Of winds, worms and mana: the traditional calendar of the Torres Islands, Vanuatu.

APPROACHING THE MELANESIAN ENVIRONMENT THROUGH TIME

Across Island Melanesia, people experience meteorological, celestial, oceanic and terrestrial phenomena as interlinked processes whose combined effects impinge in significant ways upon basic forms of production, such as subsistence horticulture and maritime tenure, by establishing various 'parameters on what human-beings can achieve, while not determining what they do within these constraints' (Sillitoe 1994:268). Not surprisingly, local groups often create associations between the patterns and perturbations of the environment and cycles of human production which constitute organizing frameworks evocative of what Western observers label 'calendars'.

The relation between time systems and production systems--in which time is generally treated as indistinct from measurement and duration--is an old topic that has been thoroughly examined, even to the point of reification, from a Euro-American perspective by several prominent authors (see Attali 1982; Cipolla 1981; Thompson 1967, Whitrow 1980, 1988, for a few classic examples). In turn, the term 'calendar' often evokes the idea of a year as a fixed number of days and months, measured as per the (Gregorian) solar year, during which a number of productive processes take place. But, as has long been demonstrated through various important contributions to the anthropology of time (e.g. Adam 1994; Bourdieu 1968; Evans-Pritchard 1939; Forth 1982; Fox 1979; Munn 1977, 1992), the experience of temporality in Melanesia and elsewhere cannot be reduced to a question of 'calendars, clocks and cyclic rhythms', pace Aveni (1989). As Hoskins cogently demonstrates in her study of the Austronesian people of Kodi, it is important to explore how different contexts and perceptions of time--as individual strategy, ritual sequence, the social life of the house, exchange cycles, cumulative effects in nature and the body--are 'related to wider conceptions that make them meaningful', and, consequently, how 'temporal categories provide both ways of thinking and ways of acting' (1993:xiv). In sum, temporality constitutes a useful point of reference that can often provide insights into concepts of agency, ontology and the deployment of particular social values. It is towards a discussion of these issues that the present article is directed.

Considering that time and calendrics often provide rich sources of material for comparative discussion across a broad range of anthropological issues, the general absence of research about temporality in relation to traditional ecological knowledge constitutes an important shortcoming in Austronesian studies (some references to time and calendars can be found in Austen 1939, 1950; Barnes 1971; Burman 1981; Malinowski 1927; Leach 1950; Damon 1981, 1990; Meggit 1958; Sturzenhrfecker 1994; Wagner 1987:24-44).This is no less true in Vanuatu, for which there is a noticeable lack of information regarding the micro-sociology of time. (1) One important exception is the recent, synthetic account that Kirch and Green have proffered about calendrical systems in Remote Oceania (2001). The present article builds on Kirch and Green's seminal contribution by elaborating on the ritual, environmental and ontological elements which are present in the construction of a particular 'calendar' in this region of Island Melanesia. (2)

As regards the seasonal phenomena of the local Melanesian environment analysed herein, the present contribution offers new data relating to the ecology and ethnographic significance of a pan-Oceanic marine creature known as the Palolo worm. This sea worm regularly appears once or twice within a given period of the lunar year and is therefore associated by many societies with other meteorological and celestial fluctuations. For this reason the Palolo has traditionally been of relevance as a temporal marker in the local calendrical frameworks of different island groups (Malinowski 1927; Leach 1954). Yet, as with calendrical devices, there is also a notable paucity of ethnographic research regarding indigenous attitudes towards the Palolo (two important references can be found in Ecklund 1977, and Johannes 1981). (3)

This article explores the significance of the Palolo worm in combination with the principal ecological phenomena--specifically, the sun and moon, winds and tides--that constitute the seasonal year of the people of the Torres Islands, in Vanuatu. For comparative purposes, it also draws on relevant material gathered from earlier fieldwork in the Vanuatu community of Atchin, off the North East coast of Malakula. In particular, it focuses on the manner in which the environment functions as a mediating sphere whereby Tones islanders express two basic social values: namely, the ontological principle of 'capacity' (L: mena) with which they believe that both humans and much of the living world are imbued, and the associated notion of 'living [and life-giving] growth' (L:vavalete), which is the product of successful or efficacious mena. (4)

Mena represents a particular Tortes term for expressing a broader pan-Pacific idea-probably of Polynesian origin--that is normally transcribed as mana and has been variously translated as 'effectiveness' (Ben Burt, pers. comm.), 'reproductive power' or 'efficacious potency" (Steven Hooper, pers. comm.), and for which Kessing and Shore have each provided important summary analyses (Keesing 1984; Shore 1989). In the Torres Islands mena signifies 'capacity', insofar as it typifies the various capabilities of powerful persons, and certain personified things (both animate and inanimate), to successfully evoke various kinds of transformative forces that lie inherent in the living world. It is not the purpose of this article to dwell on the meanings of mena, but to contextualise this category as it relates to the experience of seasonal change (temporality). I have followed local usage and taken mena as a guideline because it functions as a key point of reference by which Torres islanders relate to seasonal transformation. The present material may thereby help to advance comparative discussion of the term mena (mana) by demonstrating how it is deployed in a specific Melanesian setting.

THE ETHNOGRAPHIC SETTING

The Torres Islands, in the northern reaches of the Vanuatu archipelago, are a tightly knit group of islands that straddle the cultural boundary between Island Melanesia and the Polynesian Outliers, at the north east edge of Remote Oceania. The Tones group is made up of six small islands--Toga, Loh, Linua, Tegua, Metoma and Hiu--which currently sustain a population of approximately 950 Island Melanesians who represent three distinct Austronesian language groups (Toga, Loh and Hiu) and are distributed across half a dozen coastal settlements. (5)

The Torres Islands represent an ecological macrocosm whose two annual seasons are defined by the prevailing winds and by the periodic movement of the sun to the south and north of the thermal equator. The land and the sea constitute the living environment of the Torres Islands; together, these domains act as basic sources of subsistence. The people of these islands register the seasonal fluctuations that characterise their world by means of a lunar 'calendar' whose primary function is to correlate important environmental and celestial movements with the various phases of the horticultural cycle (see Mondragon n.d., for a description of horticulture in the Torres Islands). (6)

In combination, the regular displacement of the winds and the annual motions of the sun and moon--the basic temporal markers of the Torres 'calendar'--induce changes in the biorhythms of local plants and animals. In turn, Tortes people relate to the heavens, the weather and the biorhythms of their world by variously attributing particular forms of mena to each of these phenomena. Below, I describe Torres people's relations to each of these manifestations of mena, beginning with the appearance of the bird known as tuwia and the Palolo worm (the first two paragraphs of the following section are based on Western "scientific/biological' data, while the rest of the text derives from anthochthonous information advanced by various knowledge specialists from North-Central Vanuatu).

