ASSOCIATION FOR CONSUMER RESEARCH

Labovitz School of Business & Economics, University of Minnesota Duluth, 11 E. Superior Street, Suite 210, Duluth, MN 55802 Gender and Land Reforms Consumption, Production and Some Contradictions

Susie Jacobs, Staffordshire University, United Kingdom [to cite]:

Susie Jacobs (1993) ,"Gender and Land Reforms Consumption, Production and Some Contradictions", in GCB - Gender and

Consumer Behavior Volume 2, eds. Dr. Janeen Arnold Costa, Salt Lake City, UT : Association for Consumer Research, Pages:

133 to 146.

[url]:

http://www.acrwebsite.org/volumes/15590/gender/v02/GCB-02

[copyright notice]:

This work is copyrighted by The Association for Consumer Research. For permission to copy or use this work in whole or in

part, please contact the Copyright Clearance Center at http://www.copyright.com/.

Gender and L,and ReformsConsumption, Production and Some Contradictions

Susie Jacobs, Staffordshire University, United Kingdom

This paper attempts to apply sociological andfeminist perspectives on consumption in a "third"

world context - specifically, that of gender andland reform. Perspectives on consumptionwere,of course, developed as critiques (or sometimes,defenses) of consumer society/the culture ofadvanced capitalism (Kellner, 1983; Nava, 1990;see Hoggart,1957;Marcuse, 1964). As such,analyses of consumption do not always traveleasily. If free choice [of goods] came tosymbolize the freedom of the "free"world (Nava,L987),at least before the collapse ofbureaucraticstate socialism, then the "Third"and particularlythe "Fourth'Worlds are characterized in part bylimited existence and/or availability of many typesof consumer goods. Additionally, in many areas,and depending upon varying social norms anddegrees of male control, the availability of goodsmay be particularly restricted for women. Thequestion of the relevance of "consumption" incontexts of capitalist underdevelopment raises thatof the generally validity of analyses ofconsumption. If these are of use only in advancedsocieties, or else in strictly limited contexts in pre-capitalist societies (Appadurai, 1,986), then theirtheoretical robustness cannot be seen as great.

It should be noted that the land reforms discussedtake place in the context, not ofpre-capitalist but,of capitalist production, although in some cases atransition to capitalist production may beoccurring. Some land reforms and resettlementgparticularly from colonial governments, have beenthe result of authoritarian state action and havetaken place as part of population "management"

schemes. However,land reforms and redis-tributions'more often take place as a result ofrevolutionary struggles, or else they take the formof state actions to pre-empt such struggles. Suchland reforms are usually m€ant to benefit poorerpeasants and to create a stratum of better-off"middle"or wealthier peasants. Usually they areenacted by welfarist governments in societies inwhich the main agrarian classes have been

landlords/agrarian capitalists, a landlessproletariat and an impoverished small p€asantry.

The question implicit here is the (Marxist)problem of what, if any, priority should be givento the sphere ofproduction. The contextsdescribed are hardly similar to those of latecapitalism, in which the process of productionbecomes "swallowed,"conveniently invisible to theconsumer (Jameson, 1984). Land reforms involve,by definition,changes in control and/or ownershipof land, one of the main means of production (theprocesses of production are painfully obvious).However, these are also contexts in whichincreased commoditization takes place (or ismeant to take place) and in which the volume ofgoods and services increases, althoughto a lesserextent. Seeing land reforms as involving solely achange in production relations, as is common,gives an incomplete picfure. An outline of theaims of land reform illustrates this: suchprograms (and those who implement thern) seekto better the lives of small agricultural producersin various ways: by granting land or else landrights, by providing agricultural and welfareservices by raising incomes and particularly byensuring food security. The latter aims involvesecuring or raising standards of consumption inbasic ways.

These two aspects of economic and social life are,of course, closely intertwined. This has beennoted by various theorists, and particularly byfeminists. Galbraith, for instance, questioned anover-heavy emphasis upon production as male-oriented (1974,cited in Nava, 1990), pointing outthat women's labor in managing and administeringconsumptionwas just as important to capital asthe labor involved in production. Feminists havenoted that shopping (in the West, women's work)has been trivialized trough its association with thefeminine (Nava, 1990) as well as through itsassociation with non-productive activity. Alongother lines, feminists have criticized an exclusive

133

focus on women and "consumption" as obscurtng

*o."n't contributionsto production' I do not

*iJ to sidestep the question of the relative,''weignf'of production and consumption - they are

noi Jntirety contiguous - and will return to this

below. However, the suggestion that we

deconstruct dichotomous views of

JioOuctioni"onsumption can be useful' A good

ixample of such thinking is Alan Warde's

,ogg"ttlon (1990:3) that we take account of the

cvJfi of production and consumption, thereby

ciarirying what is meant by "consumption'" Warde

distinguishes four elements in this rycle: the

produ-ction process' the conditions of access, the

*unn"t of delivery and the environment of

enjoyment. I use this schema, as it seems

particularly fruitful for analyses of gender and

land reforms.

Discussion of the effect of land reforms on either

gendered production or consumption relations is

Ionstrained, simply because few studies of women

andior gender relations and land reforms exist' I

have attempted to locate such studies and have

found fourteen (including my own) in which

landholdings were distributed to individual

families. Most of the studies concern northern

and sub-saharan African settlements, being

located in Libya, Upper Volta (Burkina Faso)'

Ethiopia, Kenya, Tanzania, Zimbabwe and(separately) Rhodesia' Chambers (1969)

compared several African schemes from the

1960s. The non-Africanworks concern a Sri

Lankan settlement, a study of China in the early

stages of land reform, a study of Honduras, a

OetaiteO study of Peru and a comparison of Peru'

Chile and Cuba. Rogers (1981) also compares

several cases in order to make a general

argument. Anotherpiece is a hypotheticalcasestudy which is in part a composite of real cases

(Palmer, 1985). In other papers, the topic of

gender and land reform is mentioned (Mogadam'

iggZ; tvtutt" ,7992). Still other studies, or other

sections of the studies mentioned above, discuss

land reform in the context of production

cooperatives or collectives (Davison, 1988; Deere'

1983; Stacey, 1982)' I include the latter to provide

some useful points of comparison; however, I

concentrate upon studies discussing gender and

individual famity farming so that comparisons will

be meaningful.

