Mabu: Huli Subsistence and Production

by Dr. Chris Ballard

The Baruya, in addition to their empirical knowledge in considering soils as agricultural materials, attempt to explain with the help of their myths and legends the invisible origin and structure of the visible world. Their myths form the basis of agricultural practices that enable them to have a direct effect on the invisible. Thus an abundant yield of sweet potatoes is dependent not only on the skillful choice of the best black soil… but also on the efficiency of magic formulae inherited from the ancestors. The knowledge of the secret names of the sweet potatoes gives man power over them. Agriculture is thus a subtle combination of technical and magical skills and is lived and thought of as such. Oilier, Drover and Godelier 1971, p.41

The Idea of The Garden
Much as the Huli accounts of their “natural” environment outlined in Chapter B2 draw upon and reflect a culturally specific metaphysics, the ways in which subsistence practices are imbued with meaning for Huli shed light on the broader field of Huli society and its historical constitution. Pursuing a similar line, Weiner (1991:5) has defined subsistence activity as ‘all the daily intentional movements and activities through which the Foi inscribe their agency and identity upon the earth’. There flows from this perspective the possibility, which this chapter seeks to explore, that subsistence practices bear the inscription of much past activity, evident both explicitly in Huli discourse and also in ways not immediately apparent to Huli themselves. I also seek both to demonstrate the difficulty of distinguishing “techniques” from “beliefs” and to illustrate the underlying logic or logics that serve to relate the two.

This chapter documents the range of crops (B4.2), of domesticated animals and wild game (B4.3) and of subsistence techniques (B4.4) that are known and available to Huli. Running through this analysis is an emphasis on the imprint of earlier subsistence logics. Huli people of the Tari region are now heavily dependent upon a single food staple, sweet potato, the bulk of which is cultivated using a single technique, mounding. Yet, unusually within the context of the Highlands region, Huli retain a keen sense of the historical priority of other crops and other techniques and there is thus a discourse of the adoption and transformation of new crops, new techniques and new ideas. Though it is by now an ancient transformation, the historic development of an agricultural system founded on fixed gardens (mabu), understood here as the emergence of the “idea” of the garden, is traced as an illustration of the relationship between ideas and techniques. In conclusion (B4.5), the parts played by the organisation of labour and the demands of different forms of consumption in structuring the nature of Huli production are described. An account of the variable exploitation of different environmental zones within Huli territory sets the terms for an analysis of the role played within broader local economies by wetland use, which is the subject of a more specific study in Part C.

B4.2 Crops
As amongst other Highlands groups, crops are imbued by Huli-speakers with a significance that far exceeds their nutritional value. Prominent in myths, in metaphor and in general Huli discourse, the staple vegetable foods are classically “good to think with”. The register of crop species grown by Huli people in the Tari region is extensive. Appendix B7 lists all of those crops of the region known to me from gardens and local-produce markets and through conversation. Figure B14, which illustrates the frequency of occurrence of all crops in 231 garden plots at Dobani parish in Haeapugua, gives some impression of the range of crops grown in a single locality. As a comparison of these two lists suggests, not all of the species and cultivars known to Huli are to be found in any one area, but local variation in the register tends to reflect altitudinal restrictions on availability or yield rather than localised preferences or differential diffusion; these altitudinal variations apart, it is thus possible to speak of a broad “Huli” crop register.

Figures B15 and B16, drawing on the results of Bourke (n.d.) and my own observations in the Tari region, show both “usual” and “extreme” altitudinal limits for most of the crops listed in Appendix B7, set against the altitudes of the major basin and valley floors. The usual limits are those within which crops commonly grow in the Highlands region, with viable yields. The extreme limits describe observations at heights below or above the usual minimal or maximal altitudes and represent atypical, isolated plantings, either as experiments with little expectation of success, or as instances of unusual microclimates. A further distinction is made between those crops available to Huli before (Figure B15) and after contact in 1934 (Figure B16). Of the observations to be made on these figures, the most significant for this thesis is the scope introduced by sweet potato for the extension of permanent settlement to higher altitude locations such as the Margarima and Lebani valleys. Of the major pre-contact crops available to Huli, only sugarcane, tobacco, Highland pitpit, rungia, oenanthe and karuka pandanus could have been grown safely within their altitudinal limits at these two higher valleys. Other crops that may have been significant at lower altitudes, such as Pueraria lobata, lima beans, gourds, Job’s tears and, critically, taro, are all capable of producing at altitudes of between 2200 m and 2300 m, but not at viable yields or with sufficient reliability to be employed as staples. A further suite of pre-contact crops grown in the lower valleys, including banana, cucumber, Amaranthus tricolor, yams and ginger, is largely unproductive in the higher valleys. ¹

