University of Pittsburgh- Of the Commonwealth System of Higher Education

The Hamlet as Mediator Author(s): Frank Cancian Source: Ethnology, Vol. 35, No. 3, Special Issue: Mesoamerican Community Organization: Barrios and Other Customary Social Units: Part II (Summer, 1996), pp. 215-228 Published by: University of Pittsburgh- Of the Commonwealth System of Higher Education Stable URL: http://www.jstor.org/stable/3773920 Accessed: 24/03/2010 15:05
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 at http://www.jstor.org/action/showPublisher?publisherCode=upitt. 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 support@jstor.org.

University of Pittsburgh- Of the Commonwealth System of Higher Education is collaborating with JSTOR to digitize, preserve and extend access to Ethnology.

http://www.jstor.org

THE HAMLET AS MEDIATOR1

?? ~~I^^gia/"~~;~~ |vK^^^-fiMsl~ 1University >

Cancian IFrank of California, Irvine

Our focus on "the customary social unit that mediates relations between household
? . . and community in Mesoamerican Indian and rural society" (Mulhare 1996:93)

leads to a rich conceptual space. A mediator is expected to do more than carry clear messages from one party to another. A mediator must somehow shape the connection, transformthe message, and enhance or buffer the force of the exchange. And mediation itself is complex, especially because it varies with the relative power of the (minimally three) parties involved. These aspects of mediationraise interesting questions about small rural social forms and their relations to their contexts. This articleexplores some of those questionsby describinghamlets in Zinacantain, Chiapas (Mexico) over recent decades and comparing them with similar social units in Chinese peasantcommunitiesduringearlierperiods. In both places hamlets have been importantmediators between households and the larger community through which households are connected to the state. The goals here are to identify characteristicsof hamletsthat signal their mediating role, and to explore the conditions under which these characteristics change. In abstractterms, the hamlet of concern is territoriallydelimited and has a population between several dozen and about 200 households. It has two characteristicsthat are central to its role as a mediating social form. First, the hamlet is socially incomplete. That is, the social and public life of its residents extends beyond its boundaries in importantways. For example, marriage partnersmay be sought from outside the hamlet, and/or public roles taken by hamlet residents may be played out in a larger sociopolitical unit. In Mesoamerica this often means that religious offices (cargos) are served at the municipio (township) level, not in the hamlet. It means much the same thing in the world described by fat dictionaries, where a hamlet is defined as "a group of houses or a small village, esp. one without a church" (Brown 1993). Thus, the incompleteness of social and public life in hamlets is a distinctive characteristicof their social form.2 Given current ideas about the incompleteness of most aspects of life as we see them, I should emphasize that I mean to characterizehamlets relative to other social forms, not in an absolute way (Cancian 1992:205-08). For example, in Mesoamerica, the municipio is a more complete social form than the hamlet. Second, the hamlet's public life is not formally organized, and the hamlet is not fully articulatedwith the larger unit of which it is a part. This often means 1) that its most powerful leaders are called traditional from the point of view of the encompassing state, and/or 2) that relations with the outside world are mediated throughresidentswho are of low status and powerless within the hamlet. Political life in the hamlet is only loosely and/or informally connected with the larger system-in 215
ETHNOLOGYvol. 35 no. 3, Summer 1996, pp. 215-28. ETHNOLOGY, c/o Departmentof Anthropology,The University of Pittsburgh,PittsburghPA 15260 USA Copyright? 1996 The University of Pittsburgh.All rights reserved.

