Skip to navigation – Site map

HomeThemed issues20IssuePartible Houses


Partible Houses

Variants of Vicinage in Mozambique, Portugal and Brazil
João de Pina-Cabral


Vicinage is a word that emerged in Southern African ethnography in the 1960s to describe how, within a neighbourhood, some houses are more constitutionally linked with each other due to the residential and kinship history of the people who inhabit them (that is, due to the continued identities that the residents transport). In this paper, I treat houses as being in ontogeny, a constant process of self-constitution. Much as Marilyn Strathern argued for persons, houses are dividual in that their singularity comes about through an act of alliance, but they remain ever enmeshed within a set of co-presences that mean they are also partible, for their existence is ever dependent on the existence of other households in the vicinage. The essay focuses on three distinct types of vicinage, endeavouring to show what distinguishes them and what they have in common: among the Chopi of southern Mozambique, as studied by David Webster in the 1960s; in northwestern Iberia, both in the countryside and in urban contexts, as studied by myself in the 1980s; and among the periurban populations of southern Bahia with special reference to the work of Marcellin in Cachoeira in the early 2000s.

Top of page

Full text

This text was originally drafted for the Roundtable “Dinâmicas da Vicinalidade: parentesco, casa e mobilidade”, Anpocs 2011. I am grateful to Simone Frangella, Omar Ribeiro Thomaz, and Emília Pietrafesa de Godoi for their company in organizing this event. The paper is a product of the project “A trama territorial: pertença, mobilidade e trabalho no Brasil contemporâneo” (FCT/PTDC/CS-ANT/102957/2008), for which I thank the Portuguese Foundation for Science and Technology. I also want to thank Consuelo Araos, Benoît De L’Etoile for their encouragement and lively interchanges at the time I was Visiting Professor at the École Normale Supérieure in Paris, November/December 2016.

Effective neighbourhood will ever constitute a link among men

Émile Durkheim

To cohabit the earth is prior to any possible community or nation or neighbourhood

Judith Butler, Parting Ways, p. 2767

1Vicinage is a word that I brought with me from my training as an anthropologist in southern Africa and which, at the end of the 1980s, I unearthed in order to describe familial proximity among the urban bourgeoisie of Oporto (northern Portugal) (Pina-Cabral 2003). There, the existence of a conjugal household (casa, lit. house) was deeply enmeshed within a network of other conjugal households residing close by. The couple that formed each of these households contained a person (most often the wife) who, at an earlier moment of her life, had shared membership of a household of the previous generation with core members of the other households in the vicinage. Note that I am avoiding defining them as kinsmen (parents, children, siblings, cousins, aunts/nieces, affines, etc.), for that they were, of course, but the putty that transformed these vicinages into such important spaces of personal ontogeny was not “kinship”, “residence”, or “alliance” as such, but rather it was a personal experience of pastness (cf. Martins 1974) that condensed various modes of co-presence (see Sahlins 2011a and 2011b).

2I called this pastness continued identity, for what mattered was that these people carried with them a certainty of co-presence that emerged from a history of familial cohabitation and that was written into the history of their own constitution as “familial persons” (their personal ontogeny, see Toren 2012). Thus, the constitutive cohabitation that took place within the space of outwardly separate conjugal households (what sociologists wrongly persist in calling “nuclear families”) was in fact predicated on earlier moments of cohabitation and, in turn, led to dense networks of interhousehold relating.

3To that extent, these bourgeois casas were partible houses much in the same way as persons can be said to be “partible” (cf. Pina-Cabral 2013, Strathern 1988). They were dividual in that their singularity came about through an act of alliance, but they remained ever enmeshed within a set of co-presences that meant that they were also partible, for their existence was ever dependent on the existence of the other households of their vicinage and on the pastness (the continued identities) that the members of all of these households jointly transported. In time, again through alliance, this dialectic of singularity and partibility would prolong itself. Note, however, that marriages were often internal to the vicinage and, in any case, socio-economic homogamy was and remains a highly valued ideal (almost a norm) within Oporto bourgeois circles. Vicinages themselves, therefore, often came to overlap or to merge.

4The dictionary definition that best suits this meaning of vicinage is “the fact of being or living close to another or others” (OED). But, in fact the word is useful because it does much the same job by relation to “relatedness” (cf. Carsten 2000) as the word “kindred” once did to filiation and descent: that is, it allows us to describe familial sociality as a process of fuzzy constitution, rather than having to start from defined groups with well determined boundaries. The word’s history as an anthropological concept dates back to the sources of the Anglophone tradition of southern African ethnography and, most directly, to W.D. Hammond-Tooke’s description of beer drinking among the Xhosa Mpondomise of the Eastern Cape in South Africa (1963).