In this survey of empirical studies, I have

attemPted to identify asPects ofreform/resettlement which have had a particular

impact upon women. Of course, the effects of

land reform vary with specific policies and

circumstances. Programs normally contain

features which are contradictory in gender terms,

with some being beneficial to, and other

detrimental to, women in the sense of

enhancing/diminishing their power' autonomy,

access to and control over material and other

resources. Some studies also recognize that

women are not a uniform category, so that effects

of reforms may vary according to class and to

other factors, not here studied in detail, such as

ethnicity and age.

There is insufficient space here to discuss the

structure of gender relations in the societies

mentioned above, in any detail' However, Ipresent a few notes on the background of gender

diuitiont, gender ideologies, women's economic

roles in oider to provide a (minimal) context for

the observationswhich follow. It can be stated as

a general point that (with one possible exception)'

thi societies discussed are heavily male-dominated

although the exact forms and manifestations of

that domination vary from society to society'

historically, and, within societies, between

national, tribal, religious, class and othergroupings. I group the peoples and societies

discussed geograPhicallY.

Libya (the Libyan Arab Jamahriya) is culturally

fairiy homogeneous, with the majority peoples

desCended from Berber Arabs, who are, of course

Islamic and strictly patrilineal and patriarchal'

Most resettled people came from scattered areas

in the Sahara; women were engaged mainly in

herding; water collecting, farmwork and

housework (Allaghi, 1984:138)' Burkina Faso

(Upper Volta) is the only West Aftican and

ipittl*V') Francophone example given' It is

located in the Sahel, the semi-arid southern rim of

the Sahara. The area in which the resettlement

.study took place was one inhabited by the

majority MLssi people. Mossi society was' befote

colbnialism, an hierarchically-ordered kingdom;

the Mossi now are patrilineal herders and

agriculturalists. In common with other African

societies, women have some degree of economic

t34

independence, particularly in the sphere of

trading'

Ethiopia is culturally diverse, with two/thirds of

oeople being of Amharic or Galla descent (Third

woitO Guide, 1992). Other smaller ethnic groups

include the Tigreans and Niloti' Most Ethiopians

are Coptic Christians. Britreans, who live along

the Red Sea Coast, are a distinct people who are

either Roman Catholic or Muslim' Britrea has

successfully waged a liberation war against

Ethiopia, having been recognized as an

independentstate on Z4thMay, 1993. Women's

status varies between groups and religions. In the

early years of Mengistu Mariam's militarygovernment, an agrarian reform was carried out.Before this time the country's agrarianrelationships had been mainly feudal. Goody(1976) characterized highland Ethiopia astransitional between his nvo main models of pre-

industrial society. The Eurasian model is one ofstrong class differentiation, monogamy' dowry andcontrol of women primarily for their property

rights and reproductive capacities. The second,sub-saharan model is character2ed by a lowdegree of class differentiation, prevalence ofpolygyny, bridewealth and control of women for

their labor and their reproductive capacities'Tadesse (1982) notes that the status of womenvaried markedly according to their marital status,amount of (family) land owned and their classposition. For instance, those in landless and poorpeasant households assisted in most stages ofagricultural production, while women in better-offhouseholds were excluded from production.

The other sub'saharan African cases discussed(Kenya, T anzania, Mozambique, Shona'speakingareas of Southern Rhodesia/Zimbabwe) allconform broadly to Goody's model. Most of thepeoples in the societies discussed were patrilinealand were engaged in agriculture and someherding activities. In some, few, cases such as theTonga of Rhodesia/ZimbabWe, the kinship systemwas matrilineal but during colonial timespatrilineal customs such as high bridewealth andincreased rates of polygyny were adopted, possiblyin emulation of neighbors (Weinrich, L979)' Allof the societies discussed were drastically affectedby colonialism; in particular, settlers appropriatedgreat tracts of African land - sometimes the

majority - and in the most fertile areas. Majorityreligions are either African traditional religion(Mozambique, Tanzania) or Christian (KenyaZimbabwe), although all the countries mentionedexcept Zimbabwe have large Muslim minorities(e.9.,30Vo in Kenya). Althoughthese wereheavily male-dominated societies, women hadsome limited spheres of independence and ofinfluence, particularly as they gained in age and ifthey bore many children. In pre'colonial timesland was held communally and allocated byheadmen (and occasionally, headwomen) and bychiefs and this system was continued in Britishcolonial territories in distorted form. Customarilyhusbands allocated wives plots of land on which foplant crops of their choice; the wives alsocontrolled proceeds ftom these plots. However,since they were expected to feed their husband,themselves and their children, their choices were,in practice, limited. The practice of colonial lawand colonial officials eroded women's land rights,their prestige and their ability to exert informalinfluence.

Sri Lanka has a modified form of caste systemnalthough the majority religion is Buddhist, with alarge Hindu minority. The majority people,Tlcloof the population, are Sinhalese, with one-fifth ofthe population belonging to the Tamil minority,descendants of Dravidian peoples (Third WorldGuide, 1992). Rogers (1981) speculates that priorto British colonialism there was some degree ofrecognition of female land rights. The landreform discussed nationalized ten British'ownedplantations after independence' One unpublishedpaper (Mukta, !992) mentions a land reformmovement in one of the poorest Indian states,Bihar. The majority Indian population is, ofcourse, Hindu. In caste systems, women's "purity"

(and to some extent, men's) is controlled not onlythrough individual spouses and families but alsothrough the operation of the caste system itself.The kinship system usually operates throughpatrilineal extended families rather than throughlineage groups. These societies conform, broadly'to Goody's Eurasian type, and -women's visibilityin agriculture and in other realms varies accordingto class and also caste position. China,particularly in pre-revolutionary times, was one ofthe more patriarchal societies in existence. It was(and largely remains) patrilineal; female

135

infanticide was (and is) common; wives have a low

stafus and while young are subject to the authorityof husbands and of in'laws. In all the Asian

examples, women secure their family position only

through th€ birth of sons (and not always then).

