Volume 96 1987 > Volume 96, No.3 > Aspects of economic development on the Nembi Plateau, Papua New Guinea, by R. Crittenden, p 335-360
                                                                                             Previous | Next   

- 335

Tribal fighting in the highlands of Papua New Guinea is as often interpreted as due to land shortages (Allen and Giddings 1982) as to the result of a general breakdown in law and order (Wormsley 1985). On the Nembi Plateau (Figs 1 and 2) there has been fighting since 1981. However, there is no absolute shortage of clan lands, despite population densities on land close to the road running the length of the plateau and close to Government services being among the highest in the highlands. Law and order was only brought to the Nembi recently. The area was pacified in the 1960s by patrols from Nipa Patrol Post. From 1965 to 1968 Local Government Councils were established to serve the region and a system of Village Courts to administer customary law was introduced in the early 1970s. Both the Local Government Councils and the Village Courts still operate.

Tribal fighting before pacification on the Nembi was associated with intense interclan competition in the exchange network, the basis of social structure and the political economy throughout the highlands (Brown 1976). Disruption of this exchange network by the introduction of a monetary economy and the prohibition of fighting prompted the Nembi to turn to other avenues, for example business development, to compete against each other.

The 1981 outbreak of tribal fighting involved some 20 clans. The fighting centred upon a dispute that originated before the colonial administration built Nipa Patrol Post in 1959 and pacified the area. The administration interpreted the dispute as one involving the ownership of a piece of land. Indeed, three clans were contesting the right to cultivate a portion of land, but this was only the outward manifestation of an argument that centred upon the murder of a clan leader (Maruk Shep) in the 1930s, in a dispute over pig stealing.

In 1967 patrol officers at Nipa ordered the three principal clans in the conflict not to garden or build houses upon land that separated them and which had become a focus for their dispute. The matter, still interpreted

- 336
Papua New Guinea, showing the location of the Nembi Plateau.
- 337
Locations referred to in this article.
- 338

as a land dispute because of supposed shortages on the plateau, came before the District Land Court in October 1979. A settlement involved the exchange of pigs, money and pearl-shells. Two of the clans, allied with each other, were given exclusive access to the land and were ordered to pay compensation to the third clan, Sigiriba. When the two clans planted gardens on the land young men of the Sigiriba, remembering the story of the death of Maruk Shep, began to steal from the gardens. This provocation led to open warfare in April 1981.

Since 1981 the Nembi Plateau has been twice declared a “fighting zone” by the Provincial Government (as defined by the Inter-Group Fighting Act of 1977), and over 30 lives have been lost in a population of some 4,000. As a consequence, a Government-run community school has been permanently abandoned, and at various times another school has been closed; a mission station has been closed and health services, especially maternal and child health (MCH) clinics, have been disrupted. Nevertheless, the Village Court system still operates and, except for the sporadic fighting which has its own sets of rules and constraints, law and order has not broken down.

This paper is an attempt to give some explanation why the Nembi, always recognised as a particularly truculent people by the colonial administration (Patrol Report No. 4, Mendi 1952/53), after 20 years of relative acquiescence to an imposed rule have chosen to revert to their own traditional methods, i.e., fighting, to settle their differences (cf. Gordon and Meggitt 1985:145-88 for the Enga situation). In part, the fighting is a response to disillusion with the ability of the Government, be it provincial, national or local government councils, to foster economic progress on the Nembi Plateau, to provide opportunity for cash employment and other business opportunities. It is this, to an extent, that prompted renewed fighting in Enga (Gordon and Meggitt 1985:151-3) parts of the Western Highlands Province and in Simbu. In turn it is related to the Nembi's ability to come to terms with the dramatic changes imposed upon them in the last 30 years.

It is my contention that the analysis of the causes of the fighting on the Nembi Plateau (and indeed elsewhere in the highlands of Papua New Guinea) must be centred upon the nexus between the values and beliefs of two political economies: traditional values and beliefs on the one hand and introduced Western ones on the other.

It has long been realised that there is an important link between social organisation and the ecology of land use in agrarian societies. Brookfield and Brown (1964), Bowers (1968) and Watson (1977) are - 339 pioneers in this field in Papua New Guinea. Nevertheless, concentration seems to have been on the ecological and environmental components of the man-land equation, and not simply because environmental factors are interdependent and are therefore more coherent (Stocking and Abel 1981). Detrimental alterations in the man-land relations are thus expressed in terms of “land degradation” rather than of “social degradation”. Indeed, a recent study in the Southern Highlands of Papua New Guinea developed an index of land degradation (Wood 1984) where the dissonance in the man-land relationship might as usefully have been expressed as child malnutrition rates.

Additionally, the human component of the equation is often subsumed by demographic data and designated as “population pressure”. In some cases, and particularly pertinent here, social unrest, land disputes and tribal fighting thus become mere symptoms of population pressure upon scarce resources (Allen and Giddings 1982).

Recently Blaikie and Brookfield (1987) called for a larger contribution by social scientists to the study of the man-land equation. Allen and Crittenden (1987) provide an analysis of man-land relationships in the Papua New Guinea highlands emphasising the political economy side of the equation. However, as Brookfield and Blaikie point out, there remains an inherent contradiction in the equation “between the criteria used for land degradation, and those for beneficial social change or ‘development’, through time” (1987:14). This contradiction is more than apparent in the highlands of Papua New Guinea where the pace of social change in the last 50 years has been extremely rapid (for an eloquent discussion of this see Read 1986). To examine this contradiction, and to provide some balance to the equation as called for by Blaikie and Brookfield, an appropriate methodology is that commonly used in Farming Systems Research. What is required are insights into the social system/political economy, as a nexus portrayed in the aspirations and decisions of individuals, indigenous knowledge (see, for example, Chambers 1983, Richards 1985, Bdilya 1986).