5Among the Xhosa Mpondomise the constitutive core of the political system was a kind of local hospitality group the actual functioning of which could in no way be described in terms of descent. Patrilineal descent was central to their social life, of course, but the principles at play in the constitution of these groupings went well beyond it. At the time (the mid-1960s), the relevant agenda was a felt need to progress beyond so-called “descent theory” and Hammond-Tooke wanted to show that there were other principles of social coherence at play that allowed for a political complexity that could not simply be described in terms of patrilineages. What he observed was that the language of agnatic descent that people so readily brought forth in speech was “little more than an a posteriori rationalization of an accomplished fact.” (Webster 2009 [1976]: 76)

6David Webster – a colleague and student of Hammond-Tooke – used the concept of vicinage as a central analytical tool in his ethnography of the Chope of southern Mozambique in the 1970s. For him, a vicinage is “a neighbourhood group that includes various homesteads that are (usually) contiguous. Belonging to the vicinage depends on one’s loyalty to the other members or to the leader.” (2009: 90). More recently, however, in the work that I have been carrying out among fishing folk in the mangroves of coastal Bahia (Brazil, see Pina-Cabral and Silva 2013), I find that vicinage is a useful tool to describe the process of enchainment between households said to be “pulled out” (puxadas) of each other. This process is masterfully described in the ethnography that Louis Herns Marcelin produced about the casas of Afro-descendants in the Bahian city of Cachoeira (1996).

7In the present essay, I will endeavour to describe succinctly three types of vicinage. I will show that they point to very different principles both of family coherence and of narrating kinship links between persons.

The Chope chitiyana

8In most Western European languages, we normally refer to the main unit of cohabitation, where children are raised and people obtain their essential subsistence, by means of a metonymy with its building (casa, maison, house, haus). Such terms are analytically useful, since they facilitate mutual understanding (see Carsten and Hugh-Jones 1995), but they do transport with them a propensity for implicit ethnocentrism, in particular when the familial practices of the people that gave rise to these concepts are outside the comparative exercise (see Pina-Cabral 1989). Thus, we should start by making it clear that Chope modes of rural living, such as they still existed at the time of Webster’s fieldwork (1969-1976), were marked by a type of domestic environment that can only be described very approximately by words such as “house/household”, maison, or casa. The English word which Hammond-Tooke and Webster favoured to describe the elementary social unit in Southeast Africa was “homestead”, but the more common word in southern Africa to describe this kind of dwelling is an Afrikaans adaptation of the Portuguese word for cattle byre, kraal, as the central focus of the residential unit is indeed the place where cattle were kept.

9The Chope kraal (inti) is not unlike the muti of the neighbouring Tsongas that was so masterfully described by Henri-Alexandre Junod in those passages where he carries out the first ethnographic experiment with the concept of rite of passage (1962 [1912]: I, 280-327). This was also one of the first ethnographic instantiations of Spencer’s “organic analogy” and its impact on future anthropology, via the essays on the Tsonga mother’s brother and joking relationships that A.R. Radcliffe-Brown was going to publish in Structure and Function in Primitive Societies (1962), has largely been ignored by the historians of anthropology. Another superb description of how the muti/inti worked can be found in the first chapters of the autobiographical novel (Chitlangu, fils de chef) that the founder of Frelimo, Eduardo Mondlane produced in his youth under a pseudonym (see Khambane and Clerc 1946). The Southeast African kraal is essentially constituted by a cattle-pen and an ancestral tree around which are constructed a set of rounded adobe houses with large and complex straw roofs that may be moved from one location to another. These simple, mostly one-roomed huts are grouped inside a rounded palisade of thorn bushes that also encloses large ceremonial trees, granaries, a number of wooden sheds and, at its centre, the cattle byre. Junod’s description of the gendered use of the hut’s space is a lesson that, in many ways, projects forward to Bourdieu’s description of the Kabyle house (2013 [1972]).

10The homestead is inhabited by a man and his dependents, including one or more wives. In the latter case, each wife’s hut and granary constitutes a separate reproductive nucleus in the descent system. Each wife has a plot of land that the husband prepares for her and puts at her disposal. She feeds her children and contributes to the husband’s upkeep with her products. At a later moment in the cycle, if and when they move away from the father’s homestead with their new wives, the adult sons will found their own homesteads, often but not always close to the father’s.

11Here is where the process becomes interesting to our present concerns. The Chope are patrilinear in the sense that they grant primary importance to links of agnatic descent over other kinds of familial links. Nevertheless, other types of association co-exist with agnatic descent and come to compete with it. A man is free to choose where he will set up his homestead (that is, the vicinage within which he will live) according to four other principles of association: (i) matrilateral kinship; (ii) affinity; (iii) friendship (formal and non-formal, where the first type constitutes a very strong formal link among the Chope); and (iv) namesakes (see Pina-Cabral 2010 for a comparative study of this notion).

12In the latter case, some time soon after the person is born, people are given the name of another person who thus assumes rights and obligations towards the named child. More than half of the persons at the time of Webster’s fieldwork (57%, Webster 2009: 220) had been raised in the homestead of their namesake, which means that the majority of Chope persons transport with them throughout their life relations of co-presence with their namesake’s relatives that are rooted in very early and prolonged childhood cohabitation. As a consequence, at the moment of choosing a new domicile, many of them feel that their elective affinities lie closer to the namesake’s vicinage than to their father’s. Nothing prevents them from setting up their homesteads in the proximity of the namesake’s homestead.

13As it happens, in this as in so many other aspects, Chope society is very plastic and people’s freedom of choice very considerable. According to Webster’s sample, homestead heads change the location of their homestead at least 1.6 times in their lives. This means that a considerable number of Chope men decide to change their homestead and move to another vicinage (chitiyana) when they have already set up home independently and have had children. This decision is motivated by a number of factors that range from the simple desire to live closer to their friends to the search for better economic conditions or yet the search for a better situation in terms of political power.

