2;Z : : z; a :3. a, etc.Unlike our paradigmatic Arabic greeting as-salam, not all exchanges come with an automati-cally intensifiableresponse formula(forexample, the response to the pan-Arabahlan wa sahlan(literally"plain and people," but more or less equivalent to our "Hello") is ahlan bKkahlan toyou). Ifone wishes to intensify the response of the initiatingformula, one has to performtwoacts in the second turn:reply with the expected second pair-partof the exchange (and closingit), but initiate another exchange that the firstparty(A)must complete. Here is an example:

A1. ahlanwasahlan,: : ahlanbfk2 : hayyallahmanja, (MayGodgrant ong life toone who hascome)2. wa hayya-k2(Andlong life to you).



10/20

Having completed the second exchange, A may stop or reply in kind by also initiatinganotherexchange. Thischaining type isfrequentlyfound inthe one-on-one greetings,especially amongsayyid individuals.And, of course, this is not surprising, orthis discourse structurepermitstherespondentto "better"the greetingreceived from the initiator,and hence to maximize the pietyof his act. Again, the verbal act is highly iconic of the Qur'anic model of speaking.A combination of these two structuresyields a thirdtype of chaining, commonly heard as aone-on-one greeting between sadah. Foropeners, the two partieswould begin a series of re-ciprocal exchanges, followed immediately by a handshake and a kiss-on-cheek gestureaccom-panied by an embrace (forsome of these nonverbal indexes, see footnote 9). More than likely,the nextexchange would be based on an inquiryinto each other'shealth that would be indexedby the expressions:

A Bkefhal-akHowareyou) :wa!!ah, fini'mah ByGod,I'm ine)(x,) :: (x2)This will probably be repeated in another indexical form:

maI-ak TIs hereanythinghematter?) : al-hamdullilahGodbe praised)want?kefhal-akAndyou?How areyou?)(y,) :: (y2): (z,)al-hamdullilah (z2)(Note that B's second turn is more like a "back" turnthan the beginning of a new exchange.)Next might begin an extended inquiryinto the health of each other's kinsmen and friends, andso forth.

In the sayyid community some greeting routines involving friends or special guests may vi-olate the structuresgiven above in that one of the parties-he who is bent on "honoring" theother-seizes the turnof talk and monopolizes it, the effect being that he showers a cascade offormulaic sayings on the other without bothering to wait for or even expecting an exchange.The other may once in a while utteran ahlan wa sahlan in response, but does not bother tokeep up with the other's barrage. Observe that the greeter who monopolizes the turn of talkdoes not slight the other; on the contrary, he is honoring him by heaping one greeting afteranotheron him without demanding a reciprocal act. This use of greetings indexes the hierar-chical relationshipof these two interlocutors.

the construction of the sayyid person in the greetingAccording to what we have previously said, one would expect the sayyid greeting to con-structa person in which piety is ideologically dominant over honor. The way in which the

sayyid can create the image of piety is to emulate the Qur'anic model of greeting, since it ispious to follow Islamic credo. This model stipulates that the respondent should reciprocate inkindor intensifythe illocutionaryforce of the greeting received from the addressor.This intensityof greeting is iconically indexed in several ways. One way is the metaphoricuse of number categories in the noun (singular,dual, and plural). If(A)says marhabah("wel-come," feminine singular),an appropriatereply would be marhabt-en (dual form). If the takerof the first turn intensifies his greeting by using marhabt-en, the respondent can build the cres-cendo by replyingwith marahib(the broken pluralform).Anotherexample: the initiatorof thegreeting says ahlan (indefinite accusative of the singularahl- [people]) meaning "Hello" andthe respondent indexes the intensified response by the use of the dual form ahl-en (note thatthere is no use of the pluralform as a more intensified version of the dual).Another way of iconically indexing an intensified response of greeting is the use of an "in-tensifierphrase" that is added to the "blessing" mentioned in the initiator formula.This pointhas already been demonstrated in one greeting formula,but there are other examples:



11/20

sabah-tu: : sabbah-akumllah bil-xerwal-'afiyah(Goodmorning) :': (MayGod blessyourmorningwithgoodnessandprotection)mase-tu :: massa-kumllah bil-xerwal-'afiyah(Goodevening) : : (MayGodblessyoureveningwithgoodnessandprotection)

Ineach respondent formula the phrase bil-xerwal- afiyah intensifies the illocutionaryforce ofthe greeting. Note, too, that this effect is achieved by the use of the intensive FormIIof the verb(sabbah,massa) in the second pair-part,whereas the verb in the initiator formula is in the non-intensive Form I (see also below Form Ihayy and FormIIhayya). In addition, the respondentformulasare more explicitly or overtly benedictions because God's name is directly invokedand the kinds of blessing ("goodness" and "protection")mentioned.There is still another way of intensifyingthe response. Considerthe following exchanges:

'Tdmubarak : al-jamY(BlessedHoliday) (Toall)saharmubarak :: al-jam?(BlessedMonth) (Toall)kulla annahwantumbi-xer : wantumka6alik(Everyearmayyoube ingoodhealth) (Andyoulikewise)al-jamT(ITolall)Parallels of this response patterncan be found in other pairedformulas in the dialect, such asa!!ah yi'en-ak (MayGod help you) : : a!!ah yi'en al-jamrc MayGod help all [of us]). Withthe exception of the thirdone, these exchanges become intensive not by heaping the blessingonto the addressor, but by extending it beyond the immediate partyof the greeting to includethe entire Muslim community.Still more interestingis an intensification of the greeting response by focusing on the messageform-Jakobson's "poetic function" (Jakobson 1960).

hayyal-betw-ahl-ahLongivethe house and thepeoplein it) ::hayyaa!! h mandaxl-ah MayGodpreservehe one who has entered t)The anaphoric pronoun -ah in the second pair-partrhymes with the pronoun in the firstpair-part,the effect being to foreground the message form of the greeting. The intensification'( ofmessage form is an icon of the intensification of greeting. Anotherexpression illustratingthisuse of the poetic function can be found in what to myears sounds like a quintessentially Yemenigreeting used on the occasion of the Id(religiousfestival):

minal-'ayid-Tns-salim-inThecelebrants f the'id are ingoodhealth) ::a'ad-akuma!!h bis-salim-in (May God keep you among the ones in good health)The firstformula is particularly poetic with its parallelismof the Form I verb active participle(C,aC2iC,)and the -in plural ending. The parallelismof rhyme is intensified in the second pair-partby a repetitionof the -in sound.A humorous elaboration of the 'Tdgreeting is quintessentially Yemeni and deserves specialcomment. The greatersays:

hajj-in ayiri awhariw-inmugambacTahajjpilgrim ra groom o be)(a) (b) (a) (b)(Note: a)and(b) ndicate nternal hyme)

to which the addressee responds with a simple 'in sa 'a!!ah (IfGod wills), but then may wantto elaborate by initiating another exchange. There is, of course, the presupposition that theaddressee is still a bachelor and the exchange is used in order to draw out a laugh from theinterlocutor.In none of the above greetings is there an overt reference to the concept of honor. They in-voke God's name and his blessings, and so on, or refer to particularlyreligious occasions. Yet,we cannot discount honor in the greeting routine. Those sJdah who can command respectbecause of their learning, and so forth, can expect to receive highly intensified greeting re-



12/20

sponses from less prestigious men. Thus, the elaborated routines are not only an icon of theQur'anic model of speaking, which index the speakers' piety, but they are also indexing therelative status-honor of the two interlocutors,hierarchyof course being permittedin the sayyidview of social relations. We will also see how importantsignaling honor is in the greeting rou-tines heard on various special social occasions.

Next, we will consider the use of the greeting in two culturalevents, the sayyid gat chew andthe wedding.The gatT"chew is the most keenly anticipated, and in many ways important,social event inthe average day of the typical Yemeni male (and many females). Except for the month of Ra-madhan, it is ordinarilyheld in the midafternoon, lasting usually until the sunset call to prayer.Friendsconvene in the mafraj(sittingroom) of one of their houses, the host of the chew pro-viding the accoutrements that will make everyone comfortable, but usually each individual isresponsible for bringing his own gat.

There is a cultural meaning underlying this event that is significant for the analysis of thespeech greetings. The chew's purpose is to attain relaxation(the Arabic word for this is rahah)through pleasant conversation among friends in a comfortable atmosphere. Chewing gat isthought to help stimulate conversation and all the paraphernaliaof the chew (the cool water-sometimes scented, the spittoons in which to expectorate the masticated leaves, the mada'ahor waterpipe for those who like to smoke tobacco, ashtrays,comfortable mattresses and car-peting, back cushions and elbow rests,burning incense passed around in a mabxarah[brazier])are there to help ensure each guest's comfort. A solicitous host is always asking his friends antamurtah("Are you comfortable?") and periodically getting up to rearrangethe cushions forsomeone so that he can relax,refillingthe waterjugs or changing the charcoal in the waterpipe,and so on.When a person arrivesat a chew, he must decide how to greetthe assembled guests. His firstconcern is to know whether there are any "honored" guests present, in which case he mustgreet them individually. If there are none, he has a choice of whether to greet each individualpersonally (relatively rare),or greet the assembly as a whole. When invited to a special gatchew, men often ask their hosts in advance who will be there so they can determine properetiquette.One can enter the room, standing by the door so that one is facing everyone and utter as-salamcale-kum,receivingthe ritualresponse chanted in unison wa 'ale-kum as-salam. Lookingaround the room, one sees an honored guest. Now, one can either greet everyone present,starting romthe rightand elaboratingthe greeting slightly with the honored guest, or one canmore simply go up to the guest and greet him in that fashion, making sure that one's greetingis more intensive than the visitor's response.