This paper argues that a strong belief in sorcery, compounded by an increase in malaria and a sense of recent economic deprivation, is a major cause of recent tribal warfare on the Nembi Plateau. Using what is commonly called by rural sociologists the farmer first and last method (F.F.L.—see Chambers and Ghildyal 1985), this paper presents an inside view of the Nembi predicament and their warfare in which they try to do something about it. This analysis attempts to bring to the fore the neglected and undervalued beliefs and perceptions of a small sector of - 340 “the rural poor”, and to compare the insight thus gained with conventional interpretations of tribal conflicts in Papua New Guinea that emphasise the man-land relationships as one of food supply, population density, land use intensity and land degradation.


The Nembi live in small hamlets (andaa) scattered through the grass basins and on the lower slopes of the Nembi Plateau, and in the lower Nembi valley and parts of adjacent valleys. Hamlets consist of family groups (yem), including a man, his wife (or wives) and their children. In the past most men lived in a men's house (enza), usually close to a ceremonial ground, and the women lived in a women's house (tenda) among the gardens. Occasionally both men and women shared a partitioned house (enzatenda), with the rear of the house reserved for the women and pigs, and the front for men. The enzatenda has become the common form of residence in recent years.

Related families occupy adjacent territories, forming subclans which are essentially territorial descent groups. They refer to themselves as no yem (our families) and to other subclans as nemi yem (their families). A number of subclans which can claim descent from a common ancestor make up a clan whose members refer to all the families as no yem regardless of their subclans. Sorcery is rare within the clan, and marriage is prohibited between people who claim a common remembered ancestral link (Crittenden 1982:107-22).

Access to land depends on a person's ability to prove a right to it. Clan members are distinguished on the basis of the kin links that they exploit to gain land and the right of residency in a community (Crittenden 1982:107-22). Consequently clans display a degree of ambilineality. Nevertheless, where a man lives is not fortuitous for he wishes to live with kin that he trusts.

In the 1960s the Nembi economy was increasingly disrupted by colonialism. Fighting was prohibited and the administration inflated exchange. The long-distance trade network, which the Nembi relied upon to boost their own meagre resources, was destroyed (Crittenden 1982:194-255). With improved health care, population densities in some locations increased beyond the capabilities of the present systems of agriculture. Additionally, the plateau is marginal for the production of coffee, the major export crop of the highlands. Coffee was introduced to the plateau in 1965 (Patrol Report No. 3 Nipa 1964/65), but by 1971 only - 341 7.4 ha was planted. The area under the crop remains small. Chillies were tried in 1971 (Patrol Report No. 4 Poroma 1970/71) with little success. Pyrethrum was also unsuccessful (Patrol Report No. 13 Nipa 1963/64). Cash incomes are therefore low. Consequently the Nembi are isolated and relatively disadvantaged and the plateau is one of the poorest areas of the Southern Highlands. It is noted for its high rates of child malnutrition and disease, locally high population densities and the low fertility of its soils (Allen et al. 1980, Crittenden and Baines 1985 and 1986).

The lack of opportunity was keenly felt (Patrol Report No. 3 Nipa 1963/64). By late 1962 money had replaced trade and exchange items (pearl-shells, cowries, cloth and stone axes) as payment for firewood and food for patrols (Patrol Report No. 4 Nipa 1962/63). The Nipa station land was bought in 1960 with shells and axes. In 1965 the Kar Mission airstrip in the Nembi valley was paid for with money (Patrol Report No. 3 Nipa 1965/66). By the late 1960s money began to enter traditional exchange transactions (as shown by money in bride-wealth payments, Table 1).

Limited cash-earning opportunities, however, forced the Nembi to continue using traditional wealth items. But the need for cash (for school fees, vehicles, bride payments and death payments) was inescapable (Rowley 1965), especially as traditional items lost their value because of inflation (Hughes 1978). Outmigration for work on plantations in the highlands to the east, or on the coast, provided some opportunity to earn small amounts of cash, and the Highlands Labour Scheme (Harris 1972) took the first recruits from the Nembi Plateau in 1962 (Patrol Report No. 4 Nipa 1962/63 and Patrol Report No. 7 Nipa 1964/65). Many families and many more single men continue to travel to the Western Highlands for work during the coffee season every year. In June 1980 (the height of the coffee flush) 25 percent of single adult men were absent working on plantations.

The lack of local opportunity for obtaining cash, and therefore developing the Nembi economy, intensified clan rivalries. These rivalries are centred upon ceremonial exchange. Exchanges are the essence of Nembi society, and they operate through relationships connecting intermarrying clans. Desire for the more fertile plateau land and for places close to the road, coupled with the fear of sorcery at lower altitudes, encouraged the most influential men in the exchange network to manipulate traditional concepts of land tenure to suit their needs. These big-men (el hauma) are political leaders who use the exchange of wealth to reinforce group allegiances. The struggle for renown involving clan

- 342
Bride-wealth from the 1920s to the present
            Mean Number of Items Per Bride-wealth          
Period and Number of Marriages   Pigs Money (Kina) Pearl-shell Casso-wary Cowrie Salt Tree Oil Black Palm Bow Stone Axe Other
1921-25 1 5 20 3 2 1 1 1
1925-30 4 2.5 31.25 4.75 2.25 1 0.25 0.25
1931-35 5 1.6 41.2 5.8 6.4 2.4 0.5 0.2
1936-40 12 2.2 32.1 6.9 5.1 3 2 0.6
1941-45 7 1.6 48.6 4.7 3.9 1.1 1.3
1946-50 3 1.5 40 5.5 4 2 1.5
1951-55 11 1.5 26.1 3.5 2.7 2.1 8.2
1956-60 8 1.75 47.5 7.1 3.6 2.4 0.1 1.3
1961-65 12 2.1 53.3 0.1 3.9 2.4 1 0.8
1966-70 12 1.7 1.7 40 0.7 1.8 1.1 1.1 0.4 0.1
1971-75 13 2.2 146.2 76 2.1 0.1 0.1
1976-80 12 3.6 165.0 61.1 0.5 0.1

Source: Crittenden 1982

- 343

rivalry is effected by the monetary economy. New economic opportunities provide alternative avenues to gain prestige. The dearth of opportunities intensified competition and hostilities.