 THE TUWIA AND THE NUT

 Sometime around mid-Spring, after the planting of yams in the new
 gardens began, a bird that I had never previously observed made its
 appearance on the beaches of the Tones Islands. The creature
 resembled a sandpiper, and engaged itself by picking at small
 insects within the sand and grassy soil. 'It is a tuwia, David Hunt,
 a local islander, told me. We slowly approached a group of tuwiti
 and David sang out: 'Tuwiti, tuwia, nut mad gor?' [Tuwiti, tuwia,
 has the Palolo worm come yet?]. As we observed them, one of the
 birds bobbed its head back and forth a few times. Even though the
 tuwia carry out this motion repeatedly, apparently to keep watch on
 nearby movement, David smiled in satisfaction. The tuwia', he
 explained, had said yes. This heralded the arrival of the Palolo,
 and consequently meant that the winds would soon be shifting and
 that the track of the winter sun would reach its lowest point in
 the sky.

 As I would later be told, the movement elicited by David
 represented an expression of the mena of the tuwia.
 (from the author's field diary, Loh Island, Torres group, October
 1999)

Twice a year in the Torres Islands, shortly after the appearance of the tuwia, (7) towards the beginning of the hurricane season, a foetid stench blows in from the sea, announcing the arrival of what Torres people refer to as the nut.

The nut, known commonly as the Palolo worm (Annelid palolo viridis or Eunice viridis in Oceania; Leodice viridis in Island South East Asia), is a marine creature which lives in the crevices of coral reefs across the Pacific Ocean and is particularly abundant in Melanesian waters. For the greater part of its life, the Palolo remains within its coral abode and is rarely visible. But on two specific days in every lunar year, the rear end of all the Palolo worms, which can measure from one to five inches in length, simultaneously detach themselves from their parent body and rise to the surface of the Melanesian seas in order to participate in a massive breeding frenzy (the Palolo worm itself remains safely within the reef and gradually grows back to full size). In this part of Remote Oceania, the first such emergence of the Palolo begins in the morning, around five to seven days after the last full moon in October. Although it is often the case that the volume of worms can fluctuate markedly from one year to the next and one island group to another, this is the larger of the two appearances of the Palolo? The second mating event takes place almost exactly one lunar month (29.5 days) later, prior to the last quarter moon in November.

The timing of the Palolo's yearly emergence is remarkable for the precision with which it occurs, and is a consequence of the exactitude of the creature's internal biochemical 'clock'. (9) This internal 'clock' allows each individual Palolo to independently register the exact sequence of lunar cycles which have transpired since the last mating season, thus allowing the worm to perfectly synchronise its moment of detachment with that of all the other worms (according to Aveni, "the dim lunar illumination is the zeitgeber that entrains the animal to a lunar periodicity', 1989:174, but see note 10). After the worms have finished mating, large masses of green epitokes--packets of eggs and sperm that the worms have exuded throughout the day--remain drifting on the waves. As the day wanes, the epitokes accumulate in enormous, tangled masses near the shores of the Melanesian islands. It is at this time that a putrid stench fills the air, alerting island peoples that 'the worm' has arrived. (10)

In Remote Oceania, the social relevance of the Palolo varies widely. During the two years in which I was present at the time of the Palolo's emergence in Vanuatu (1998 and 1999), it was evident that local reactions to the sea-worm elicited long-standing ritual and social activities. Across North and Central Vanuatu, the rarity of the worm has guaranteed its continuing value as a special object of exchange. Thus, the Palolo is generally cooked within ground ovens, constituting the key ingredient in ceremonial yam-puddings (B: laplap). In the Tortes Islands the worm is used as a filling for laplap, while in Atchin, Malakula, it is spread as a topping and then oven-baked. Across North East Malakula these laplap 'bakes' are considered a prized delicacy and were traditionally employed as part of an annual series of exchanges and feasts which were (and, to a limited degree, continue to be) carried out with neighbouring island groups in order to strengthen kinship ties or promote peace-making ceremonies. According to people in the small islands of Atchin and Wala, these exchanges were traditionally related to the changing phases of the moon, as well as the new horticultural year.

At present, it is clear that there has been a gradual decrease in the overall scale of traditional activities associated with the Palolo. Nevertheless, its continuing social relevance is powerfully present, for example, in the bulletins which are sometimes carried on national radio prior to its appearance. For over a decade, Radio Vanuatu has transmitted special broadcasts announcing the probable timing of the Palolo's emergence (one of the most recent such reports appeared in the website http://www.fnsvanuatu.com/pvpbin/pvponline.exe, under the headline 'Wan taem long wan yia, worn blong solwota long wiken ikam!'). By influencing local people's awareness of the broader scale of the Palolo's appearance, the radio reiterates the importance of the Palolo as both an indicator of seasonal time and as a truly massive and communal phenomenon. 'Something that is bigger than our vision,' as a Tortes elder explained to me, referring to the fact that the Palolo's influence extended well beyond the visible horizon and the bounds of the local geographical imagination. Perhaps more importantly, to judge by local accounts, these radio bulletins have expanded awareness of cultural diversity nationwide by presenting descriptions of the many different methods by which local communities gather the worm and prepare it for human consumption. Through these presentations, which have rendered it a 'modemised' environmental event, the Palolo has become an important pan-insular phenomenon by which distant Melanesian peoples establish imagined connections and shared cultural/environmental values with each other (see Bolton 1999a and 1999b for comments regarding the influence of radio and the expansion of scale in the "imagined geographies' of Vanuatu societies).

As indicated at the beginning of this section, Torres islanders appear to interpret the mena of the tuwia as its capacity to herald the coming of the Palolo. The reason why the tuwia's premonition is held to be significant is because it comes at a time of increased horticultural activity, when new yam gardens are being planted (I refer to the primary yam crop, Dioscorea alata, because less important crop types are often cultivated at other times of the year). The emergence of the nut is significant for much the same reason: it is directly associated with environmental transformations that will have a direct impact on the health and eventual fate of new gardens. The main such transformations of this period are the shifting of the sun and the winds. Below, I explore how Torres islanders employ the nut and the sun as a mediating sphere through which they attempt to extend their own (human) form of mena to the environment in order to facilitate the progress of seasonal time and the success of 'living growth', or 'bloom' (vavalete), in their food-producing gardens.