Arrival Guestahlanwasahlan, : ahlanbT-k2hayya!!ahman a,(Welcome) (Welcome o you. Long ife to the one who hascome)wahayya-kum/ ah2 Marhabah(And onglifeto you.Welcome)

Note that the guest could replyto the last formula with wa marhabahbi-kum (and welcome toyou) or marhabt-en(welcome [duall), and so on, but he knows that if this round robin is tocease, he must allow the arrivalto have the last word.

Ideally, of course, one should stand for the arrival and shake his hand (see footnote 9), butanyone who has chewed gat Yemeni-style knows how awkward standing under such circum-stances can be. Chewable gat leaves are usually strippedfrom the branch and placed in one'slap, so one has to be careful to gather them up in the folds of the futah (the male garment)before rising from the sitting position. Some guests, therefore, might prefer to remain seated



13/20

when shakingthe arrival'shand, but how can this be done when standing signifies respect ofthe other's honor?The way out of the dilemma is the following exchange:Guest seatedparty) Arrivalal-gayim Cazz : wal-jalis la yahan orwal-jalis afOal(The tanding artysdearIbelovedl) (And heseatedpartysnotdespicable)or(The eatedpartysbetter)

Here we have an example of how the verbal indexical system interactswith the gesturalone.The indexical system of speech can intervene to indicate that no slightof the other's honor wasintended, and that no disrespect was taken.

Suppose the arrival,after having uttered as-salam 'alekum, decides he need not greet anyperson present as an honored guest. Now he encounters another choice of strategy. He hasacquitted himself honorably to the group and he can confidently take his seat, but he can elab-orate the greetingfurtherwith the following exchange:

Arrival Assemblywa riyyih-hum wa huh(And(mayGodi grant hem re- (andhimlalsol)laxation)

Note that riyyih(FormII) s derived from the same root as rahah(rest)and murtah comfortable)whose meanings are connected with the notion of relaxation. The exchange in other wordsalludes to the culturalpurpose of the gat chew.Yemeni weddings are quite extravagant, all-day affairs lasting from 2 to 7 days and some-times longer, with countless relatives, friends, and outside guests in attendance. One of themore importantcelebrations for the groom is the samrahheld in his honor on the letat ad-daxlah(the nightof the consummation), when his family, in-laws, and friendscome to chat and whileaway the evening in pleasantentertainment,before he is taken on his procession zaffah to meethis bride for the first time and to sleep with her that night. The samrah is held in a large mafrajthat is usually packed with guests squeezed shoulder-to-shoulder along the walls, the groomor harFw eated in his wedding clothes at the far end of the room in the place of the honoredguest.What is the greeting strategy to be employed by a guest in such a situation? The room isusually too crowded for the arrival to greet each individual personally, with an elaboratedgreetingfor the groom (though I have in rare instances seen it done). It is far more practicablefor him to use an one-on-group greeting. He enters the room loudly addressing the assemblywith as-salam aale-kum nd receives the ritualresponse. Having now fulfilled his duty towardsthe assembled group, he can honor the groom. The guest walks up to him and the groom standsto receive him. They shake hands and the guest may kiss him once on both cheeks, afterwhichhe says to the groom: ahlan wa sahlan (and receives the ritualresponse).At this point the guestmay tell the groom a briefjoke or a pleasantrythat will draw a laugh, this speech event beingconsidered appropriateat the wedding samrahbecause it is the guest's function to entertainthegroom and put him at ease on the night he is to consummate his marriage.

The next time that the guest sees the groom after the marriage has been consummated, hegreets him with the salutation dam a!!ah as-sarur(May God make your joys everlasting) towhich the groom responds wa sarurakdayim (And may your joys be everlasting). The use ofthe formula indexes the period of the marriage subsequent to its consummation. Note that theinitiator tem is this time in the intensive formrelative to the response (given that it is in the formof an explicit benediction). Again, it is not difficult to explain this apparent exception to theformof the greetingexchange, for it is the groomwho is being honored and therefore he shouldreceive the more intensive partof the exchange.



14/20

the construction of the tribal person in the greetingBefore we go on to talk about the formulaic exchanges and their chaining in discourse, itshould be pointed out thatadult male tribesmen often use creakyvoice and a high tenor, almosta falsetto in pronouncing the greetings, which might be interpretedas a stylized way of ren-

dering "manhood" and "virility."Loudness or force is also an importantfeature of the perfor-mance.'2 By contrast, the voice of the sayyid is more "natural" and softer, sometimes eventrailingoff into a barelyaudible whisper. These differences in voice quality iconically index the"aggressive" and "warlike" person as opposed to the man of "quiet contemplation" and"peace."Most tribalone-on-one greetings involve reciprocal exchanges (that is, where the responseformuladoes not intensify the illocutionary force of the initiatingformula) linked together inwhat I have previously called a Type 1 chain. Here is a typical routine:

A B1. guwT-tYou'vebeenstrengthened : najT-tYou'vebeensavedIbyGod]) IbyGod])2. kef al-hal (How are you?) : sallamhal-ak(MayGod bless you)3. hayy aI-lahyah Longlife to you; :: hayy(a) ahyat-ak Longlife to you;literally, Longife to thebeard") literally, Longifeto yourbeard")Like heirsayyid counterparts,these exchanges take the formof religiousbenedictions; yet notethat, as a result of the ideology of honor englobing piety, the exchanges are symmetrical, thebalance in the give-and-take of communication being an icon of the balance and equality ofsocial relationships. They are never, to my knowledge, otherwise. Whereas it would be piousforthe respondent to "heap" the blessing on the other, this would compromise his status as anhonorable man.