The Latin American cases discussed are, in themain, heavily influenced by Spanish colonial-derived culture and its religious manifestation,Roman Catholicism. In many areas early colonialproc€sses entailed the death from either diseaseor genocide of the indigenous population, so thatthe remaining population is partly Spanish (orother European) descended and is RomanCatholic and Spanish-speaking. The dominantgender ideology is what is often termed the" machismo / Morianismo" complex, emphasizingmale aggression and sexual prowess and femalepurity and vulnerableness. Ideally, according tothis ideology, women do not work outside thehome except in cases of economic necessity. Inpractice, women contribute heavily to agriculturalproduction but this is often overlooked (Dixon-Mueller, L985). Among native American peopleswho have survived, other gender systems mayexist. An example discussed further below is thatof Peru, where half of the population is of nativeAmerican descent and where 30Vo of. thepopulation speak no Spanish. The Quechua ofhighland peru have a more egalitarian system ofgender relations than is common in l,atinAmerica. Women inherited land bilaterally, and a"complementary"system existed (Huber, 1991) inwhich women and men perform different tasks;women have a high degree of power andautonomy.

TWo further notes are relevant here. Firstly,"indigenous,"colonial or world-religious ideologiesare not the only ones relevant to gender relationsin the cases discussed. "Western"(including

socialist and feminist) ideas of gender equality areknown in most societies and may have a strongimpact. Secondly, many of the studies discussedtook place at time of war or large-scaleinsurgency. A more thorough discussion ofgender and land reforms would entail analysis ofthis contexts of violent upheavals, themselvesoften precipitating land reforms.

Given the cultural and regional variations ingender systems, it is notable that the ways inwhich land reforms have affected women andgender relations are similar in many, although notall, respecfs.

In discussing the case studies, I have kept Warde'suseful framework in mind. I write "keptin mind"rather than "used"because in the field of "Third"

World studies, much is written about theproduction process, somewhat less aboutconditions of access, and still less about mannerof delivery and little at all about the environmentof "enjoyment"(in many cases, hardly anappropriate word). In the case of gender andland reform, as noted, much of what is writtenfocuses on productiveresources and on theproduction process in the form of allocation oflandholdings, the gender division of labor andwomen's work burdens. Irss of a focus arefeatures which relate more to the consumptionprocess, such as women's access to their ownholdings and crops; provision of services andaccess to incomes, both household incomes andwomen's own. Other factors, such aspower/autonomy over decision-making, areimportant for both production and consumption.Still others such as the rise of the nuclear familymodelrelate less directly to theproduction/consumption cycle but do haveimportant repercussions for it.

In the comparison of studies of land reform andgender, I compare the above-mentioned aspects.

1. Allocation of landholdines:This is a basic criterion, from which many effectsand repercussions stem. The question ofallocation of land holdings relates to "accessto

resources" in Warde's schema. In most programs,landholdings (or fitles) are granted exclusively ormainly to men. Such allocation is not madeexplicitly throughgender criteria. In general, onlypeople deemed to be "household heads"-areallotted land titles. Except in the case of widowsor divorcees, "household heads"are implicitlyassumed to be male. Widows/divorcees are notallowed to hold land in all cases, but even in theschemes in which they are (e.g., Honduras,Tanzania, Zimbabwe), in practice relatively fewwomen benefit (McCall, 1987:205;Safilios-

136

Rothschild, 1988:217). Palmer (1985:30) notes

that a [married] woman's access is akin to that of

a bonded laborer.

TWo cases of reform along individual family lines

were found in which married women were

allocated land or titles on an individual basis. The

first, mentioned in a paper by Mukra (1992)' was

that of a low-caste movement located in Bodhgaya

village in Bihar, one of the poorest states in the

country. Encouraged by socialist'feministmernbers, the movement demanded land

redistribution in women's names. At first, womenthemselves were not in favor of this demand.However, by 1979/80 the landlord (a religiousinstitution) had been demoted and land wasredistributed to women. It appears that womenhave benefitted from this radical change. Thesecond case is a much larger-scale one' In areasof China under Communist Party control afterWorld War II, the CCP consolidated its socialbase through land reform to individual smallpeasant families. Although, as elsewhere, the landwas initially allotted to household "heads." In1948, the Central Committee recognized thediscrepancy between the promise of land rights towomen and this practice, and instructed cadres toallocate land to women even where their landswere included in family holdings (whether in natalfamilies or families of marriage). In practice, thispolicy was only rarely implemented, being easilycircumvented both by male officials and familymembers. However, it was of symbolic importanceand even de jure rights for women materiallybettered their (lowly) positions within families(Stacey, 1982:262).

In other cases land has been redistributed on acollective or cooperative basis, so that noindividual owns land or titles (e.9., CubaMozambique; Zimbabwean "Model B" sch€mes;Peruvian cooperatives). In these, women usuallyhave membership on an individualbasis regardlessof marital status. However, they may find it-difficult to become accepted as full cooperativemembers since they are often regarded as beingunable to work as hard as men. Where they arefull members, they are formally granted the samerights as men. Despite granting of land rights, thevarious studies of collective land reforms indicatethat these have not been ofunambiguousbenefit

to women, for reasons discussed further below.However, these collectives do have the merit ofnot disenfranchisingwomen from the outset.

Rogers (1981) argues forcefully that variousreforms (except those along collective lines), invarious cultures, whether implemented from thepolitical "right"or the "left,"have eroded collectiveland rights, have assigned land to men on anindividual basis and have thereby underminedwomen's customary rights. Most writers agreewith this assessment. However, one author,Huber (1991) argues (for the Peruvian case) thatthe effects of legal factors upon women have beenoverestimated. She writes that factors such asprior gender ideologies and the gender division oflabor determined the ways in which womenparticipated in reforms. In some cases, womenwere able to use strategies to compensate for anydiscrimination they suffered. However, Huberdoes not disagree with the point that the landreform tended to be disadvantageousfor women.She merely argues that reforms were, in the end,weak and that these disadvantages could beovercome.