A part of these interclan hostilities is sorcery, especially that originating in the warmer, humid and forested lowlands. It is used to remove the influence of competing and aspiring rival big-men, and is much feared (cf. Lederman 1981). There are many kinds of sorcery on the Nembi Plateau and in the surrounding region, and all are practised to cause death. It is important to relatives of a dead person to know the cause of death, whether it is caused by sorcery or some action of spirits, so the death may be avenged: the sorcerer killed or the spirit appeased. Ritual autopsy is performed to determine the cause of death (Alkebi or Alpuhande) similar to that performed in the Tari Basin to the north (Glasse 1968). Tom, Membenggi, Nomp and Tensogol are common forms of sorcery. Pur ink or Webi Yan describes the symptoms of shivering and fever caused by sorcery, especially Tensogol. Such rituals as Kaebel (ancestor worship), or Porke, Ekom, Tomberg, Yegil or Esowe (which also describes the particular spirit) are performed to mitigate the action of those spirits and, if appropriate, the sorcerers who have invoked them. Webi Yan describes the sickness, malaria, that has only begun to occur with frequency throughout the region in the last 30-40 years.

Warfare is not a direct response by the clansmen of the Nembi to either an absolute shortage of land, degradation of land, loss of soil fertility, declining yields or increased soil erosion. These factors are coincidental and are part of the problem of clan warfare only in as much as they are the result of intense interclan competition and associated sorcery.


The changing epidemiology of malaria in the Papua New Guinea highlands over the last 30 years gives some rationale to the Nembi belief that sorcery has increased (Radford et al. 1976). Malaria endemicity in Papua New Guinea ranges from holoendemic in some lowland coastal areas (including the Purari and Gulf lowlands to the south of the Southern Highlands Province) to complete absence in some parts of the highlands. Before Europeans entered the highlands, traditional links between the peripheral highlands people and those of the coast provided avenues for the disease to enter parts of the highlands; raiding war - 344 parties (as with the Kukukuku and the Vailala people), trade or bride exchange expeditions (as with the Samberigi and Bara people of the Kikori hinterland) often entailed overnight stops in lowland areas by highlanders, who subsequently returned with the malaria parasites in their blood.

The Nembi, because of their central position in the regional trade network, were frequent overnight visitors to trade partners in the Lake Kutubu and Mubi River region—a region of endemic malaria, and the Kutubu ol (men of Kutubu) gained a reputation as powerful sorcerers. Nembi men returning from the Kutubu and Mubi lowlands invariably became sick (feverish). Other parts of the highlands, notably the swampy lowlands of the Wahgi Valley, had a similar reputation and were feared as places inhabited by evil spirits that afflicted people with ague and fever (Black 1954).

European penetration of the highlands facilitated the spread of malaria (Peters et al. 1958, Black 1955, Radford et al. 1976). Passive case detection, through the examination of blood slides collected by aid post orderlies, and the results of specific surveys give evidence of the spread and presence of malaria in the highland valleys. The Kewa people immediately to the south of the Nembi call malaria pole yanya—a sickness from Erave (Franklin and Franklin 1978), suggesting that trade routes to the south of the plateau followed by early administration patrols, as well as the establishment of Erave Station in 1953, spread malaria into the region. A survey at Erave in 1956 suggested that malaria had spread outwards from the station and diminished in proportion to distance from the station (Peters et al. 1958). The building of the airstrip provided breeding sites for A. punctulatus. Imported labourers and police were the infective reservoir. Patrols in the mid-1950s also reported a high prevalence of mosquitoes, splenomegaly and positive blood slides in the Kutubu lowlands (A. Speer, personal comment in Radford et al. 1976).

Although Nembi warriors contracted malaria in the adjacent lower lying valleys to the south and west of the plateau, the lack of a vector population and sites suitable for mosquito breeding prevented a reservoir of malaria (infective reservoir) being established on the plateau. Immediately to the south of the plateau in the Erave valley (1450m) the drains and ditches of the garden fringes provided abundant potential breeding sites for mosquitoes. In this valley temperatures and rainfall are conducive to malaria transmission. In the highland valleys to the east, the period May to October can be appreciably colder and drier than - 345 November to April—indeed, this is the case on the Nembi Plateau—and malaria transmission is considerably reduced or impaired.

As malaria spread (30-40 years ago), so people migrated to higher altitudes, accounting for the expanses of kunai grassland beneath which old cultivations are distinguishable in the valleys of the Erave and Wage Rivers bordering the Nembi Plateau. Competition for locations on the higher slopes and basins of the plateau increased and added to the belief that sorcery was rampant. Thus, only with the movement of relatively large numbers of people (police and carriers) in conjunction with the development of favourable breeding sites for mosquitoes (wheel tracks and road-side ditches) did the spread of malaria increase. Such a process is not unique to the Nembi and the Erave valleys, and similar stories of sorcery associated with a febrile sickness (malaria) are found throughout the highlands (Meggitt 1981).

Various other epidemic diseases also occurred on the Nembi Plateau but were not associated with sorcery. For example, in September 1965 dysentery broke out in the Enip census unit area. It killed 5 percent of the population in three weeks. In September and October 1969 an influenza epidemic killed 112 people in the Nipa Basin and killed 50 people on the Nembi Plateau. The disease was introduced by an administrative patrol from Mendi (Patrol Report No. 3 Nipa 1969/70). In 1979 and 1980 measles epidemics occurred in the central Nembi Plateau and, while only three children died, 123 were hospitalised at one time or another at the ‘Ol’ Health Centre (Crittenden 1982:353-8).

Buchbinder (1973) reported widespread influenza epidemics throughout the highlands in the 1960s. Influenza epidemics were also reported in the 1930s. These have sometimes been confused with malaria. McCarthy (1970) reported meeting a missionary doctor near the Dutch border in 1937 who explained there was a widespread influenza epidemic in the area. In the Koge area of Sina-Sina (Simbu Province), Hide, interviewing people in the late 1960s, found they could recollect a “big sick” in the 1930s (1981). Hide cited an unpublished volume by Bergmann (1971), who reported it as a severe epidemic of influenza.