216

ETHNOLOGY

the Mesoamerican case with officialsat the municipio level andbeyond. Thiskindof hamlet fromZinacantan between the 1960sand virtually disappeared the 1980s. At the beginning of theperiodZinacantan's hamlets the closelyresembled idealized onejustcharacterized. the were much more end, By they sociallycomplete, with the world outsidethem. Indepenformallyorganized,and closely articulated dence based on distancefrom the municipio'spolitical and religious center and detachment from higherlevels of government was replaced by full formalstatusin a local systemorganized local officials and to rules dictated from above. by subject Seen from the local and from the household of the no longer hamlet point view, mediated relationswith the worldoutside.In manyways it becamea local outpost of thatworld. in Zinacantian To document this transition I will 1) give a brief overviewof the earlierform of hamletsandof the changesandtheirproximate causes;2) reviewin detailthe recenthistoryof formalroles in one hamlet;and3) summarize a survey of similartransitions in all the otherhamletsof the municipio.These sectionsare basedon my fieldwork at lengthin Cancian reported (1992)andon theworkof many otherscited there. Comparisons with Chineseruralsociety based on a few recent studiesof earlierperiodsfollow, andthe essayconcludes withthe tentative published to whichthesecomparisons lead. generalizations RECENTCHANGES IN ZINACANTAN: OVERVIEW AND CONTEXT In 1960, andfor at leasttwo decadesbeforethat, Zinacantian fit the modelthat Mulhare (1996) and Nutini (1976:14) describe. The hamlet (paraje) was the customary social unit mediating between the household and the community Eachof abouta dozenhamlets hadformallyappointed represented by the municipio.3 officialswho collectedtaxes for municipio fiestasandtook themto officials in the Pueblo(thepoliticalandreligiouscenterof the municipio). Shamans in each resident hamletperformedannualritualsfor the hamletat the municipio'ssacredplaces. Hamlet membership did not changewith change of residence;it was more like national to e.g., a manwho wentto live in his wife's hamletcontinued citizenship; pay taxes in his homehamlet. The hamletswere left unmolestedand informalas long as they paid their contributions to fiestasandotheractivitiesorganized at the municipio level. Before the 1960s most formalactivities were concentrated in the Pueblo.Manyindividuals and families from the hamletsvisited the Puebloto use the formalcourts or the Catholicchurches,and some movedtheretemporarily to serve in formalcivil or of the municipio religiousoffices. But they did so as members community,not as of theirhamlets. representatives of individuals andfamiliesto the municipiohad By the 1980s this relationship several hamlets their own churchesand officials, and many more changed: got Zinacantecos did formalpublicbusinessandpublicservicein theirhamlets.Hamlets offeredmorecomplete sociallivesto theirresidents. The informality of the idealized

AS MEDIATOR 217 THE HAMLET hamletwas gone. The proximatecauses of this transformation were Mexican governmentand of new schools, roads, water thatpaid for construction CatholicChurchprograms and light systems, community buildings,jails, and churchesin the hamlets, and the construction. of new formalroles thatparalleled The fosteredthe development broader contextof the changeinvolvedindirectcausesthatrangedfrompopulation hamletsize to regionalandnational economicexpansion based growththatincreased in the mid-1970s. in andnearthe stateof Chiapas discovered on petroleum resources the end of the activistEcheverria This economicexpansion,in its turn,cametoward on of the politicalactivismof the itself came the heels (1970-76)presidency-which late 1960s. Fromthe international pointof view, the beginningof this activismwas in clasheswith MexicoCity police whenthe 1968 marked by the killingof students Olympicswere held there. Thus, duringthe early 1970s, the nationalstate had of Zinacantan. were Itsprograms reasonsto extendits reachintothehamlets multiple the of form of loans to Thirdfacilitated by the worldwidemovements capital(in thateventually contributed to the economiccrisis of 1982. Worldcountries) IN ONE HAMLET: NACHIG THE TRANSFORMATION has a uniquehistory,but the generaldirectionand Each hamletin Zinacantain the were similarfor most of them. Nachig, of the of transformation many specifics in 1940. The nationalcensus listed its the hamletI know best, was medium-sized that three hamlets were larger,two were aboutthe at and showed population 269, samesize, andthreewere a little smaller.At thattime one couldwalk from Nachig or the marketcity of San Crist6balde las Cases, to the Pueblo of Zinacantan, conductbusiness, and get home the same day. Both the Puebloand San Crist6bal in thewestern hamlets wereless accessible to the moredistant (see partof Zinacantan Figure 1). In 1940 Nachig's status as a hamletwas markedby its Principales.These officialscollectedcontributions (taxes)forpublicworksandfiestasin thePueblo,and andtook the of the Ladino for the monthlysalary Secretary, Municipal (non-Indian) in thePueblo.Ontheirweeklytrips,they (mayor) Municipal moneyto thePresidente accounts fromconsultants' Presidente. the to and from alsocarried Judging messages were young morerecently,Principales observed of the period,andfromincumbents men, often bachelors, who served for one-yearterms. They were responsible servants but, in keepingwiththeiryouth,they hadno authority (see also community the entireformalcontactof Nachig Tax et al. 1947). At thattime they represented with the municipio. Therewere few otherformalroles in the hamletin the 1940s. In the late 1930s to havea school.Thus,therewas at least Nachighadbecomeone of the few hamlets de Educaci6n a Presidente one official(probably [Taxet al. 1947:40])for the school, members. from with one or two additional a schoolcommittee andprobably Judging activitiesof recent school committees,most of their duties concernedpromoting andmaintaining attendance buildings.In the early 1940s, manyNachigmen

218

ETHNOLOGY

Figure 1: Hamlets of ZinacantAn (Chiapas), 1983

....

10
SALINAS O ...............' 7 PUEI PATOSIL 4 5 / JQIGEL OIGEL NAVENCHAUC * CHAINATIC . x ZEQUENTIC * Y?ALENTAY

? PIG

.
* o

JOCCHENOM POTOVTIC

APAZ

1I : * ELANVO . HAMLETS ESTABLISHED AFTER 1965

..

1I 2. 3. 4. 5. 6.