The aspect of allocation of landholdings, the mainmeans of production in these agrarian settings, iscrucial to the outcomes of such reforms forwomen, men and gender relations. The relativelyfavorable outcome (for women) of cases wherewomen do receive land titles emphasizes thispoint. Wives "start out" structurally disadvantagedvis-a-vis husbands and other men. Other,potentially advantageous changes which occur,remain within this framework of femalesubordination.

2. Gender division(s) of labor:This topic has to do with labor as a factor ofproduction; that is, in most cases, women's laboris one of rrent productive assets. Hypothetically,both male and female labor could contribute to ajoint productive enterprise in which eventualconsumption of goods and services was also joint,although this scenario does not usually occur inpractice. Not all cases discuss this matter andevidence with regard to it appears contradictory.For instance, in the Sri Lankan case and in ascheme in Mwea, Kenya, women took over taskspreviously allocated to men (De Silva, 1982:

L37

Hangar and Moris, 1973) although men did notreciprocate by taking on women's work. In myown study in northeastern areas of Zimbabwe, Ifound that the gender division of labor became"blurred" in favor of women: that is, husbandsparticipated more in fieldwork than they hadpreviouslY.

3. Women's work burdens:The amount of work women perform relates bothto relations of production ("theprocess ofproduction")and of consumption. In generaLterms, women's burdens of work increased withland reform. This occurred for a variety ofreasons. Women's labor as a force of productionin fieldwork is nearly always used moreintensively, even in cases where men alsoparticipate more, because households usually havemore land to cultivate and because initialpreparatory work (e.g., clearing land forhomesteads and fields) is particularly arduous.Women may be expected to contribute more laborto (men's) cash crop gardens, as in the Mweascheme. And women may work harder due tochanges in the above-mentioned gender division oflabor (Lund, 1978). Women's roles as biologicalreproducers may also become more ta:ring if fheyare expected to reproduce additional householdlabor in the form of children.

Another reason for the increase in women's workburdens is the increase associated with changes inprovision of services (detailed below). Forinstance, Conti, writing of the Upper Voltascheme (1979), noted that it presupposed agreater burden on women's work than had beencommon. Wells were fewer in number and faraway, so that women had to carry water fordrinking, cooking and washing on their heads.Customarily, mills for grinding millet wereavailable in villages but at the time of study wereonly at the planning stage, so women had to grindgrain individually. In general, any "spare"time

women had previously enjoyed disappeared.

4. Provision of services:This criterion primarily has to do with the"manner of delivery" [of consumption]. However,service provision is a topic in which the linkbetween consumption and production becomesappar€nt. Without roads, access to marketing

points or outlets and so on, productive accesscannot be translated into consumption via thecash nexus.

As noted above, provision of transport was amajor problem in many programs and this servedto increase the burden of work for peasants,particularly women. Access to shops and marketsemerged as a problem; the former is commonlydeemed to be a female sphere. Marketingservices are commonly seen as a male sphere(although important exceptions exist, such as WestAfrican female market traders). There areexceptions to this role assignment in somesocieties. however. In Islamic societies menusually shop in order to preserve female seclusionand honor. In Honduras, Safilios-Rothschild(7989:225) found that poor peasant men shoppedfor food in order to preserve control overhousehold decision-making. In T anzania, wherewomen are responsible for food provision, theyhad to purchase items such as sugar, salt,kerosene, clothing, cooking pots which werepreviously exchanged or made at home. Mostother areas studied here had undergone a greatetdegree of commoditization at an earlier period, sothat women's tasks were simply made moredifficult rather than dramatically altered. Anassociated problem, the provision of clean, safewater supplies, was of major concern for women,who are seen as responsible for health issues inmost societies.

More positively, government welfare services suchas schools, clinics and agricultural extension wereoften available in land reform schemes (Geza,1986; Hangar and Moris, 1973; McCall, 1987;Mogadam, 1992). In such cases (e.g.,Zimbabwean resettlement) schools and clinicswere provided but were located at a great distanceaway from scheme or from particular villageswithin schemes. Educational provision and accessappears to be a matter of some gender conflict.In the Afghani example referred to by Mogadam(1992), the whole scheme faltered because thegovernment included the teaching of femaleliteracy as part of the program. Men revolted,seeing their hegemony (and women's "modesty")

threatened and many turned to Islamistmovements. In another example of the Frei,Christian Democratic reform in Chile (Deere,

138

1986), women were excluded from running the

schemes. Male scheme leaders concentrated

solely on technical and economic problems.

Matters such as schooling, housing and health

care, which are relevant economically and in

which women might have assisted, were ignored.

The whole scheme was weakened by women's

mar ginalization in "Centros de Madres."

TWo points can be made concerning serviceprovision and land reforms. Firstly, the record

was somewhat contradictory although overall, lackof adequate service provision was detrimental for

women. Secondly, provision of services,puticularly welfare services, emerged as animportant site of gender struggle'

5. Household income:This criterion relates to "conditions of access" andto "the environment of enjoyment" ofconsumption. Most studies indicate that landreforms have raised overall "household" incomes(more accurately, seen as husbands'incomes);indeed this is one of the main aims of land reformpolicy.

In Zimbabwe, this strategy has been partlysuccessful, at least until the droughts of the late1980s/early 1990s. In the mid-1980s, there was a"boom"in peasant production, including that inResettlement Areas, due to the improvedprovision of credit and extension, the provision ofhigh-yielding seeds and the removal of pricediscrimination against black farmers. The benefitsof such policies ar€ uneven: they are mostmarked in fertile areas and go disproportionatelyto men. However, in my research I found thatmany men redistributed at least a share of theirgains to wives. De Silva (1982:145) argues thatfor Sri Lanka, both sexes benefit from economic"success." TWo other sfudies, generally critical ofgender aspects of land reform policies also notesome degree of economic betterment for women.Palmer's (1985:33) composite study indicates thatwomen had improved diets, better clothing andmore consumer goods, albeit at the expense ofincreased workloads. Tadesse (1982) reports thatin Ethiopia, women felt freer to ask husbands formoney for clothing and for household items.More importantly, they were greatly relieved that

[at the time] they did not have to worry about

food shortages. For many women, especially onesin drought-proneareas, this aspect of reformoverrides all others: they feel more prosperousand less fearful of the prospects of hunger fortheir children and themselves.