Nelson (1971) reported a “big sick” in the Nebilyer valley (Western Highlands) and dated it to the 1930s. He reported it as the introduction of malaria. This is unlikely for he reports the introduction of sweet potato into the area at this time. Sweet potato is known to have been introduced much earlier (Golson 1980). It was probably part of the same influenza epidemic reported by Bergmann and Hide. Because of its similar altitude to the Wahgi valley, the Nebilyer was probably - 346 malarious before the 1930s. In the nearby Kaugel valley, Bowers (1968) reported real depletions of the age cohorts born between 1930 and 1945 because of these epidemics. Age cohorts on the Nembi Plateau show similar depletions (Crittenden 1982:301-15). There was also a widespread dysentery epidemic in the highlands during the Second World War (Burton 1983), reputedly brought into the highlands by Canadian airmen who had crash-landed in the Eastern Highlands (ANGAU 1944). This outbreak was reported in the Enga and Tari region in the diary of Danny Leahy (Ashton 1979), and was also reported by Feachem (1977).

Ongka, Strathern's informant from the Melpa, to the north-east of Mt Hagen, also tells of a big sick when he was a child (Strathern 1979). Big sicks are also reported by Ivinskis et al. in the oral history of the Chimbu (1956). These reports of a “big sick” refer either to influenza or dysentery. The spread of malaria until very recently was not of epidemic proportions. Lack of an infective reservoir limited its incidence to individuals who travelled to areas where the disease was endemic. In most highland areas it was attributed to sorcery originating in the lower lying valleys and practised upon individuals. The Nembi have no recollection of “big sicks”, except their belief that declining yields, small sweet potato tubers, malnourished children and the failure of economic opportunities to materalise are associated with sorcerers and a sickness in the ground developed since the arrival of Europeans.

Other than the invidious spread of malaria, direct intervention in the Nembi way of life by the administration also contributed to the Nembi's fear of the valleys to the south of the plateau as areas of virulent sorcery. In the past the Nembi, like the Mae Enga, preferred to retaliate to sorcery and to take revenge by killing with axes (Meggitt 1977). Prohibition of clan warfare thus removed the accepted response to sorcery. Unable to retaliate, the Nembi's preoccupation with sorcery further developed. At the same time the configuration of the major alliances of the Nembi Plateau and its bounding valleys highlights the valleys as zones of conflict (Crittenden 1982:136), with the lower Nembi valley alliance, the “Aeron”, continually at war with the “Karinj” of the central plateau to the north-west.

A recent malaria epidemic in the region, associated with construction of the highlands highway through the Nembi valley which started in 1976, demonstrates the point. Unusually heavy rainfall in late 1978 and early 1979 provided conditions suitable for the breeding of A. punctulatus in the wheel tracks and ruts. In May and June 1979 13 children in - 347 the middle Nembi valley died from malaria in one month. There was a total of 60 deaths in three months of a total population affected in the area of about 1000 (Karanakaran 1980). The Nembi and Erave valleys, once regions of mesoendemic malaria, were in 1979-80 hyperendemic (Department of Health 1980). In 1974-75 the malaria parasite rate in the Nembi valley was 8.7 per cent; in April 1979 the malaria parasite rate of children between ages of 2-9 was 92 to 99 percent. Spraying with DDT and mass drug administration took place in August and September 1979. Within three months the epidemic was controlled.

This brief epidemic focused the valley as an area of sorcery. As the epidemic developed the population at the southern end of the Nembi valley (the Aeron), protected by some immunity and with access to health care (Det Health Centre), showed an increase in morbidity but not mortality; the people higher up the valley and on the Nembi Plateau (Karinj), as they had little or no immunity and did not have access to antimalarial drugs, showed a very high mortality.


Warfare takes on renewed significance to the situation of the Nembi when interpreted as the result of the belief in sorcery, the incidence of malaria and the lack of economic opportunity, rather than the result of gradually growing populations and increasing population densities over a long period of time (Barnett 1977, Boserup 1965, Allen et al. 1980). The locally extremely high population densities on the plateau are misleading and need to be interpreted carefully; not the least because population growth in the highlands is augmented by an intensely competitive system of social exchange (centred upon the exchange of pigs), which also makes heavy demands on production from the land (Allen and Crittenden 1987). Differences in the inherent fertility and capacity of the land for sustained production have always been important, but for production of a social surplus and not simply for food. The Nembi, in an area of poor soils and therefore low productivity, maintained their standing in the regional exchange economy because they were astride an important trade route from the lowlands to the highlands. With disruption of the regional trade network, the few opportunities available in the monetary economy whereby men could gain renown and prestige caused intense competition. As a part of this struggle land favourable in location with respect to roads, access to Government services and freedom of - 348 malaria was placed under pressure. Other more remote parts of clan territories were correspondingly abandoned.

Density of population on total Pubi clan land is 62 persons per km2, among the highest in Papua New Guinea, but rising to 272 persons per km2 on land on the plateau itself. This is a very high density in comparison with other highlands groups (Table 2). These locally high densities are reflected in the ratio of cultivated land to fallow (in the “used” clan territory), perhaps the simplest method of expressing intensity of land use. That for the Pubi is 1:1.8 (excluding primary forest). Reworking Waddell's figures (1972), a ratio of 1:3.1 is obtained for the Raiapu Enga, a people reputed to be at the heartland of intensive sweet potato cultivation (Brookfield 1964).

The high density is also expressed in the area under cultivation per head. The Pubi have an average of 0.092 ha cultivated per head and the Raiapu Enga have between 0.16 and 0.19 ha. It would seem that, from the conventional viewpoint, the Nembi are certainly under stress (Barnett 1977), for yields are low, soil fertility is low, fallow periods are short, population densities are locally high and rates of child malnutrition are high. But, as yet, the Nembi response has been through their belief in sorcery rather than by reaction to these symptoms and modification of their cultivation techniques.

Unlike the Raiapu Enga, only a very few of Nembi farmers mulch or compost to maintain soil fertility. The Nembi still cultivate their land as they did 30 to 40 years ago, although the fallow periods have become shorter. This suggests that the high population densities and high cultivated to fallow ratios on the Nembi Plateau developed only recently, and that the Nembi do not yet perceive the problem as one where they must change their agricultural techniques.