The contrast between the productivity of taro and that of sweet potato between 2000 m and 2700 m is critical for the history of high-altitude settlement. ²

Table B7 summarizes data on maturation rates for taro and sweet potato in the Highlands region; although taro generally matures at between 7 and 12 months in the intermediate valleys below 2000 m, estimates for the rate at which it matures above 2000 m range from a minimum of 12 to a maximum of 28 months. ³ Given the strategies described for sweet potato production at high altitudes, where regular crop-destroying frosts require a period of replanting with a crucial lapse before the first crops after a frost (Clarke 1989; see also B5.2), the long maturation rate for taro effectively rules it out as a primary staple candidate above 2000 m. There is also some evidence for a decline in taro yield at the highest altitudes, though relevant data is very limited: yields of between 10.4 and 24.8 tonnes per hectare per year (t!ha/yr) have been reported from altitudes between 1400 m and 1550 m, but Bayliss-Smith’s (1985a: Table 5) experimental plots at about 2200 min the Tambul basin yielded only between 2.2 and 5.2 t/ha/yr; Goodbody (in press: Table 35) records yields in Simbu Province of 17.8 t/ha at 1500 m, compared with only 11.4 t/ha at 2400 m. While pre-Ipomoean agriculture that employed taro as one of a suite of staples would have been feasible in the higher basins such as Margarima and Lebani, Bayliss-Smith (1985a:313) concludes from his Tambul study that, at such altitudes, it ‘seems unlikely that a taro-dominated subsistence economy based on wetland drainage would be viable’. Bourke (n.d.) suggests a mean limit of 2250 m for pre-Ipomoean taro subsistence, but even this renders taro-based settlement of the Margarima (2200 m) and Lebani (2300 m) basins as marginal propositions.

Sweet potato generally matures approximately twice as quickly as taro; Huli people, when describing garden sequences, frequently identify the second sweet potato crop in a given stage as the appropriate time to harvest the first crop of taro (see B4.4). On the basis of the figures in Table B7, sweet potato appears to mature at between 5 and 8 months between 1500 m and 2000 m, and between 7 and 12 months above 2000 m; the consequences of this quicker rate of maturation for sweet potato by comparison with taro for high valley settlement have been interpreted by Clarke (1977:161) as an expansion in the scope for high-altitude production, if not an extension of settlement. Since contact, the crop register at higher altitudes has been significantly enhanced through the introduction of the white potato, cabbage, Chinese cabbage, corn, pea and carrot, a revolution that has also had consequences for the security of food supply in the intermediate basins (B5.2).

My own surveys of the crop registers of the intermediate Huli basins (Haeapugua, Mogoropugua, Dalipugua, Benalia, Paijaka, Tari and Komo) suggest that they are fairly similar. However, the altitudinal range of these intermediate basins, between about 1500 m and 2000 m, also contains usual limits for a number of crops, introducing the possibility of signficant variations in the productivity of these species within and between basins. Yam (Dioscorea alata,), lowland pitpit, diploid bananas and ginger of the pre-contact crops, and winged bean, pineapple, orange, peanut, cassava and aibika of the post-contact crops, all reach their upper altitudinal limits between 1700 m and 1900 m, a range of immediate relevance to settlements in the Dalipugua and Mogoropugua basins and on the Paijaka plateau. Another possibly significant source of variation appears to be in the distribution of cultivars of the same species, though the extent to which this is a function of local synonymic differences in identification, discussed below, is not clear. At the lower end of the altitudinal range of Huli settlement, the most significant threshold is that for sago; the staple for all Huli neighbours to the south, sago has been transplanted to lower-lying areas of the Lower Tagali valley where it is marginally above its usual altitudinal maximum. Sago is also harvested by fringe Huli groups scattered to the south and east of Benalia, towards Lake Kutubu. ⁴