MOVING THE SUN

Throughout much of the solar year, people in the Torres islands do not entertain any particular concerns about the changing path of the sun across the sky. However, towards the end of the Austral Spring, when the new yam gardens have been seeded and the sun approaches the southernmost point of its annual migration, islanders begin to take heed of its position at dawn upon the eastern horizon. (11)

There is a small beach of white sand on the small island of Loh (in the Tortes group) known as Peliauluwo that lies tucked away in the innermost corner of a large semi-circular bay that faces the south eastern horizon. From this place, one can sometimes discern the silhouette of Ureparapara, eighty miles away. Once a year at dawn (on November 15th by the Gregorian calendar), if the sky is clear, the people of Loh gather in Peliauluwo to observe the rising sun as it emerges just to the south of the outline of Ureparapara. This, they claim, is a signal that the Palolo shall emerge from the ocean later that day. Although there has never been a tradition of horizon-based astronomical observation in the Tortes, Loh islanders have long employed the peculiar solar alignment as a key indicator of the emergence of the Palolo and the approach of the summer solstice. (12)

Torres islanders maintain that the return journey of the sun towards the north immediately following the summer solstice--which, at the time of the Palolo's appearance, is still about six weeks into the future--is contingent on a special ritual act that must be performed towards the end of the harvest of the nut. This act is of critical importance to Torres people's perception of seasonal change, because they believe that it ensures the continuity of the sun's motion and the renewal of its annual cycle. The ritual itself consists of a brief and, to untrained eyes, rather uninteresting motion, but knowledgeable Torres elders describe it as a subtle procedure which includes sotto voce magical chanting and can only be carried out at the moment when the harvested nut is being transferred from the container in which it was initially gathered to the woven Pandanus basket in which it will be carried into the kitchen house. This ritual represents one of the most singular expressions of mena in the Torres Islands, which only certain humans are capable of attempting, and it cannot be understood without reference to the context in which the Palolo is harvested.

The traditional Torres method for collecting the Palolo is an extended procedure that begins with the coming of the migrating tuwia several weeks prior to the worm's appear ance. Thereafter, the sun-Ureparapara alignment serves as an additional reminder of the nut's imminent arrival. By keeping track of the rising sun during this particular period, people can (if they are interested, because not everybody is motivated to gather Palolo) begin to prepare for the harvest of the nut.

For ritual specialists from the islands of Loh and Toga one of the most important preparations that is set in motion up to two weeks prior to the worm's arrival is the manufacture of long-handled 'worm-catching' nets known locally as geprir. These instruments consist of a long (up to 6 or 7 foot) bamboo pole topped by a round, flat 'net' that is woven from thin strands of bark obtained from the trunk of a particular type of Beach Pandanus (L: vog; B: gras tri). Importantly, the geprir is understood by most Torres people to be one of several artefacts whose form distinguishes the material culture and social identity of the Loh and Toga people from those of Hiu, the northernmost socio-linguistic group of the Tones. But the manufacture of geprir is a dying art and only a handful of Torres elders now remember how to make them. Consequently, for most people any sturdy container, such as a bucket, suffices in order to gather the newly harvested sea worm. However, the 'other' relevant container, that is the basket in which the worm is carried into the kitchen, must be made of woven Pandanus, for reasons that are explained forthwith.

On the morning of the sun's alignment with Ureparapara, those Loh islanders who intend to harvest the nut begin to prepare their ground ovens, and gather yams--preferably of the type known in Bislama as wovile--and leaves from the Cottonwood tree (L: duv: B: burao). The yams and leaves are used as ingredients for baking the worm.

As explained previously, there is no way of measuring the exact day and evening when the Palolo are to appear (see note 12), but islanders become alert for any trace of its smell after the last full moon in October. Upon first perceiving the stench of the worm, which is the final and immediate precedent to its emergence, people wait out the remainder of the day preparing the ingredients and the yam pudding with which the Palolo will be cooked. As the day wanes, if no passing rains materialise (they are said to scare away the worm), those Torres islanders who have an interest in feasting on Palolo gradually arrive at the reef and prepare to gather nut. In contrast with other island groups, such as those of North East Malakula, where islanders paddle their outrigger canoes beyond the reef and collect the worm from the sea, Torres people stand on the rocky shoreline and gather the worm with a geprir because it is said to be taboo to intentionally come into contact with the water while gathering the Palolo; this, they say, would induce the nut to withdraw towards the open sea. One Torres man explained that this taboo probably derived from the common-sense fact that if one steps into the ocean at this time it is possible to get bitten by the numerous fish, including barracuda and moray eels, that come close to shore in order to feed on the masses of epitokes. Collection of the worm is gender neutral, and both men and women participate.

Shortly after they have positioned themselves upon the rocky seaside, people begin to call the worm to shore (L: rav mu, 'pull it in,' like a canoe,). They do this in two ways: by singing a short magical song of abundance, called tutuwa, and by carrying torches fashioned from dry coconut leaves. The root tuwa is a verbal evocation ('pulling in', or attraction) for animate things, such as fish, Palolo worms or even spirits. Thus, the tutuwa is a song that is intended to increase the volume of nut that come towards the shallows, but tutuwa can also be employed to evoke abundance, plenty and/or fertility in various living entities, such as food-producing gardens or in the wombs of pregnant women. (13) Although tutuwa are an important indicator of mena, in the sense of a person's capacity to elicit abundance from the living world, the key object in this procedure appears to be the torch.

As they gather Palolo throughout the course of the night and into the dawn of the following day, Torres people burn a great number of coconut-leaf torches, scattering copious amounts of ash and cinders in the process. According to Torres accounts, torches are employed simply because their light is believed to attract the worm--a belief that was also mentioned to me by people in Malakula. However, after conversing about Palolo-catching techniques with persons from other parts of the Northern islands, it emerged that torches are used by many groups across Vanuatu and that their specific make and meaning are often directly related to garden fertility. A clear example of these associations is visible in Atchin, where torches are fashioned from the sticks of wood that have been previously employed to direct the growth of yam vines. In the Torres, the relevance of using torches fashioned from coconut leaves resides in the abundant ashes that this dry leaf produces as it burns, because these ashes then alight on the reef, on the beach and on the baskets of worms in which the harvested Palolo are stored.

At daybreak, as the harvest draws to a close, the ashes which have gathered on top of the baskets-full of nut are mixed in with the green mass of worms. Tones islanders carry this mixing out by turning the worm over as they transfer it into a second container--the Pandanus basket in which the nut will be transported into the kitchen house. This manoeuvre is carried out with a large wooden fork that is deemed to be magically charged and should therefore be specifically fashioned for the event. The turning should ideally be performed as the sun begins to rise.