More than justthe balance of exchange signals the honor of the actors. Note that in the thirdformulaic pair there is an explicit allusion to the concept of honor. Lahyahrefers to a man'sbeard, butone must bear in mind that in tribalsociety facial hair is interpretedas a sign of malehonor."1 nconnection with this meaning of the beard, consider a widespread gesture amongBedouin and sedentarytribalpopulations where the man graspshis goatee with his righthandand points it towards the man he is challenging, saying 'cb cale-k (Shame on you). One pre-supposes that his own honor has been impugned and that he is now shaming the honor of hisinterlocutor.

There is another one-on-one greetingthat consists of only one exchange that illustratessomeof the same points:A B

guwk-twa cilm-ak :: cilm-anawa salamat-ak(May oube strengthened (Ournewsisyoursafety)[byGod].Andwhat'syournews?) orcilm-akmasarr Your ews is[thathere s]noevil)There is again reciprocity in the response formulawithout intensification.Moreover, the literaltranslations obscure the fact that A is really asking about B's intentions (that is, whether theyare hostile or not), on the presupposition that all tribal relations are potentially hostile due tolong-standingand unresolved feuds. Of course, one feuds in defense of one's honor; therefore,the above dialogue creates, as itwere, the image of someone always wary and always readytoprotecthis honor.

Inthe case of eitherof the above greetings, ifone of the interlocutorsis invitingthe other intohis home, he will use a special formula:

marhabah og al-cen war-raswa marhabas-sot(Weicomeon thehead andeyes,andwelcome of thevoice)



15/20

The expression fog al- 'en war-ras(on the head and eye) is an intensive formof welcoming theother. (Several informants interpretedit metonymically; that is, the head and eyes are consid-ered among the most prized partsof the body and by contiguity with them the welcome takeson added worth.) The second formula marhabas-sot (welcome of the voice) means that thehost (thespeakerof this invitation)will sacrifice a sheep in honor of his guest, the voice referringmetonymically to the bleat of the animal when its throat is cut. The fact that the balance ofexchanges has suddenly been offset is not a contradiction of the argument that an honorableperson is being constructed, for the formula is not so much a greeting as it is an invitation toaccept the speaker's hospitality, one of the "glorious deeds" by which honor accrues to anindividual. The speaker has made himself into an honorable person by inviting the other intohis home for a meal, and not just any meal, but one at which meat will be served (makingthedeed even more "glorious").The tribesmen also have a greeting exchange that is unique in its rule of use.

salam ahiyah :: ablag-t(Greetingsf longlife) (Iamfulfilled orsatisfiedl)Likeprevious exchanges we have come across (forexample, salam cale-kum),this is used byan individual in saluting a group (the response formula being spoken in unison by the assem-bly),14but unlike these other exchanges one cannot also use it to greet an individual.Fieldworkprovided an instructive,if embarrassing lesson. Ionce greeted an importanttribalsheikh who was a famous poet of the region with salam tahiyah. His response was not what Iexpected; instead of hearingablag-t,he greeted me with ahlan wa sahlan. A titterrose up frommyfriends,and when Iafterwardsasked them to explain my mistake,they tactfullypointed out,"Itwas an interestingidea you had to greet an importantsheikh with salam tahiyah, the wayone would with salam cale-kum. But, unfortunately,you can only use it when greeting morethan one individual."

The second importantpoint connected with this exchange's rule of use is that once it hasbeen uttered, iteffectively cuts off any greeting with individuals of that group. The same is nottrue,of course, of salamale-kum. Another anecdote fromthe field can verifythis feature of therule of use. While I was strolling in a wadi outside a hamlet, I came across a partyof youngmen of my acquaintance. Not being in a gregarious mood, I simply hailed them with a salamtahTyah nd was about to proceed on my way when I noticed among them the son of one ofmy good friends about whom I was anxious to hear news. I stopped to inquire about himthrough this young man, but was momentarily confused about what to do. I knew that anytransactionto be conducted between persons who have just encountered each other must bepreceded by the greeting. Itcould not hurtto greet him again and did so, but the young menlaughinglycut me off, explaining, "No, once you have greeted us with salam tahiyah,you can'tthen say a special "Hello" to Abdullah."On special occasions such as gat chews or wedding celebrations, an individual greeting agroup must decide which formulaic exchange to use, depending on such factors as the size ofthe group(ifvery large,then salam tahJyahs probably more practicable),the importance of theoccasion orany special guests who mightbe present (dignitaries,forinstance, or honored mem-bers of the group such as guests, grooms, and so on). Even in very largeassemblies, however,where special guests are present,the arrivalwould probablyfeel compelled to greetthe specialguests on an individual basis. There are basically two options open to him. One of these is toshake each person's hand, but linger a little at the side of the special guests whom he honorswith a longer and more elaborate greeting. This is by way of performinga "glorious deed" ofhospitality. By this strategyhe has managed to greet every individual in the assembly. He mayachieve the same end in more abbreviated fashion by addressingthe whole assembly with as-salam (or some equivalent) and then approach the individual he knows he must single out forspecial attention. What he cannot do is use the exchange salam tahiyah, for this cuts him offfromgreeting individuals.