14In fact, the vicinage (chitiyana)–a group of six or seven homesteads including around 10 adult men–is both the basic unit of land ownership and the minimal unit of the political system. The chitiyana comes into being when a big man (wahombe) manages to unite around him a series of clients. Most likely, the initial nucleus is constituted by a set of agnatic relatives (brothers, sons, or patrilateral nephews) but, as the Chope insist: “people are power.” So, big men seek to reinforce the agnatic nucleus of their vicinage with a series of other clients whose logic of association is more likely to be one of the other four kinds described above.

15The chitiyana, thus, is the site par excellence of the relations of proximity and distance that bring the political system into being. The land of the vicinage that the big man distributes among his clients is granted to him by the sub-chief who, in turn, receives territorial rights from the chief. The big man surrounds himself with people who love him. Yet it is precisely among such people (brothers, uncles, agnatic nephews, uterine relatives, affines, old friends, namesakes and even, in the case of famous healers or musicians, apprentices) that the factors of fission and political conflict emerge that ultimately convince people to move their homestead somewhere else in the course of their adult life.

16The capacity to cope with this possibility of movement is what explains the greater or lesser success of a Chope big man (wahombe). If to this we add the fact that there is a relatively high rate of divorce, a mode of living emerges in which the singular person (man or woman) has a considerable margin for negotiation as to where they wish to live and with whom. In the end, agnatic descent turns out to have had less structuring influence then it would appear from people’s own accounts of their solidarities. In such a society, where there was plenty of land available for everyone still in the 1970s, the power that each man managed to yield was completely dependent on his capacity to convince other people to stay close to him.

17The vicinage was the fundamental locus of social engagement because it was there that interpersonal relations of various kinds based on continued identities of a diverse nature were mobilized in order to produce the relations of formal power that ultimately structure the political system. That is how we have to interpret Webster’s insistence that this was “a society of friends”. For that same reason, as Omar Ribeiro Thomaz highlights in his introduction to Webster’s ethnography (2009: i-v), the supposedly “socialist” reforms introduced by Frelimo in the decade following on Webster’s fieldwork, obliged people who were not friends to live together. This gave rise to a kind of systemic enmity that contributed decisively towards setting up the atmosphere of distrust that fuelled the civil war, which ravaged the country throughout the 1980s.

North-western Iberian house and family

18Almost a millennium has passed since 1017, when a royal charter was first granted to the city of Léon by Alfonse V who, at the time, was the suzerain of the County of Portucalem, roughly corresponding to today’s northern Portugal. This was a decree that regulated municipal administration, determined the limits of the territory of the city, and established the privileges of its resident households. This document is at the source of a very long tradition of royal charters that, whilst establishing royal authority, also consolidated the customary rights of communities. While granting the right to use their lands to a territorially determined group of households, the foro or foral (charter) also granted rights of political citizenship and self-rule via municipal institutions. Households, as personified by their heads (the vecinos/vizinhos), were the elements of community, not persons. In the centuries that followed, such charters were promulgated by the Kings of Castille and Léon and also by the first three dynasties of Portuguese Kings, well beyond the limits of the Iberian Peninsula, around the world from Asia to the Americas (cf. Boxer 1965).

19This was a form of social organization where royal administration and religious administration imposed themselves on household members, granting them rights and privileges but only to the extent that they became members of parishes and municipalities. Now, the latter are territorially determined social units, where the logic of belonging is not measured by means of a language of descent but rather by a language of sedentariness, of autochthony. Further still, belonging to a municipality (concelho) or to its composing parishes is not a personal right but a domestic right. That is, the minimal unit of the system is the household and, as it was still the case in the 1970s when I carried out fieldwork in rural Alto Minho (NW Portugal), if you did not belong to a landed household you were not a “neighbour” (vizinho), that is, your rights of local citizenship were somehow diminished, much as you might have been born locally and have resided there your whole life. To that extent, extradomestic familial links did not transmit rights of local belonging but were seen as purely personal and informal. Political life was the life of the territorialized community: household (casa), parish (freguesia), municipality (concelho) (cf. Pina-Cabral 1986: 37-81). In short, a condition for such a system to operate was that descent should not be operational as a principle of granting political and territorial rights. Mixing politics with kinship came to be seen as unethical, the mark of “corruption” par excellence.

20This “Iberian sociocentrism”, as the ethnologist Julio Caro Baroja called it (1946), is especially marked in the northwest of the Peninsula (Galiza and Minho) where scattered settlement dominates. A “parish” (parroquia/freguesia, respectively) is a kind of community of neighbouring but not contiguous landed households disposed more or less around a centre constituted by a church, the vicar’s residence and a cemetery (cf. Pina-Cabral 1986, Lisón Tolosana 1983). The sedentariness of these Galaico-Portuguese parishes is so marked that parish names and parish borders survived unchanged for nearly a millennium to the nineteenth century bureaucratic reforms that ended the Ancient Regime. Often, as was the case in the parish where I resided in the Alto Minho in the 1970s, the remains of the founding origin of the Christian community–the third century Roman cemetery–were regularly being unearthed in the vicar’s garden plot. Contrary to East African examples, where villages moved at the death of their leader, as Junod so masterfully described, here the community is literally rooted in the land.