In the past, the 'turning' of the nut was accompanied by a magical song which has been forgotten by most present-day Torres people and was only remembered in fragments by some of the elder islanders with whom I spoke. The song is relevant because it appears to facilitate the transformation of the nut from marine creature into consumable foodstuff as it is transferred from one basket to another. However, some islanders report that the 'song of transformation' is actually directed at the emerging sun, rather than the worm, as one might suppose--because it is the sun and not the worm which commands the greatest attention at this time. This is the reason for which sunrise is the critical moment at which the ritual must take place: the 'turning' worm is directly associated with the 'turning' of the sun's annual movement, from south to north, immediately following the summer solstice--which, as previously stated, only actually takes place from six to eight weeks later, on 22 December. Failure to perform this ritual 'turning' is said to be potentially catastrophic because the sun would not be imbued with mena by the ritual performers and might thereby remain 'rooted' to its position upon arriving at the summer solstice. This would cause it to rise and set indefinitely over the same horizon, thus halting seasonal time and the progression of the horticultural cycle.

The process described above is evocative of Tambiah's theory of 'persuasive analogy' (1973), in which 'traditional' societies employ rituals that encourage future action (by contrast with modern scientific discourse, which, according to Tambiah, deploys "empirical analogies' in order to predict future action). But such a theoretical distinction can only function at a broader level of comparison between 'traditional' ritual and 'modern' discourse. In this case, it is worthwhile to consider that Torres people understand the production of such things as 'knowledge', action and 'capacity' to be the outcome of special processes of containment and ontological transformation.

These processes hinge on what I have referred to elsewhere as the 'aesthetics' of knowledge production in the Torres Islands (see Mondragon 2003, Chapter 7). These are principles which have their basis on the transfer of things (objects, persons) from one enclosing space to another, and their subsequent transformation into something different. Significantly, the thing being submitted to transformation may consist of something as simple as a yam (when cooked) or a worm (in a basket), or be as complicated as a human person (e.g. in the context of initiation ceremonies)--but in all cases mena is cited as a necessary element for the outcome of such procedures to be deemed successful. Ideally, the containing spaces wherein such transformations take place should be circumscribed by 'fibrous' enclosures (L: kukwek), of which an iconic example is woven Pandanus baskets. Other possible 'transformative' containers include such different objects as stone ovens, ritual enclosures and even human organs like the eyes and the stomach; the indigenous terms for these organs, which translate as 'containers made of flesh', are variations of the expression 'fibrous baskets' and, consequently, they are believed to act as the fundamental containers by which the multifarious images and substances of the outside world enter the human body, are transformed ('digested'), and eventually externalised as personalised knowledge, action or 'capacity' (see Rubinstein 1981:154, for strikingly similar metaphors as observed on the island of Malo, North-Central Vanuatu).

While it is not the purpose of this discussion to elaborate on the above imagery of symbolic containment and ontological transformation, I believe that these principles may be significant to a better understanding of the Palolo-sun ritual. Although similar formulae of 'ontological transformation' have rarely been described in other parts of Remote Oceania, the following description made long ago by Leenhardt suggests that the notion of enclosure and transformation within the context of human-environmental relations may still be widespread across the region:

 In the Loyalty Islands [Kanaky/New Caledonia], the terms tenge and
 cenga refer to all fibrous containers: the visceral bags--stomach,
 bladder, womb and heart-as well as the visceral fibres of a basket.
 Thought is an object, a given, an inspiration, a revelation, a
 lucky find sensed inside a visceral container, or glimpsed in a
 woven container like a basket. It must be remembered that an object
 in a basket ceases to be a part of the whole. The object
 circumscribed by the woven fibres is separated from the ensemble; it
 is other. On Lifou, the principal of the Loyalty group, sand from
 the beach and the same sand in a basket are two categories of sand,
 each has a different name (1979:9, brackets added).

Clearly, these are complicated principles that transcend the main themes of this article. But while they do not clarify how the symbolic transformation of the nut into consumable matter might influence the motion of the sun, these ideas do appear to play a role within the sun-turning ritual that renders it a unique medium by which Torres islanders employ the harvest of an oceanic creature in order to project their mena towards the sky and ensure the positive outcome of vavalete for the newly planted yam gardens.

The movements of the sun and the emergence of the Palolo are ecological phenomena which Torres islanders experience as part of a broader complex of seasonal transformations that include the predominant winds, weather regimes, lunar cycles and tidal phases. Below, I describe each of these phenomena, highlighting some of the other forms that the expression of mena can take in the context of human-environmental relations.

WINDS, CLOUDS AND WEATHER

The influence of the winds on the character of local weather and biorhythms is considerable, and is important to the daily activities, senses of geographical orientation and cosmovision of most indigenous groups across Vanuatu and Oceania (Dwyer and Minnegal 2000; Finney 1998; Sillitoe 1994). Not surprisingly, the calendrical relevance of the nut is strongly dependent on the fact that the Palolo's mating season coincides with the twice-yearly shift of the prevailing winds across Island Melanesia, which unfolds within the interval of the lunar month occuring between October and November. Importantly, winds do not occur in isolation and they are associated with various 'forces' that are attributed to the cardinal directions. In addition, winds are inseparable from the creation, movement and potential mena ascribed to clouds, and play a significant role in local rituals of weather magic.

Wind is not simply wind to the Torres islanders. Each wind, known generically as lang (sometimes pronounced leng) possesses a certain mena (in the sense of transformative power over living things) and is associated with a cardinal 'area' (as opposed to a 'vector', which is the common way of representing wind directions in Western wind-roses). (14) The areas from where each fang arrives constitute part of a complex (absolute, or horizon based) system of orientation by which Torres islanders express movement within the regional land and seascape. (15) Each wind-related area is further associated with specific meteorological and spiritual forces that lie beyond the visible horizon from whence the winds come.

The two predominant winds in Vanuatu are the south east trade winds, which last from May to early November, and the monsoon, which is the period of greatest cyclonic activity and runs from November to April. The trade winds originate in the southerly subtropical anticyclone centres of the South Pacific, while the northerly winds of the 'cyclone season' are characteristic of the regional monsoonal wind system of the Western Pacific. Together, these winds function as the main regulators of seasonal variation across Melanesia (Curry and Armstrong 1959, Brookfield and Hart 1971:4-21).

