Housing after World War II
With the easing of restrictions on the movement of indigenes into urban areas in the aftermath of World War II, migrants began to trickle steadily into Port Moresby. The colonial administration had no general long-term program for housing the newcomers, in what had become in many respects an Australian small town. Before the war, indigenous workers recruited from beyond the immediate villages had been accommodated in barrack and dormitory conditions. The only habitats in Port Moresby not fully controlled by European interests in that period were the local villages of the Motu people, the traditional coastal inhabitants of the area who had intermarried to a degree with the Koita, who lived inland. A few of the early postwar migrants managed to establish themselves in these villages, on the basis of old trading relationships and sometimes through intermarriage with the Motu-Koita. Apart from this, migrant housing in the decade after the war was principally of three types: workers’ ‘compounds’, established by government departments and companies such as the traders Burns Philp, Carpenters and Steamships; domestic quarters attached to European residences (commonly referred to by Europeans in kitchen pidgin as ‘boyhouses’); and so-called settlements.
The establishment of the first postwar settlements was achieved not by illicit squatting but by arrangement with local traditional landowners, as in the well-documented case of Rabia Camp at Kaugere, and occasionally with town authorities. The distinction between these two types of liaison was less clear in practice than on paper, since the Motu-Koita regarded the Port Moresby urban area as essentially their land, despite some areas having been acquired by the administration through early transactions classified as purchase by the latter. For example, a downtown settlement, Ranuguri, which is partially on government land and partially on customary land, began when migrants from the Gulf district were invited to move into buildings vacated by the army at the end of the war. According to some commentators, the arrangement was intended as a temporary measure by the administration while it built proper housing for employees. While this suggests that the administration itself told the settlers to move into the buildings, there is equivocation among the settlers themselves over who issued the invitation. I was told, for instance, by a long-term settler that the customary landowners invited his kin to take over the buildings. This interpretation, which ignores official classification of the piece of ground on which the buildings stood as government land, is predicated on the old trading relationships between Motu and Gulf people, and on the established Motuan practice of inviting individuals from their trading-partner groups to settle on their land.
The settlements grew in size and number as the first-comers were joined by near and distant kin from their places of origin. Their growth caused concern to traditional landholders and to the administration. For the Motu-Koita landholders, it became increasingly difficult to collect the rent-in-kind or continue the personal exchange relationships that had been integral in the negotiations with the first small groups of settlers, while the administration feared the prospect of overcrowding and the gradual transformation of small but tidy groups of makeshift houses into large and unruly conglomerates. When the settlements first appeared they were tolerated by the administration, since they provided a solution to the problem of housing workers during postwar reconstruction. But, after a few years, officials became fearful that a migrant underclass would develop. Some settlements were at one point denied water supplies and garbage and sanitary services in an attempt to discourage the migrants, according to Ryan, or to remove the settlers from land required for other purposes, according to Oram. Other strategies, such as plans by the Housing Commission (established in 1968) to resettle the occupants in better-organised cheap housing estates, came to only partial fruition.
By the 1970s, the administration had accepted the inevitable presence of settlements and had moved from attempts to discourage them to various assistance schemes, a transition that mirrored similar changes in government policies elsewhere in the Third World. In 1973, the flow of people into town was beyond the capacity of the Housing Commission’s resettlement scheme and new self-help policies were introduced. Some existing settlements were provided with paved footpaths and basic services such as water taps, and blocks of land were allocated on selected sites around the town on which people could build their own shelter (theoretically in accordance with building regulations). The scheme was elaborated to include loans to enable tenants to build adequate housing for themselves. One product of the combination of the Housing Commission’s resettlement plan and the self-help policies was the development of Morata suburb on the edge of the Port Moresby town area, a planned operation begun in 1971, which involved rental and self-help housing. In addition, community welfare groups attempted to assist particular settlement populations in their attempts to get basic services such as water supplies and to develop handcrafts and other income-generating activities.
