The Sun by Night

Huli Moral Topography and Myths of a Time of Darkness

by Dr. Chris Ballard, Australian National University

An atom of fire in certain cosmological dreams is sufficient to set a whole world ablaze.
Bachelard, The Psychoanalysis of Fire, p. 72

In his remarkable disquisition on fire, Gaston Bachelard explores the mythic substructure of science, seeking to show how reverie, ”the unconscious of the scientific mind” (1987: 10), constitutes the limits to imagination and describes the boundaries of scientific endeavor. His intention is to reveal the “secret persistence of [an] idolatry of fire” (1987: 4), which extends, through the process of reverie, from a primitive phenomenology of affectivity via medieval alchemy to modern science. Bachelard’s proposition of a universal role for fire in reverie, in dream and in the unconscious, finds expression in the cosmology of Huli speakers of the Southern Highlands of Papua New Guinea, for whom fire, as fertile substance in its exterior and most potent form, invokes a register of fluids that fuel the processes of transformation, of growth and decline in the earth and in people. This chapter is one in a series of essays (Ballard 1992, 1994 and in press) in which I attempt to map the immense edifice of Huli cosmological beliefs both through time, as a distinct but changing sense of historicity, and through space, as a grounded cosmography. Other ethnographers of the Huli have indicated something of the scale, coherence and astonishing complexity of Huli cosmology (Goldman 1983; Frankel 1986; Glasse 1995); and my own work quite clearly follows their different leads. The principal concern of this contribution is to describe how myth is implicated in ll-ie (sic) elaboration and transformation of Huli cosmography, the topography of cosmological belief. Huli myths of a time of darkness and the threat of a fiery transformation of the universe provide the means of uncovering the workings of a moral economy of fluids, which in tum enable us to appreciate the essentially hydraulic contours of Huli cosmography.

Where previous works have addressed the transformations in Huli cosmology in its encompassment of the novel narratives and historical trajectories supplied by Christianity (Ballard 1992, in press; see also Frankel 1986: ch. 2) and “development” in the form of mining and petroleum companies (Ballard 1992; see also Clark 1993), this chapter represents a preliminary attempt to describe the fundamental ground, the moral topography, of Huli cosmology. Topography, as J. Hillis Miller (1995: 3) reminds us, extends beyond the simple definition of a written description of place to embrace the sense of “the art of mapping by graphic signs” while simultaneously naming what is mapped; topography is thus a conscious cartographic act that contains within it an unconscious process of identification between language and location. What I want to identify, under these terms, as a ”moral topography” is an account of the moral foundations for a community’s engagement with the land in which topology, the knowledge of landscape, is inflected by an unconscious sense of the atopical, the ”unplace­ able place … that is everywhere and nowhere, a place you cannot get to from here … another name for the ground of things, the pre-original ground of the ground, something other to any activity of mapping” (Miller 1995: 7). A moral topography extends to daily practice, without the mediation of exegesis and beyond the capacity of imagination, as an intimation of the fundamental ground for existence; Huli myths of a time of darkness offer a portal to this place that you cannot get to from here.

A Time of Darkness

Myths that tell of a time of darkness are widespread across much of the north coast and central highlands of mainland Papua New Guinea. While these myths had been recorded independently by individual ethnographers and missionaries, the first attempt to comprehend them as a genre was made by the geographer Russell Blong (1979, 1982) who circulated a questionnaire enquiring after time­ of-darkness myths, collating and comparing the results from 99 different sources in Papua New Guinea. Commonly, time of darkness myths describe an event in the ancient or mythic past during which the sun was obscured and a sand or ash fell from the sky, covering gardens and killing crops and wild animals. At the end of this event, the sun reappeared, and heavy rains washed away the cover of sandy ash.

For volcanologists and archaeologists, the temptation has been to engage these myths fairly literally as testimony to a seventeenth-century fall of volcanic ash, tentatively associated with an eruption at Long Island off the north coast of mainland New Guinea (Blong 1982). Blong’s analysis deals somewhat unprob­lematically with the different narratives (many of them reduced to a series of statements by anthropologist and missionary respondents to Blong’s questionnaire) as a discrete and broadly coherent category; little attention is paid to the cosmological contexts for these myths, to the form of their narration or to their particular salience within the different communities. For Blong, the coherence of the genre derives from the putative association of these myths with a single historical event, rather than the various roles they play as localized mythic narratives. Many respondents to Blong’s questionnaire repeatedly express the opinion that these narratives hold little continuing interest for the communities of which they write and, consequently, scarcely feature in their own ethnographies. The narratives themselves are largely factual, and few communities apparently seek to account for the time of darkness either in terms of cosmological beliefs or any other wider context, such as the expression of local enmities. But the number of respondents who, after years of living with communities, had not heard time­ of-darkness stories until prompted by Blong’s questionnaire to enquire after them, is itself intriguing and suggestive of unexplored possibilities. One of the few regions where time-of-darkness myths have been the subject of some local elaboration is the Duna-Huli-Ipili area of the Southern Highlands and Enga provinces (see Preface, this volume). It is no coincidence that eth­nographers of the Huli and the Ipili, such as Glasse (1963) and Meggitt (1973), were among the few observers to both document time-of-darkness narratives and attempt to couch their analyses in terms of local cosmologies; more recent analyses of Huli versions of these myths have been offered by Goldman (1983), Frankel (1986) and Allen and Wood (1980) and for the neighboring Ipili by Biersack (1991). This localized elaboration is intriguing, for it confounds any of the materialist explanations offered by Blong, bearing no obvious relation to the presumed distribution of the physical effects of the Long Island tephra: Duna, Ipili and Huli all lie close to the extreme margins of the reconstructed distribution of the ashfall and its effects, where one might expect its significance to have been comparably diminished (Blong 1982: 62, fig. 29). In the analysis that follows, Huli accounts of a time of darkness, referred to as mbingi, are engaged not as the means of access to a historical event (though Blong’ s emphasis on history is curiously matched by a Huli insistence on the facticity of myth and its physical signs), but rather for the way in which they illuminate the architecture of Huli cosmology from within. I am more interested to consider how the Huli have taken an event, whatever its historical status, and worked upon it as an element of a wider cosmology, both disclosing and trans­ forming the conceptual structure of their own universe.