From the conventional point of view the situation is worse when it is considered that as much as 25 percent to 40 percent of clan land on the plateau is unusable (Allen et al. 1980, Crittenden 1982:140), either because of terrain or because the soils are not deep enough or fertile enough for sweet potato cultivation. A man usually clears a new garden from secondary growth—cane grass and shrubby regrowth with some tree growth (Dodonea viscosa)—upon marriage. Only rarely does he clear new land for sweet potato gardens thereafter. An average household of the Pubi clan has 0.538 ha of land under sweet potato and 0.29 ha under mixed garden, supporting a family of 4 adults, 2 children and 3 pigs. Although portions of individual gardens are occasionally allowed to revert to short grass fallow, the cultivation periods for the

- 349
Gross Population Densities of the Clans of the Central Nembi Plateau compared with Other Highlands Groups
Nembi Plateau Clans Gross Density for Total Clan Land (Per Km) 1 Gross Density for Land on the Nembi Plateau (Per Km) 2
Mul-Terrel 30 120
Purt-Purre 50 150
Koin-Iomo 83 249
Ib ) 136
Sigiriba (lower Pwe) ) 80-120(c) 252
Sigiriba (upper Pwe) ) 269
Murupa 80 80
Penarop 29 152
Pubi 62 272
Epi-Yokul 30 69
Porelep 31 31 3
Other Highland Groups 4    
Central Simbu 89  
Mae Enga 62  
Gururumba 44  
Huli 44  
Dei (Mt Hagen) 38  
Gahuku Gama 35  
Raiapu Enga 35  
Bena Bena 34  
Gadsup 30  
Siane 29  
Kakoli 25  
Kuma 19  
Maring 15  
Kyaka Enga 13  
South Fore 12  

gardens may be up to 25-30 years, one generation, or until a man's sons marry and clear new gardens. In the past, high mobility, associated with clan warfare, ensured that some fallows (to secondary regrowth) were longer and that cultivation periods were also not too long. With the cessation of warfare the nutrient status of the soils has rapidly declined - 350 (Wood 1984a) with continuous cultivation. In many cases sons are not clearing new gardens upon marriage but continuing to cultivate their fathers' gardens because of their favourable location close to the roads and mission stations. The result is seen in the erosion of topsoil to the lower slopes and to the bottom of dolines. On the higher slopes topsoil is virtually absent. The Nembi do not take preventative and conservative measures to avoid soil erosion. Although soil erosion in Papua New Guinea is low by international standards (Bleeker 1983), it is a significant problem on the Nembi Plateau. Steep slopes, extremely rugged terrain, high population densities, short fallow periods and a lack of ameliorating techniques have resulted in the yields of sweet potato that are among the lowest in the Papua New Guinea highlands.

The average yield of sweet potato on the Nembi Plateau is 7.1 tonnes per hectare (compared with 17-21 tonnes in Enga, Waddell 1972). The production of sweet potato per capita per day (1.7 kg) and the amounts consumed by people (1.07 kg) are correspondingly low (Enga 1.3 kg). Although the amount varies according to availability, pigs on average are fed 1.3 kg of sweet potato per day, accounting for approximately 38 percent (1.423 tonnes) of the production of an average family. Kimber (1972) considered that 31 percent was a fair estimate of the proportion of the production of highland sweet potato gardens fed to pigs. The Nembi, although producing very little, still “invest” over a third of their production into social exchange—pigs—and it is this pressure that is more significant than population growth per se to the situation of the Nembi. Having lost their position on the long-distance trade routes with which they augmented their own production, they are striving to maintain their position in the regional economy, without the same opportunities to generate cash that exist in other areas. Comparison of the Nembi with the Tsembaga Maring and the Enga emphasises the Nembi's struggle.

The Tsembaga Maring feed 27 percent of their sweet potato to pigs (Rappaport 1968) and the Raiapu Enga feed 64 percent of their sweet potato to pigs—1.4 kg/day/pig (Waddell 1972:118). Both practice different methods of pig husbandry. The Tsembaga Maring leave their pigs to forage in the forest and feed them a supplement of sweet potato. This is characteristic of the societies of the forested highland fringe. The Raiapu Enga, having transformed their environment almost entirely to grassland and gardens, reputedly have the most intensive system of pig management in the highlands. They fodder their pigs almost exclusively on sweet potato and their social exchange is correspondingly more elaborate and intensive than the Tsembaga Maring. The pig/human - 351 ratio for the Tsembaga Maring varies between 0.8:1 and 0.3:1 while that for the Raiapu Enga varies between 2.3:1 and 3.3:1 (Morren 1977, Rappaport 1968, Waddell 1972). The Nembi ratio of 0.5:1, for a society with a social exchange system as elaborate as the Enga and which values the pig very highly, is very low. Moreover the Nembi, in their pig husbandry, are as firmly at the intensive end of the spectrum, and their pig killing ceremonies (Mok-ink) are as elaborate as those of the Enga (Tee).

Despite the smaller amount of sweet potato available for human consumption on the Nembi Plateau, the average ration fed to pigs is similar to that in Enga. Declining yields have thus placed pressure on both the production of pigs and the staple diets of the Nembi. Indeed, the average size of tuber harvested and consumed on the Nembi Plateau in September to December 1980 was 98 grammes, tubers that elsewhere in the highlands would have been only fed to pigs.

The Nembi have not adopted new and better technologies to increase the marginal productivity of their labour. Analysis of 33 farm families, disaggregated into 52 hearth units based upon a woman and her children, revealed very little difference in the level of intensity of production between them (Crittenden 1982:452-68). Consequently women, who are responsible for cultivating sweet potato, are working extremely hard and are forced to garden areas where yields are extremely low. The returns for their labour are low and, with mothers working hard, child care suffers and children do not thrive (Crittenden and Baines 1986). But, with only minor modifications to their farming technology, the Nembi could appreciably raise the productivity of their farming.