CHAJTOJ CHIQUINIBALVO JECHCHENTIC JECHTOCH JOIGELTO LASELVA

7 8. 9. 10. 11. 12.

PETZTOJ SAN NICOLAS SHULVO TIERRABLANCA VOCHOJVOALTO VOCHOJVOBAJO

Kilometers(approx.)

AfterVogt 1969:viiiand 156, with additions. (probably about half) were members of the ejido (land reform) movement, and there was probably at least one Nachig man on the central organizingcommittee.4During those years the leader of the ejido movement that included Nachig and all the major

hamletsaroundit (includingthe Pueblo)was beginningtwo decadesas political Thusthe ejidodelegates,or the informal caciqueof the municipio. politicalleaders behindthem, probably were the mostpowerful in leaders the hamlet. political the late 1940s these three roles school By committee,and ejido (Principal, involved more as as delegate) men,perhaps many eightor nineat a time.5Whilethe school and ejido officials were important to many people, they did not directly concernevery householdin Nachig,for not every household had a child in school or a member of the ejidomovement. connected Onlythe Principales everyhousehold to the municipio. Theformalunityof thehamlet was "ritually ceremonies expressed by two annual all the shamans performed by livingwithinit" (Vogt 1969:148).Theseceremonies

THE HAMLET AS MEDIATOR 219 thehamlet's to "ancestral ceremonial relationship godsin the [municipio] emphasized serveas mayordomos andplayhost center.Significantly enough,the ... principales circuit"(Vogt 1969:148-49). for the ritualmealswhichbeginandendtheceremonial the importance of municipio-wide The rituals unified the hamlet and reaffirmed cultureto its participants. betweenthe hamletandthe religiouscargo systemin Therewas no connection the 1940s. Service in religiousoffices, which was the key to the municipio-wide statussystem (Cancian1965, 1967), was not partof public life in Nachig, for the hamlethadno church.Nachigresidents who wanted to establish themselves through ritualservicehadto go to the Puebloto serve, for threeof the four churches where cargos could be served were located there, and all but two of the roughly 40 of fiestaswereservedentirelyin thePueblo. to sponsorship expensive cargosdevoted The fourthchurchand two cargosthat connectedthe hamletto the Puebloevery otherSunday(Cancian1965:221)were in Salinas,a hamleton the old routefrom centralMexicothrough the statecapitalandthe Puebloof Zinacantain to the market de las Cases.At thattimethe Pueblowas also hometo less than city of SanCrist6bal a dozencivil or politicaloffices, including the Presidente andmunicipio Municipal for Both and civil were offices served terms (Tax et al. judges. religious year-long from 1947:41, 52), often by people who, like Nachigpeople, moved temporarily theirhamletsto the Pueblo.In sum, in the 1940s Nachigwas a place of residence that looked like my idealized hamlet. Its populationwas small, its leadership or powerless,and social andpublic life therewas incomplete,for the "informal," religiouscargosandpowerful,formalpoliticaloffices were in the Pueblo.Though many other changes came to Nachig in the next two decades, its formal roles remained essentiallythe sameinto the 1960s(Cancian1992:109). The Pan American of the changesthat concernus Highwaywas the harbinger in about1950 andmostlypavedby the end of the decade,the new here. Completed highwaypassedthroughNachigandotherhamletsthathadbeen remote.It brought Nachigto withinfifteenminutesof San Crist6bal by car. When a churchwas built nearthe highwayin the late 1960s, important new came to The financed roles construction was and public encouraged Nachig. by an orderof Catholicnuns and led by an important Nachigman. At first a committee andtwo religiouscargocaredfor the church,butby 1976therewerefoursacristans holderswho sponsored fiestas. Theirdutieswere modeledon those of men serving in the Pueblo,6 as first-levelservicein the Pueblo's andthe religiouscargoscounted four-levelhierarchy.Duringthe same period, as part of the reformsinitiatedby to serve the hamlet.In VaticanCouncilII, two Nachigmen were namedcatechists sum, in a decadeNachigwent from a placewith no formalCatholicritualroles to one whereeight local men servedin suchroles at all times. Nachig's formal political roles expandedin the mid-1970s. The Presidente to settle anAgenteMunicipal-anofficialwithformalauthority Municipal appointed (Casadel Pueblo)in which building disputes.The Agentesoonhada newcommunity to work, anda jail to holdprisoners.Beforethosechanges,disputeswere mediated andwithoutterm(Collier 1973), andcases by hamletelderswho servedinformally