Here a contrast between land reforms organizedalong collective and along indMdual family linescan be made. For a variety of reasons -

undercapitalization; initial poverty of participants;(possibly) lack of incentives; corruption; lack ofstate support - collective farms and cooperativesoften fail to raise peasant incomes and to "deliver

the [consumption] goods." This phenomenon isparticularly unfortunate for peasant women whoare at least formally, entitled to income on anindividual basis. However, in my 1984 researchon six Zimbabwean state agricultural productioncooperatives, I found two factors preventedwomen from realizing sufficient income. One wasthat women on five of the farms studied, like men,remained poor because of the cooperatives'poorproductionrecords. Women's disillusionmentatthe continued general poverty was compounded bythe fact that husbands often appropriated wives'incomes even though they were not "entitled" todo so, and male cooperative leaders rarelyintervened. Despite their anger about suchmatters, women cooperative members Iinterviewed experienced the right to individualentitlement as a positive feature. At a minimum,it constituted a focus for struggle. Likewise,Deere (1986) found that Cuban women benefittedas members of collectives because theirmembership was independent of men. However,the aspect of increased formal rights failed tocompensate for lack of real "betterment,"in termsof rights, of income and of food security,

6. Class differentiation:Several studies pay attention to this importantaspect, which straddles relations of production (animportant basis of class) and of consumption,through which "markers"of class and of status canbe asserted (Bourdieu,1984). I have argued thatwomenls labor is used most intensively in the"middle" peasant stratum (Jacobs, 1991).However, Tadesse (1982:207) argues that inEthiopia, the "higher"a woman's class position,the less agricultural work she does. Regardless ofdisagreement concerning in which peasant stratum

r39

rvomen's labor is most essential, several studies

IJJ tft"t wealthier women's labor is least essential

t"'tft" nout"hold, since their husbands can afford

iJ iiir. in labor. Here, withdrawal of female labor

ir "*a as a status marker, and women are pushed

into tft" sphere of consumption' In terms of

uUility ,o tontum", then, these women are better

off than poor or "rniddle" peasant women'

However, they may also lose a measure of power

through their separation ftom production'

Conversely, poor peasant women, who are worst

off in tonturnption terms, may be more powerful

within their own households'

This matter may vary according to gender norms

as well as according to women's productive

contribution. Safilios-Rothschild's Honduran

sfudy concentrates on class differentiation with

regard to women's control over income' Because

of the norm of machismo, it was found thatpoorer peasant women had to hide their own

inro."i; wealthier women's incomes posed less of

a threat to husbands; women were allowed to be

more open about the existence of income and

they had more control over household income,

including that of men.

7. Women's own landholdinqs:This topic refers to land that women may have

held as a customary right before land reform: the

criterion relates to "access to resources" as well as

to "mannerof delivery" and to "environment of

enjoyment." As noted earlier, in some areas(notably, southeast Asia and some sub-SaharanAfrican peoples) women held land through a

matrilineal system, but more commonly, husbandsallocated plots of land to wives. In Africa, it was

common for women to plant subsistencefoodstuffssuch as groundnutsrather than cash

crops. Thus, women's autonomy over landoperated within strict limits and according to prior

ginder expectations. Nevertheless, land rights

constituted one of the few spheres of autonomyavailable for women and had great subjectiveimportance as well as "objective"effects inimproving women's and children's diets. In areas

where such rights operated, they provided an

opportunity for female control over production as

well as consumPtion'

In many of the schemes surveyed, women lost any

customary rights held previously through land

reforms (Chambers, 1969; Colson, 1960; Conti,

!979;Deere,1986; Hangar and Moris, 1973;

McCall, 1987; Rogers, 1981). McCall mentions,

for instance, with regard to Tanzanian ujamaavillages, that women considered the loss of land

on which to grow their "own"crops to be their

most serious problem. In the Mwea scheme,Kenya, and among the Tonga of SouthernRhodesia/Zimbabwe, women became completely

dependentupon husbands due to loss of their own

land and crops and had to rely upon the men's"generosity." In other societies no previous land

rights existed for women, or else they had long

been eroded or were no longer practiced, as in

some areas of contemporary Zimbabwe(Pankhurst and Jacobs, 1988). In my research I

found that a large minority of women, between

207o-33%,were allocated a small garden plot by

husbands. However, no study mentions that

women's access to land is acrually betteredthroughland reform: at best, as in my study, the

situation does not deteriorate.

8. Women's own incomes:The situation with regard to women's own incomeparallels that of land rights. Married womeniommonly lose out while the husbandAtouseholdhead gains income (Colson, 1960; Deere, 1986;

Hangar and Moris, 1973; McCall' 1987;Palmer,1985; Rogers, 1981; Safilios-Rothschild' 1988)'

Colson noted that women often lose access to

income sources in sub-saharan African schem€s,through loss of access to raw materials, through

the move itself and through loss of opportunitiesfor trading and marketing. In the Upper Volta

example given, the effects upon women were

dramitic. Through loss of such opportunities they

were no longer able to exchange gifts. They were

therefore not able to participate in ceremoniesand lost contact with their own extended families(Conti, 1979:89). Husbands commonly acquire

monopolies of new economic opportunities and so

their power is enhanced. Similarly, Safilios-Rothichild (1988) found that in Honduras,landreform depressed women's incomes while raising

men's. In the schemes I studied it does not seem

that women lost access to income, since mostpreviously lacked control over income in any case'

Relatively high wives' incomes were reported; it

140

may be that the Zimbabwean case is unusual in

itrii t"tp".t and that jn- purely financial-terms

*orrn are better off than previously. Men's

trigtt"r incomes appear to have "trickleddown'"

However, disposal of any income women do

Dossess ' the "enjoyment of consumption"is

another matter. Husbands may refuse to allocate

any "extra"income to wives, if they so choose

(Dwyer and Bruce, 1988). Perhaps particularly in

Aftica, the enjoyment of consumption emerges as

I rnale prerogative. Conti notes that in Upper

Volta, goods such as radios and bicycles could be

afforded by men after 4-5 years saving; however,,no such goods were available for women, who

Wgre dependent upon "nic€"husbands for gifts. In

Tanzania, women had some limited marketing

opportunitie s in ui amaa villages. However, as is

cornmon, they were expected to spend any cashon household reproduction. Men, on the otherhand, were free to spend money on bicycles, beerond women (McCall, 1987:207). Likewise, men'so:rpenditure patterns - again, on beer andwomen - are a common and bitter complaint ofZimbabwean women (Pankhurst and Jacobs,1988), since needed resources are thereby divertedaway from the household.