In the Torres Islands the predominant seasonal winds occur within a weather regime that is fairly equable throughout the year. Torres people sometimes speak of 'cool' as opposed to 'hot' periods--tuketakwe, 'when a fresh wind is blowing', and vetenatung, 'when the sun is strong'. These categories follow the two main climate divisions in Vanuatu, which are termed the rainy or hot season, that lasts from November to April, and the dry season, which extends from June to December (Weightman 1989:2). In turn, both seasons are affected by anomalous shifts in the macro-regional climate, characterised by phenomena such as the El Nino Southern Oscillation, which provoked severe drought across Melanesia during 1997-1998 (Allen and Bourke 1997, Ballard 1998) and, most recently, in late 2003.

The south east trades are known as langmehelavu in the language of Loh, and even though they do not appear to be associated with powerful magic, the effects of the langmehelavu on the everyday lives of people ensure that they occupy a prominent position in the Torres 'wind-rose', alongside the more dangerous and powerful north and north-west winds. By contrast, following a pattern that is common in Vanuatu, there is no particular importance attached to the infrequent winds that arise from the south and south-west, and consequently I was unable to elicit terms for these features in either Atchin (Malakula) or the Torres. (16)

The trade winds affect the movements and activities of Torres islanders in various important ways. If hunting for wild fowl or flying foxes (fruit bats) during the period of the trades, people will generally head for the leeward sides of the islands, where these flying creatures roost in protective cover from the winds. Conversely, flying foxes become a nuisance when the south east trades give way to the months of summer calm and erratic northerlies, as they migrate to the windward sides of the islands and are able to feed on the fruit trees of eastward-facing gardens. The power of the trades also makes the ocean surface choppy, provoking endless whitecaps and making maritime travel a risky venture, or at the very least a highly arduous and uncomfortable undertaking. Which explains why most inter-island voyaging, as well as deep-sea fishing, is kept to a minimum during this period.

Due to the many ways in which the trade winds affect Torres peoples' perceptions of the environment, and given these winds' uninterrupted presence for six or seven months of the year, few aspects of the weather are more dramatic than the sudden calm which marks the end of this 'windy' season. This is followed by the beginning of the 'cyclone' season, which is characterised by periods of complete quiet that are interrupted by erratic winds from the north west and the north, and unpredictable hurricane activity. The winds of this season arise in the north and north west and are known in Loh as langrevot and langmetegelau. (17)

As they do with the winds, the people of Loh distinguish between different kinds of clouds. Indeed, drawing on detailed comments provided by people from Atchin, Raga and the Torres, it appears that across North and Central Vanuatu clouds may play a more prominent role than the winds in people's relations with the environment. This is probably due to the fact that each type of cloud is believed to possess particular 'forces' (often man-made) that preclude generalising terminologies; a peculiarity that is made evident by the absence of a generic term for 'cloud' in the language of Loh. The manipulation of the winds, and especially of the clouds, is thought to be critical to a person's ability to manipulate the elements. Clouds are therefore associated with specific meteorological phenomena that are often believed to be the object, or desired outcome, of weather magic (L: n'ah). In cases where a cloud becomes the object of human attention, it is usually interpreted as a sign that magic--whether good or evil, powerful or quotidian--has recently been carried out in the vicinity of the observer. (18)

According to various Torres people who are familiar with different types of weather magic, the ability 'to observe' (L: tulung), and by thus observing to manipulate the weather is the key to successful magic. Whatever the type of cloud or wind that is being manipulated, the critical faculty that is necessary for a man to extend his mena towards the environment--and, through the environment, towards other persons--is vision. Consequently, the sign of a magician's efficacy over the weather should also be visible for all to see, after the magic has been cast, as a particular formation at the head of large cumulus clouds. This sign is known as her (also har; neher when paired with its pronominal) and is synonymous with a kind of plant that goes by the same name and is a basic material component in rainmaking techniques (see Durrad 1940c: 186-190 for a detailed account of rain-making magic as recounted to him by the people of Lob one hundred years ago). (19)

The two 'wind' seasons described above provide the main meteorological divisions of the Torres horticultural cycle, which comprise the clearing, planting and seeding of new gardens (approximately September to February), and the subsequent period of growth, bloom and harvest (February to June). But while this binary division of the seasons creates a general meteorological backdrop for the horticultural cycle, it is the lunar year--which can last between eleven and twelve months--that constitutes the basic temporal marker for the 'calendar' proper. This is why the moon, through its periodicity and its effect on the oceans, is of greater practical and temporal relevance to the Torres people.

THE MOON AND THE TIDES

The moon (L: megage) is of particular concern to small island societies such as those of the Torres because the regularity of its synodic (monthly) cycle stands in direct relation to tidal events and nocturnal activity, which influence the productive efforts--both terrestrial and maritime--of human beings. The importance of lunar rhythms is underscored by the highly detailed descriptions and categories that Torres people have developed in order to define lunar phases and, in particular, the tides.

Torres people do not employ terms for specific day-to-day changes in the moon's appearance, exception being made for the term nemegage leu, which denotes the first waxing and the last waning crescents. Rather, they divide the phases of the moon into three general categories: nemegage mad den (lit. 'it has become round'), indicates those phases which take place between the first quarter and the full moon; nemegage mad du metou (lit. 'it has become ripe/ready'), denotes the phases that take place between full moon and the last quarter; and nemegage lereleaere, which refers to the phases that occur between the last quarter and the new moon (the meaning of this name appears to be forgotten). The terms given to the first and last crescent moons of a month are interesting in that they incorporate the word leu, which signifies the first appearance of objects which are undergoing the beginnings of 'living growth', such as the first appearance of shoots and fruits on the branches of a tree (e.g. netawe leu, nialweh leu, indicating growth on two different kinds of tree--the tawe and the nialweh).

The location of the moon (whether on the horizon or in the sky) is described in terms that refer to the quality of moonlight, rather than the actual position of the moon itself. This may or may not be related to the fact that, according to the aesthetics of Torres perception, the existence of a given object is contingent on the quality of the light which makes it visible. (20) In this case, nehurhurinu refers to the halo of light which appears on the horizon just prior to the emergence of the rising moon. Conversely, nehirher describes the light which surrounds the moon when it is visible in its entirety in the sky. Unfortunately, the specific meaning of both terms, as well as any broader relationship that they may have had with other human or non-human activity, appears to have been forgotten.