21As Julio Caro Baroja also noted, each parish is the conglomerating centre of a set of social relations that carries with itself strong endogamic tendencies: local group endogamy remains a marked characteristic of the region to this day. (For comparative studies of the issue of marital differentiation in the Iberian Peninsula, see Bestard 1998; Pina-Cabral 1991; Rowland 2011.) Indeed, in the two parishes I studied the percentage of marriages that took place within the parish sub-units (the hamlets that are called lugares) was in the order of 25% for hamlets sized on average around 30 households. What this means is that nearly all the neighbours were close relatives and were quite well aware of this, even though they preferred to describe their relations in terms of neighbourliness. Marriages often occur within the bounds of Church imposed prohibitions. In spite of the fact that there is a whole folklore concerning the evils of endogamy, the demand for canonical dispensations has always been very high and the Church commonly yields to them. A certain patrimonial continuity thus emerges across generations, in spite of the practice of what Goody would have called diverging devolution—that is, divisible, bilateral inheritance (see Goody 1969 and Caro Baroja 1946). Diverging devolution, of course, is the norm in the NW of the Peninsula but not elsewhere in the northern and Pyrenean regions, where more linear modes of household reproduction are historically dominant. (For a comparative analysis of the structures of household reproduction in Iberia, see Pina-Cabral 1991.)

22In contrast to the Chope case–where a language of agnatic descent screened and validated rhetorically a complex set of modes of social aggregation–in the Alto Minho, there is a propensity to transform kinship association into neighbourly association (vizinhança). Effective community relations are presented as relations between neighbours, because they are relations between households and not between individual persons. Mutual help in moments of crisis and spontaneousnacts of amity are preferentially formulated in terms of neighbourly relations rather than kinship. For example, during funerary rites, whilst the bereaved are defined as those that were raised in the house (whether or not they still live there), help in organizing funerary ceremonies is strictly a matter of neighbourliness (and this is symbolised by the opening of all doors as soon as someone dies–see Pina-Cabral, 1986 214-225). Another example can be drawn from witchcraft beliefs: the common way to describe those who may attack one is “the enemy behind the door” (o inimigo detrás da porta)–that is, close neighbours, whether or not they are also close relatives (ibid.: 175-185). I conclude that the tendency in north-western Iberia for kinsmen to be described as neighbours and for mutual help to be formulated in terms of neighbourly cooperation is fully part of Caro Baroja’s “sociocentric” tradition, the roots of which are so distant in time and the political implications of which so deeply set in Iberian institutional history.

23In such a system, the logic of neighbourhood and the logic of kinship tend towards isomorphism, thus corroborating the ideology of sedentariness that has characterised historically the broader political system since the days of Alfonse V. Unlike what happens in the Basque country or Catalonia, in northwestern Iberia houses can be divided, but their survival from generation to generation is strongly valued and is ritually celebrated in the parish cemetery and at Easter time. As I described at length in Sons of Adam, Daughters of Eve (1986), ever since the end of the Black Plague in the fourteenth century, the whole system has depended for its functioning, on the one hand, in the symbolic exclusion of those who have no land and, on the other hand, on the shedding off at each generation of excess population through migration.

24When, at the end of the 1980’s, I shifted my ethnographic attention to the study of bourgeois families in the large city of Oporto, I was convinced that I would not encounter any of that, as this modern city presented itself as an impersonal environment, where independent “nuclear families” were the norm. The dominant notion of landed household that I had worked with in the rural area was necessarily absent and local endogamy seemed to be out of the question in a big city. Surprisingly, however, as the study evolved, I came to see that the family histories of bourgeois households that I was collecting suggested that they were fully part of the same long-term history (Pina-Cabral 1991, 2003). On the one hand, socio-professional homogamy was strongly enforced and its breach strongly penalised and, on the other hand, as suggested above, I observed the centrality of a process that I could best describe using the concept of vicinage. That is, people did live in conjugal homes (apartments or small houses) spread throughout the city, but their residential choices were strongly marked by a search for elective affinities. They made marked and conscious efforts to find residence in close proximity to those from whom they could expect mutual help, particularly in the matter of raising children. Therefore, as in Chopeland, there were no delimited spaces (neighbourhoods) were all households were related; rather, there was a very marked tendency for the emergence of greater densities of related households (vicinages) in residential quarters that were also open to unrelated people. It has been suggested that (Vanda Aparecida da Silva, personal communication) a structural feature of this nature might also be observable in large Brazilian cities, a hypothesis that is certainly deserving of further test.

25In the Chope case, the vicinage was structured by the centripetal presence of the big man (the wahombe), whose political aspirations ensured its cohesion and was characteristically narrated in the language of agnatic descent. In Oporto the centripetal force was the aggregative role of the “mother-in-law” (sogra)–that is, women with married daughters and granddaughters. I had already observed this strong tendency for favouring uxorilateral links in the rural region and was now surprised to find that it was equally evident among the urban bourgeoisie. This was the case even though the discourse of male hegemony was ideologically far more prevalent among the bourgeois families than in the countryside. Uxorilateral vicinages constituted the backbone of family life and played a dominant role in personal ontogeny; they were the space where early personal ontogeny occurred. The small conjugal apartments where children in theory resided were hardly sufficient to describe the histories of cohabitation of people who, from their earliest days in life, circulated daily among a set of uxorilaterally related homes.