The Mana of Mbingi

The Huli term mbingi refers precisely to a “time of darkness,” combining as it does the terms mbi, for “dark” or “night,” and the temporal adverb angi, for “when” or “time of.” Narratives that relate the events associated. with mbingi and prescribe appropriate behavior defy ready identification with any of the standard speech genres described for Huli (Goldman 1983: 62-63), straddling a number of different conventional formats. The narratives themselves are relatively standardized and readily identified as belonging to a specific genre, but rather than being labeled either as folk stories (bite), clan genealogies (malu) or origin stories (tene te), though they exhibit elements of each of these speech forms, they represent instead the exposition or delivery of a form of knowledge or custom (mana). Mbingi knowledge is thus referred to as mbingi mana rather than mbingi bi te or mbingi tene te. Allowing for localized variation in the specific genealogies that are recounted to trace the historical connections be­ tween the living and those present at the time of darkness, mbingi mana rep­ resents an unusually coherent and homogeneous tradition across most of the central Huli area. ¹

As a form of mana, mbingi narrative traditions possess their own genealogies; a holder (yi) of mbingi mana, literally a mbingi manayi, will often identify the individual from whom he learned this knowledge, underscoring both the validity of his own status as a manayi and the “truth” of his particular account-a nice illustration of Goldman’s (1983) argument concerning the pervasive importance for Huli of identifying sources (tene). ² While the senior agnates of the clans that controlled the major Huli ritual centers all held mbingi mana and were crucial to the successful implementation of this mana, prominent men in many other clans were also mbingi manayi. As a form of knowledge, mbingi mana was unusual in that it was neither communicated among men in return for services or items of wealth such as cowrie shells, as in the case of mana relating to intercourse with women, nor reserved solely for men on the basis of a particular descent status, as in the case of mana relating to clan rituals. Narratives describ­ing mbingi were formerly recounted in both private and semipublic contexts and among groups mixed in terms of age and gender. While the details of this mana were known to very few, the broad outlines of the narratives and the general gist of appropriate procedures were known to almost all. Narrative 1 (at the end of this chapter) is broadly typical of mbingi narratives recounted in the central Huli valleys. It was recounted at my request in 1991, in the company of a group of men of varying ages, by an older man from a clan with no particular links to the major ritual center such as Gelote or Be­ benite. The narrator establishes his credentials as a manayi and then proceeds to recount the basic features of mbingi, to list the ilili taboos or restrictions on behavior such as intercourse and to identify those living individuals with whom a “full” understanding of this mana properly resides.

Mbingi is presaged by a mighty thunder and lightning and the gathering of clouds on the northern and eastern horizons. Under the instruction of local man­ayi, people build houses on high, sloping ground. The particular form of house required (the “mbingi house”) is unique to this event and consists of a platform raised on low stilts, with split-timber walls and roofs. ³ Snakes and other wild animals gather around and underneath the houses for protection. Firewood and sweet potato are gathered and stored in the houses. Sweet potato mounds with near-ripe tubers are covered with grass and marked with tall wooden stakes. Women are sent back to their natal parishes, and wooden divisions are placed within the houses to separate men and boys from women and girls; a strict proscription on all sexual intercourse is observed. The skulls of the ancestors are retrieved from their ossuaries and brought into the houses. When mbingi comes, it does so, in the words of a man from the Koroba valley, “like a spear, like a great rock” (Magaya, Ballard Tape 91/lB: 86-113).

The day sky darkens until it is as black as night, and then the sky breaks and falls like hail in the form of ash and sand, identified as da pindu (“sky stuff”) or mbi dindi (“earth of darkness”), which covers the ground but slides off the sloping roofs of the houses. The rivers rise, and as the streams block up with ash, they overflow. For as many days as the darkness persists, the majority of people remain secluded indoors, consuming their stocks of food.

On the first day of darkness, the only people permitted to leave the houses are singletons, those with no other siblings; ⁴ they first peer through cracks in the house walls before venturing out into the dark, harvesting sweet potato from the marked mounds, which they take back and feed to the others. On the second day of darkness, those with only one sibling are also permitted to leave the house. On each successive day of darkness, sibling sets of increasingly larger sizes leave the houses for the first time. After either four or eight days (indi­vidual narratives differ at this level of detail), the clouds clear, the sky grows light and the rivers revert to their proper courses. ⁵

In the aftermath of mbingi, a time of plenty ensues, as crops planted in the rich ash grow exceptionally well: “The taro and banana were huge. The pitpit grew as thick as a man’s calf, and the rungia leaves were the size of men’s feet. Pigs grew as large as cows are now” (Maiya-Gane, author’s field notes, 1991). Despite its fearsome manifestation and a collective anxiety over the nature of its consequences, mbingi is thus potentially the harbinger of a time of plenty-an event to be endured, but one also to be welcomed and even sought. The terms used to describe this newly fecund landscape are precisely those conventionally employed in bi te mythic narratives about the subterranean lands of plenty pre­ sided over by Baya Haro cannibal ogres (see Goldman, in press); the cucumbers there are so large that they are supported by stakes, groves of lowland pandanus hang heavy with fruit and there is an abundance both of game and of domes­ticated pigs. These lands of plenty are held to be found not only to the south of Huli territory, in the direction of the land of the dead, which lies beyond the territory of the Duguba of the Papuan Plateau, but also directly beneath the Huli landscape. This orientation presents something of a conundrum, for the south is also the ultimate terminus for the rivers that carry the spirits of the dead to Humbirini Andaga, the land of the dead; ⁶ and though the analogues for the post­ mbingi landscape are all subterranean and southerly, the immediate source of the restorative mbingi is from clouds in the sky that gather initially on the horizon.

A second conundrum relates to the role of the sky as a source of mbingi. With the exception of myths relating to the sun and the moon, the sky is a relatively unelaborated field in Huli cosmography, particularly in comparison with the more complex celestial cosmographies of the neighboring Enga, Ipili and Duna (see Modjeska 1977: 96; Wiessner and Tumu, in press; Biersack, this volume). Sky people, daluyali, residing in a sky world, daluya anda, occasion­ ally feature in bite myths (see Goldman, this volume) but did not figure prom­inently either in ritual or in myths of origin (an exception to this last claim being origin myths for some of the Huli clans immediately adjacent to Duna and Ipili communities, and myths relating to the Ogoanda tower of the Irari ritual site; see Goldman, this volume). Moreover, the celestial origins of mbingi are not certain, for the Huli terms for “sky” (heiri) and “mountain” (hari) are only tonally distinguished, and exegeses proffered by different narrators alternate be­tween sky and mountain as the sources of the sandy mbi dindi soil: ⁷ ”Ground and sky mixed together, and the waters poured down outside [the mbingi houses]” (Mangobe, 30.12.90, Ballard Tape, 90/lB: 388-509). It is as though, at the horizons of Huli cosmography, the distinctions between sky and skyline become increasingly blurred.