9. Power and autonomy: decision-makiugThis, "political"factot relates to all points in theproduction/consumption cycle as well morewidely. Few sfudies explicitly examine the area ofdecision-making, although this is mentioned inpassing in several. The studies which did discussthis found women's scope for decision-making,particularly over decisions concerning production,to be unchanged or narrowed throughlandreform.

Allaghi (1984: 139) who examines daughters ratherthan wives, found participation in family decisionmaking not to be increased as a result of skillstraining received in the Libyan project. On theMwea scheme, Kenya, Hangar and Moris (1973)found women's scope for decision-making, narrowin any case, reduced in a de facto *n since theycould no longer rely upon their own resources toprovide food. Additionally,men took all decisionsover expenditure.

With regard to decisions over consumption ofincome, evidence is patchy. In Safilios'Rothschild's Honduran sample, as noted, womenhad to defer to their husbands authority andmachismo to the extent of making incomes"inyisible" whenever they seemed to threatenmale status. It is uncommon for wives to be ableto make decisions about disposal of incomewithout husbands'consent, even in societies withdifferent gender norms. Where women areempowered to make decisions, even trivial onessuch as aboutwhat food to buy (see section 4above), they may be chastised if the husband doesnot agree with the decision (that is, if he does not"enjoy"his dinner!).

Often women have to resort to seeking to exertmore influence upon husbands and other malehousehold members since they lack effectivepower. In Sri Lanka, Lund (1978) reported thatsettler wives took little part in economic decisions,although they had gained more scope forassertiveness. Stacey (1982) implies thata similarprocess took place in the early Chinese reform.In Zimbabwe, many female settlers felt they hadgained more say in decisions, although thisphenomenon varied according to class position.One aspect of Zimbabwean settler women'sincreased influence was that men were s€en asbeing "betterhusbands:" i.e., husbands who wereless likely to spend money on the extra-householdactivities mentioned. Their behavior was also dueto lack of proximity of beerhalls and to theinfluence of Resettlement Officers (Jacobs, 1989).Anotherway to view the latter phenomenon isthat women approved of men's investment inproduction rather than consumption: perhaps thiswas partly in hope of future, shared consumption.

10. The nuclear familv model:Suchcontradictoryeffects are related to thepredominance of the nuclear family in manyschemes, either in reality or as a model,particularly for policy:makers of "ideal" life(Allaghi, 1984; Fapohunda, 1987; I-und, 1978).This may continue the same structure withinwhich immediate production and consumptionrelations occur, or it may create a new context.Nuclear family life may give wives more say in thefamily and more influence over husbands (DeSilva, 1982). This may be the result where

141

resettlement involves relocation to an unknown

ntace wittr unfamiliar people, so that settler men'may

have to rely more heavily on wives for

sujport. But even this "favorable"outcome can

ie double'eaged. The decreased importance of

krn and neighbors (which occurs for a variety of

reasons) can be detrimental even while raising

wivqs' status. Various studies mention that

wonien suffer particularly from isolation due to

loss of community. In Africa, if the official model

is not only one of a nuclear family but is also of a

monogamous tlpe, this may have devastating

effects. A man may register one wife, while

leaving others without any access to land(Tadesse, 1982:214).

As ftequently noted in feminist writings, a wife's

dependence upon the husband is integral to the

nutlear family. This feature emerges so often in

the studies cited here that it hardly requires

further emphasis. This seems to me to be the

case, even where ' as is common - women activelydesire to lie in such units. There seems littlereason to think that women have gained inpersonal or material autonomy as a result of

family changes. Indeed, they may have lost a

degree of autonomy for an increase in materialwell-being. It is possible that the inverse is alsothe case. Deere (cited in Huber, 1991:L82) notes

that in the Peruvian reform when nuclear familystrucnrre was weakened through male migrationwomen gained in independence and prestige but

at the cost of more work and of immiserisation.The nuclear family modelcan entail mufualinterdependence and equal exchange within ahousehold. Land reforms have encouraged astructure in which women's dependence upon menis emphasized. This one'sided and ideologically-construed dependence (which masks men's actualdependence upon women) may be one of themost important indirect effects of land reform,and one with far-reaching, ambiguousconsequences for women.

CONCLUSION

What can be said about the general relationshipbetween gender, production and consumption inland reform programs?

Here I consider briefly, some general themes withregard to gender and consumption' As noted,most early writers on consumption/consumerismregarded the phenomenon as a symptom of massmanipulation. Betty Friedan, in her analysis ofthe "housewife-consumer," continued this line ofanalysis. As Nava (1990) points out, it was forless well-known writers such as E. Willis toforeshadow later work in cultural studies. Willis(1974) questioned the prevailing view of womenconsumers as cultural dupes or victims, andemphasized the rational and rewarding aspect ofconsumerism. Both views are "correct"and

contain insights.

I return to some of the questions posed earlier:

To what extent can Production andconsumption be viewed seParatelY?

Should one sphere be privileged over theother?

Does a perspective taking gender andconsumption into account, have any relevancefor "Third"/'Fourth'World studies?

Would such a perspective be altered throughfeminist input?

Lastly, is the relation of women to the sphereof consumption different in theunderdeveloped and advanced capitalistworlds?