Apart from the impact of sweet potato and subsequently of the post-contact crop introductions on the permanence of settlement above 2000 m, the other important variation in the altitudinal range of pre-contact staples relates to the distribution of the different pandanus species. The two key pandanus crops are marita pandanus (abare: Pandanus conoideus), and karuka pandanus (anga: Pandanus julianettii). The usual mean maximum for marita pandanus is about 1700 m, extending on occasion to an extreme maximum of 1980 m; this renders marita available to the lower-lying Huli settlements of the Lower Tagali, Benalia, Komo, Haeapugua and the southern part of the Tari basin, but not (at least in significant numbers) to Dalipugua, Mogoropugua, the Paijaka plateau or the higher altitude valleys. By contrast, the distribution of karuka pandanus, with a usual minimum of only 1800 m (extended in extreme instances to 1450 m), is effectively restricted to the upper slopes around the Paijaka plateau and Mogoropugua and to the higher altitude valleys of Margarima and Lebani. In these higher valleys, karuka is cited as the primary attraction for settlement; indeed the history of pre-Ipomoean settlement above 2000m can perhaps be viewed effectively as a record of the fluctuating significance over time of karuka pandanus.
As most of the basins of intermediate altitude lie close to, or within, the gap left between the usual maximum for marita and the usual minimum for karuka, the vast bulk of the Huli population has to rely upon extended kin and affinal ties to gain access to the centres of production for both pandanus species. Marita and karuka harvests are of considerable significance for most Huli, but entail either trade or temporary migration for the majority, both of which involve activation of the necessary kin and affinal ties or trade links with the lowland Etoro and other duguba groups (Dwyer 1990:63). For Huli at Haeapugua, located within the “pandanus gap”, the need for such links, both above and below, is keenly apparent:
They call out to those who don’t have karuka pandanus [anga] to come and eat. Those in the headwater regions [manemane] call out to us at Haeapugua. Those in the lower reaches [wabiwabi] call out to us at Haeapugua to come and eat marita pandanus [abare], or they bring it to us. Pudaya, 3.11.92, 92/4B:199-238

Thus, even though many Huli lack direct access to marita and karuka pandanus, the nature of local social structure permits their distribution throughout Huli territory and further reinforces the proposition of a single Huli-wide crop register. As for other plants or animals, Huli taxonomies for crops tend to be shallow, with little emphasis on higher-order groupingS. There are no terms for unique beginners, for example, such as “plant” or “animal”. With only a handful of exceptions, Huli “folk” generic labels correspond instead to scientific species terms: hina thus refers to Ipomoea batatas and would not be incorporated under any more inclusive term, other than pindu (“thing”) or tomo (“food”). The few exceptions to this general observation are listed in Table B8, where there is some suggestion that the most common Huli generic identifications that incorporate more than one scientific species are those that group both domesticated and wild species under a single generic label. In place of a complex higer-order taxonomy, there is a wide proliferation in Huli of varietal, or more accurately, cultivar terms which are commonly combined with the superordinate Huli generic category (dama hina for the dama sweet potato cultivar, for example). The profusion of cultivar terms, with over a hundred terms for sweet potato cultivars and more than thirty each for taro, karuka pandanus and banana (Appendix B7), certainly reflects local variation in nomenclature. This is illustrated below with reference to sweet potato cultivars. But the task of matching Huli and scientific taxonomies is probably less rewarding than the insights afforded by the ways in which crops and crop terms are deployed by Huli, the role of crops in the Huli imagination. Certain crops, for example, are accorded praise terms (kai. mini) which are used in formal narratives and gamu spells. Table B9 lists a number of praise terms for crops either identified in narrative transcripts or elicited through interviews. Without exception, praise terms are reserved for crops known to the Huli before contact in the 1930s, reflecting the keen sense of historical precedence or tene discussed in B3.3. A “covert” taxonomic system for crops emerges in Huli discourse and praxis, founded on the relative antiquity of different species and expressed in terms of a moral gradient from earliest (ala or bamba ore: “before-truly”) to most recent (ayu ore: “now-truly”).

While post-contact crops have, in the more recent past, clearly been excluded from ceremonial contexts, finer distinctions between pre-contact crops of differing antiquity were formerly evident in ritual performances. The gebe hagama rites performed at major gebeanda ritual sites such as Gelote involved the reconstruction of what was deemed to have been the lifestyle of the earliest female ancestor, Memeleme. ⁵ This rite required the performing ritual leaders to refrain from eating historically recent foods such as sweet potato and subsist instead upon taro and sago, the latter imported for the rite from the Papuan Plateau. This identification of historical sequences for the introduction of different species and cultivars is a common element in discussion about crops. Usually, differences in crop antiquity are expressed in relative terms through reference to crop pairs: taro precedes sweet potato, cucumber precedes pumpkin. A broader historical framework is in fact articulated by Huli people, drawing on these simpler temporal oppositions to construct a sequence running from the earliest time, ira goba naga (“wood-rotten-time” or “wood-rotten-eaten”) or dama angi (“the time of dama spirits”), a period when ancestral dama are held to have eaten the decayed woods found deep in the main swamps, together with both karuka and “wild” pandanus, nano mushrooms and an unidentified wild tuber (homa bawi; Dioscorea nummularia I Pueraria sp.) ⁶ These crops, all considered to grow without human assistance, are designated as wild (pariwali) and are explicitly associated with dama spirits8•