That mbingi has occurred in the past is held by Huli to be an unquestionable fact. Many people can still find patches of the light grey-green sandy soil buried in their gardens, and it is frequently uncovered during ditch-digging.’ The stumps of two ancient trees said to have been growing when mbingi fell are pointed out in the Haeapugua basin; and in most clans, individuals who were alive during mbingi or born during the event are named and located genealog­ically. But, while they are essentially accounts of a historic event, the narratives are couched in a variety of tenses and, in providing a catalog of portents and prescriptions for appropriate behavior, invoke the specter of a future recurrence of mbingi. ⁸ Although the widespread adoption of Christianity since the 1950s has seen an end to traditional ritual performance, many elements of the mana for mbingi persist in Huli attitudes toward the apocalypse envisioned in the Bible (Ballard 1992).

What does it mean to envisage the recurrence of an event such as mbingi and to legislate codes for moral and practical behavior in the event of such a recur­rence? Although Blong (1982) limits the distribution of notions of possible recurrence of the time of darkness to communities of the Southern Highlands and Enga regions, such as the Kaluli, Kewa. Enga. Duna, Huli and Paiela, his own compilation of myths lists numerous other instances of communities that retain the knowledge necessary to prepare for repeat events while insisting on the uniqueness of the original event (Blong 1979: Kalarn, Goroka, Dreikikir, Akuna and so on). A single event, therefore, is sufficient to generate the expectation of recurrence.

Whether Huli perceive a past replete with multiple mbingi events is a moot point. Glasse’s (1965: 45) early assertion that “Huli believe that bingi [sic] has occurred several times” was later retracted by him in a personal communication to Blong (Blong 1979: 73). Frankel’s (1986: 17ff) more detailed analysis fairly clearly identifies a single event giving rise to the idea of mbingi and the pos­sibility of future events. My own observation is that, while Huli subscribe to the notion of multiple mbingi events in the past and to the possibility of recur­rence in the future, they do not in fact describe the effects of more than one fall of the mbi dindi sand. Nor, as keen observers of soil stratigraphy, do Huli identify more than one layer of mbi dindi sand in the ditches of their gardens. Conceptions of the recurrence of a time of darkness cannot be equated simply with a history of multiple ashfall events, even if it were possible to demonstrate such a history. ⁹ Instead, the possibility of recurrence derives from a much more fundamental conception of the potential in the land and among humans for regeneration, for the renewal of a landscape under permanent threat of decline.

While what might be regarded as the archetypal mbingi event, as described in Narrative 1, apparently refers to the fall of ash from the sky, mbingi as a category of events for Huli actually denotes a far wider range of cataclysmic events, all of which play a similar role in Huli cosmology and temporality. Accordingly, past floods, earthquakes, mudflows and eclipses are all described as mbingi events, cosmographic convulsions that transform the ground for hu­ man agency, eradicating the irredeemable and reconstituting the circulation of fertile substance. Mbingi events mark the meter or rhythm of Huli temporality and, as chronological markers of decline in the state of the land and of the behavior of men and women, serve to morally invigorate the trajectory of Huli historicity.

Growth, Decline and Fluids

I have argued elsewhere (Ballard 1995) that Huli temporality is structured not only around the occurrence of major natural phenomena, such as drought and frost, but also around human acts, such as war, as the surface expressions or tokens of the land’s well-being. In this respect Huli temporality stands in strong contrast to that described for the Foi by Weiner (1991: 34-37). Where the symp­toms of seasonality are further embellished by the Foi in their distinction be­ tween wet and dry modes of subsistence and sociality, the markedly less pronounced seasonality of the Huli climate is still further diminished by a cor­ responding diminution in the significance accorded to the available evidence for annual cycles.

Instead, Huli temporality and, more broadly, Huli historicity are expressed in terms of much longer cycles of entropic decline in growth, interspersed with dramatic regenerative events. A pivotal role in cosmological thought for notions of entropy, understood here as a general belief in decline and dissipation rather than increase and accumulation, is common to several communities of the west­ ern Papua New Guinea Highlands (Biersack 1995). ¹⁰ It is important to stress that this sense of entropic decline, manifest for Huli in the declining fertility of the land, extends also to human mana, which is itself continually under threat of dispersal, disruption and depletion: ”The long memories of yesteryear have grown short…. Now young men refuse to look after pigs, people marry too early and women bear children too young” (Walubu-Mabira, 23.10.92, Ballard Tape 92/lB: 546-574); “All that our fathers told us not to eat we now eat. All that our fathers told us not to see we now see. All that our fathers told us not to say we now say” (Pudaya, 3.1.92, Ballard Tape 92/4B: 199-238). Vicious wars, suicides and other breaches of conventional mana are also tokens of en­tropic decline and formerly elicited ritual responses similar to those performed in the event of “natural” disasters. The depletion of mana, which is held to account for the imperfect performance of ritual and for the continual decline in moral standards, is critical to understanding the way in which fertility is morally constituted and hence negotiable in practice for the Huli.

Huli men describe a relationship between entropy, gender and fluid substances whose defining moment is expressed in a widely known cosmogonic myth. A woman whose child is crying for water declines a gourd of fertile fluid offered to her by the sun, proffering her breast to the child and introducing mortality to humans in the form of her breast milk. The fluid from the discarded gourd is consumed by a python, which consequently gains immortality. Another common myth recounted by both the Huli and many of their neighbors (see Biersack and Josephides, this volume) tells of a woman, located by the Huli version in the Koroba Basin, whose husband leaves on a trading expedition to the lpili salt springs; though instructed not to remove the flying-fox bone stopper inserted in a blackpalm tree behind their house, she does so when their child begins to cry for water. A torrent issues from the hole in the tree, flooding the basin and drowning her and the child. The two myths are structurally similar, with failure to observe an interdiction or a command introducing mortality to the universe but, despite the mythical privileging of the male voice (see Goldman, this volume), what is at issue is not so much the caprice of women as the entropic effects of any action that subverts the moral order.