Such a small study as this is cannot answer suchbroad questions. I pose them to draw tentativeconclusions from the cases discussed and to

indicate possible directions for further research.

Concerning the question of the separateness ofproduction and consumption: the examples cited

in this paper indicate the difficulty in making a

sharp distinction between the two spheres. Inmany cas€s (examples?) these overlap so that it is

not always clear what is "production" or"consumption." However, I would argue againstcollapsing the two conc€pts completely. Unlike inadvanced capitalist societies, where productiveactMties tend to be hidden (they may, in any case,

take place in Third World enclaves...) so that only

142

"consumption'ls visible, in rural sectors of

underdeveloped countries production is highly

visible. Althoughl have used Warde's concept of

the production/consumption cycle whichemphasizes overlapping processes, it also

acknowledges differences between the two.

The example of land reform underlines thepredominance of "production'bver "consumption."

The fact that women have lost, or at least havenot secured land rights is fundamental to theirwhole economic position. Where land reform hasbeen along collective lines, women's status is oftenimproved.

At the same time, women in "individual family"schernes have made gains (usually, alongsiderather than instead of men) through rnanyreforms, and these have tended to be in thesphere of consumption: improved serviceprovision, higfier incomes, food security and (attimes) more consumer goods. For some women,such gains outweigh any negative effects.Conversely, most collective schemes have at bestimproved food security but have not usuallyimproved incomes; for this reason many womenbecome dissatisfied. These outcomes emphasizethe relevance of taking consumption into accountwhen analyzing the Third World process€s.

What about the relevance of feminist perspectivesto the relationship(s) between land reforms,consumption and production? Here I concentrateon the example of sub-Saharan Africa, as itcontrasts with the "West." In the West, womendo, of course, take part in productive activity, bothinside and outside the home. Nevertheless, theyare stereotyped as consumers and shopping mayform a prime outlet - a sphere of "enjoyment"- aswell as an important element of gender identities.In fact there is something curious about this, sincein the West as in most societies women arepoorer than men and so have less to "consume."

Nevertheless, the stereotype remains. In much ofAfrica, women are clearly producers and are themain farmers. However, they are not in generalseen as, or allowed to be, consumers. This ispartly because women are conc€ntrated in ruralareas and there is more access to consumer goodsin cities. But there is a more important reason:consumption is seen as a male prerogative.

Women are not debarred from purchasing goodsbut decision making over consumption usuallyrests with men. Women are often seeR as entitledonly to a narrow range of goods (cookingpots,clothing, blankets'perhaps a small number oflivestock) on their own accounts. Hence genderstruggles at both the social and personal levels aremore likely to take place around the sphere ofconsumption - e.g., women's right to spendmoney, to make decisions, and to be seen topossess consumer goods.

This situation may account for the fact thatwomen are often satisfied, on balance, withindividual family schemes even though they are insome respects detrimental to them. Betterconsumption"prospects"seem to be a trade-offfor lessened autonomy over production anddecision-making, including that over the sphere ofconsumption itself. This may herald greatersimilarity to the contradictoryposition of Westernwomen: consumerism both enhances women'spower and influences and acts as an instrument ofsubordination (Nava, L990).

Perhaps there is room for a note of optimism.Bauman (1983) traces the history of consumptionas a compensation gained [by the lower-middleand working classes] for the encroachment ofdisciplinary power [of the state, at the point ofproduction]. If any aspects of consumerisrn in thewest can be seen as attempting to resist suchdisciplinary power (e.g., through the creation ofcounter-cultures, not all of which becomeincorporated), perhaps in some Third Worldcontexts women's desire to become "consumers"is

one attempt to resist PatriarchY.

Women have lost out, in general, through landreforml and this remains fundamental' However,using a gender perspective highlights theimportance of the sphere of consumption,wherewomen have sometimes made gains. Taking bothgender and consumption seriously yields a morenuancedview ofland reform processes, and onewhich gives a fuller picture of contradictoryoutcomes.

143

i:ii

NOTES

I. It should be said that I do not make theseobsenations as part of any general argumentagainst land reform. Writers such as El-Ghonemy (1990) have demonstrated that landreforms are beneficial to peasant producers intprms of increasing equify, even where it is alsotrue that not all sectors of peasantry arebenefitted equally in class terms (that is,where, better-off peasants benefitdisproportionately). This observation iscommonly made in terms of peasant classdifferentiation. However, "equity"must alsoapply to women, who are increasinglyrecognized as important (or the main)producers even in "non-femalefarming" areas.

If land reforms are not to further increasegender differentiation among peasantries andother rural producers, women's needs must bemore specifically taken into account by policy'makers. In particular, without land titles heldon the same basis as men (whatever that basisis: outright ownership; titles or deeds;allocation through communal authorities),women are likely to become increasinglysubordinated to men in the name of household"unity."

REFERENCES

Allaghi, H. (1984), "RuralWomen in aResettlement Project: The Case of the LibyanArab Jamhiriya" International Labor Office, RuralDevelopment and Women in Africa, InternafionalLabor Office, Geneva.

Appadurai, A. (1986), The Social LW of Things,Cambridge University Press, Cambridge.

Barraclough, S. (1991),,,4n End to Hunger? Zed,London.

Bauman, Z. (1983), "Industrialism, Consumerismand Power," Theory, Culure and Society, vol. 1,no .3 .

Bourdieu, P. (1984), Dlsanction: A Social Citiqueof the ludgernent af Tastes, Routledge and KeganPaul, London.

Bush, R. and Cliffe, L. (1984), "LandReform orTransformation in Zimbabwe," Review of AfricanPo litic al Ec onomy, 29 :7 7 -9 4.

Chambers, R. (1969), Settlement Schemes inTropical Africa, Routledge and Kegan Paul,London.

Colson, E. (1960), The Social Oryanization af theGwembe Tonga: vol. 1: The HumanConsequences of Reseftlemenf, University ofZambia Press, Lusaka.

Conti, A. (1979), "CapitalistOrganization ofProduction through non-Capitalist Relations:Women's Role in a Pilot Resettlement in UpperYolta," Review of Afican Political Econonry 15-16:75-92.