220

ETHNOLOGY

in the they could not settle were takento the judges andthe Presidente Municipal the new systemincluded threeAgentes,not one, andeachof the Pueblo.In practice threehad an assistant.This madeit possiblefor the pairsto serve two-weekterms andthen have four weeks free to earntheir livings. Thus, in just a few years, the andlocaladministration formalcivil rolesconnected to dispute settlement wentfrom noneto six. of the firstelectricandpiped-water Finally,the 1960ssaw construction systems in Nachig and creationof a committeeto administer them and collect fees for of a deliveryto federaloffices in San Crist6bal,and the 1970s saw construction secondschool, creationof its school committee,andappointment of local officials in chargeof administration of communal lands.Overall,the number of religiousand public service roles in Nachig abouttripledbetween 1960 and 1977 (while the aboutdoubled). Morethanhalfof thenewroleswereadded between1974 population and 1977 (Cancian1992:109). Eventsset off by the new pressures to servepublicoffices in the hamletmade Nachig'sstatuseven moreformal.The most important changecamewhenmenwho servedin hamletoffices tired of additional demands from municipioofficialswho wanted themfor serviceatthemunicipio level. Finally,onemanrefused thedemands of a Nachig nominating unless he was from the demandsof meeting protected In municipioofficials. response,Nachig officials drafteda resolutionand got it by municipioofficials: it exempteda man who served a year in either approved or him Nachig the Pueblofromtaxes for thatandthe followingyearandexempted from other service at both levels for the same period. Service in Nachig was officiallyon a parwith servicein the Pueblo. These changesmade Nachig a more socially completeplace and a formally civil andreligiousservicecouldbe done, organized place, a placewherehonorable and fiestas disputesofficially settled, enjoyed. It was a place with a church, a in frontof it, and,just acrossthe highway,a publicparkanda community bandstand buildingwith a jail behindit. WhilemanyNachigpeoplestill wentto the Pueblofor big service,big disputes,andbig fiestas,for sometherewas no longera needto do to higherlevelsof the religioushierarchy, so-especially for thosewho didnot aspire encountered no majortrouble,andwishedto avoidthe extraexpenseof the fiestas at the churchesin the Pueblo.At the sametime, therewere new local fiestataxesto be paid, hamletserviceto be done, andalmostno placeto avoidthe gaze of people and its role valley. For all practical purposes,Nachighadbecomethe community, as a mediating level of social organization hadreceded,at leasttemporarily. In addition, by the end of the 1970sNachighadsplit intothreeofficialhamlets. The separation of Jechtoch,an area on the easternside of Nachig, was relatively simple. The areawas definedin the late 1960swhenthe Nachigwatersystemwas built. Jechtochwas on higher groundthan the rest of the hamlet, and was not included in the water system for technicalreasons. Later some of its leaders the peopleto takeadvantage of government aimedat smallrural organized programs
in official roles, for they were your neighbors in a large hamlet tucked into a single

THE HAMLET AS MEDIATOR

221

on thewesternside of Nachig,was morecomplex,for it involvedold of Jechchentic, withthe growingrivalries between divisions combined politicalparties(PRI political andPAN) at the municipiolevel. in theearly1980s theirpopulation weresmall.Together Jechtoch andJechchentic While both had schools, neitherhad an was about20 per cent of "old Nachig."7 or at that buttheydidhavetheirownPrincipales. a time, building Agente community as as hamlets located on a major were hamlets; independent They independent schools andwaterandelectricsystems and served highway by government-sponsored can be. were formedin otherpartsof Zinacantain duringthis period. Manynew hamlets the next section In variousways they recreated as hamlet But, local, autonomy. in hamlets with schools, of Zinacantecos lived 1980s the vast the shows, by majority like that churches,cargos,community buildings,andAgentes.Theirindependence, that dependedon of people in Nachig, was formal independence, independence had basedon distanceanddetachment connection andrules.The olderindependence of the reach the been undoneby the reachof the church,andto a greater extent,by state. IN OTHER HAMLETS THE TRANSFORMATION Most otherhamletswent throughchangeslike those in Nachig. Schools were built in almost all of them. In the larger ones governmentgrants supported of plazasbordered construction by publicbuildings,and formalreligiousandcivil roles were created.Table 1 gives an overviewof the change. livedneara school.Somewhat all children later,as girlsjoined By 1960virtually school. Most new schools built boys in classes, more and more childrenattended between 1940 and 1960 were part of the INI (InstitutoNacional Indigenista) educationalprograms that expandedgreatly in the mid-1950s. Later, school reachedotherhamletsand, in hamletsthatalreadyhad schools, many construction of gradesincreased. new classrooms were addedandthe number By the 1980smost at least a few gradesneartheirhomes. childrenattended Zinacanteco of hamletreligioncame later. The Nachig church,built in The formalization nearthe Pueblo hamlets built outsidethe Center 1969,was the fourth (whichincludes thatused its services;see the note to Table 1). As the table shows, by 1983 more than 80 per cent of Zinacantecos living outsidethe Centerwere in hamletswith with local cargosthatcounted churches,andmorethan60 per cent lived in hamlets as first-levelservicein the Pueblo.