26In this urban environment, the dominant centrifugal factor that prevented these vicinages from extending beyond around five to six households was the vagaries of class belonging. When a relative ascended or descended in class condition in a marked way, a tendency emerged to start another nucleus of related homes elsewhere. The other relatives whose class condition evolved in a similar fashion tended towards moving closer to the new nucleus, particularly but not exclusively if they were uxorilaterally related. Strong pressure was exerted by the previous matriarch to prevent the movement, but this eventually failed due to the need to establish appropriate contexts for the socio-professional homogamy of the children of those who had been promoted. (The fact that crèches and secondary schools are predominantly private and very expensive, and they are the central loci of class identity, only reinforces this tendency.)

27I never encountered explicit formulations either of the existence of vicinages or of the dynamic of class-belonging that they reflected because that would have countered two central ideological principles in this society: firstly, the notion that kinship links must be stronger than economic interest; and secondly, the language of male hegemony. If the prevalent patriarchal ideology was not challenged by the uxorilateral preference, it was because there was no principle of descent.

28The household is the only corporate kinship unit and its reproductive nucleus (contrary to what happened among the Chope, where a man could have various wives) was unitary, since Tridentine notions of marriage dominate: there is a strong supposition that marriage is exclusive and stable between two persons and that it is immoral for two couples of the same generation to reside within the same household.

29The bourgeois home is a household where the man is both the definitional bread-winner and the unchallenged head, but always in collaboration with a single wife. Nevertheless, the moment that I tried to look at these homes from the perspective of family history I found out that, much like with the language of descent in East Africa, the patriarchal discourse lost its relevance and the centrality of the language of the casa turned out to screen the fact that the process of cohabitation leading to personal ontogeny took place within uxorilateral vicinages.

30Further still, the growing matrimonial instability that accompanied the growth in divorce rates that occurred in the 1970s and 1980s turned out to have the effect of reinforcing the centrality of these uxorilateral vicinages. The homes of newly divorced women (often middle-class working women) were only viable due to their positioning within strongly supportive uxorilateral vicinages. Divorced men would typically reposition themselves by being taken into a new vicinage. The way in which, in situations of matrimonial rupture, the Portuguese law and the courts favour the rights of maternity over those of paternity turns out to protect uxorilateral vicinages against the possible disruption of conjugal separation and divorce (see Pina-Cabral 2003).

Porous houses in Bahia

31In the mid 2000s, I moved to the study of fishing folk in the mangroves of coastal Bahia – where there is a blending of historical influences due to Portuguese colonization, African slavery, and Ameridian ancestry. There, I encountered a form of family life where descent is also not a relevant principle. Bahian households were more like those of the landless poor of north-western Iberia–farmhands and fishing folk (see Cole 1991)–than like the landed peasant houses I had described in Sons of Adam, Daughters of Eve (1986). The distinctive aspects of Bahian family life that I will summarily identify below are very similar to those described by Louis Herns Marcelin in his doctoral dissertation on the Recôncavo town of Cachoeira (1996). (For a more detailed account, see Pina-Cabral and Silva 2013.)

32Firstly, these people’s lives are economically very precarious, which means that young adults often find it difficult to meet the demands of a stable domestic environment. At the same time, personal freedom is a strongly emphasized value. The personal mobility of young adults, therefore, presents itself almost as a moral imperative. People are strongly expected to virar-se (lit. turn oneself around, work out solutions on one’s own). As a result, the tone of family life is set more by interpersonal relations than by the collective duties of household belonging. Household borders being permeable, each household’s existence is predicated on the existence of the households that surround it.

33This is reinforced by a regime of land use where the rights of property over the physical house and the land that is associated to it always remain somewhat ambiguous. In this region of Brazil, access to land is mostly based on customary possession rights (posse) rather than property rights and this gives rise to considerable uncertainty over long term residence (cf. Pina-Cabral e Silva 2013). Once we associate this to a systemic lack of capital, leading to a dependence on complex systems of local microcredit, we see that there are very few rights or interests that are transmitted through inheritance. This means that the continuity between generations (what we might call the principle of linearity) is somewhat less important than intragenerational association (the principle of laterality). In the words of Marcelin, “to the actors themselves and in their everyday practice, intragenerational experience rather than intergenerational experience is what is perceived as the central symbol in the construction of relations between them.” (1996: 190-1)

34Secondly, therefore, there is a strongly perceived continuity between related households. The houses belonging to siblings are literally “pulled out” (puxadas) from the parental household (they are built behind or above the parental house). Since conjugal relations tend to be unstable during the first period of adult life when women are more fertile, relations with siblings, half-siblings and residentially close cousins residing nearby assume enormous importance. To the residents of Cachoeira, Marcelin notes, “the house [...] does not come to mind as an absolutely autonomous unit or as circumscribed in a delimited space: the house [household] is inseparable from the network of people and houses within which it finds its definition.” (1996: 99)

35Borders between households are permeable and the daily life of close relatives is one of constant interaction. The members of the casa (house, household) are not exclusively those who sleep there daily and one can be a member of more than one household. For example, a man can share a house with a woman, where their children cohabit and, at the same time, also be a nuclear member of his mother’s household, which he helps to sustain, construct, and repair and where his own children by an earlier partner or his sister’s children are being raised. At the same time, he might also claim rights of belonging to the house of his aunt or of his grandmother where, in turn, he was raised. To try to determine with precision who belongs to which house within a nucleus of related houses can be difficult and, most of all, it would be an error of judgement, since circulation and co-belonging are dominant principles. Again in Marcelin’s words, “the house [household] is a place that is structured by the convergence and sedimentation of family relations in perpetual construction within a configuration of houses.” (1996: 126)

36As a matter of fact, the way people are raised in infancy contributes towards this tendency to emphasize both personal mobility and belonging to multiple domestic contexts. Susana de Matos Viegas (2007) has called our attention to the way in which, in this region, the raising/nurturing of children by people other than their parents (criar, criação) is more than a merely convenient option but constitutes a structural inevitability. Marcelin too stresses that to leave one’s child to a relative in another house is a “right” felt by all the actors present within a vicinage (that he calls a “configuration of houses”, 1996: 138).