16/20

The one exception that I know of to this rule of use occurs at the groom's wedding samrah(the celebration held on the night he is to consummate his marriage).On this occasion, thegatheringis often enormous and the arrivalmay very well forego the nicety of honoring eachindividual with a one-on-one greeting. If he therefore utterssalim tah.yah, he will have ac-quitted himself of the duty to salute the assembly, but he will have failed to have singled outthe groom forspecial acknowledgment. How, then, does our hapless guest solve the problem?He adds to the formula salam tahFyahhe qualifier min dun (gr) al-harFwwith the exceptionof the groom) and then goes up to honor him with a special greeting. What is going on here?We said that salam tah.yahprecludes a greeting of an individual, yet an exception is made inthe case of the groom. The breaking of the norm on his wedding night has the significance ofhonoring him, because for once the individual takes precedence over the group in public rec-ognition of honor. This act puts him in brackets,so to speak, and sets him apart from the restof the company.

As a final example of elaborated routines, let us examine the greeting known as the hal thatis unique to tribal interactions.It s unique in comparison with all the previousroutines becauseit is used by a group in unison saluting another group.

Groups might arriveat any number of different social affairs.At a wedding, it is customaryfor hamlets surroundingthe groom's locale to visit on the wedding night, each hamlet sendinga delegation of anywhere froma couple to perhaps a dozen or more men. At the cid religiousfestivals,groups representinghamletsscatteredalong a wadFwill convene at some centralpointand greet each other with hal, afterwhich they participate in a joint celebration. It is rarethatgroupscome en masse to gat chews, but ifthey do, they will greetwith the hal.The same appliesto funeralsas well as dispute mediations.The basic halgreeting isas follows. When the arrivinggroupenters the meeting place (usuallya large hall called a dFwan), he people already assembled there jump to their feet, forming acircle around the room, their backs hugging the wall. Space is then made in the circle's circum-ference for the new guests and the host orsenior person of the assembly (forexample, a sheikh)shouts in the creaky tone of voice described earlier:

1. hawwil-humariijal Give hemthehal,O men)The host group then initiates the greetingwith the following formulaic exchange:

kefahwal-akumarijjal : kufi-t(Howareyou,O men) (IamprotectedbyGod])Afterthis exchange, the hosts initiate another round:

2. 'anacabd-akum : wa musan(I'myourslave) (and II m]safeguarded)The meaning of this exchange needs to be clarified, because it would appear to contradict theegalitarianideology so importantto the tribalconception of the person. The hosts appear to behumbling themselves before their guests, but the formula musan uttered by the guests is usedfor a person who is undera sheikh's protection(andwho would blacken his honor if he allowedanything to happen to his charge). Therefore,both partieshave mutually lowered themselvesand paritybetween them is thereby established.

Having now greeted the assembly as a whole, the group can break up into individuals, whoapproach the groom and exchange with him the following:

Guest Groomdamallah sarur-akum sallamahwal-akumminas-sirrw-aslah llahsan-akumwa mahal-akum :: w-allahyidTms-sarur



17/20

(MayGodmakeyour oyseverlasting. (MayGodpreserve oufromharm.MayGodblessyouraffairs ndhealth) Godprolongs oys)

conclusionOne reason that the speech event of greeting in Yemeni society is importantis that it signalsconcepts of honor and piety that are central values in the cultural system of the sadah and thetribes. Why should such values be alluded to in the firstplace? I argue that a certain type of

public person is being created in the speech event of greeting(it is performativein an Austiniansense). We have said that honor requiresa demonstration of respect to the other and a demandfor respect of the self, a demonstration that is accomplished, as we have seen, in the greeting.We also said that piety entails acting in certain ways prescribed by Islam, one of these piousactions being the greeting. It ollows that the greetingcreates "honorable" and "pious" personsin the course of its performance.What is interestingfrom a lingusitic point of view is how the person is created by the use ofsigns in the greeting. By uttering a formula that is in the form of a blessing, the first speakerdemonstratespiety and respect towards the other. The addressee in the sayyid community will

tryto intensifyhis response by the use, basically, of linguistic icons: a longer response, a moreintensive message form, higher numbercategories, and so forth.Alternatively,he may intensifythe response simply by performingtwo acts instead of one: closing one exchange and openinganother, which is yet another icon of intensification. In short, the pragmaticsof the greetinghinge crucially on iconic indexes.In the tribal greeting the formulaic exchange is usually balanced, iconically reflecting theconcept of equality among "honorable" men. Some of the indexes symbolically refer to thingsin the speech event that allude to male honor or presuppose social relationships,such as feud-ing motivated by values of honor. And many greetings index the coexistence of a group to behonored, the group figuring prominently in the tribalcode.