populations and, shortly after the formal break from Nachig in 1977, Jechtoch was able to get its own water system and its own school. Political divisions at the municipio level contributedto the split, and people from the remainder of Nachig played their part by refusing to contribute labor for projects in Jechtoch. The case

222

ETHNOLOGY
Table 1: Hamlets Outside the Center {a}

1940 Number with School Per cent population there Number with Church Per cent population there Number with Cargos Per cent population there Number with Building Per cent population there Number with Agente M. Per cent population there Municipio Population Outside the Center Population Number of hamlets Largest hamlet Smallest hamlet 5 69 1 7 1 7 4509 3427 9 819 182

1960 10 100 2 28 3 39 7650 5997 10 1227 121

1983 18 {b} 100 10 81 6 63 11 83 11 81 18000 {c} 13500 21 3100 {d} 110

Sources: Based on data from Cancian 1992, especially Tables 7.2, 7.3, B.2 (for 1940 and 1960), B.4 (col. Al for 1960, col. F' for 1983). a. The Center is defined in Cancian 1992:106. In 1983 it included the Pueblo, La Selva, San Nicolas, Vochojvo Alto, Vochojvo Bajo, andpartsof Patosil. For these calculationsall Patosil population is included in the Center. b. Three new, small hamlets (totaling less than 5 per cent of the population)had no schools. Their children had to go to adjoining hamlets. c. The 1983 population figure is an estimate (see Cancian 1992:214 and Table B.2 for discussion and details). The figures below it are rounded. d. The population of the 11 hamlets with an Agente in 1983 was: largest-3100, 10< 1600, 7 < 1200, 5 < 500, 3 < 300.

THE HAMLET AS MEDIATOR 223 arrivedin the mid-1970sas the Formalcivil offices (e.g., AgenteMunicipal) thatreached intothecountryside. increased government Manyof thepublic programs amountsof state and buildingswere built in the early 1980s when extraordinary federalmoney were funneledinto Zinacantan as part of two major initiatives:a effortto decentralize and effortto controlunrest, national administration; a state-wide in otherpartsof Chiapas. especiallyguerrillaactivityreported of Zinacantecos Overall,by the early 1980sthe greatmajority living outsidethe andformalreligiousandcivil roles. Of Center were in hamlets withpublicbuildings the nine hamletsoutsidethe Centerlisted in the 1960 nationalcensus, eight had without churches andhamletcivil buildings by the early 1980s.8Mostof the hamlets the fission of the formalroles andpublicbuildings hadbeencreated through recently With few were still small. nine. a exceptions,they original In sum, most of the people living in the hamletsof Zinacantain experienced changessimilarto those that occurredin Nachig. As the hamletsgrew in formal citizens. Many people, to ordinary the Pueblobecameless important importance which could out the life to most, they aspiredwithoutleaving perhaps play public theirhamlets.Onthe otherhand,the demands of hamlet publiclife hadbecomevery in All over the municipio, the muchmorea partof everyday Zinacantan. experience protectionprovidedby the hamletas the social unit that mediatedbetween the domesticgroupandthe community was gone. IN CHINA HAMLETS Studiesof Chinesevillages andtheirrelationto the largersystem revealmany Of course, China is huge, interesting parallelswith the situationin Zinacantan. I with which am the and the subtleties and of both basics diverse, complex. ignorant andguidesthat Chinaspecialistswork. Thus,the parallels arebasesfor speculation I do not intendto generalizeabout in Zinacantan. help clarifywhat has happened China. G. WilliamSkinner'sclassic marketpapersframethis discussion.In the first areasas systemsof villages around (Skinner1964), he identifiesstandard marketing thatthey could be seen as culture-bearing a markettown, and arguesconvincingly Skinner(1964:32)says: communities." unitsthathe labels "standard marketing
Insofar as the Chinese peasant can be said to live in a self-containedworld, that world is not the village but the standard marketing community. The effective social field of the peasant, I will argue, is delimited not by the narrow horizons of his village but rather by the boundaries of his standard marketing area.

area/comBut the standard Thus, life in the village was incomplete. marketing had service It lacked. much that the specialists,provided village munityprovided relationsandform rotatingcreditsocietiesthat to build patron-client opportunities it was the unit within extendedbeyondthe village, and, perhapsmost important, Themarket townwas also for theirchildren. whichfamiliessoughtmarriage partners