In retrospect, the attempts by the colonial administration to deal with the continual arrival of migrants through the development of a variety of habitats in the late colonial period, documented by Oram, can be seen as a process of piecemeal reaction, rather than systematic planning. As Oram has pointed out, the distribution of housing areas in Port Moresby by the end of the colonial period did not conform to any recognisable spatial theory. There were ‘high-covenant’ housing areas occupied almost exclusively by Europeans, low-covenant planned estates such as the ‘partially integrated’ suburb of Hohola with a few Europeans but mostly housing Papua New Guineans, cheaper rental housing, self-help (‘no-covenant’) housing areas, the original settlements, company compounds and domestic quarters. While some cheap indigenous housing areas established by the administration were deliberately sited away from high-covenant European residential areas, they were not hidden but lined major roads and were not far distant from European housing, while some workers’ compounds and all domestic quarters were integrated into European-dominated areas.
The difference between high-covenant housing for Europeans and low-covenant and other types of housing for Papua New Guineans in the late colonial era is conventionally related in the literature to other racially segregatory practices of the period. But, importantly, the variety of indigenous urban housing itself engendered another type of distinction. There was a perceptual dichotomisation by Europeans of migrant Papua New Guinean urban habitats based on a concatenation of the spatial relationship between European and indigenous housing and the degree of orderliness of the latter. The low-covenant houses, company compounds and domestic quarters were reasonably tidy in appearance, and their comparative physical proximity to European housing implied the relative social proximity of their inhabitants, as workmates, subordinates or domestic servants. This was not necessarily manifest in friendly relations; in fact, discriminatory attitudes and cultural misunderstanding were common and regular occupational contact and interaction did not guarantee friendly socialising outside working hours. But the image of social proximity, however limited, was in contrast with that engendered for Europeans by the migrant settlements on customary and government land and the self-help housing areas, a little more removed from the elite European housing estates and rarely entered by whites. The initial small settlements that appeared in the immediate postwar period had been neatly laid out, but they took on an increasingly untidy appearance as they grew. Similarly, in the self-help housing areas, an unruly impression was created by the improvised nature of houses made from whatever material settlers could obtain and clustered asymmetrically on land blocks that had originally been surveyor-planned. A social researcher of the period, who bemoaned the ‘appalling symmetry’ of the rows of neat low-covenant houses, and preferred the creative approach of self-help housing areas, conceded of the latter that ‘To European eyes the results may not be as aesthetically pleasing as are the serried ranks of little boxes’.
Literature on Third-World settlements, and slums in general, commonly makes the point that such habitats are unfairly judged by outsiders on their untidy appearance. The aesthetic judgment of settlements in urban Papua New Guinea was not made simply in a comparison with European housing styles, however, but was contextualised in a dualism of the latter and an idealised ‘traditional’ rural village. In this respect, settlements and architecturally changing urban villages were both seen as aesthetically deviant. For example, in a discussion of the growth of Lae, a city at the mouth of the Bumbu River on Papua New Guinea’s north-eastern coast, Willis comments on the five ‘traditional’ villages immediately across the river mouth:
Despite their proximity to the city they have not yet been assimilated into it and are still a separate entity. The village houses, often built from reclaimed scraps of timber, iron and fibro-cement, seem dilapidated even by comparison with the houses in the adjacent ‘squatter’ camps. They stand in strong contrast to the neat bungalows of the city and to the other traditional villages to east and west where houses are still built of native materials and have a neat, post-card picturesqueness.
Under the labels ‘settlements’ or ‘squatter settlements’ or sometimes ‘shantytowns’ (the latter term has now almost disappeared), the customary and government land settlements and the no-covenant housing areas were discursively collapsed together, reinforcing the image of a socially distant population. The connection of a spatial and aesthetic dichotomy of urban indigenous habitats with a relative degree of social distance of their inhabitants from the economically dominant white population was reflected in journalistic literature of the period. For example, Peter Hastings wrote in 1969 of an ‘urban proletariat, unskilled, living on the breadline in shanties erected in increasing numbers on the bare brown hills of that disagreeable town … a polyglot force of men without jobs, women or land — the dispossessed of the new dispensation’. Osmar White in 1965 similarly sequestered in print those migrants to Port Moresby who ‘set up squalid shantytowns on its perimeters, subsisting as best they could on wages for casual, unskilled labour’.