Following this period is a second, ma naga (“taro-time”), in which people now recognized as fully human ate taro, cucumber (bambo; Cucumis sativus), bottle gourd (mbagua; Lagenaria siceraria) and yam (nandi; Dioscorea alata), but dug no gana ditches for proper mabu gardens (see B4.4) and raised no pigs9. ⁷ The relative antiquity of taro is a recurrent theme in Huli culture and the term for taro, ma, is possibly the root of an important lexicon which includes such words as mabu (garden), malu (genealogy, charter), mamali (ancestor), marne (father), mamabuni (mother’s father) and mana (custom, precedent) (see Goldman 1981a:65). Most of these terms, evidently, refer to qualities or states associated with maleness or to knowledge that is notionally the preserve of men; there is a tendency for sons to inherit their father’s taro stocks and men will sometimes refer to their older stocks as their father’s taro. Taro itself was used in significant rituals, such as the gebe hagama rite mentioned above and the ma hiraga rite performed to ensure the strength of newly born infants (Frankel 1986:54). By comparison with other Highlands groups where taro has more recently been a dominant staple, such as the Bimin from whom Bayliss-Smith (1985b) has recorded at least 108 cultivar terms, the knowledge of fewer than 40 cultivar terms for taro by HuH-speakers is probably an accurate reflection of the diminished importance of taro as a subsistence staple in the Tari region. Nevertheless, many people nominate the tirima and gihagua cultivars, which are often identified in myths as being associated with early ancestors, as the two earliest cultivars of taro. Specific directions of origin are also recalled: tirima is said to have been introduced from the Duguba of the Papuan Plateau, and bogaya, as its name suggests, from the Bogaya of the lower Strickland Gorge area. Two cultivars, miti and simbu (“Chimbu”) or dandayi (“police”), are explicitly identified as post-contact introductions.

M a is often referred to in discussion as an earlier counterpoint to the current Huli staple, sweet potato (hina). As the earlier staple, taro rather than sweet potato was employed in ritual contexts, but the historical and current significance of sweet potato is such that it is well incorporated in figurative speech (e.g. Goldman 1993:388) and has its own praise term for use at formal occasions. ⁸ So important to the Huli economy is sweet potato that no meal is considered complete without it, and it has assumed the status of a generic term for all food. Famine is thus described as hina gari (“sweet potato-lack/hunger”); as with similar terms describing famine in terms of temporary shortages of preferred staples such as taro among the Telefolmin or at Wamira, hina gari refers more accurately to a deficiency in the supply of the one crop. ⁹ Appropriately, the concept of ancestral famine is referred to in terms of a lack of taro (ma gari).

Hina is clearly understood to have emerged or to have been introduced to the Tari region within what might be described as the temporal scope of Huli history (see Appendix B6). Informal recitations of genealogies are often punctuated with references to the first individual to have “held” sweet potato vines. Although ma serves in some genealogies to denote the emergence of human, as distinct from dama ancestors, many genealogies explicitly equate human-ness with the advent of hina, and the emergence of successively newer sweet potato cultivars is identified with the passage of different generations: ‘hina mbira, daba mbira’ (“for each sweet potato [cultivar], a generation”). While I have encountered no myths of any length that account for the origins of ma, the origins of hina in the Tari region are widely associated with the ancestors of Digima hameigini ¹⁰ In Narrative B5, a Digima tene sketches the outlines of this myth. No external origin is identified for this first cultivar, which is known either as muguba (hence the praise term “alu muguba” for all hina) or digi hina (after Digima clan); Wood (1984, Vol.I:232), however, records Huli origin myths for sweet potato that trace its diffusion into the Tari basin from the north and west (see Chapter D2). Yam (Dioscorea alata), an unspecified bean and the bottle gourd are all identified as having been present at this time, though the narrative intention of this scheme is clearly to provide a context that explains why people should have attempted initially to trail sweet potato vines up sticks, following a practice appropriate to the other three crops. ¹¹ As another man put it, sweet potato was only recognised as such, in a sense only became hina, when it was correctly planted in the soil:

Dindi dugu dambi hayagola, ba timbuni, digi hina, digi hina lenego When it was covered with soil, it grew large and it was said that this was [really] digi hina
Digi-Malingi, 13.6.91, 19/9A:382-end

In narratives of varying degrees of complexity, most other clans then trace the acquisition by their ancestors of digi hina from Digima clan. At Haeapugua, Miniba and Wenani clans are held to have been the first to receive digi hina from Digima; the other clans of the basin acknowledge their receipt of the first hina through these two clans. Narrative B6 tells of a Miniba woman who returned from Digima carrying digi hina, and comments incidentally on the preceding staples and the ensuing sequence of introductions of other sweet potato cultivars.