224

ETHNOLOGY

the meeting place for secret societies whose membership was drawn from the standardmarketing area surroundingthe town, and in North China it was often the

site of lineage halls. All this supportsSkinner's(1964:35-37)point: to live a hadto go beyondthevillage, butnot beyond completesocial life the Chinese peasant the standard area.In Skinner's marketing (1964:35-36)words, "a peasantdevelops a fairlygood social map of his standard areawhereasthe terrain marketing beyond
it is largely unexplored." These standardmarketing areas typically covered 50 square kilometers, while those in mountainousregions often extended over more than 100 square kilometers, making mountainous Zinacantanat 117 square kilometers a good fit. Their mean populationat "somewhatover 7,000" (Skinner 1964:34) parallels Zinacantan's7,650 in 1960, and the typical total of about 1,500 householdsclustered in eighteen villages completes the picture of a marketingcommunity as the municipio and the village as the hamlet. In what follows I substitute village for hamlet because "village" is standardin the literatureon China. Huang (1985), in explicit counterpointto Skinner, emphasizes the insularity of villages.9 Huang's description of local leadership and village-state relations speak directly to my second criterion of an ideal hamlet; viz. the informal organizationof public life. Using county archives, Huang (1985:224-25) shows the limits of state/bureaucraticintrusion into village life.
In the nineteenth century, Baodi county had 19 li, 46 bao, and 900 villages, which were theoretically divided in jia of 100 households and 10 pai [of 10 households] each. . . . In theory, the county government appointed subcounty officials all the way down to the paitou, the man responsible for ten households. But in practice, it did not try to extend its influence down the hierarchy farther than the xiangbao, who oversaw a group of 20-odd villages. A xiangbao was required to appear before the county yamen to take an oath of office, but there was no such requirementfor the village-level posts.

andoften nonexistent, thatthe effective(real)village leadersavoidedofficialties to the statebureaucracy thatthe xiangbao, the lowest-levelleader and,mostimportant, tied to the bureaucracy, was usually not a powerfulperson. "More often, the xiangbaowas one of the locality'slesser lights, propped up by the real leadersto serve as a bufferbetweenthemselvesand statepower"(Huang1985:227).Huang in power (1985:231)concludes,"[T]heBaodiexample suggestsa kindof equilibrium betweenstateandlocalsociety,in whichtaxescouldgenerally be leviedto theextent thatlocal leadersandvillagecommunities considered tolerable." In sum, these studiesof premodern Chinadocument with the situation parallels in Zinacantan the 1940-1960 Skinner shows that the during period. village in China was an incompletesocial unit that was in many ways similar to the hamlet in Zinacantan. was Huangdocuments ways in whichthe effectivelocal-levelleadership not directlyresponsible to the bureaucratic center(thestate),andspecifically how it averted directties by putting contact withthe powerless peoplein rolesthatmediated outsideworld. Thus, from the point of view of the state, village leadership was

Huang shows that the official bureaucraticlevels below the xiangbaowere ineffective

THE HAMLET AS MEDIATOR 225 informal.Fromthe pointof view of local leaders,thatinformality was local power unfettered the state. relatively by Some parallelsbetweenchangesin Zinacantain and changesin Chinaare also in Zinacantan. usefulin understanding whathappened Huangshowsthatin the early twentiethcentury,duringthe end of the Qingdynasty(to 1911) andthe beginning of the Republican period, the state createdmilitaryunits, expandedschools, and establisheda modernpolice force-all institutions that increasedits presencein taxes.Thesechanges"fundamentally altered the relationship villages-and increased betweenstateandvillage"(Huang1985:275). intothecountryside, Duara(1988) Focusingon the laterperiodof stateexpansion in documents as thestatesoughtto extract changes the rolesof villageleaders greater revenues frompeasant villages.He showsthat,just as the state'sneedfor fundswas landtax revenues became becausemorefrequent landsales increasing, undependable the landregistrieson which taxation undermined was based. Supplementary levies were assessedon the village as a unit (as opposedto the individual As landowner). a consequence, village leaderswere forcedto give up their roles as protectorsof localinterests,as honored, relations with the outside. powerful peoplewho brokered Duara(1988:249)saystheyhadto "sideeitherwiththe stateor withthecommunities they led. No village leaderwho cared about his status in the communitycould surviveundersuch circumstances." office, and yieldedto selfManyrelinquished interestedindividuals,often tax farmers,who had no social ties to the village. Especially while war lords dominatedin the 1920s, the traditionalsystem of in the countryside was in disarray.10 leadership By the end of this period,the role of the village as a mediatorbetweenrural Chinese householdsand the state had been transformed. A system that taxed landowners a network of obscure and through negotiations powerlessintermediaries hadbecomemoredemanding, reached morepeople, andwas no longerbuffered by that limitedthe powerfulpeople with local ties or powerlesslocal intermediaries state'sreach.Schools,police, andotherexpressions of modernization increased direct contactwith the state. In Chinatoo it was harderto avoid observation formal by people. AND CONCLUSIONS SUMMARY This articlewas intended1) to explorethe role of hamletsas social formsthat mediate betweenhouseholds andlarger,moreformalcommunities, and2) to discuss the demise of hamletsunderpressures createdby expanding states. Whenhamlets that are socially incomplete from the point of flourishthey have small populations view of their residentsandpoliticallyinformal from the point of view of the state. andpoliticalinformalityThesecharacteristics-small size, social incompleteness, makethe mediating roles of hamletspossible. In 1940to 1960 in Zinacantan andin the latenineteenth andvery earlytwentieth were an important social form." Their social centuriesin China,hamlets/villages was marked in Zinacantan by the roles residents took in the incompleteness