Davison, J. (1988), "Land Redistribution inMozambique and its Effects upon Women'sCollective Production: Case Studies from SofalaProvince," Davison (ed) Agriculture, Women andLand: The, frican Expeience, Westview, Boulder,Colorado.

Deere, C.D. (1985), "RuralWomen and StatePolicy: The Latin American Reform Experience,"World Development, 13:10337 -53,

Deere, C.D. (1986), "RuralWomen and AgrarianReform in Peru, Chile and Cuba" Nash, J. andSafa, H. (eds) ll/omen and Change in LatinAmeica, Bergin and Garvey, New York.

De Silva D. (1982), "Womenand SocialAdaptationin a Pioneer Mahaweli Settlement,"Wimalaharma, K. (ed) Land SettlementExpeiences in Si Lanka, Ministry of Lands,Colombo.

Dixon-Mueller, R. (1985),Women's Wo* in ThirdWorld Agriculture: Concepts and Indicators,International Labor Office, Geneva.

Dwyer, D. and Bruce, J. (eds) (1988),/ HomeDivided, Stanford University Press, Sanford,California.

]l

IM

El-Ghonemy, M.R' (1990), The Political Economy

of Rwal Pover$: The Case for Land Reform,

Routledge, London'

Fapohunda, E. (1987), "The Nuclear Household

Uoaet in Nigeria: Public and Private Sector

?olicy," Development and Change,vol. 18, no.

2:281-94.

Friedan, B. (1965), The Feminine Mystique'

Penguin, Harmondsworth.

Galbraith, J.K. (1974), "Howthe Economy Hangs

on her APron Strings,"Ms., MaY.

Geza, S. (1986), "The Role of Resettlement in the

Social Development of Zimbabwe," Joumal of

Soeial Development in Africa,l (1)35'a2.

Goody, l. (197 6), hoduction and Reproduction,Cambridge: Cambridge University Press.

Hall, S. and Jameson, F. (1990), "Clingingto theWreckage: A Conversation," Marxism Today,September.

Hanger, J. and Moris, J. (L973),Mwea: AnIrigated Rice Settlement in Kenya, Welfforum,Munich.

Hoggart, R. (1957), The Uses of Literaq, Penguin,Harmondsworth.

Huber, S. (1991), "ThePeruvian Land Reformand the Position of Rural Women,"/oumal ofLegal Pluralism and Unofficial Law, numbers 30-31.

Jacobs, S. (1989a), "Zimbabwe: State, Class andGendered Models of Land Resettlement,"Parpart,J. and Staudt, K. (eds) Women and the State inAfrica, L. Reinner, Boulder.

Jacobs, S. (1989b), "GenderDivisions and LandResettlement in Zimbabwe," unpublished D.Phil.Thesis, Institute of Development Studies at theUniversity of Sussex, Falmer, Brighton.

Jacobs, S. (1991), Production, Consumption andGender Conflicts in Rural Zimbabwe," Genderand Consumer Behavior Conference hoceedings,

Janeen Arnold Costa, ed. Salt Lake City, UT:University of Utah Printing Services, 253-262'.

Jameson, F. (1984), "Postmodernism,or theCultural Logic of Late Capitalism," New LeftReview, No. 146, July/August.

Kellner, D. (1983), "CulturalTheory,

Commodities and the Consumer Society," Theory,Culture and Society, vol. 1, no.3:66-83.

Lund, R. (1978),Prospeity to Mahaweli: A Surveyon Wornen's Wo*ing and Living Conditions in aSettlement Area, People's Bank Research Division,Colombo, Sri Lanka.

Marcuse, H. (1964), One Dimensional Man.

McCall, M. (1987), "CarryingHeavier Burdens butCarrying kss Weight: Some Implications ofVillagization for Women in Tanzania" Mommsen,J. and Townsend,J. (eds) Ifte Geography ofGendcr in the Third World, Hutchinson, London'

Mogadam, V. (1992), "Patriarchy and the Politicsof Gender in Modernizing Societies: Iran,Pakistan and Afghani stan," Intemational Sociologl,vol.7, no. 1.

Mukta, P. (L992), "Accommodation,Resistance

and Cultural Life in India: CapitalistTransformations,"School of Arts Forum,Staffordshire University, Stoke-on-Trent,December.

Nava, M. (1987), "Consumerismand itsContradictio ns," Cultural Studies, no. 2.

Nava, M. (1990), "Consumerism Reconsidered:Buying and Power,"paper to British SociologicalAssociation Annual Conference, University ofSurrey, Guildford, unpublished.

Palmer, L (1985), Women's Roles and GenderDffirences in Development: The Nemow Case,Kumarian Press, W. Hartford, Connecticut.

Pankhurst,D. and Jacobs, S. (1988), "Land

Tenure, Agricultural Production and Women:The Case of Zimbabwe's Peasantry,"Davison, J.(ed) op.cit.

145

Rogers, B. (1981), "LandReform- the Solution or

the Problem?" Human Rights Quarterly, vol. 3, pt.

2:96-102.

Safilios-Rothschild, C. (1988)' "The Impact of

Agraian Reform on Men's and Women's

Intomes in Rural Honduras,"Dwyer and Bruce,

oP.cit.

Stacey, J. (1982), "People'sWar and the New

Democratic Patriarchy in China"""Ioumal of

Comparative Famity Studies, vol. 13, no. 3:255-7 6.

Tadesse, Z. (1982), "The Impact of Land Reformon Women: The Case of Ethiopi4"Beneria, L.(ed) Women and Developmenf, Praeger, NewYork.

Third World Guide, 1993 /4 (L992),Instituto delTercer Mundo, Montevideo.

Warde, A. (1990), "Introductionto the Sociologyof Consumption,"Sociologt, vol. 24, no. L.

Weinrich, A.ICH. (1979) Women and RacialDiscimination in Rhdesia, UNESCO, Paris.

Willis, E. (1970), "'Consumerism' and Women," L.Tanner (ed) Voices from Women's Liberation,Mentor.

146