Agronomic research on the Nembi Plateau has produced recommendations based upon farming technology used elsewhere in the highlands, especially green mulching and composting. Research by a team of agronomists (Agricultural Field Trials, Studies and Extension Monitoring Unit—AFTSEMU), as part of the Southern Highlands Rural Development Project, attempted to strike a balance between the farmer first and last model (FFL) of research, and that model of research generally concerned with a transfer of high technology. It conducted research into traditional practices to better understand the present system of agricultural production, by establishing whether traditional techniques were sound in terms of Western scientific principles. Although establishing the efficacy of composting and mulching (Floyd, Lefroy and D'Souza 1987), the acceptability of the techniques, or of their modification, was not examined. It was recognised that the techniques were, however, labour intensive and increased rates of compost - 352 application by those already composting would increase the workload of women. On the Nembi Plateau, with women working close to their limit already, it is unlikely that such labour intensive techniques will be widely adopted.

The alternative is to bring into cultivation land that has lain fallow, or has reverted to primary forest, and to rest existing gardens that have been cultivated for 20 to 30 years or more. The division of labour would require that men clear the regrowth vegetation and for families to move back to the more remote areas of their clan territories. Movement across the whole clan territory to cultivate gardens was the norm 30 to 40 years ago.


Nembi clan lands extend into the Waga valley to the west of the plateau. Heavily forested, it is used for hunting, gathering wild fruits and nuts, and obtaining house building and other materials. Abandoned clearings in the forest and extensive kunai grasslands (Imperata sp) in the southern portion of the Waga valley are evidence of expanding settlement that was taking place as recently as 40 years ago. One of the first patrols across the Nembi Plateau after the Second World War remarked that “. . . over the total patrol many migrations were noted. Areas populated pre-war were now deserted” (Patrol Report No. 1 Lake Kutubu, 1952/53). In part, this was due to the spread of malaria but the Administration also encouraged these movements. Economic opportunity was another attraction.

When the Administration stopped clan fighting in the early 1960s, many clans reclaimed land they previously owned but had been driven from in warfare (Patrol Report No. 6 Nipa 1959/60). The pattern of clan alliances and associated claims to territory were complex. The colonial Administration rationalised the chaotic maze of claims and counter-claims by recognising the clan boundaries and land occupied at contact and pacification as the de facto and de jure situation. The unequal distribution of land in the more fertile basins and corridors of the plateau became the status quo. These basins were the scene of the most intense pre-pacification warfare.

The Administration also encouraged people living in the Waga valley and to the west to relocate on to the plateau, thus exacerbating the situation. Patrol officers suggested that economic opportunity was greater on the plateau (Augu Village Book 1962-1966, Patrol Reports Nipa No. 2 - 353 (1959/60, No. 11 1965/66, No. 1 1967/68), but, in fact, administration was easier if people were concentrated into contiguous village groups close to the main patrol route across the plateau. The road traversing the plateau now follows this route and access to the monetary economy via this artery attracted people to resettle close to it (Patrol Report No. 11 Nipa 1971/72).

The census figures between 1962 and 1966, and for some census units into the early 1970s, show this movement of people back to the plateau from the Waga valley to the west and south-west (Crittenden 1982:311). Less than half of the Nembi people were enumerated in the first census in 1962 and the majority of people were not recorded until 1966. In 1966 the number of people counted in the Penarop unit was still 36 percent more than in the previous year as people continued to move back to the plateau from the Waga valley beguiled by the promise of economic gain. The recorded population of the Mul census unit increased by 15 percent over the same period.

By 1967 most of the large-scale movements of people back to the plateau were over, but in favoured areas densities further increased as people exploited kin ties to settle in those areas. Similar movements were reported for the Tari Basin (Wood 1984b). With economic opportunities largely ephemeral and illusory, the increased population densities added to the tension between clans. Regional alliances became more entrenched and are reflected not only in the present fighting but also in election results for the Local Government Council wards, the provincial and national parliaments.

Reversing this movement on to the plateau by providing incentives for families to return to the Waga valley and further to the west would be a satisfactory solution to the population pressure on the Nembi Plateau. It would relieve interclan tension. It is here that the Nembi's understanding of malaria, sorcery, fighting and economic development complements the more conventional interpretations of man-land relationships that are basically agronomic and centred upon food production, population densities and land shortages. Malaria must be removed from the valleys and economic opportunities and services provided.

Conventional wisdom is that problems of land shortage are the result of growing population and increasing population densities. On the Nembi Plateau, although population has grown, migration to favourable areas - 354 of land is a more significant cause of locally high population densities and these movements are a direct consequence of intervention in the Nembi way of life by outside forces. There is no land shortage per se. It has also been assumed that rural people are unlikely to change their subsistence agriculture techniques until they find themselves under stress. The Nembi response to the outside forces disrupting their way of life and placing their economy under stress is different. They believe that sorcery is a major cause of their situation. Rather than reacting to the symptoms (relative land shortage, declining yields and soil fertility) and changing their subsistence agriculture, they have reacted to the lack of economic opportunity and increased competition and neglect by the Government by “fighting”, a traditional method of settling disputes. This is not to say, however, that there has been a total breakdown in “law and order” for the Village Courts still operate to settle minor disputes, to a large extent, according to customary law.

The Nembi have a social system that places great emphasis on obtaining prowess and in the ordering of society upon the individual rather than on any corporate group. An underlying web of kinship provides some structure and, combined with beliefs in sorcery and the supernatural, is central to intergroup competition and hostility. Disruption of the Nembi economy by colonialism—the cessation of fighting and inflation of exchange, as well as destruction of long-distance trade—contributed to sudden increases in population densities, loss of soil fertility and a relative shortage of land. In themselves these factors have not caused conflicts of individual interests, for land is not owned by individuals but by the clan, and power and prestige, although accruing to individual big-men and often gained by the use of sorcery, is more properly accredited to clans.

The competition between clans is not primarily concerned with land but with group prestige—the distribution of wealth and influence in the exchange network. Warfare over land, when it occurs, is thus not a direct response to any shortage of land, degradation of land, loss of soil fertility, declining yields or increased soil erosion. These factors are coincidental and are part of the problem of clan warfare only in as much as they are the result of clan competition and its associated sorcery. Cessation of fighting and the imposition of an alien law in the 1960s, although bringing obvious benefits, did not bring with them any change in attitudes and beliefs. Moreover, the vision of economic gain, wealth and prestige was ephemeral and, to a large extent, illusory. A relative shortage of favourably located land has, thus, only become indirectly a - 355 political-economic issue because it is a facet of the rapid change which the Nembi have experienced in the last two and one half decades, and the intense competition which has developed between groups to obtain access to the cash economy.