226

ETHNOLOGY

ritual associations municipio-wide system,andin China by bothsupravillage religious In and Skinner 1985; 1964, (Duara1988) 1971). bothplaces villageexogamy(Huang the state's formal reach faded before it touchedthe hamlets, and there was a andformaloffices.Powerful localpeoplewere betweenhamlet disjunction leadership traditional fromthe pointof view of the community and informal and (Zinacantan) the state (China). In both places the appointment of low-statuslocal people to to the larger, formal system buffered officially representthe hamlets/villages powerfullocal peopleagainstdirectmanipulation by outsidepowers,andpermitted themto morefreelyinfluencelocalaffairs.Thus,hamletsas a socialformmediated in a number betweenhouseholds andthe community/state of ways. In theseperiods hamletswere an important level of social organization for whatthey madepossible for residentsandwhatthey madeimpossible for nonresidents. andto the partsof Chinadiscussedhere. Changelatercameto both Zinacantan Theexpansions of the stateintothecountryside, in the 1970sin Zinacantan especially andin the earlydecadesof this centuryin China,wereparallelin manyways. They followedthe routetakenby manyagrarian societiesin this century:schools (along withthe schoolteacher who is an outsider), andmore police,bettercommunications, roads. The apparent contrasts betweenZinacantan andChinaare also interesting. The Chinadescribed and Duara was an so state by Huang agrarian economy, expansion extraction fromvillage-based The Mexicanstate dependedon increased producers. did not havethe sameeconomicneeds.In recentdecades commerce andmanufacturhave tax revenues the value-added tax after the economic ing provided (like imposed crisis of 1982);andsincethe late 1970spetroleum andhydroelectric powerfromthe stateof Chiapas haveaddedto national wealth.Directtaxation of poor ruralpeople was unnecessary. It was the desireto maintain the state politicalcontrolthatrequired to increasedirectcontactwith the countryside. differences betweenZinacantan andChinathe resultwas Despitethe important the same:the statereached into hamletsandtouchedhouseholds moredirectly.The socialincompleteness andpoliticalinformality thatmadehamlets effectivemediators between householdsand larger, more formal social forms disappeared, at least temporarily.
NOTES 1. I am indebted to Francesca Cancian for comments on this essay. 2. The incompleteness of social life in the hamlet is one of the features that distinguishes it from the closed corporate peasant community (Wolf 1957). 3. Vogt's (1969, Chapters7 and 8) extensive discussion of social groupings and settlementpatterns in Zinacantan identifies 1) the household; 2) the SNA (Tzotzil), a localized patrilineage; 3) the waterhole group formed by multiple SNAs that took water from the same waterhole (before the installation of piped water in the last two or three decades); and 4) the hamlet. There are no generic names in Tzotzil for the first three units. The hamlet is paraje in Spanish or parahel in Tzotzil. Vogt's (1969) description is mainly built on detailed knowledge of Paste and other hamlets in that area of Zinacantan, and he (Vogt 1969:149) notes that over time hamlets within the municipio have become differentiatedin culture. Collier (1975:79-82), working with detailedknowledge of Apaz, notes that the

THE HAMLET AS MEDIATOR

227

SNA and the waterhole group worked somewhat differentlythere. Most important,"only about half the households are included in [SNAs, and] . . . they differ from their counterpartsin other Zinacanteco hamlets in having no corporateritualfunctions" (Collier 1975:81). That is, SNAs were not consistently importantto the majority of the people of Apaz when Collier made his observations. Since change, like that brought to waterhole groups by piped water systems, is ubiquitous, observed differences between hamlets may reflect the time of observation, the individualhistory of the hamlet, or both. Wasserstrom (1983, Chapters 5 and 6) and Collier (1975) are especially good on historical sources of differences between hamlets. Many other publications on Zinacantaninclude observations on settlement patterns and life in hamlets (see Vogt 1978; Bricker and Gossen 1989 for references). In this article the term municipio is used in its Spanish form because the municipio is so politically, and often ethnically, importantin Mexico. Since the customary social unit is, as Mulhare shows, much more diverse than the municipio, I will use the abstraction"hamlets"in Zinacantanto mean units that Zinacantecos see as "parajes."The next section summarizes the more detailed description in Cancian (1992) and adds some details from Tax et al. (1947). 4. Not all Nachig men were members of the ejido movement. When land was distributedin 1961, only 43 per cent (n=98) of Nachig families received land (Wasserstrom1983:171). In my 1983 census (Cancian 1992, Appendix C), 103 (32 per cent) had full shares and a few more (n= 19, 6.0 per cent) had informal partial shares of ejido land. 5. Figures in Table 7.1 in Cancian (1992) are for the 1980s. Tax et al. (1947:52) shows two Principales for Nachig; Vogt (1969:148n) shows three. 6. In 1976 there were also cargos in three other hamlets (Cancian 1965). All the roles discussed here were served by men. About this time women began to take official roles in political party organizations thatbrought nationalnorms to Zinacantan.Roles in the cargo system filled by women before this period are described in Cancian (1965). 7. In 1983 Nachig had 210 households, Jechtoch27, and Jechchentic29 (Cancian 1992:220, column
F').