The possibility of dating the arrival of sweet potato through estimates of the antiquity of those ancestors who are said to have been the first to “hold” sweet potato was explored during the documentation of clan genealogies. The Digima ancestor, Digiwa, who first cultivated sweet potato, lies ten generations above the narrator, DigiMalingi. Given Malingi’s estimated date of birth (EB) in 1925, Digiwa’s genealogically estimated date of birth (GEB; see Appendix B6), assuming average generational lengths of between 30 and 40 years, would appear to lie between ?1625 and ?1725 AD, suggesting that his active adult life spanned between an absolute maximum of ? 1645 and an absolute minimum of ?1780. At these generational depths it is difficult to be more accurate than this, but the records from other clan genealogies of generational depth above the adult narrator show a considerable degree of concordance with the estimate of ten generations’ depth for Digiwa. At Dalipugua, estimates ranged from 8 to 11 generations (from 10 clans), at Mogoropugua from 8 to 10 generations (5 clans), at Lebani from 8 to 10 generations (7 clans), at Haeapugua from 7 to 10 generations (8 clans), at Margarima 9 generations (1 clan) and at Komo 10 generations (1 clan).

In his review of Wiessner and Tumu’s (in prep.) attempts to date the arrival of sweet potato using Enga genealogies, Jorgensen (in press:27-28) has wisely cautioned that the location of sweet potato towards the end of the mythological period may function semiotically to mark the assumption of culture and the passage from non-human to human … in this way Enga history may be said to begin with the sweet potato, but this is probably best understood in a broadly metaphorical sense.

Huli clan origin narratives differ as to whether possession of sweet potato marks the historical transition from dama to human ancestors, or whether sweet potato was adopted by taro-eating ancestors who had already been human for some generations. All are in agreement, however, on the fact that the adoption of sweet potato was instrumental in the transformation of ancestors into “modern” humans, with “modern” forms of exchange such as the use of pigs in bridewealth. ¹² Jorgenson’s argument develops a circularity under these conditions: sweet potato marks the transition to modernity and its adoption is thus correctly identified as a metaphor for the emergence of human modernity; yet, as Huli have it, it is only with the adoption of sweet potato that their ancestors were able to engage in intensive pig production and to transact and otherwise behave as modern humans. Similarly, the elaboration of Huli genealogies from ten generations above the current adults could be taken either to reflect the maximum temporal extent of Huli history or a post-Ipomoean increase in population and the creation of the need for detailed genealogical recall.

The list of sweet potato cultivars given in Appendix B7 contains cultivar terms recorded throughout Huli territory. It should be stressed that no individual would be able to recall all of the terms listed here, let alone identify the different named cultivars. In practice, a far narrower range of cultivars is actually employed in any one area: in a 1991 survey of 231 garden plots at Dobani in the Haeapugua basin, four cultivars were found to dominate overwhelmingly the range of planted cultivars, and only a further seventeen named cultivars were present (Figure B17). As Heider (1969) has pointed out, the range of cultivar terms for sweet potato in the Highlands, while reflecting the importance of the crop and its propensity for somatic mutation (Yen 1974), owes much to the proliferation of synonyms for the same cultivars. Experiments by both Heider (1969) and Sillitoe (1983:141) suggest that there is little agreement in the application of cultivar terms, either between people or by the same individual on different occasions. ¹³ Powell (Powell with Harrison 1982: Table 4) has described cultivar characteristics for sweet potato in the Haeapugua basin, but the limited success of my attempts, in conjunction with Jill Clapin at replicating these descriptions even when working with some of the same people who originally assisted Powell, would appear to support the conclusions of Sillitoe and Heider. This is not to imply that cultivar terminology is totally inconsistent in the region; many people are aware of the existence of synonyms for the same cultivar, but the problem may have been compounded since contact by the extensive replacement of pre-contact cultivars by a large number of higher-yielding cultivars introduced to the Highlands region by European plantation owners in the Wahgi valley and directly to Tari by government departments, and will obviously require a more intensive study than any attempted so far. ¹⁴

A more certain distinction, at least in Huli discourse, is made between earlier and sweet potato cultivars. Table B10 lists the responses of nine older men from four erent basins to questions about the relative antiquity of different cultivars. The onses generally distinguished between four gross phases of introduction: the earliest ivar (muguba); those introduced subsequently, but prior to contact with the inistration; a handful of new cultivars introduced either deliberately or incidentally early administration patrols prior to 1950 (see B3.2); and those cultivars introduced ugh government agencies, markets or the travels of Huli people to other areas of Papua New Guinea. Finer distinctions are made within each of the three latter phases, ally in connection with specific individuals in clan genealogies. ))

These three major historical “staple phases” identified by Huli (ira goba naga, ma a and hina naga) surface in a much wider range of contexts than the discussion of ps, invoking or articulating connections with options of temporal progression, such entropy (Chapter B5), and change in such matters as agricultural technique, the status pigs and pork and the nature of sociality. In each case, the perceived staple comes to d for a specific complex of techniques, forms of exchange and other, less significant. Most of the other crops listed in Appendix B7 are also located temporally within scheme; Table B 11, a compilation of observations made in both formal and informal contexts, provides a historical register of these crops, with the addition of a temporal division employed by Huli in describing changes to the crop register contact.