37In Bahia, therefore, we can point to two central processes concerning the way in which cohabitation functions in the constitution of persons that mark a strong difference with the other ethnographic contexts referred above. The first of these is the occurrence of a lack of synchronization between the cycle of fertility and the cycle of domestic reproduction. To put it simply, people tend to have children during the first part of their adult lives, at a time when they lack the social and economic means to create viable domestic environments. The life of young adults is marked by an impetus for mobility in search of socio-professional success, which is not compatible with residential stability. But, at the same time, they do not lack the means to raise their children, for these are taken up to be raised by other people within their parental vicinage.

38Only later on in life, when people have come to discover the limits of their mobility and have accumulated some capital or job security, do they feel they can set up home in a more permanent fashion. The entry into the cycle of household reproduction, therefore, tends to occur later on in one’s adult life. It is wrong to think that Bahian houses do not include men as fathers/homeowners. This is one of the equivoques associated to the more incompetent approaches to the debate on matrifocality. (For a singularly ill-informed example, see Blackwood 2005.) All attentive observers have noted that, among the people of low income in Bahia, the consolidation of conjugal relations of homeownership tend to occur later on in life and the formalization of conjugality is seen as a recognition of, not a precondition for the occurrence of that process (e.g. Marcelin 1996: 266). Typically the actual building that is the hypostatisation of the house is made, repaired and sustained by men, either in their quality of husband/father or in their quality as son/uncle.

39The second process is that of the constitution of vicinages. These must not be confused with neighbourhoods, as Marcelin shows by the use of plans where the logic of enchainment between houses that I described above is clearly demonstrated (1996: 140-1). I have encountered a similar situation among mangrove dwellers (cf. Pina-Cabral and Silva 2013). Much like in the case of the rural Chope or of the bourgeois families of Oporto, these vicinages do not constitute areas wholly occupied by a set of relatives–which would be a neighbourhood, a quarter or a compound. Rather, they are zones of residential dwelling where there is a greater density of occupation by houses of related people (which are also related houses). The borders of the vicinage are fuzzy and mobile but they are characteristically rooted around one or more central households of the older generation. The conjugal instability of the younger adults gives rise to frequent reconfigurations, accompanying and reflecting the constant process of house construction, repair and renewal. Poor people’s houses in Bahia are easy to build and they are often made of adobe, which needs to be renewed every ten years or so. They are covered with tiles or zinc plates that are routinely stored and re-used. Changes in use are easily accommodated. As Marcelin again concurs, “To trace the borders of ‘domestic groups’ within a configuration of houses [a vicinage] would be a purely futile task due to the mobility of the actors. The diverse and dynamic circuits that are constructed among the units of residence would derail any hope of drawing a static portrait of these units.” (1996: 130)

40The pessoal or galera (generally referring to a group of friends) are the ambivalent collective nouns that describe those who belong to a vicinage. This word, describing the strong bonds of interdependence that arose among the convicts that formed a work group (thus, galera, lit. galley), is particularly evocative of the sharing of a fateful condition, leading to active, non-compulsive mutuality (see Pina-Cabral 2012). Among the members of the group there is intense interaction and constant circulation of food, services, and microcredit but, contrary to what happens in the Alto Minho, these exchanges are always accounted for in terms of singular persons, never in terms of debts between households or between groups of kinsmen. The process of association remains ever open-ended and not everyone has a similar relation to everyone else. In fact, many of the more permanent conjugal relations that we observed were between people who, after a history of personal and conjugal mobility, had come back together within their vicinage of orientation, so to speak. A great importance is given to the history of friendship and everyone’s relation to everyone else is constantly being reassessed in terms of “consideration”, that is, the marks that one gives of being attentive to another’s emotional needs.

41The houses of people of low income in Bahia are not independent of the vicinages to which they belong and people’s membership of houses is related to their own histories of interpersonal relations. As among the Chope or among the bourgeois of Oporto, vicinage must be seen as a broader context of cohabitation that comes to define the meaning of domestic cohabitation.

42If there are similarities between these four examples, however, divergences can also be identified. The different modes of structuring of vicinages identified above can be seen to be related to distinct ways of connecting familial life to the broader realm of political relations, on the one hand, and to conjugality, on the other.

Conclusion: diverging vicinage

43In East Africa, men can have more than one wife and their homesteads can integrate their sons’ wives. Therefore, the homestead (the inti) already includes within itself the seeds of its future dispersion. It is the wahombe’s task to construct out of his own varied and personal ties, the centripetal movement that aggregates homesteads, granting to them their political significance. Thus, at the level of the relations between vicinages (chitiyana), the language of agnatic descent becomes dominant. But not because it is the source of all the relevant links between men that give rise to the vicinage. Rather, because it operates as a screen (a favoured narrative tool) for the actual existence of diverse modes of relatedness that include uxorilateral relations and relations with people who are not kin. Thus, agnatic descent functions as a mediator between the complexities of personal life (where friendship, namesake relations, affinity, matrilaterality, etc. are all equally present) and political organization (namely the political role of the “chiefs”, the so-called “traditional authorities”).