“Pollution” beliefs articulated by Huli men are in many ways typical of similar beliefs held widely across the Highlands (Strathern and Strathern 1971; Meigs 1984; Poole 1984; Biersack 1987) in which the particular valence of fluid substances (such as semen, breast milk, vaginal excretions, corpse fluids and afterbirth) is not an inherent quality. Instead, it is the morality of individual actions in relation to a substance, and the balance between male and female action in particular, that defines the meaning of that substance and its potential to induce fertility or mortality. Hence both semen and breast milk can, under different conditions, be either beneficial or inimical to the growth of their re­cipients. ¹¹ are thus negotiable processes over which humans seek a degree of purchase through appropriate moral behavior. The deployment and observance of mana knowledge (itself subject to processes of increase and decline) and ilili proscriptions or taboos are the means whereby humans deter­ mine the meaning of substances and gain a degree of control over their lives and environments.

For Huli, those ‘substances most expressive of the processes of growth and decline are fluids. Iba, the term for water, forms the root for a lexicon of fertile­ fluid terms such as ibane (“grease, fat” ), wi ibane (“penis-grease,” semen), andu ibane (“breast-grease,” breast milk), dindi ibane (“soil-grease” ) and ira ibane (“tree-grease,” sap). In their fertile state these fluids introduce moisture or grease to people and to the land. Conversely, infertile fluid substances act as desiccants, inducing a state of gabu, or dryness. Thus the poison technique of tomia (literally, “to give an emission” ) required the application of dried menstrual blood to the victim’s food or belongings; contact with this tomia would then drain the victim’s body of moisture, resulting in death by, desiccation. The equation of dryness with death and decay pervades Huli belief more widely: Much attention is paid to the level of moisture in the many swamps of the Huli landscape, and the invasive spread of dryland grasses over the surfaces of these swamps is taken as a clear indication of the general decline of the land’s fertility. ¹²

Notions of enhousement and containment play a central role in this moral economy of fluids. The suffix -ne denotes a substance in its contained form, fluid in flow (iba) contrasting with its contained and controlled (and hence fer­ tile) form, ibane. Huli beliefs are closely matched by Weiner’s account of the anxiety expressed by Foi men over uncontrolled flow: “What kara’o [tree oil], blood and petroleum and all sexual substances … have in common is that they are poison only in their un-held-back state-when they are allowed to flow, to make unmediated suture with other substances and other bodies. Held-back blood becomes the fetus, held-back semen a vital, healthful adult man” (1995: 171). Thus, while free-flowing menstrual blood and the exuviae of childbirth are among the substances deemed most inimical to the growth of Huli men, menstrual blood contained in bamboo tubes and ritually capped with bird-of­ paradise feathers bound by the umbilical cords of children was a key component in fertility rituals performed at Gelote and other major ritual centers, where the tubes were buried in ponds and mud-soaks to replenish the depleted fertility of the land (see Stiirzenhofecker’s [1995: 94-96] description of Duna contributions to this rite): “From this menstrual blood and the child’s umbilical cord, many children would be born” (Yaliduma-Dai, 12.4.91, Ballard Tape 91/5B: 61-88). If women and female fluids are identified in Huli myth as the source of substances such as fire, mineral oil, sago and pandanus (Goldman 1981: 51- 52), it is through the directing and domesticating agency of men that the fertile possibilities of these substances are realized: The men of these myths tame the potentially destructive powers of women’s heat (pobo), open women’s vaginas to release the buildup of polluting blood and to allow for the passage of children, and harness the latent fertility of the female spirit of the haroali bachelor cult through the correct exercise of mana (Ballard 1995: 135). This engineering role for men in myth is matched in Huli cosmography, which identifies male iba tiri spirits as “cosmic plumbers” (see Biersack and Goldman, this volume), cease­lessly maintaining the free flow of rivers and spirits of the dead toward the south and guaranteeing the cyclicity of Huli life.

Insofar as Huli cosmography appears to be composed of “hydraulic” flows of power, it is modeled upon more mundane conceptions of flow in daily forms of exchange. ¹³ Accounts of the regulation of rivers by iba tiri spirits take their cue from the quotidian practices of Huli gardeners. One of the principal activities of men in relation to agricultural subsistence is the maintenance of a balance in the supply of water to their gardens, channeling excess rainfall or swamp waters away while carefully ponding water around their plots of moisture-loving taro. Men also conceive a social role for themselves as controllers of the flows of women and wealth. The minutiae of daily existence and the arcane details of cosmology find echoes in each other, and it is small surprise to find that the concept of mana, of the role for ancestral knowledge in the practice and control of the flow of social life, is seated foundationally in the basic structure of the cosmos.

The Root of the Earth

The core expression of fluid fertility in the Huli cosmos is dindi pongone (“earth”-“knot, root”), the root of the earth. Goldman (1983) and Frankel (1986) have outlined much of what is documented about dindi pongone, which can briefly be described as a root or a vine that runs beneath the earth, consisting of a python (puya) bound around with cane (gewa) and capped by a layer of stone. Where it surfaces, the stone cap of the root becomes visible in the form of outcrops of rock, within which caves are identified as mouths of the python. Each of these surface indications of the presence of the root is regarded as a major ritual site (gebeanda), and the names of these sites and of the ritual leaders (gebeali) associated with them are common knowledge. Crucially, the direction identified for the flow of the root of the earth reverses the regional pattern of surface drainage toward the south (Frankel 1986: 19-23; Ballard 1994). Instead, the root of the earth flows upward from the Papuan Plateau, to the south and west, toward the Enga and the Ipili in the north and east. In some places, the root travels underground; in others, it enters river channels and flows upstream within the visible river. Running against the current of time, against the southerly passage of the life cycle, the root of the earth represents a fundamentally antientropic flow.