Although there are many studies in the ethnography of communication demonstratingthe"rule-governed"natureof speaking and thattryto reveal the "emic" models of such behavior,there is still relatively little that has been done on what might be called the "cultural interpre-tation" of speaking, that is, the interpretationof the meanings speech events create that arecentral to social interaction. And while interpretiveanthropology has appreciated the signifi-cance of studying language, it greatly needs to go beyond the lexicon in order to solve its in-terpretiveproblems.

notesAcknowledgments. Fieldwork was funded by grants from Fulbright-Haysand the Social Science Re-search Council Doctoral Programs.I am gratefulfor the generous assistance of both these organizations. Iwould also like to thank Taha Hamudi for his help with the translations,as well as the Yemeni teachers inthe American Peace Corps. Partsof this article were given as talks to mixed American-Yemeni audiences,whoseinsightfulemarks reatly nhancedmyunderstandingf localusage.Iwish to alsoacknowledgethe helpful suggestions for article revisions of Najwa Adra,Dan Varisco, Paul Friedrich,and the readers ofthe American Ethnologist.'A note on the phonemic transcription system is in order:

LABIAL DENTAL PALATAL VELAR PHARYNGEAL GLOTTAL-v/+v -v/+v -v/+v -v/+v -v/+v -v/+vSTOPS b t, t/d k/gFRICATIVES e/6. j x h/c hSIBILANTS s, /z sNASALS m nLATERALS 1,1TRILLS r gCONTINUANTS w yNote: Emphaticconsonants are indicated with a dot below the segment. The one exception is h.



18/20

FRONT BACKi (1) u (u) HIGH(e) (6) MIDa (a) LOWCENTRALNote: A shwa epenthetic vowel appears in certain phonetically conditioned environments. Long vowelsare indicated by a macron.

2SeeSerjeant(1977) for more information on the hijrahvillage in South Arabia. Inmy experience, otherelements of the populations including tribesmen and the low-status xaddam (servant)could, and oftenwould, live in the sanctuaries, but the sadah were always the dominant group. Itshould also be pointedout thatonly in principlewas violence abhorredwithin the sanctuary;in reality, Iheard of, and personallyexperienced, such violence occasionally, though it was morallyfrowned on by all.'That is, in the hijraha tribesman addressing a sayyid would more than likely use the sayyid form ofgreeting, whereas a sayyid visiting a tribal hamlet would more than likely greet a tribesman in his tribalspeech pattern.If, however, a tribesmanwere greeting anothertribesman,or a sayyid were greeting anothersayyid, regardlessof their location, theirown greetingforms would be employed. Only once did Iobservethe sJdah break these rules of thumb. This was at a time when the hijrahwas having serious political prob-lems with the surroundingtribes who were deeply offended by the actions of a young sayyid man. Thesadah would often employ elaborategreetingsin theirhijrahwith the tribesmen in order to show how muchthey "respected" them (this, at least, was the reason given to me). Inother words, the greeting became animportantstrategyof showing the tribes that the sadah held their customs in esteem and perhaps to implythatthey were "one" with them.In Middle Eastethnography a fairlyextensive discussion has emerged on the "person" and the "self"but it is not always clear in what sense(s) these terms are to be understood.A critical review of this literatureto clarifythe different uses of the term "person" would help sharpen the analytical formulationsof variousethnographic problems, but this obviously cannot be undertakenin the scope of this article.'To put it somewhat simplistically, forwe do not have space here to elaborate on this complex theme, awoman is considered a highly passionate creature who-and this is where the ideology says she is differentfrom a man-has not enough willpower to curb her unruly appetites. Therefore, it is believed that con-trols-veiling, public avoidance, and male relativesfor escorts-must be imposed on her behavior in orderto ensure the purityof her'ir (chastity). Given the onus on the man's honor of the woman's' ir6, it followsthat strangemen and women ordinarily avoid encountering or acknowledging each other. The woman isnot a "public person" to strange men. The rule of thumb, as far as the man's conduct is concerned, is toremainsilent, refrain romeye contact, and maintaina space of several feet between himself and this anon-ymous "object" on the street. Needless to say, and in spite of his "iron will," the male lapses in this strictconduct once in a while, at which point the woman is perfectly within her rightsto publicly berate theshameless scoundrel, if only to protect her 'iro. Forreadings on the cultural concepts of gender and nor-mative behavior related to them, see Antoun (1968), Abu-Zahra(1970), Dwyer (1974), Beck and Keddie(1978), Davis (1983), and others.