8. Salinas, which was near the Pueblo, had only the churchlocated there since the nineteenthcentury, and Chianatic, the smallest of the nine, had only a civil building and an Agente. 9. Huang (1985:220ff.) says that villages on the north China plain were larger than those in the Chengdu plain studied by Skinner, that villagers went to market less frequently, and that when they were there they interacted less with people from other villages. He argues that the north China plain villages were quite insular, thatmen seldom talked with men from othervillages, even when the villages were adjacent and the men as youngsters had been classmates in a joint school. Yet, though he makes little of it, Huang (1985:222n) notes that north China plain villages are typically exogamous. How people find mates without talking is hard to explain, unless, of course, the women, who are not the players in most of this peasant talk, get it done. But then, their connections across village boundaries must count. 10. Duara seeks to revise both Skinner and Huang. He argues that their work leads to an oversimplification of the situationin the Chinese countrysidein the early decades of this century, and he shows that extravillage relationshipssuch as those in irrigationassociations were often very importantto local life. He sees rural people as embedded in a "cultural nexus" that includes religious cults that are not localized, and he asserts that "an exclusive analytical focus-whether on the village or on the market town-is an arbitraryand abstractprocedure" (Duara 1988:247). 11. As Skinner (1971) makes clear for China, this form may represent a recurrentstage in cycles of changing village-state relations, not a point on a unilinear evolutionary path. BIBLIOGRAPHY Bricker, V. R., and G. H. Gossen. (eds.) 1989. EthnographicEncounters in Southern Mesoamerica: Essays in Honor of Evon ZartmanVogt. Austin. Brown, L. (ed.) 1993. The New Shorter Oxford English Dictionary. Oxford. Cancian, F. 1965. Economics and Prestige in a Maya Community: The Religious Cargo System in

228

ETHNOLOGY

Zinacantan. Stanford. 1967. Political and Religious Organizations.Handbookof Middle American Indians, Vol. 6, eds. R. Wauchope and M. Nash, pp. 283-98. Austin. 1992. The Decline of Community in Zinacantan: Economy, Public Life, and Social Stratification, 1960-1987. Stanford. Collier, G. A. 1975. Fields of the Tzotzil: The Ecological Bases of Traditionin Highland Chiapas. Austin. Collier, J. F. 1973. Law and Social Change in Zinacantan.Stanford. Duara, P. 1988. Culture, Power, and the State: Rural North China, 1900-1942. Stanford. Huang, P. C. C. 1985. The Peasant Economy and Social Change in North China. Stanford. Mulhare, E. 1996. Barrio Matters: Toward an Ethnology of Mesoamerican Customary Social Units. Ethnology 35(2):93-106. Nutini, H. G. 1976. Introduction:The Nature and Treatmentof Kinship in Mesoamerica. Essays on Mexican Kinship, eds. H. G. Nutini, P. Carrasco, and J. M. Taggert, pp. 3-27. Pittsburgh. Skinner, G. W. 1964. Marketing and Social Structurein Rural China, Part I. The Journal of Asian Studies 24:3-43. 1971. Chinese Peasants and the Closed Community:An Open and Shut Case. Comparative Studies in Society and History 13:270-81. Tax, S., et al. 1947. Notas Sobre Zinacantan,Chiapaspor Miembros de la Expedici6n a Zinacantan1942-3 Bajo la Direcci6n de Sol Tax. Microfilm Collection of Manuscripts on Middle American CulturalAnthropology, No. 20. Vogt, E. Z. 1969. Zinacantan:A Maya Communityin the Highlands of Chiapas. Cambridge MA. 1978. Bibliography of the Harvard Chiapas Project: The First Twenty Years, 1957-1977. Cambridge MA. Wasserstrom, R. 1983. Class and Society in Central Chiapas. Berkeley. Wolf, E. R. 1957. Closed Corporate Peasant Communities in Mesoamerica and Central Java. Southwestern Journal of Anthropology 13:1-18.