44In north-western Iberia, to the contrary, the principle of residence (the casa) is dominant and a fundamental breach is drawn between family life and the primary levels of political organization, the parish (freguesia) and the municipality (concelho). The principle of descent is absent and kinship links are theoretically bilateral–in spite of a very strong leaning towards uxorilaterality. This means that supradomestic relations cannot be described as giving rise to any form of collective action and are subsumed under the highly polysemic category of “family”. The sense of unicity that characterises the house (casa), making it the elementary unit within a system of territorialized political organization, is dependent on a mode of marriage where sexual access, filiation and household management are fully shared by a couple that is conceived as being permanent (the famous Tridentine stabilitas, see Ariès 1962).

45Finally, in Bahia, the importance given to the mobility of the singular person, and his or her corresponding “independence”, leads to processes of asynchronicity between the cycle of personal fertility and the cycle of constitution of households. Households are seen as manifestations of relations between independent people and they do not survive the end of the conjugal relation. The vicinages that emerge among these co-dependent households (and which are central to local political and economic life) are described as optional associations, as “friendship” based on histories of “consideration” (Pina-Cabral and Silva 2013).

46Agnatic descent in East Africa; belonging to territorial communities in north-western Iberia; or interpersonal friendship in Bahia, respectively, are all local narrative moulds for family life, which are also constitutive principles both of personal ontogeny and of the broader political system. Furthermore, they are co-dependent on the operation of distinct modes of conjugality/alliance. Thus, by bringing together the three regions in a comparative approach that takes recourse to the “intermediate category” of vicinage, I aim to link together anthropological theory and ethnographic description at a moment in which the epistemological bases of the anthropological pursuit are undergoing a process of radical reconsideration (see Pina-Cabral 2017)

47As an intermediate category, vicinage aims to describe the way in which cohabitation as a process associated to ontogeny prolongs itself temporally to later moments of the relational cycle through modes of aggregation that rely on pastness (continued identities)–that is, the continuation in later moments of the implications of each person’s primordial experiences of constitutive intersubjectivity. Through co-presence, the personal ontogeny of children, siblings, and grandchildren transports in itself continuities with the ontogeny of parents, aunts and cousins and these continuities (these “participations”, see Pina-Cabral 2018) are a central mould for the forms of broader relatedness, bearing important implications for the local modes of political and economic life.

48In each of these regional cases, the preference for specific forms of describing relations (the existence of ideological moulds) limits the plurality and negotiated complexity of the relations between persons presenting them as relations of a specifically recognizable kind; this is what allows for a connection between domestic life and broader processes of political and economic association. Thus, in the case of the Chope there is a preference for transforming vicinal relatedness into agnatic relations; in the case of north-western Iberia, vicinal relatedness is moulded by a separation between family life and territorially demarcated political entities that is perceived as ethically compulsive; finally, in coastal Bahia, vicinal relatedness is presented as a mode of friendship based on histories of interpersonal “consideration”.

49To conclude, I would like to note that, incest prohibitions apart, relations of alliance appear to be affected by the operation of vicinage, which is not surprising if we see vicinages as aggregates of partible houses and partible persons. In the Chope case lineage exogamy is practiced, but Webster notes that vicinage endogamy occurs often between people who are related by one of the other four principles of association. In the Bahian case, conjugal instability allows for unions to be made and unmade, either reinforcing vicinage membership or easily integrating strangers. Finally, among the bourgeoisie of Oporto, socio-professional homogamy is almost compulsory, which gives rise to situations where vicinages are recreated as a result of processes of linkage between previously existent and spatially overlapping vicinages.

Top of page


Ariès, Philippe. 1962. Centuries of Childhood: A Social History of Family Life. Trans. Robert Baldick. New York: Vintage.

Bestard Camps, Joan. 1998. Parentesco y Modernidad. Madrid: Paidos.

Blackwood, Evelyn. 2005. “Wedding Bell Blues: Marriage, Missing Men, and Matrifocal Families”, American Ethnologist 32 (1), pp. 3-19.

Bourdieu, Pierre. 2013 [1972]. Outline of a Theory of Practice. Cambridge: CUP.

Boxer, Charles R. 1965. Portuguese Society in the Tropics: The Municipal Councils of Goa, Macao, Bahia, and Luanda 1510-1800, Madison, University of Wisconsin Press.

Caro Baroja, J. 1946. Los Pueblos de España: ensayo de etnologia. Barcelona, Barna (2ª ed., Madrid, Istmo, 1976).

Carsten, Janet. 2000. Cultures of Relatedness: New Approaches to the Study of Kinship (edited). Cambridge: CUP.

Carsten, Janet and S. Hugh-Jones (ed.s). 1995. About the House: Lévi-Strauss and Beyond. Cambridge: CUP.

Cole, Sally. 1991. Women of the Praia: Work and Lives in a Portuguese Coastal Community. Princeton, NJ: YUP

Cortés, Hernán. 2001. Letters from Mexico, trad. And annot. Anthony Pagden e Hon H. Eliott. Princeton, NJ: YUP.