Sociological research did not in fact support this image: Richard Jackson commented that while there might be a ‘grain of truth’ in the view of settlements as repositories of ill-adjusted transients, ‘a strong case can be made in support of the opposite view: that the settlements are frequently the homes of people less transitory, no more unemployed and just as urbanized as other sectors of the population’. A survey in another town, Popondetta, had shown that nearly two-thirds of the administration’s labourers, and many tradesmen, lived in settlements, as did ‘business managers, teachers, and two members of the national Parliament’. An examination of the ‘Six-Mile’ settlement in Port Moresby in the 1970s indicated that about 70 per cent of the adults were employed and in a sample of settlements in several urban centres, it was found that households (built largely around nuclear families) contained an average of between one and two wage-earners. In this respect, Papua New Guinea’s settlement-dwellers were socioeconomically similar to those in many other Third-World settlement and slum communities. Yet the image of dispossessed squatters in settlements, contrasting with a more socially proximate working urban indigenous population in other types of housing, prevailed in popular European discourse. While this simple dichotomy misrepresented the complex demography of Port Moresby, it became institutionalised discursively, and has survived the colonial period and a continuing process of change in the nature and variety of habitats in Port Moresby.
In the 1960s, some settlers who could more justifiably be called squatters were appearing in Port Moresby. The earlier settlements were enclaves of people sharing a region of origin, such as the Eastern Gulf people at Ranuguri settlement, or Purari people at Rabia camp. Later migrants from other areas, particularly the Highlands, were naturally excluded from building houses in these existing micro-ethnic enclaves. Lacking traditional exchange links with Motu-Koita landholders, or the wherewithal to obtain no-covenant land blocks from the administration, those who were unable to negotiate some kind of arrangement with customary landholders were often obliged to find a niche on unused government land. An example is Gordons Ridge settlement, where migrants from the Simbu and Goilala districts developed discrete groups of shelters on a hill overlooking one of the most exclusive European housing estates of the period. Their appearance caused concern among the Europeans below, and many of them were persuaded to resettle at Morata, the newly developed suburb containing a mixture of low-covenant and no-covenant housing. Some squatters, however, remained and the Gordons Ridge settlement gradually developed, with the inhabitants improving their dwellings, planting trees and building an access road, consolidating the habitat and gaining legitimacy, if not absolute legality, as it grew in the 1980s.
By the 1970s, a broad range of housing, from unauthorised buildings on government land, through various types of customary or administration-authorised dwellings on unused land, to planned self-help housing on leased blocks, was being referred to in common European discourse as ‘the settlements’ or ‘squatter settlements’. The broad and indiscriminate use of such terms, together with the difficulty of precisely categorising the variety of indigenous housing that had developed in Port Moresby, could have been responsible for the disparity in academic literature of the period (among empirically careful researchers) in representations of the actual numbers of ‘settlements’ in existence at different points in time from the 1950s until the 1970s. At any rate, in the face of continuing changes in the nature of the town’s habitats, the institutionalised notion of settlements (incorporating the notion of squatters), was to prove resilient in the following years.
Housing after independence
When Papua New Guinea became an independent nation in 1975, and the European population began to dwindle, the housing areas it had inhabited were taken over by indigenes. This substitution, in combination with continuing housing projects, increasing urban migration and the demands of the extended kin systems typical of Melanesian societies, began to change the spatial distinction between the housing types discussed above, without significantly altering the institutionalised notion of a dichotomy of ‘settlements’ and formal (ie, low- and high-covenant) housing. Building projects were extending some of the formal housing estates in the outer suburbs to the point where they began to connect with self-help housing areas that had originally been discrete and with enclaves of squatters who had established themselves quietly at the periphery of the urban area since the 1960s. At the same time, the inhabitants of these latter areas were building extensions to, or upgrading, their houses, if they were able. Consequently, while some settlements remained demarcated by boundary fencing, the overall visual and spatial distinction between squatter settlements, legitimate settlements and planned low-cost housing estates became less clear in some outer suburbs of Port Moresby after independence. In addition, the untidy development of suburbs with basic infrastructure provided the opportunity for individuals or families to establish illicit dwellings in the interstices of growing communities of migrants from a variety of ethnic backgrounds, where few questions of legitimacy were asked unless serious trouble arose. In this respect, for example, parts of the formerly European-dominated suburb of Boroko and of the low-covenant suburb of Hohola had a significantly different mixture of housing types by the 1990s than they had in the early 1970s.