In the context of Huli cosmology, the root of the earth functions as the ultimate source and repository of all fertile substance, providing fertile fluid (ibane) to the land and the people. However, the root is itself subject to entropic decline, constituting a cosmic barometer of the morality of human agency, and conse­ quently requiring the intervention of humans in the form of repeated acts of ritual replenishment. Usually, replenishment of the root took the form of “earth­ spell” (dindi gamu) rituals, during which gourds of mineral oil and tree oil were poured within caves at the gebeanda ritual sites; in the more complex forms of the dindi baya baya fertility rite (Ballard, in press). performed at the sites of Gelote and Bebenite, it was the menstrual blood of younger women, the blood taken from the cut fingers of young boys or the blood and fatty “grease” of substituted pigs that symbolized the replenishment of the fertile core of the Huli cosmos.

The python in the root of the earth, which stands iconically for the root as a whole, is itself composed of ibane-indeed, it is the quintessential form of controlled and fertile fluid, bound around by the cane with which men construct the girdles that bind their waists: “The cane and the python hold the land together” (Hiluwa-Irugua, Ballard Tape 91/18B: 46-81). At the heart of the Koroba version of the myth of the woman who removes the stopper from the tree is the revelation, made by the ritual leader at the Iba Gunu site in the Koroba Basin, that the fluid that issues forth from the tree comes from the python: “There is a blackpalm tree [at the Iba Gunu site] and at its base lies the python. The tree grows up from the belly of the python” (Dali-Andago, 8.2.91, Ballard field notes); the fluid that gushes from the tree is the ibane of the python but, released through improper action, it inflicts death rather than life. ¹⁴ The gewa cane that contains the python is a manifestation of the mana implemented by men in the governance of fertility; the statement that “Gebeali is gewa, dindi pongone is puya” (Hiluwa-Irugua, 25.8.91, Ballard Tape 91/18B: 0-24) iden­tifies cane with the fence-making and domesticating functions of male ancestors (gebeali) and confirms the necessary link between observation of the mana of the ancestors and the continuing fertility of the root of the earth.

How, then, given this summary account of Huli cosmology, are we to situate the mbingi narratives in which the logic of a restitution of cosmic fertility, from the skies above and from the northern horizons, seemingly runs against the cosmographic current of fertile substance flowing underground and from south to north?

The Essence of Fire

In order to obtain the essence of fire one must go to its source, to its reserve, where it husbands its strength and concentrates itself, that is to say, to the mineral. Bachelard 1987: 73

The role of mineral oil and other fluids in rituals performed at the gebeanda ritual sites is critical to an understanding of the link between the root of the earth and the cataclysm of mbingi. The coincidence of the occurrence of seeps of natural mineral oil in the vicinity of most of the major gebeanda can now be established as an important element in the location of these sites (Ballard 1995: 143-144). ¹⁵ Mineral oil and tree oil appear to have been used almost interchangeably at the ritual sites; both types of oil are explicitly held to issue directly from the root of the earth and serve to identify oil as tile essential form of the subterranean python. Ritual replenishment of the root was thus an act of humanly assisted auto-regeneration, restoring to the essential core its own scat­tered substance. Poole has documented a strikingly similar cosmological status for mineral oil on the western banks of the Strickland River, where Bimin­ Kuskusmin ritual leaders talk of oil as “our blood, our semen, our bone, our heritage from our ancestors … our life” (1986: 169). For the Bimin, too, the oil represents ”the source of fertility, growth, renewal, and change” (1986: 171) and was used in ritual in attempts to restore fertility to a failing landscape. But, in keeping with the model of morally negotiable valence described earlier, oil is not as “stable” in its significance for the Huli as it appears to be for the Bimin.

At the southern limits of the root of the earth, in its Huli conception, lies the Onabasulu ritual site of Malaiya, described to Schieffelin (1991: 65) as “not a place, but a fire that burns eternally by the water at Dobanifofa Ifthe fire dies out … so all the fires in the world will extinguish, together with the sun, and the world will die. On the other hand, if the fire ever flares out of control, it will roar out over the plateau and burn up the world.” The obvious propensity of mineral oil to convert into flame imparts an unstable quality to the root of the earth, rendering it a force of ambiguous potential, balanced finely upon the morality of human actions. But this dual manifestation of the python, as oil and as fire, finally alerts us to the continuity of python and sun, closing the circle that describes the limits of Huli cosmography.

From a Huli perspective, the sun (Ni) rises into the sky each morning above Mt. Ambua, crossing over the sky-bridge, Da Togo, ¹⁶ toward distant Mt. Mbi­ duba, above the Lebani valley to the west of Huli territory, before entering the earth at dusk. ¹⁷ By night, the sun travels along the root of the earth, to surface again at dawn from Ambua. “He [the sun] crosses the cane and the python as a bridge. At night he goes across this bridge, by day he comes out amongst us here. At night when we sleep he goes inside over there at Mbiduba and goes along burning beneath the ground (dama nana pora dindigoha)” (Ngaori­ Mandiga, 1.8.91, Ballard Tape 91/14A: 333-363). One of the gebeali ritual leaders at the site of Bebenite makes the final link: “By day he [the sun] is a man; by night he is the python” (Hubi-Morali, 4.11.92, Ballard Tape 92/5B: 0-186). The sun is thus the python, the python is oil, and oil is latent flame, the sun by night.

Mbingi, in this context, is an act of replenishment writ large. If the model of negotiable fertility is extended to this cosmic scale, it is the morality of human behavior under the precepts of mbingi mana that determines the quality of mbingi, either as devastation or as rejuvenation. I have addressed the issue of direct human involvement in the provocation of mbingi events elsewhere (Ballard 1992); the point to be stressed here is the sense of anxiety that attends the observation of the mana for mbingi, for the precision of the narratives and the severity of the injunctions of this mana reflect the instability of mbingi as a manifestation of tl1e sun/python and the perilous existence of a micral topography grounded upon latent flame.