Although the continuing importance of hina remains unquestioned, there is little sence for any broad current commitment to other pre-contact crops of the kind described for taro amongst Telefolmin by
Jorgensen (in press). A few older men and men persist in growing early cultivars of sweet potato, claiming that they prefer their iliar taste, but the introduction of new cultivars and new crop species since contact radically altered the emphasis placed on other “traditional” crops. Seed was berately carried on early patrols of the 1930s and 1940s to Tari (see Appendix B1) other crops spread rapidly from the government posts established during this period ake Kutubu and Wabag. Com, in particular, made a notable early impact, ounted in a rather quaint “myth” by a patrol officer in the 1950s ¹⁵

These three major historical “staple phases” identified by Huli (ira goba naga, ma a and hina naga) surface in a much wider range of contexts than the discussion of ps (?), invoking or articulating connections with notions of temporal progression, such entropy (Chapter B5), and change in such matters as agricultural technique, the status igs and pork and the nature of sociality. In each case, the perceived staple comes to d for a specific complex of techniques, forms of exchange and other, less significant. Most of the other crops listed in Appendix B7 are also located temporally within scheme; Table B 11, a compilation of observations made in both formal and rmal contexts, provides a historical register of these crops, with the addition of a r temporal division employed by Huli in describing changes to the crop register e contact.

Although the continuing importance of hina remains unquestioned, there is little sence for any broad current commitment to other pre-contact crops of the kind cribed for taro amongst Telefolmin by
Jorgensen (in press). A few older men and men persist in growing early cultivars of sweet potato, claiming that they prefer their iliar taste, but the introduction of new cultivars and new crop species since contact radically altered the emphasis placed on other “traditional” crops. Seed was berately carried on early patrols of the 1930s and 1940s to Tari (see Appendix B1) other crops spread rapidly from the government posts established during this period ake Kutubu and Wabag. Com, in particular, made a notable early impact, counted in a rather quaint “myth” by a patrol officer in the 1950s ¹⁶

The frequency of occurrence of crops in a 1991 survey of 239 Dobani garden plots (in 52 gardens) gives a very rough indication of the relative contemporary importance of different crops (Figure B14). The nature of the data severely understates the actual dominance of sweet potato, by number, weight of yield or area planted, but may give a more accurate impression of the increasing popularity of some of the post-contact crops: Xanthosoma taro appears as frequently as the pre-contact Colocasia taro, and pumpkin, corn and choko have also been widely adopted. There is also some evidence in Figure B 14 for the decline of Rorippa, cucumber and yam, with the last two crops each appearing in only one of the 239 plots surveyed. Though coffee is probably the most significant cash crop in the Tari region, problems of distribution and the small size of holdings have prevented the development of an industry of the scale evident in other Highlands provinces. ¹⁷ A 1965 survey counted 5293 coffee trees in the Tari region, of which only 180 were in the Haeapugua area (Hunter 1964/65: Appendix A). The coffee “boom” of the late 1970s and early 1980s, and the establishment of a road link between Tari and the rest of the Highlands, saw a rapid increase in coffee production in the Tari region. Almost all coffee production has been at the level of individual households; a map of land use in the Tari and Haeapugua basins shows clearly the fragmented distribution of coffee gardens in 1978 (Wood and Allen 1982). Subsequent attempts to establish larger plantations on communal land, as in the swamp at Mogoropugua, have all failed. In 1991 at Haeapugua, there were 2232 coffee trees at Dobani parish (62 trees per hectare of dryland garden, 15 trees per person), though most were poorly tended and owners were reluctant to harvest given the depressed coffee prices of the time.

While there is a pragmatism about innovations, and perhaps even a willingness to experiment in the Tari region which might account for the speed of diffusion of novel crops and the corresponding decline of former staples, successive famines appear as watersheds in the broader history of the adoption of new crops by Huli. The causes of these famines are considered in more detail in B5.2, but their consequences for the crop register are clear. In the course of a major drought and famine during 1941 and 1942, two new sweet potato cultivars, bo and dambera, are said to have been passed rapidly throughout Huli territory and to have yielded spectacularly well for a few years; men and women walked between the various Huli basins carrying bundles of the new vines back to their gardens (see Narrative B6; also Agiru, 10.3.91, Mogoropugua Fieldnotes).