"There s, of course a difference between muhtarim respectful)and muhtaram respected)and both eval-uative terms may be used in describing a person.7Note that I say "adults" are evaluated for their respect; children are ambiguous in the game of honor.

Theyare, rather ike women, to be ignored in public if they are not partof one's family. On the other hand,they may be required to recognize an adult's honor and demonstrate respect for it in certain encounters,and the adult may not necessarily be expected to reciprocate.8Translationsromthe Qur'an are my own (Abdallah 1934).9Besides the formulaic exchanges, a speech event of greeting involves hand gestures; kissing; facialexpressions such as smiling and direct eye contact; bodily movements like head-nodding, standing, andin some cases droppingto one's knees in prostration; patial arrangementof actors,and so forth.A greetingmay, in fact, be performed without anything being said, as long as the appropriate gestures, facial expres-sions, and bodily movements are maintained by the actors in the scene. Gestures are usually thought to bean obligatory accompaniment of the verbal exchange. I often have saluted people in words only to becalled out by my interlocutorin a mock-serious tone of reprimand: yadd-ak, yadd-ak (Yourhand! yourhand!)(note the geminate formof yad in Yemeni Arabic)which I then immediatelyextended. It is an insultto give someone a soiled or wet hand to shake, but rather han ask to be excused, the person may simplyhold out his wrist or forearmwhich the other clasps. Proprietydictates that under most circumstances onestands when greeting another person (though there are ways of being excused from this obligation, as weshall see, which involve a strategicuse of speech).Thedistinctionof right/lefthand is, of course, important n such acts as eating and handingover anythinglike a letter, a book, money, tea, and so on. In all these acts the right hand is prescribed, the left handproscribed.The difference is captured perfectly in the Yemeni proverb al-yadd al-yaman lama saruf, al-yadd al-yasarlama xabue(The righthand when one is honored, the left hand when one is maligned). Evenspace to the rightside of ego's body is categorized as "sacred" and "honorable" so that when greetingothers, one not only does so with one's righthand, buy also startswith the person on one's rightside and



19/20

moves counterclockwise until everyone has been greeted in a given space. If a man's honor is deservingof special regard,he is usually seated to the rightof the host in the main of place of honor."'For he idea of "intensification" in poetry, see Friedrich 1979).1G1t (orqat as it is pronounced in classical Arabic)is a leaf chewed for its mild narcotic effect.121wish to thankGreg Urbanfordrawing my attention to this interpretation.3Apious Muslim also sportsa beard and in the context of sayyid society it has religious connotations.

141first heard this greeting in the hijrahunderpeculiarcircumstances. An intelligentand highly articulatesayyid youngsterwho talked to me at length about a varietyof ethnographic topics would parodythe tribalgreeting with this exchange. Besides revealing something about sayyid attitudes toward tribal patternsofspeaking, the parody underscored what in sayyid perception is quintessentially "tribal."

references citedAbdallah, YousufAli, transl.

1934 The Glorious Kur'an.Beirut: Dar al-Fikr.Abu-Zahra,Nadia M.1970 On the Modesty of Women in Arab Muslim Villages: A Reply. American Anthropologist72:1079-1088.Antoun, Richard1968 On the Modesty of Women in Arab Muslim Villages: A Study in the Accommodation of Tradi-tions. American Anthropologist70:671-697.Austin,J.L.1962 How to do Thingswith Words. New York:Oxford UniversityPress.Barth,Frederik1981 Featuresof Personand Society in Swat: Collected Essaysof FrederikBarth.Vol. 2. London: Rout-ledge & KeganPaul.Basso, Keith

1979 Portraitsof "The Whiteman." Cambridge: Cambridge University Press.Bauman, Richard,and Joel Scherzer, eds.1974 Explorations n the Ethnographyof Speaking. Cambridge:Cambridge University Press.Beck, Lois,and Nikki Keddie, eds.1978 Women in the Muslim World. Cambridge,MA: HarvardUniversityPress.Bourdieu, Pierre1965 The Sentiment of Honor in Kabyle Society. In Honor and Shame. J.G. Peristiany,ed. pp. 191-241. Chicago: Universityof Chicago Press.Davis, Susan S.1983 Patience and Power: Women's Livesin RuralMorocco. Cambridge, MA: Schenkman.Doughty, CharlesM.1921 Travels in Arabia Deserta. London:JonathanCape.Dumont, Louis1966 Homo Hierarchicus.Paris: Gallimard.Dwyer, Daisy Hilse1974 Imagesand Self-images: Male and Female in Morocco. Cambridge:CambridgeUniversityPress.Eickelman, Dale F.1976 Moroccan Islam.Austin: Universityof Texas Press.1985 Knowledge and Power in Morocco. Princeton: Princeton UniversityPress.Ferguson,CharlesA.1967 Root-echo Responses in SyrianArabic Politeness Formulas. In LinguisticStudies in Memory ofRichard Slade Harrell. Don GrahamStuart,ed. pp. 37-45. Washington, DC: Georgetown UniversityPress.Friedrich,Paul1979 Poetic Languageand the Imagination:A Radical Reformulationof the Sapir-WhorfHypothesis.