Fire is the ultra-living element. It is intimate and it is universal. It lives in our heart. It lives in the sky. It rises from the depths of the substance and offers itself with the warmth of love. Or it can go back down into the substance and hide there, latent and pent­ up, like hate and vengeance. Among all phenomena. it is really the only one to which there can be so definitely attributed the opposing values of good and evil. (Bachelard 1987: 7)

Reverie and Atopos

Huli narratives of the time of darkness present a public myth whose details were known to few but whose broader themes were much more widely under­ stood, finding resonance in the manner in which mana is observed in all things. Though an understanding of the link between the sun and the root of the earth underpins much of Huli cosmology, this “interior” identification was a highly restricted revelation. Knowledge, Huli people say, resides within like the ira pubu grub enclosed within the trunk of the sago palm. Casual distribution or the uncontrolled release of knowledge diminishes its effect. and research into a topic such as Huli cosmology repeatedly confronts an understandable reluctance to impart detail; some declare the limits of the knowledge that they are prepared to disclose, and others are not even to be approached. Respect for a system of knowledge must also respect its rules; where analysis of Huli cosmology is closed off from indigenous exegesis, through the withholding of knowledge or the loss of key individuals, there is perhaps a point of departure for a Bache­ lardian reverie that seeks to extend that analysis beyond what can be demon­ started, to recast lines of enquiry and reintroduce uncertainty to the attempted neatness of one’s conclusions.

The cyclical continuity between the sun and the root of the earth raises questions relating to the themes of interiority and containment addressed throughout this volume. While the gewa cane, representing the mana of the ancestors, might be figured as the means of containing the elementally unstable ibane of the python, it does not provide the Huli root of the earth with its essential form, as Biersack (this volume) reports for the Ipili. Rather than the encircling cane, it is perhaps the continuous cycling of the python as sun, or flame, by day and as oil, or latent flame, by night that describes the full form of the root of the earth and en-houses the Huli cosmos. The path described by the passage of the sun and the python is itself a form of containment, a systemic flow whose closure implies an ending, an ultimate stilling of movement.

A cosmogonic myth recounted by the lqwaye of the eastern Papua New Guinea Highlands describes a creator, Omalyce, who is himself the cosmos in its original, complete form (Mimica 1988: 75): the figure of a man with his penis in his own mouth, a perfect cycle of fluid that has to be severed for the world to come into being. There is, in the completeness of Omalyce and of the – Huli sun/python, an intimation of finitude that, in turn, invokes a final event or eschaton (Ballard 1992). In reversing Mimica’s (1988: 93) observation that every end “is also the renewal of the cosmos,” it is possible to propose that this act of renewal or making-whole must also portend its own demise; that the desire expressed by the Huli for a mana-made-whole and a world renewed through mbingi is itself the conceptual seed for the fundamentally entropic basis of their existence.

What the mbingi narratives further introduce to this model of an encircling root is a sense of the dynamics of transformation that requires us to reconfigure the root as a sheath or membrane rather than a simple network of passages. Mbingi is not the world turned upside down (Knauft, this volume), nor is it the descent from above of a new landscape. The sense captured in the mbingi nar­ ratives, where day turns to night, is of a movement between interiority and exteriority, a turning-inside-out of the cosmos (see Strathem. this volume); the image of the python, which exchanges interior for exterior in renewing its skin, supplies a metaphor for the larger act of world transformation, implying a pro­cess of cosmic introversion that exceeds the capacity of imagination. The cir­cularity of sun and python suggest that it is not just the subterranean that is the source of ground renewal (Biersack, this volume) but rather an unseen interior, an atopos that is at once above and below, whose extroversion exchanges an inner heat and moisture or vitality for a dry and exhausted outer skin.

Narative 1: Mbingi Mana

(Narrator: Walubu-Mabira Hege, Walete Parish, 5.1.91, Tape 90-91/2A: 36-75)

Mbinginaga ai mbingi peneyagoni larogoni, obeneni i hale haru dadagua o larogoni About mbingi: I will talk of how mbingi comes.
I will talk of the middle part of which I heard.

ogoni anduane, Ngoari-Hebaria, Ngoari-Hebaria ibugo, Ngoari-Hebaria igini, o Ngoari­ Gira, Ngoari-Gira igini, Ngoari-Perela, Ngoari-Perela igini, Ngoari-Tege
The owner of this [mana] was Ngoari-Hebaria. Ngoari-Hebaria’s son was Ngoari-Gira, whose son was Ngoari-Perela, whose son was Ngoari-Tege [i.e., transmission of the mana passed through this line of first-born sons].

anigo, ogoni ibugua mana manda binigo, o [agali dongone] mbingi pora angi
This was his mana: when mbingi came

ina aba handa hari ligo kilikulu lole birago anda timbuni ina bilimu lene
our fathers said the sky would thunder so large houses should be built
.
dindi hanuni ede Yalimali wiaru uyu Taibaanda hamaga wiaru dagua bialu ke go la We should build them at places such as Yalimali and Taibaanda [both raised ground locations].

o hari tugu lalu kane, o, ege maria agima polebira
The sky would thunder until the fourth month.

ege ogoni piai kagola, anda bu maro birago hondowa
When this [fourth] month finished, they would complete the houses.

o hari ligu podolebira, ai iba nigu Ii ibulebira
The sky would break and fall down and down there the rivers would rise.

au bule birago i nogonaga, waneigininaga
It would do thus, so houses had to be made for the pigs, the children,

o anda hobane hanguhangu helowa, wali agalila libu, wali i hangu, agali i hangu
and the houses divided for men and women to be separate, women on one side, men on the other.

ogoni mebia haribiyagua [hondo], libu kebigo homai holebira, mebia hari aria hameigini haruago aria homai
If this law was broken, all would die. Whoever broke the law, all their clan would die;

Tanime mebia heneyagua [hondo] Tani, Dobanime mebia heneyagua [hondo] Dobani if Tani [clan] broke this law, all Tani would die, if Dobani broke the law, all Dobani would die.

nde ibu ka dagua hanguhangu haruago homanwgoni
Whuever stayed outside would die.

ni nabi hayagua [hondo], mbingi hina mulene kirali hangu Ii tagira holebiragoni
Two only would stay outside to give mbingi sweet potato [to the others].

wali agalila, walinaga mulene mende, agalinaga mulene mende
To men and to women, one to give to women, one to give to men.

ti unu nogo hayaru bo baanda howa o balu nalu bedagoni
All would stay inside, kill their pigs, put them inside and eat them.

ani bidagola howa hari uyugu podalu dindi ogoria ibira hole biragoni
Having done thus, the sky would break and fall.

iba payule biragoni, iba Tagali nigu ibalu ogoria payule biragoni
The rivers would overflow. The Tagali River would flood this area.

hari ligu podalu ogoria ibira hole biragoni
Up there the sky/mountains would break and fall down.