Again, during the 1972 frosts and famine (B5.2), crop failures led to the wholesale replacement of the existing range of sweet potato cultivars by new cultivars and by European potatoes flown in by the colonial government (Oberia, 15.3.91, Lebani Fieldnotes; J.M.Powell pers.comm.) 19. Great emphasis is placed on the selection of new cultivars, and Huli continue to travel long distances, by foot prior to contact and more recently by vehicle, in order to acquire new cultivars and new crop species. Although narratives about events such as the 1941/42 famine assert that the new food crop species and cultivars only emerged (literally “came out”: tagira) during the famine, it is conceivable that they were already present beforehand but only assumed their new significance when the existing staples failed. This adoption of new cultivars after the 1941/42 drought may serve as a model for the initial adoption of sweet potato in the Highlands region. Corlett (1984: 109-110) has documented a major event of forest clearance at high altitudes in the vicinity of Mt Wilhelm at about 300 BP; he relates this to increased human activity at higher altitudes during an exceptional drought event at that date but notes the absence of similar impacts during pre-300 BP drought events. This situation, Corlett suggests, reflects the adoption of sweet potato immediately before the 300 BP event and the establishment of permanent human settlement at higher altitudes. The threat posed to taro by drought is well-documented, Morren and Hyndman (1987:312) recording the Mountain Ok practice of transferring taro stock to swamps during drought events. It is possible to speculate that an extreme drought event following the initial introduction and availability of sweet potato could have promoted its prospects as a staple across much of the Highlands region almost instantaneously (R.M.Bourke pers.comm.) ¹⁸ Indeed, Wiessner and Tumu (in prep.) have recorded several Enga oral traditions in which the initial adoption and widespread distribution of sweet potato is said to have occurred during a great famine, when the existing staples failed.

In much the same way that crops are employed as temporal markers in Huli history, they are also deployed in the process of negotiating the structure of Huli society more generally. The concept of tene, which informs Huli historicity through the privilege accorded to precedence, is also implicated in gender distinction. Ideally, each domesticated crop is opposed to, but also preceded historically by, a wild or undomesticated counterpart: anga mundiya (Pandanus brosimos) thus came before “true” anga pandanus (Pandanus hima or gili (Saccharum edule) before du (Saccharum officinarum); hai garo preceded edible bananas; and “wild” taro (tumbu: Alocasia spp.) was present before people began to eat “real” taro (ma: Colocasia esculenta). ¹⁹ These historical distinctions correspond to a strongly gendered contrast between domesticated crops, equated with the mythic domesticating acts of men, and their wild, feminine counterparts. These wild crops, which are held to have first been planted by the originary ancestress, Memeleme, are known by a collective term, pariwali (“?-woman”). In an important sense, then, all wild forms of plant food are feminine versions of the domesticated male species.

(Reprinted with permission of Dr. Chris Ballard. The Death of A Great Land: Ritual, History and Subsistence Revolution in Southern Highlands of Papua New Guinea. A thesis submitted for the degree of Doctor of Philosophy of The Australian National University, Canberra, January 1995. pp. 69-76)

1. Single banana trees are present in individual gardens of the Lebani and Margarima valleys, but their poor productivity stresses the marginal nature of the “extreme” limits listed in Figure B15. Much of the wide range of uses for banana leaves and stems common at lower altitudes is assumed in these two valleys by cordyline. [↩]
2. Bourke (n.d.) nominates 2700 mas the upper limit for arable agriculture in Papua New Guinea, an altitude which covers all of the major valley basins under discussion here (listed in Table B 1). [↩]
3. The range of these estimates is explained in part by the results of a brief survey in the Porgera valley, to the north of the Tari basin, which suggest that local variations in soil quality and in micro-climatic conditions can exercise a considerable influence over maturation periods in taro and sweet potato, with the full maturity of crops from sites at similar altitudes varying by as much as six months (Hughes and Sullivan 1990:273). [↩]
4. There has been extensive trade in sago between Komo Huli and Etoro-speakers of the Papuan plateau, described by Dwyer (1990:68). [↩]
5. Memeleme is a synonym used widely in the Haeapugua area for the Huli ancestress, Tia Nangume or Dindi Ainya (see Narrative Bl). [↩]
6. Another temporal sequence constructed around crop staples employs the metaphor and vocabulary of a genealogy, with the fern yagua bearing an unidentified wild tuber (homa bawi: ?Dioscorea nummularia I Pueraria sp.), which bears yam (nandi: Dioscorea .a.l.alli), which in tum bears sweet potato (hina: Ipomoea batatas). Taro (rna: Colocasia esculenta), on this account, is “descended” from wild taro (tumbu: Alocasia spp. [↩]