The richest variety of terms related to the motions and influence of the moon occurs in those which apply to the timing and type of various local tides. I have thus far been able to collect at least seventeen different categories for various tides or tidal processes, many of which refer to the diurnal or nocturnal position and particular phase of the moon in the sky. Most of the names for tides can be slotted into one of two general categories, which indicate type (high tide, liavu; or low tide, met), and the general time of day (morning, evening, midday) at which they occur. I was able to identify six different categories of high tide: liavunelivuliavu, is generic for the beginning of any given high tide; liavumuren, high tide that occurs during the day; liavumutave a high tide in the morning; liavrevrev, a high tide that occurs in the evening; liavulilowate a high tide at midday; liavuhar, the highest tide of the year. Conversely, there are at least four types of low tide: numetmet, the beginning of a low tide; metmetavu, low tide in the morning; metmuren, low tide at midday, metrevrev, evening low tide. In addition to these high/low tide categories, there are a number of other categories about which I was never able to obtain more than incomplete information, which correlate the time of day, tidal type and lunar phase. Some of these are: metmuligu, a low tide which occurs during the time of the new moon; metmegage low tide during full moon; metmegagemetavutoten, low tide during the day while there is a waning quarter moon visible in the morning sky.

The highly organised body of authochthonous environmental knowledge presented here demonstrates that Torres people pay acute attention to those aspects of the environment that most closely influence their lives--for tides affect the freshwater lens under their islands and the soil moisture content of their lowland (coastal) yam gardens, in addition to determining the various methods and times that are dedicated to harvesting the sea (see Mondragon n.d. and 2003, Chapter 5, for a full description of coastal and hilltop horticulture in the Torres Islands). In turn, terrestrial and maritime activities are both considered to be different aspects of the elicitation of abundance, and their outcomes are often correlated to the successful deployment of mena. Finally, at present there is apparently no trace of weather rituals or garden magic associated directly with the moon and the tides; it appears that 'capacity' is measured in this case by the quality of the person's practical knowledge of these phenomena--and the subsequent advantages that derive therewith. Nevertheless, it is striking that few mythical or symbolic kinds of narratives are available regarding the moon and the tides. (21)

THE TORRES 'CALENDAR'

The lunar month, or synodic cycle, continues to be the rough unit of measurement for the traditional Torres 'calendar', which is divided into eleven (possibly twelve) such lunations. Only a handful of people now remember the traditional names for these months, following their gradual displacement by the initial imposition, and eventual adoption, of the Western (Gregorian) calendar and the necessities derived from relating to extra-local agents and institutions.

The 'traditional' months of greatest importance correspond to the most demanding phases of the horticultural cycle, namely, those during which the harvest of the primary yam crop (Dioscorea alata) and the planting of new gardens take place (mid-August, September, October, November, December and early January). Ecological transformations during this period include the annual shift of the winds, the appearance of the Palolo worm, the arrival of the tuwia and the sun's movement towards and away from its position at the time of the summer solstice--all in combination with the ongoing effects of the moon and the tides. Another significant event that symbolically marks the end of the planting season is the appearance of the vuhoro (also pronounced vuhorow), a flying ant, in the first or second week of January. (22) In addition, there is evidence that some asterisms (constellations) are recognised by the Torres islanders and integrated into the above framework (due to the limits of this article I have not discussed the imagination of the heavens, another topic that has long been marginalised in Melanesian studies, but am currently preparing an article about ethnoastronomy in the Torres Islands).

The reconstruction of this 'calendar' as I present it below would have been impossible on the basis of oral remembrances alone and is not a format which Torres people employ. It is an alien model which exhibits a useful correlation between lunations, phases of horticultural activity and Gregorian calendar months--but it is based on a comparison that does not arise in Torres people's minds. For the most part, the chart below is based on Durrad's original model (1940a). (23)

The lunar months that constitute this annual calendrical framework are not a celestial idiom' for the horticultural cycle, so much as a pragmatic representation of the extensive and accumulated relations towards different environmental spheres--sea, soil, weather, biorhythms--that are focused around the process of 'making/working' (L: rag) a garden (L: raglete). In this sense, it resembles those of other Island Melanesians, such as the Trobriand islanders, of whose calendar Malinowski stated that 'the real framework of native [calendrical] division, as well as their picture of the year, is represented by a succession of activities in garden work' (1927:211). (24)

CONCLUSION

The concepts of the living world that give form to the Torres 'calendar' derive from what Biersack refers to as the 'life-world of praxis and not just signification' (1990:67). This life-world of praxis speaks directly to broader senses of place, ontological principles and social values, which render seasonal change and the perception of temporality meaningful. In various parts of Melanesia some of the values derived from human-environmental relations include 'pollution' and 'purity', 'growth' and 'decay', are often gendered, and are variously related to different features of the land, the sky and the sea (e.g. Biersack 1982; Sorum 1984; Kahn 1986, 1990; Munn 1986). Herein, I have taken the subject of temporality in order to approach the values of capacity and productivity that are evoked by the term mena.

As expressed by the Torres islanders, mena is a kind of supernatural power, or capacity, that efficacious persons can evoke with the purpose of directing it at an almost unlimited range of animate and inanimate objects. Because it is primarily expressed through ritual action, mena can be described as an acting agent in chants, invocations, and other forms of evocation directed at natural or supernatural spheres. In this regard, the purpose of mena is to facilitate the powers or capacities of the objects or spheres towards which it is directed. The present article has described the presence of mena within objects or phenomena that are directly related to the seasonal and environmental milieu of the Torres Islands. Importantly, these objects and phenomena are in direct relation to cycles of food-production and human reproduction, and encompass things as complex as the sun or as simple as the soil and crops of a garden.

It would, however, be a mistake to consider mena simply as a self-standing or 'facilitating agent'--as an external force which is channelled and manipulated by humans. In this sense, I believe it is important to rectify that mena, as understood by Torres and Banks islanders, and probably also by neighbouring Melanesian groups in the South West and Central Solomon Islands, is somewhat unlike the 'sacred power' of Central and Eastern Polynesian societies. Rather, in North Vanuatu mena is a power that appears to derive originally from human qualities, from ways of being, knowing and acting that the efficacious individual must cultivate and perfect throughout a lifetime of relations with other persons and things. For this reason, in the Torres Islands the successful evocation of mena is interpreted as being an important expression of a person's ability to present him or herself as a powerful, and powerfully rooted, individual with respect to the local social and natural milieu.