ani bialu kagola ogoni ne nalu karia
When they were inside, people didn’t eat just small amounts [of food],

tomo ogoni ha emene ndo, baanda hayago hangu nalu bedagoni
they threw a large amount of food inside and ate it.

ogoni nalu bedaria o dindi harila bago pilo wiaabo hole biragoni
While they sat and ate, the sky and ground would become dark.

halini angi, wa lolebira halini angi wa laruagola, o mbi lowa honowinigo
After eight days, the light would come and at this time there was born,

o Tani Hebaria ogoni aria ibu mbalinime honowinigo, o Baru-Mbiyago
among Tani Hebaria’s kin, his sister bore Baru-Mbiyago.

mbilo wuwa unu tamuni biruwa taba henenego
When it was dark inside and all were sitting there, he was born.

amu hariga tagira puwa egabi layagola honowinigo, Ngoari-Gaea
Those who went outside and were born when the light came included Ngoari-Gaea of Dindiago clan.

ai, Ngoari-Gaea igini, Yula, ai ogoni ibu ani hama pedagoni
Ngoari-Gaea’s son, Yula, he is now here.

ani bigi binigo mbingi biago ani bulebira lalu ina hendene ndo, bi ani layaria, hale hene
No-one has seen mbingi do thus, only talk of it has been heard.

bi hale hene dagoni ina hame mana ogoni, o ya/i hene ogoni
This talk that we heard was the mana of our ancestors. This is how it was.

(Reprinted with permission of Dr. Chris Ballard.)

REFERENCES

Allen, B.J. and Wood, A.W. (1980). “Legendary volcanic eruptions and the Huli, Papua New Guinea.” Journal of the Polynesian Society 89(3): 341-347.

Bachelard, G. (1987). The Psychoanalysis of Fire. London: Quartet Books. Translated by A. Ross, 1964. Original edition 1938.

Ballard, C. (1992). “The fire next time: British Petroleum, the Book of Revelations and Huli ritual.” Paper presented at the First European Colloquium on Pacific Studies, Nijmegen, December 18, 1992.

—. (1994). “The centre cannot hold: Trade networks and sacred geography in the Papua New Guinea Highlands.” Archaeology in Oceania 29(3): 130-148.

—. (1995). “The death of a great land: Ritual, history, and subsistence revolution in the Southern Highlands of Papua New Guinea.” Unpublished Ph.D. thesis, Research School of Pacific and Asian Studies, Australian National University.

—. (in press). “Immaculate misconceptions: The many fates and resurrections of Baya Baya.” In C. Ballard and J. Clark (eds.), Blurred Boundaries and Trans­ formed Identities: Myth and Ritual in the Southern Highlands of Papua New Guinea. Canberra: Department of Anthropology, Australian National University.

Biersack, A. (1987). “Moonlight: Negative images of transcendence in Paiela pollution.”
Oceania 57(3): 178-194.

—.(1990). “Histories in the making: Paiela and historical anthropology.” History and Anthropology 5(1): 63-85.

—. (1991). “Prisoners of time: Millenarian praxis in a Melanesian valley.” In A. Biersack (ed.), Clio in Oceania: Toward a Historical Anthropology. Washington, D.C.: Smithsonian Institution Press, pp. 231-295.

—.(1995). “Introduction: The Huli. Duna, aqd Ipili peoples yesterday and today.” In A. Biersack (ed.), Papuan Borderlands: Huli, Duna, and lpili Perspectives on the Papua New Guinea Highlands. Ann Arbor: University of Michigan Press, pp. 1-54.
Blong, R.J. (ed.). (1979). “Time of darkness legends from Papua New Guinea.” Oral History 7(10).

—. (1982). The Time of Darkness: Local Legends and Volcanic Reality in Papua New Guinea. Canberra: Australian National University Press.

Bourke, R.M. (1988). “Taim Hangre: Variation in subsistence food supply in the Papua New Guinea Highlands.” Unpublished Ph.D. thesis, Research School of Pacific Studies, Australian National University.

Clark. J. (1993). “Gold, sex, and pollution: Male illness and myth at Mt. Kare, Papua New Guinea.” American Ethnologist 20(4): 742-757.
Frankel, S.J. (1986). The Huli Response to Illness. Cambridge: Cambridge University Press.

Franklin, K.J. and Franklin, J. (1978). “A Kewa dictionary with supplemental grammatical and anthropological materials.” Pacific Linguistics C-53. Canberra: Pacific Linguistics, Research School of Pacific Studies, Australian National University Press.

Glasse, R.M. (1963). “Bingi at Tari.” Journal of the Polynesian Society 72(3): 270– 271.
—.(1965). “The Hull of the Southern Highlands.” In P. Lawrence and M.J. Meggitt (eds.), Gods, Ghosts, and Men in Melanesia. London: Oxford University Press, pp. 27-49.

—.(1995). “Time belong mbingi: Religious syncretism and the pacification of the Huli.” In A. Biersack (ed.), Papuan Borderlands:’ Huli, Duna, and Ipili Per­ spectives on the Papua New Guinea Highlands. Ann Arbor: University of Mich­ igan Press, pp. 57-86.
Goldman, L.R. (1981). “Talk never dies: An analysis of disputes among the Huli.” Unpublished Ph.D. thesis, University College, London.

—.(1983). Talk Never Dies: The Language of Huli Disputes. London: Tavistock.

—.(in press). “Melanesian cyclops: The Huli Ogre Baya Horo.” In C. Ballard and
J. Clark (eds.), Blurred Boundaries and Transformed Identities: Myth and Ritual in the Southern Highlands of Papua New Guinea. Canberra: Department of An­ thropology, Australian National University.

Jorgensen, D. (1981). “Taro and arrows: Order, entropy, and religion among the Tde­ folmin.” Unpublished Ph.D. thesis, University of British Columbia.

Meggitt, M.J. (1973). “The sun and the shakers: A millenarian cult and its transforma­ tions in the New Guinea Highlands.” Oceania 44(1): 1-37; (2): 109-162.

Meigs, A.S. (1984). Food, Sex and Pollution: A New Guinea Religion. New Brunswick, N.J.: Rutgers University Press.

Miller, J.H. (1995). Topographies. Stanford: Stanford University Press.