7. Ma naga is also described as rna angi (“the time of taro”) or rna nga (“[when] taro [was] present”). A further epochal distinction, intermediate between ira goba naga and rna naga, is occasionally made:
   this is anga naga (“karuka time”), a time when karuka pandanus was eaten, an observation that reflects
   the symbolic and dietary significance of the crop, but one which is seldom elaborated upon within epochal schemes (Pudaya, 27.9.89, 89/3A:32-51). [↩]

8. Sweet potato was excluded from ritual sites such as Gelote, within which officiants ate only taro or sago, the latter in commemoration of the lowland origins of the earliest ancestress, Memeleme. [↩]
9. ‘Hungry times [for the Telefolmin] are those in which one eats sweet potato because the supply of taro is short’ (Jorgensen 1981:55); ‘the difference between times of abundance … and times of scarcity … hinges upon the amount of taro available’ despite the relative dominance at Wamira of sweet potato and cassava (Kahn 1986:34); ‘hunger for the Wola starts with a shortage of staple sweet potato, whatever happens to the yields of other crops’ (Sillitoe 1993a:175). [↩]
10. L.Goldman (pers.comm.) reports that he too, has never head of an origin myth (tene te) for taro [↩]
11. Marilyn Strathem (1969: 197) has recorded a very similar myth from the North Melpa area in the Wahgi valley. The bean said to have been present prior to the introduction of sweet potato in the Tari region is invariably named as paboro (Paboro ibu ala: “paboro is ancient”); this is currently the term used by Huli to identify the common bean (Phaseolus vulgaris), which is thought to be a post-contact introduction (R.M.Bourke pers.comm.). Other candidates include the hyacinth bean, though this is usually termed wiru in Huli, and the winged bean (Psophocarpus tetragonolobus), which has been described as present prior to contact in some parts of the Papua New Guinea Highlands (Khan 1976) but which does not appear to have been known or widely used in the Tari area before contact. [↩]
12. Modjeska (1991) has already introduced the term “post-Ipomoean modernism” to describe the emergence of essentially modem forms of Duna exchange and leadership. [↩]
13. Kocher-Schmid (1991:96) suggests that even the terms for modem cultivars, such as the apparently ubiquitous wanmun, are not applied consistently to the same cultivar between regions. [↩]
14. Appendix B7 lists some of the terms identified as synonyms. A nice illustration of the origins of a cultivar synonym is aliga hina, the term given for a cultivar said to have been introduced to Haeapugua through the Wabia area to the east, which is a synonym for barabia hina, the same cultivar introduced from the Duna to the west [↩]
15. A more certain distinction, at least in Huli discourse, is made between earlier and sweet potato cultivars. Table B10 lists the responses of nine older men from four erent basins to questions about the relative antiquity of different cultivars. The onses generally distinguished between four gross phases of introduction: the earliest ivar (muguba); those introduced subsequently, but prior to contact with the inistration; a handful of new cultivars introduced either deliberately or incidentally early administration patrols prior to 1950 (see B3.2); and those cultivars introduced ugh government agencies, markets or the travels of Huli people to other areas of Papua New Guinea. Finer distinctions are made within each of the three latter phases, ally in connection with specific individuals in clan genealogies. [↩]
16. ‘Within the past ten years several European crops have been introduced to the area, down the Wabaga trail, and of these, the natives have shewn increasing interest. The first new crop to be introduced was com, which, from a small start, has developed into a second major crop. The story goes that the first com was thrown away by Mr Taylor’s party during his visit to Hoiyevia [Hoyabia, referring to the 1938 Taylor/Black patrol]. On his departure a solitary stalk sprang up. Being naturally superstitious, this was placed under a “taboo” and until the time it carne to fruition, it was left alone. At this stage, however, one native, more forward than his friends, tried a small piece of corn, found it sweet and good – and lived. From this meagre start developed the growth of corn throughout the nearby area. On the return of natives from the Wabaga area, more corn was imported, and spreading, is now available in greater or lesser quantities, as far south as Pai [Bai]. As mentioned before, this is now being grown on a major scale, being planted along the edges of the potato gardens, and is gaining an assured place in the native diet. Gardens of some 5- 10 acres have been seen, not as sweet potato patches, but as fields of waving com.’ (A.T.Carey 1952:46). [↩]
17. The major cash crop of the 1970s and 1980s in the Tari region has been coffee, notwithstanding the efforts of successive agencies to promote such varied projects as silkworms and cardamom (French and Walter (eds.) 1984). [↩]
18. Brookfield (1989:313) has documented a series of dates for drought events during the 18th and 19th centuries, demonstrating the frequency with which major droughts are likely to have occurred, then and earlier. [↩]
19. Frequent discoveries, at the base of deep swamp ditches, of matted layers of organic material containing what are identified as the seeds and leaves of mundiya and tawa (Pandanus antaresensis), are referred to as proof of the precedence of these wild pandanus species. [↩]