My analysis of malaria, sorcery, fighting and economic development on the Nembi Plateau is not without significance to proposals by the provincial and national governments to ameliorate the economic situation. If, as I have argued, the problems of the Nembi Plateau, including renewed fighting, are the result of interaction with the wider monetary economy, then measures designed solely to increase soil fertility and productivity will not alter the root cause of the problem. Efforts must be made to more fully integrate the Nembi economy with that of the province and enable the Nembi to participate in the cash economy without having to labour on distant plantations. Research programmes must recognise this process of integration in their design. In addition, the Nembi need to be encouraged to continue making decisions for themselves to solve these problems. For example, the eradication of malaria (or at least its control) and provision of aid posts and access to the Waga valley will prompt a return to land previously under cultivation and will start a process of westward migration that was taking place as recently as 25 years ago in some parts of the Waga. Spraying in the Erave and Nembi valleys has already all but eradicated malaria from them (Crane 1985).

The removal of sorcery from the valleys will encourage people to move back to them, will relieve the pressures of living on the Nembi Plateau and will help defuse the intense competition over limited opportunities of access to the cash economy. For as long as economic opportunities are limited, interclan fighting is an added burden that will hinder further economic development. If relatively cheap and simple measures, road building, malaria eradication and opening up clan lands for productive settlement, can defuse the fighting, then perhaps the entrepreneurial abilities of the Nembi will flourish.


I wish to thank Dr Bryant Allen and Mr Robin Mearns for discussion of the broad themes in the paper, and an anonymous referee for comments.