From a social perspective, mena stands for 'capacity' or 'efficacy' insofar as it reflects the condition of a successful provider and nurturer, of someone who can elicit 'living [and life-nurturing] growth' (vavalete)--which is iconically, but not exclusively, represented by healthy and plentiful food-producing gardens. Importantly, the process of becoming an efficacious 'provider and nurturer' implies that the individual has successfully negotiated the key phases of life which give form to one's sense of social and geographical belonging. Among other things, this implies that the individual is able--or perceived as able--to create an effective network of kin relations, which in turn define the spaces that the person inhabits and cultivates (houses and gardens), and the persons and environs towards which production and nurture are directed. Consequently, the ability to evoke and project mena is described by Torres people as representative of an individual's capacity to manage their territory and social networks in a morally positive way. A healthy garden is an embodiment of such capacity, but the capacity that derives from mena touches on many other aspects of a person's social status--such as the qualities of productivity and success that function as an important standard by which men and women are judged when aspiring to elevate their rank within particular (gendered) status-alteration systems (L: hukwe), or when attempting to achieve membership within specialised ritual sects (L: temetsie). (25)

Following from the above, it is clear that Torres islanders' perceptions of the environment, and of the values that they deploy in relating to seasonal and climatological phenomena, can provide a more nuanced understanding of time and 'place' in Vanuatu. For this reason it is worthwhile to briefly reconsider existing ideas about both concepts.

As regards the specific problem of temporality, the Torres 'calendar' clearly cannot be defined as a formal system of time-reckoning. With the exception of the six to eight week period that begins when the rising sun transits through the silhouette of Ureparapara and ends after the summer solstice, Torres islanders have no interest in 'measuring' time on a scale of years, months or even days. This does not mean that they have no concept of the passage of time, but merely that they do not necessarily equate time with 'measurement' or 'duration'. By the same token, the temporal organisation of the horticultural cycle and its broader seasonal framework is not an attempt to control or define 'time'--i.e. after a Heideggerian or a politico-economic rationale--so much as it is an effort at organising social activities and human-environmental relations which in turn provide specific and relevant contexts for the expression of mena. In this regard, it is important to point to the absence of a self-standing term for 'time' in the language and expressions of Torres societies. Rather, the Torres people have a term (L: gwe), which appears to condense the notion of 'place' and 'time', insofar as it refers to 'layers' of human generations as being intrinsic to the accumulated soil and surface area of every island (a similar term is reported for the Sa speakers of South Pentecost by Jolly, 1999:284). (26)

What of the effects of modern systems of time-reckoning? It has not been within the limits of this article to discuss the various ways in which such systems (calendars, school and church clocks, plane schedules, wristwatches) have been variously adopted by the Tortes islanders. Nevertheless, it is important to stress that, as in much of rural Vanuatu, two of these systems have become firmly ensconced in the rhythms of daily life. One is the combination of secular and Christian calendars of important dates, as determined by the State (Independence Day and New Year being the most important ceremonies) and the Anglican Church (which has made of Christmas and Easter its most important dates). The other is the school and church attendance schedules, which are also external but largely self-imposed systems, although their imposition over the time-perception of daily life can only be said to possess varying degrees of success. In this context, what is important is that the nature of the Torres 'calendar', as part and product of the subsistence systems of the islands, as well as a medium for the expression of human-environmental relations and values, explains its persistence and practical renewal--which is mostly effected by a majority of people who are not necessarily privy to the specialised knowledge of, for instance, the traditional names of the months--long after external systems of time-keeping were superimposed on lives of Torres islanders.

As regards the idea of 'place' (B: ples), this is a theme that has figured prominently in the ethnographic record of Vanuatu (Bonnemaison, 1994, 1997[1993]; Bolton 1999b; Hyslop 1999; Jolly 1994, 1999). However, most existing explorations have tended to focus on the terrestrial dimensions of 'place', primarily taking garden spaces and villages as coherent points of reference in the discussion about territorial identity (e.g. Jolly 1982; Rodman 1987). Such perspectives have been relevant insofar as they attempt to reflect the importance allocated to 'ground' (B: kraon) and toponymy as pervasive terms of reference for group identity and senses of belonging in Vanuatu, but they have done so by eschewing the meteorological, oceanic and celestial contexts within which Island Melanesians imagine themselves and play out their notions of agency. As the present article has sought to demonstrate, for a majority of ni-Vanuatu 'place' is clearly perceived to constitute part of a broader, fluctuating field of environmental events and phenomena, and for this reason alone it is important that Vanuatu researchers interested in the study of 'geographical identity' (to borrow a less terrestrially-biased term from Bonnemaison) accord greater significance to seasonal change and human-environmental relations.

The Traditional Calendar of Loh and Toga, Torres Group

Lunation Meaning Associated activity

nomwegot goto ? Clearing of new gardens

wo gwona ? Planting of yam begins

wo memo ? Planting of yam and other
 garden crops

nut Palolo worm Planting of gardens continues

nut lavu Great Palolo Yam poles erected

nut melugehe Green Palolo Garden left to grow, 'bloom'

nut memudaru Red Palolo Yam vines are in full (green)
 leaf

Numwegot meta Wet yam Yam leaves begin to turn red

Numwegot reng Dry yam First fruits of yam crop are
 lifted

Numwegot gilgil To 'dig out' wovile Harvest of the wovile
temeg (wild yam)

Numwegot gilgil To 'dig out' taro Harvest of the taro
kweta

? Harvest in full swing

Lunation Seasonal event(s) Gregorian

nomwegot goto August

wo gwona Arrival of tuwia September

wo memo Beginning of monsoon, first appearance October
 of Palolo

nut The sun at Ureparapara; second Palolo November

nut lavu Summer solstice December

nut melugehe Appearance of the vehoro (see note 24) January

nut memudaru February

Numwegot meta March

Numwegot reng End of monsoon, beginning of S.E. trades April

Numwegot gilgil May
temeg

Numwegot gilgil Winter solstice June
kweta

? July *

* The fact that there is apparently no name for the month of July might
be indicative of the lack of significant horticultural activity
associated with this period of the year.

ACKNOWLEDGEMENTS

This article is based on a period of eighteen months of ethnographic fieldwork in Vanuatu that was made possible by a postgraduate research scholarship (No. 111214) awarded by the Consejo Nacional de Ciencia y Tecnologia (CONACYT) in Mexico, as well as a grant from the William Wyse Fund for Anthropological Fieldwork (administered by Trinity College and the Department of Social Anthropology, University of Cambridge). I would like to thank Professor Dame Marilyn Strathern for having first brought to my attention Prof. Fred Damon's work regarding Melanesian calendars, and Prof. Damon himself for his critical reading of this article. Thanks also to Mme Sophie Caillon and Mr Thorgeir Storesund Kolshus, as well as the anonymous reviewers for Oceania, for valuable criticism and suggestions. I am deeply indebted to the people of the Torres Islands, and especially to jif Peter Wotekwo, Mrs Mebel Ramevetang, Mr Philip Tagaro, Mr Bretin Wokmagene Mweiu and Mr Titus Joel, for freely sharing their environmental knowledges with me.