In Language,Context and the Imagination: EssaysSelected and Introducedby Anwar S. Dil. pp. 441-512. Palo Alto, CA: StanfordUniversityPress.Geertz, Clifford1971 InSearch of North Africa. New YorkReview of Books, April22, pp. 20-24.1973 The Interpretationof Cultures. New York: Basic Books.1979 Suq: The BazaarEconomy of Sefrou. In Meaning and Order in Moroccan Society: Three Essaysin CulturalAnalysis. CliffordGeertz, HildredGeertz, and Lawrence Rosen, eds. pp. 123-313. Cam-bridge: Cambridge University Press.1983 Art as a CulturalSystem. In Local Knowledge. CliffordGeertz, ed. pp. 94-120. New York: BasicBooks.

greetings fromthe highlands of Yemen 307


20/20

Geertz, Hildred1979 The Meaningof FamilyTies. InMeaningand OrderinMoroccan Society: ThreeEssays n CulturalAnalysis. CliffordGeertz, HildredGeertz, and LawrenceRosen, eds. pp. 315-391. Cambridge: Cam-bridge University Press.Gellner, Ernest1969 Saints of the Atlas. Chicago: Universityof Chicago Press.Goffman, Erving1981 Formsof Talk. Philadelphia: Universityof Pennsylvania Press.Gumperz,John,and Dell Hymes, eds.1972 Directions in Sociolinguistics. New York:Holt, Rinehart& Winston.Hancher,Michael1979 The Classification of Cooperative IllocutionaryActs. Languagein Society 8:1-14.Heath, ShirleyBrice1983 Ways with Words. Cambridge:Cambridge University Press.Hymes, Dell1974 Toward Ethnographiesof Communication. In Foundations in Sociolinguistics: An EthnographicApproach. pp. 3-27. Philadelphia: Universityof Pennsylvania Press.Irvine,JudithT.1974 Strategiesof Status Manipulation in the Wolof Greeting. In Explorationsin the EthnographyofSpeaking. Richard Bauman and Joel Sherzer, eds. pp. 167-191. Cambridge:Cambridge UniversityPress.

Jakobson,Roman1960 Concluding Statement:Linguisticsand Poetics. In Style in Language.T. A. Sebeok, ed. pp. 350-377. Cambridge, MA:M.I.T. Press.Jamous,Raymond1981 Honneur et Baraka.Cambridge: CambridgeUniversityPress.Khuri,F.1968 The Etiquetteof Bargainingin the Middle East.American Anthropologist70:698-706.Labov,William1972 Languagein the InnerCity. Philadelphia: Universityof Pennsylvania Press.Mauss, Marcel196711925] The Gift. lan Cunnison, transl. New York: Norton.Meeker, Michael E.1976 Meaning and Society in the Near East:Examples Fromthe Black Sea Turksand the LevantineArabs. InternationalJournalof Middle EastStudies 7:243-270.1979 Literature nd Violence in Northern Arabia.Cambridge: Cambridge University Press.Peirce, Charles S.1932 Collected Papers.Vol. 2. Cambridge, MA: HarvardUniversityPress.Rosen, Lawrence1972 The Social and Conceptual Frameworkof Arab Berber Relations in CentralMorocco. In Arabsand Berbers. ErnestGellner and CharlesMicaud, eds. pp. 155-173. London: Duckworth.1984 Bargaining or Reality. Chicago: Universityof Chicago Press.Rossi, Etttore1939 L'AraboParlatoa San a'. Rome: Institutoper I'Oriente.

Sacks, Harvey,EmanuelSchegloff, and Gail Jefferson1974 A Simplest Systematics for the Organizationof Turn-taking or Conversation. Language50:696-735.Schegloff, Emanuel1968 Sequencing in Conversation Openings. American Anthropologist70:1075-1095.Searle,John1969 Speech Acts: An Essayin the Philosophy of Language. Cambridge:Cambridge University Press.Serjeant,R. B.1977 South Arabia.InCommoners, Climbers and Notables: A Samplerof Studies of Social Rankinginthe Middle East.C. A. 0. van Niewenhuijze, ed. pp. 226-248. Leiden, Holland: Brill.Silverstein,Michael1976 Shifters,LinguisticCategories and CulturalDescription. InMeaning inAnthropology. KeithBassoand HenryA. Selby, eds. pp. 1-55. Albuquerque: Universityof New Mexico Press.Youssouf, IbrahimAG, Allen D. Grimshaw, and Charles S. Bird1976 Greetings in the Desert. American Ethnologist3:797-824.

Submitted 6 December 1984Revised version submitted 15 October 1985Accepted 14 January1986Final revisions received 22 January1986


caton greetings from yemen.pdf

Documents