- 356
  • Allen, B. J., et al., 1980. Child Malnutrition and Agriculture on the Nembi Plateau, Southern Highlands Province, Papua New Guinea. Social Science and Medicine, 14:127-32.
  • —— and R. Giddings, 1982. Land Disputes and Violence in Enga, in B. Carrod, D. Lea and K. Talyaga (eds), Enga: Foundations for Development. Armidale, University of New England Press.
  • —— and R. Crittenden, 1987. Degradation and a Pre-Capitalist Political Economy: The Case of the New Guinea Highlands, in P. Blaikie and H. C. Brookfield (eds), Land Degradation and Society. London, Methuen.
  • Ashton, C., 1979. The Leahy Family, in J. Griffin (ed.), Papua New Guinea Portraits: the Expatriate Experience. Canberra, Australian National University Press.
  • Australian New Guinea Administrating Unit, MS. 1944. Diary (entry January). War Memorial Library Collection, Canberra.
  • Barnett, T., 1977. A Note on the Man-Land Relationship in Papua New Guinea. Oceania, 48:141-5.
  • Bdilya, H. H., 1986. The Relevance of Western Land Evaluation Methods to Land Appraisal of Indigenous Farmers in Bio Borno State, Nigeria. Department of Geography, University of Salford (Discussion paper No.28).
  • Bergmann, 1971. The Kamanuka. Mutaapilly, Queensland. (privately distributed)
  • Black, R. H., 1954. A Malaria Survey of the People Living on the Minj River in the Western Highlands of New Guinea. The Medical Journal of Australia, Nov. 13, 1954:782-90.
  • —— 1955-56. The Epidemiology of Malaria in the S.W. Pacific: Changes Associated with Increasing European Contact. Oceania, 27:136-42.
  • Blaikie, P. and H. C. Brookfield (eds), 1987. Land Degradation and Society. London, Methuen.
  • Bleeker, P., 1983. Soils of Papua New Guinea. Canberra, Commonwealth Scientific and Industrial Research Organization and Australian National University Press.
  • Boserup, E., 1965. The Conditions of Agricultural Growth: The Economics of Agrarian Change under Population Pressure. London, Allen and Unwin.
  • Bowers, N., 1968. The Ascending Grasslands: An Anthropological Study of Ecological Succession. Ph.D. thesis, Columbia University.
  • Brookfield, H. C., 1964. The Ecology of Highland Settlement: Some Suggestions, in J. B. Watson (ed.)., New Guinea: The Central Highlands. Special Publication. American Anthropologist, 66:20-38.
  • —— and P. Brown, 1963. Struggle for Land: Agriculture and Group Territories among the Chimbu of the New Guinea Highlands. Melbourne, Oxford University Press.
  • Brown, P., 1978. New Guinea: Ecology, Society and Culture. Annual Review of Anthropology, 7:263-91.
- 357
  • —— and A. Podelefsky, 1978. Population Density, Agricultural Intensity, Land Tenure and Group Size in the New Guinea Highlands. Ethnology, 15:211-38.
  • Buchbinder, G., 1973. Maring Micro-Adaptation: A Study of Demographic, Nutritional, Genetic and Phenotypic Variation in a Highland New Guinea Population. Ph.D. dissertation, Columbia University.
  • Burton, J., 1983. A Dysentery Epidemic in New Guinea and its Mortality. Journal of Pacific History, 18:236-62.
  • Chambers, R., 1983. Rural Development: Putting the Last First. London, Longmans.
  • —— and B. P. Ghildyal, 1985. Agricultural Research for Resource—Poor Farmers: The Farmer—First and Last Model. Institute of Development Studies, University of Sussex, discussion paper 203.
  • Crane, G., 1985. Report to the Southern Highlands Research Committee on Malaria Surveys of the Lake Kopiago, Erave and Poroma Districts of the Province. Mendi, Office of Project Co-ordination. (mimeo.)
  • Crittenden, R., 1982. Sustenance, Seasonality and Social Cycles on the Nembi Plateau, Papua New Guinea. Ph.D. thesis, Australian National University.
  • —— and J. Baines, 1985. Assessments of the Nutritional Status of Children on the Nembi Plateau in 1978 and 1980. Ecology of Food and Nutrition, 17:131-47.
  • —— and J. Baines, 1986. Seasonal Dimensions to Child Malnutrition on the Nembi Plateau, Southern Highlands Province, Papua New Guinea. Human Ecology, 14:191-224.
  • Department of Health, Mendi, 1980. Health Problems of the Southern Highlands. (Unpublished report.)
  • Feachem, P. G., 1977. Environmental Health Engineering as Human Ecology: An Example from New Guinea, in T. P. Bayliss-Smith and R. G. Feachem (eds), Subsistence and Survival: Rural Ecology in the Pacific. London, Academic Press.
  • Floyd, C. N., R. D. B. Lefroy and E. J. D'Souza, 1987. Composting and Crop Production on Volcanic Ash Soils in the Southern Highlands of Papua New Guinea. Mendi, Department of Finance, Southern Highlands Rural Development Project, AFTSEMU Technical Report No.12. (mimeo.)
  • Franklin, K. J., and J. Franklin, 1978. A Kewa Dictionary with Supplementary Grammatical and Anthropological Materials. Canberra, Department of Linguistics, Australian National University.
  • Glasse, R. M., 1968. Huli of Papua New Guinea: A Cognatic Descent System. The Hague, Mouton.
  • Golson, J. 1980. The Ipomoean Revolution Revisited: Society and the Sweet Potato in the Upper Wahgi Valley, in A. Strathern (ed.), Inequality in New Guinea Highlands Societies. Cambridge, Cambridge University Press.
  • Gordon, R., and M. J. Meggitt, 1985. Law and Order in the New Guinea
- 358
  • Highlands: Encounters with Enga. Hanover and London, University Press of New England.
  • Harris, G. T., 1972. Labour Supply and Economic Development in the Southern Highlands of Papua New Guinea. Oceania, 43:123-39.
  • Hide, R. L., 1981. Aspects of Pig Production and Use in Colonial Sina-Sina, Papua New Guinea. Ph.D. dissertation, Columbia University.
  • Hughes, I., 1978. Good Money and Bad: Inflation and Devaluation in the Colonial Process. Mankind, 11:308-18.
  • Ivinskis, V., et al., 1956. A Medical and Anthropological Study of the Chimbu Natives in the Central Highlands of New Guinea. Oceania, 27:143-57.
  • Karanakaran, K., 1980. Malaria in the Nembi Valley, Southern Highlands Province. Mendi, Department of Health. (mimeo.)
  • Kimber, A. J., 1972. The Sweet Potato in Subsistence Agriculture. Papua New Guinea Agricultural Journal, 23:80-102.
  • Lederman, R., 1981. Sorcery and Social Change in Mendi. Social Analysis, 8:15-27.
  • McCarthy, J., 1970. New Guinea Journeys. Adelaide, Rigby.
  • Meggitt, M. J., 1977. Blood is their Argument: Warfare Among the Mae Enga Tribesmen of the New Guinea Highlands. Palo Alto (California), Mayfield.
  • —— 1981. Sorcery and Social Change among the Mae Enga of Papua New Guinea. Social Analysis, 8:28-41.
  • Morren, G. E. B., 1977. From Hunting to Herding: Pigs and the Control of Energy in Montane New Guinea, in T. Bayliss-Smith and R. G. Feachem (eds), Subsistence and Survival: Rural Ecology in the Pacific. London, Academic Press.
  • Nelson, H. E., 1971. Disease, Demography and the Evolution of Social Structure in Highland New Guinea. Journal of the Polynesian Society, 80:204-16.
  • Peters, W., S. H. Christian and J. L. Jameson, 1958. Malaria in the Highlands of Papua New Guinea. The Medical Journal of Australia, September 27, 1958:409-16.
  • Radford, A. J., Van Leenwen and S. H. Christian, 1976. Social Aspects in the Changing Epidemiology of Malaria in the Highlands of New Guinea. Annals of Tropical Medicine and Parasitology, 70:11-23.
  • Rappaport, R., 1968. Pigs for the Ancestors: Ritual in the Ecology of New Guinea. New Haven, Yale University Press.
  • Read, K. E., 1986. Return to the High Valley: Coming Full Circle. Berkeley and Los Angeles, University of California Press.
  • Richards, P., 1985. Indigenous Agricultural Revolution. London, Hutchinson.
  • Rowley, C. D., 1965. The New Guinea Villager: A Retrospect from 1964. Melbourne, Cheshire.
  • Stocking, M., and N. Abel, 1981. Ecological and Environmental Factors for the Rapid Appraisal of Natural Resources. Agricultural Administration, 8:473-84.
- 359
  • Strathern, A. J. (ed.), Ongka: a Self-Account by a New Guinea Big-Man. London, Duckworth.
  • Waddell, E., 1972. The Mound Builders: Agricultural Practices, Environment and Society in the Central Highlands of New Guinea. Seattle, University of Washington Press.
  • Watson, J., 1977. Pigs Fodder and the Jones Effect. Ethnology, 16:57-70.
  • Wood, A. W., 1984a. Soils, in B. Allen (ed.), Agricultural and Nutritional Studies on the Nembi Plateau, Southern Highlands. Department of Geography Occasional Paper No.4. Mendi, University of Papua New Guinea and the Southern Highlands Rural Development Project.
  • —— 1984b. Land for Tomorrow. Subsistence Agriculture, Soil Fertility and Ecosystem Stability in the New Guinea Highlands. Ph.D. thesis, University of Papua New Guinea.
  • Wormsley, W., 1985. Report to the Enga Rural Development Project (Enga Yaka Lasemana) on the Results of the Research Project into Tribal Fighting. Wabag, Enga Yaka Lasemana. (mimeo.)

- 360 Page is blank

1   Includes all territory claimed by the clan or clan-pair as taken from air photographs (1:40,000) and survey sheets (1:100,000).
2   Land on the plateau in 1980 neither primary nor well-established secondary growth—that is, only land currently cultivated and fallow, and associated pockets of primary and secondary forest, and unusable land.
3   These clans have unsurveyed clan lands to the west in the Waga valley and the figures are estimates.
4   Figures based on census books and 1973 village directory and therefore subject to error (Brown and Podelefsky 1978).