Mimica, J. (1988). Intimations of Infinity: The Mythopoeia of the Iqwaye Counting System and Numbers. Oxford: Berg.

Modjeska, C.J.N. (1977). “Production among the Duna: Aspects of horticultural intensification in Central New Guinea.” Unpublished Ph.D. thesis, Australian National University.

Poole. F.J.P. (1984). “Symbols of substance: Bimin-Kuskusmin models of procreation,
death, and personhood.” Mankind 14(3): 191-216.

—. (1986). “‘The erosion of a sacred landscape: European exploration and cultural ecology among the Bimin-Kuskusmin of Papua New Guinea.” In M. Tobias (ed.), Mountain People. Norman: University of Oklahoma Press, pp. 169-182.

Schieffelin. E.L. (1991). “The Great Papuan Plateau.” In E.L. Schieffelin and R. Crit­ tenden (eds.). Like People You See in a Dream: First Contact in Six Papuan Societies. Stanford: Stanford University Press, pp. 58-87.

Strathem, A.J. and Strathern, M. (1971). Self-Decoration in Mount Hagen. London: Gerald Duckworth.

Stilrzenhofecker, G. (1995). “Sacrificial bodies and the cyclicity of substance.” Journal of Polynesian Studies 104(1): 89-109.

Wagner, R. (1972). Habu: The Innovation of Meaning in Daribi Religion. Chicago: University of Chicago Press.

Weiner, J.F. (1991). The Empty Place: Poetry, Space and Being among the Foi of Papua New Guinea. Bloomington: Indiana University Press.
—. (1995). The Lost Drum: The Myth of Sexuality in Papua New Guinea and Beyond. Madison: University of Wisconsin Press.

Wiessner, P. and Turnu, A. (in press). Historical Vines: The Formation of Regional Networks of Exchange, Rinial and Warfare among the Enga of Papua New Guinea. Washington, D.C.: Smithsonian Institution Press.

(Photo courtesy of PNG Tourism Promotion Authority.)

1. The one significant variant tradition, referred to as da pindu mana (“the knowl­edge of the sky stuff’), differs from mbingi mana only in its identification of a unique form of portent, consisting of the flight of birds across Huli territory from south to north. [↩]
2. Although other forms of mana are held by women, I have neither met nor heard of women who claim detailed knowledge of mbingi mana. [↩]
3. Mbingi anda have been constructed on several occasions since colonial contact, most recently in 1991; other than the homali anda burial platform, they were the only form of house formerly constructed on stilts by the Huli. [↩]
4. The implication commonly understood by Hull audiences is that singletons are unable during these forays to engage in illicit sex with siblings-the most dire of moral transgressions. This role for singletons is a theme common to time-of-darkness traditions among Ipili, Enga, Duna, Huli, Nipa and possibly Kewa people (Blong 1979)-all languages that derive from the same proto-Engan language community (Franklin and Frank­lin 1978). [↩]
5. Though I am unsure of the relative benefits of mbingi lasting either four or eight days, sexual intercourse on more than four consecutive days is held to be harmful for women, and eight months is the conventional period of pregnancy, of which the first four are regarded as particularly dangerous and the second four as increasingly safe (Frankel 1986: 100-101). [↩]
6. The notion of a southernly land of the dead toward which rivers carry the souls of the deceased is common to a number of communities along the southern fall of the Central Highlands such as the Foi (Weiner 1991: 4) and the Daribi (Wagner 1972: 111- 112). [↩]
7. Men still use this “sand of darkness” to shave, rubbing it into their beards to form a tight plastic coil, which they then tear off. [↩]
8. In this respect, mbingi mana serves much the same function as the mana for frosts, which details the possible courses for action in the event of crop-devastating frost events, drawing on the lessons learned from previous frost-induced famines. [↩]
9. A simple explanation for the concept of recurrence would be that Huli ancestors experienced more than one ashfall; but the only other ashfall confirmed for the Highlands region over the past 50,000 years is the Olgaboli ash, thinner and presumably less dev­astating in its effects but otherwise similar to the seventeenth-century Tibito ash (Blong 1982). This Olgaboli ash, probably also from a north-coast source, is dated to about 1,200 years ago, a temporal span that would demand an exceptional oral tradition of historical recall, even by the unusual standards of the Huli. [↩]
10. Jorgensen’s (1981) description of the Telefornin concept of biniman (meaning “to finish, to run out, to dissipate, to become nothing”) tallies closely with the concept of entropy that I have described for the Huli. The sense captured in Huli phrases such as dindi moko hayada (“the land has been made bad”), dindi gabu hayada (“the land has dried up”) or dindi le hayada (“the land is exhausted”) most closely equates to the Telefol concept of biniman. [↩]
11. Andrew and Marilyn Strathern make much the same observation for communities of the Mt. Hagen area. concluding that we “cannot simply argue. then, that semen is regarded as pure and menstrual blood is polluting. Each can be polluting, in particular contexts” (1971: 163). [↩]
12. Shortages of food are also conventionally attributed by the Huli to drought and a lack of soil moisture. despite the fact that wet conditions are more often the principal cause of crop failure (Bourke 1988). [↩]
13. Gabrielle Stiirzenhofecker (1995: 103) has recently described Duna cosmography in terms of a similar circulation of tini, or life force; I have not heard Hulis volunteer accounts of their cosmology that are structured around a notion of life force. but neither have I enquired along these lines. [↩]
14. A specific link between pythons and the effluence of fluids was suggested by several participants at the Brisbane conference, who noted that pythons are observed by Duna and Ok communities to contain vast quantities of digestive fluid, which they oc­casionally disgorge together with semi-digested prey. [↩]
15. This “discovery” in 1991 of a link between mineral oil seeps and ritual sites was made jointly in discussion with Akii Tumu of the Enga Cultural Centre and Polly Wiessner. [↩]
16. Or dawe toga (see Biersack, this volume, on the cognate Ipili term tawe toko). [↩]
17. Aletta Biersack (1991: 259) documents a similar path for the Paiela sun: “At dawn he [the Sun] ‘comes outside,’ brightening the outer ground; but at dusk he ‘goes inside,’ journeying to the underworld and bringing daylight to the dead.” How this exegesis sits with the apparent Paiela identification of the ground with the moon and of the sky with the sun is not clear (Biersack 1990: 77). [↩]