Fonseca, Cláudia. 2000. Família, Fofoca e Honra: Etnografia de Relações de Gênero e Violência em Grupos Populares, Porto Alegre, Ed. da UFRGS.

Goody, Jack. 1969. “Inheritance, Property and Marriage in Africa and Eurasia” in Sociology 3 (1): 55-76.

Hammond-Tooke, W. D. 1963 Kinship, Locality, and Association: Hospitality Groups among the Cape Nguni” Ethnology 2 (3), pp. 302-319.

Junod, Henri-Alexandre. 1962 [1912, 1926/7] The Life of a South African Tribe. London, Macmillan.

Khambane, Chitlangou and André-Daniel Clerc. 1946. Chitlangou, fils de chef. Lausanne: Mission Suisse de L’Afrique du Sud.

Lisón Tolosana, Carmelo. 1971. Antropologia Cultural de Galicia. Madrid: Siglo XXI.

Lisón Tolosana, Carmelo. 1983. Belmonte de los Caballeros: Anthropology and History in na Aragonese Community. Princeton: PUP.

Marcelin, Louis Herns. 1996 A Invenção da Família Afro-Americana: Família, Parentesco e Domesticidade entre os Negros do Recôncavo da Bahia, Brasil. Doctoral thesisUniversidade Federal do Rio de Janeiro, PPGAS Museu Nacional, Rio de Janeiro.

Marcelin, Louis Herns. 1999 “A linguagem da casa entre os negros no Recôncavo Baiano”, Mana 5 (2), pp. 31-60.

Marques, Ana Cláudia. 2009 “Notas sobre fundadores e povoamento do Brasil: os casos do sertão pernambucano e do norte mato-grossense”, Travessia 65, pp. 59-70.

Martins, Herminio 1974. “Time and theory in sociology” in Approaches to Sociology (ed.) John Rex. London: Routledge & Kegan Paul.

Matos Viegas, Susana. 2007. Terra Calada. Rio de Janeiro, 7Letras.

Pina-Cabral, João. 1986. Sons of Adam, Daughters of Eve: The Peasant Worldview of the Alto Minho (NW Portugal). Oxford: Clarendon Press.

Pina-Cabral, João. 1989. “L'Héritage de Maine: L'érosion des categories d'analyse dans l'étude des phénomènes familiaux en Europe”. Ethnologie Française 19: 329-340.

Pina-Cabral, João. 1991 Os Contextos da Antropologia, Lisboa, Difel.

Pina-Cabral, João. 2003 O Homem na Família: Cinco Ensaios de Antropologia, Lisboa, Imprensa de Ciências Sociais.

Pina-Cabral, João. 2009 “The all-or-nothing syndrome and the human condition”, Social Analysis 53 (2), pp. 163-176.

Pina-Cabral, João. 2010 “Xará: Namesakes in Southern Mozambique and Bahia (Brazil)”, Ethnos 73 (3), pp. 323-345.

Pina-Cabral, João. 2013. “The two faces of mutuality: contemporary themes in anthropology”, Anthropological Quarterly 86 (1), pp. 257-274.

Pina-Cabral, João. 2017. World: An anthropological examination. Malinowski Monograph Prize. Chicago: HAU Books.

Pina-Cabral, João. 2018. “Modes of Particiption” Anthropological Theory, in print.

Pina-Cabral, João e Vanda Aparecida da Silva. 2013. São Paulo: Terceiro Nome.

Radcliffe-Brown, A.R. 1952 [1940]. Structure and Function in Primitive Society: Essays and Addresses. Glencoe, Ill.: The Free Press.

Rowland, Robert. 2011. “Família y Transición Demográfica” in Hácon Jimenéz, F. and Bestard Camps, Joan (ed.s), Famílias. História de la sociedad española (del final de la Edad Media a nuestros días), pp. 605-666. Madrid: Cátedra.

Sahlins, Marshall. 2011a. “What is kinship (part one).” Journal of the Royal Anthropological Institute (N.S.) 17 (1): 2–19.

Sahlins, Marshall. 2011b. “What is kinship (part two).” Journal of the Royal Anthropological Institute (N.S.) 17 (2): 227–42.

Strathern, Marilyn. 1988. The gender of the gift. Berkeley: University of

California Press.

Toren, Christina. 2012. “Imagining the world that warrants ourimagination—The revelation of ontogeny.” Cambridge Anthropology 30 (1): 64–79.

Webster, David J. 2009 A Sociedade Chope: Indíviduo e Aliança no Sul de Moçambique (1969-1976). Ed. João de Pina-Cabral. Lisbon: Imprensa de Ciências Sociais.

Van Gennep, Arnold. 1960 [1909] The Rites of Passage. London: Routledge & Kegan Paul.

Top of page


Electronic reference

João de Pina-Cabral, “Partible Houses”Articulo - Journal of Urban Research [Online], 20 | 2019, Online since 01 December 2019, connection on 27 January 2021. URL:; DOI:

Top of page

About the author

João de Pina-Cabral

School of Anthropology and Conservation, University of Kent.

Top of page


Creative Commons 3.0 – by-nc-nd, except for those images whose rights are reserved.

Top of page
  • Logo DOAJ - Directory of Open Access Journals
  • OpenEdition Journals
Search OpenEdition Search

You will be redirected to OpenEdition Search