A further change after independence was that the homologic relationship between two- or three-bedroomed housing and European nuclear families was displaced under Melanesian kinship sensibilities. The exigencies of an unavoidable obligation to provide assistance to members of one’s extended kin-group — referred to with ambivalent sentiments as the ‘wantok system’ by urban Papua New Guineans (wantok is a Tok Pisin term used to refer to kin, both near and distant) — meant that the urban house became a locus of the nominal occupants’ kindred. It has been argued by Keith Barber that it is analytically useful to regard urban Papua New Guinean households not as physical units of co-residence but more flexibly as sets of changing social relations, acknowledging the social context in which they are embedded. This perspective avoids positing households as fixed sets of people, and more accurately reflects the change in the nature of householding when European residents were replaced by indigenes. Household numbers fluctuated with the movement of relatives from home areas into town and back, and among wantoks’ residences in town. This did not escalate into major overcrowding of individual houses, however, although domestic quarters attached to high-covenant houses were now often used as accommodation for migrant kin.
An associated phenomenon was a new kind of squatter, the ‘illegal’ occupant of formal housing. As industry grew in Papua New Guinea’s towns, large-scale employers rented houses, or blocks of houses, for employees. ‘Compounds’ now included all types of housing, up to relatively luxurious dwellings for the new Papua New Guinean elite as well as for highly paid foreigners. Where colonial employers had enforced the expulsion and exclusion of illicit occupants of compound and rental housing (mostly the wantoks of employees) relatively efficiently, Melanesian sensibilities made the task far more difficult for indigenous bosses, even though the policy of evicting such occupants remained. For example, when I lived in the large housing compound of the University of Papua New Guinea in the early 1990s, constant complaints were made by legitimate inhabitants and those employees waiting to be allocated housing about the number of squatters occupying dwellings of various kinds. The latter were often relatives of employees who had gained access to houses through their kinship connections, and sometimes were ex-employees of the university who had simply failed to move out. Despite the occasional flurry of memoranda and threats of forcible eviction, there was little real action against the squatters. Administrative staff (including some who were complainants) were constrained as a body by the dense kinship networks under which the squatters had established themselves in the first place.
Developments such as these indicated a continuing housing problem in the capital city, which was a legacy of the colonial administration’s slow response to postwar urban migration, an absence of adequate long-term town planning, and a lack of affordable housing for non-elite workers. By the late 1980s, a new belt of self-help settlements, on leased land blocks, was beginning to establish itself at a few kilometres’ remove from the expanding city’s edge in an area where there had previously been only one or two small settlements containing single micro-ethnic groups. As well as containing recent migrants from rural areas, these habitats were the recourse of former city-dwellers seeking refuge from the financial problems of living in town and the increasing limitations on personal freedoms presented by Port Moresby’s growth. Moving to locations such as ‘Eight-Mile’ (the name refers to the distance from downtown Port Moresby) was a strategy to avoid the city’s ‘law and order’ problems, obtain access to gardening land, yet remain close enough to take advantage of employment opportunities in town.
In contemporary Port Moresby then, while most settlements established during the late colonial period still exist, and illicit housing is still built by squatters, the distinction between different types of urban habitat and the associated socioeconomic status of inhabitants is even less clear than it was by the early 1970s. Squatters can be found in many types of housing, from high-covenant to illicit lean-to shelters, and, while some self-help areas have boundary fencing, there is no clear spatial separation overall between various types of urban ‘settlements’ and other types of housing. Despite the increasing complexity in the mixture of housing in the contemporary city, however, the discursive dichotomisation of the city’s habitats into legitimate housing and ‘settlements’ has survived from colonial times.