NOTES

(1.) The only existing descriptions of calendrical frameworks that I have found for Vanuatu can be consulted in Codrington 1891:348-51 ; Deacon 1934; Durrad 1940a:400; Layard 1942:6-7, 288; and Weightman 1989:42-3, 82. However, these descriptions are succinct and schematic, and fail to provide insights into the broader links between indigenous understandings of ecological phenomena, agency and social production.

(2.) For further ethnographic materials regarding time and temporality in Near and Remote Oceania, Eastern Indonesia and South East Asia, see Cuisinler 1956; Fox 1979; Forth 1982, 1983; Haley 1996; Hoskins 1993; Jolly 1999; Schroter 2000; Senft 1997; Suggs 1997; Tannenbaum 1988.

(3.) In addition, as I learned during the course of my research on the ecology of the Palolo, there is also a notable dearth of published zoological data regarding the worm's life-cycle and habits. For a relevant but slightly dated overview see Palmer and Goodenough 1978.

(4.) All indigenous language terms in this article are in the language of Loh island and in Bislama, the neo-Melanesian pidgin of Vanuatu. Loh is one of the three socio-linguistic communities that comprise the population of the Torres group (the others are those of Toga and Hiu). Loh is a language that has never previously been the subject of expert linguistic reconnaissance, and it appears for this reason to have never been recognised as different from the language of Toga; however, the people of Lob consider their language as different to that of Toga, and since there appear to be some variations at a supra-dialectical level between the two I have tentatively classified it as a separate language. Throughout this article, all non-English language terms are indicated in brackets and preceded by an L if they belong to the Loh language, or by a B if they are in Bislama. With the exception of those Loh terms that appear in the opening quote taken from Durrad all translations and transliterations have been made by the author.

(5.) Until my arrival in this small island group, the societies of the Torres Islands had not yet been the subject of an extensive ethnographic field study. In the past, the only relevant printed material regarding the Torres was provided by the Rev. W. J. Durrad, who lived on the island of Loh for six years at the beginning of the 20th century and summarised some important aspects of the kinship arrangements, subsistence activities and ritual life of the peoples of Lob, Toga and (to a lesser degree) Hiu in his 'Notes on the Torres Islands' (Durrad 1940a, 1940b, 1940c). My research in the Torres Islands was carried out over a period of ten continuous months between 1999 and 2000, and constituted the bulk of my PhD thesis (Mondragon 2003), from which I have drawn for the present article. Some relevant archaeological work was previously carried out in the islands of Toga and Metoma by Jean-Christophe Galipand (1996), and more recently there have been brief visits by anthropology and archaeology students from French and American institutions.

(6.) Several decades ago, the Rev. WJ. Durrad provided a basic description of the monthly divisions of this calendar in his 'Notes on the Torres Islands' (Durrad 1940a, 1940c), but failed to relate the calendar to Torres people's broader understandings of, and relations towards, the physical world.

(7.) Heteroscelus incanus, commonly known as the Wandering Tattler, is a migrating bird which moves south during the northern Winter. The tuwia begins its long journey in the North West coast of North America and passes through Vanuatu during the months of October, November, and early December, on its way to New Zealand, where it nests throughout the Austral summer. Its first appearance in mid-October precedes (and heralds) that of the Palolo worm by several weeks.

(8.) As the well-known work of Malinowski and Leach about the milamala demonstrate, across Melanesia the appearance of the Palolo appears to occur at slightly different times even within a specific group of islands such as the Trobriands (Leach 1954).

(9.) 'Most animals and plants--probably all living cells--have internal clocks buried deep in their biochemistry. These biological clocks manifest themselves in all kinds of physiological and behavioural rhythms. They are linked to external astronomical clocks, and normally synchronised to them. But the interesting thing is that if the biological clocks are separated from the outside world, they carry on regardless. They truly are internal clocks.' (from Richard Dawkins, Opening Lecture for the 'About Time' Exhibition, 6th February-1 April 2001 at The Ashmolean Museum of Art and Archaeology, Oxford; see http://www.ashmol.ox.ac.uk/ash/news/news020.html for a transcript of this lecture).

(10.) Strictly speaking, humans come into contact only with the epitokes. For the sake of simplicity I refer generically to the epitokes, the rear-ends of the Palolo, and the body of the Palolo itself, as 'the Palolo worm,' or 'the worm'. The epitokes of the Palolo resemble long strands of green string (thus the viridis in the worm's name) and are of a grainy texture and taste; once they have been taken out of the water, epitokes are fragile to the touch, and if not handled with extreme care quickly break down into a soft green mass. It is this mass which is considered a delicacy and utilised as a key ingredient in local dishes.

(11). See Ammarell, 1988, for examples of other Austronesian calendars in which the measurement of time by the movement of celestial bodies is undertaken only during specific periods of the solar year.

(12.) Every year, as observed from Peliauluwo, the rising sun invariably 'touches' the northern end of Ureparapara on 29 October, and transits the length of the silhouette of that island for two weeks, emerging to the south of it on 15 November. However, the phases of the moon (i.e. the lunar year) and the precise date of the Palolo's emergence fluctuate by a few days every solar year--thus rendering the fixed solar alignment of 15 November a relative point of reference. For the people of Loh the presence of the rising sun near the shadow of Ureparapara, occurring as it does within the same month during which the two regular appearances of the Palolo take place, serves as a temporal 'window' against which they can bracket the moment of the sea-worm's arrival. This "bracketing', however, was not clear from the oral accounts that I had been given prior to the event. In fact, for a majority of people on Loh, the correlation between this sunrise event and the emergence of Palolo represents an exact fit, rather than a 'window' during which different events are linked by relative temporal proximity. See Wagner 1987:41-2, for a similar 'bracketing' of astronomical events and ambiguity in the measurement of calendar time in relation to horticultural practices in New Ireland.

(13.) One of the men from the island of Toga who provided valuable information regarding the nut, reported that his great-grandmother was named Retuwaut, "she who calls forth the Palolo', in recognition of her capacity to sing songs that could 'pull in' the Palolo. The woman's name derives from the female pronominal re, 'woman/mother', tuwa, 'to draw forth abundance', and ut, the root for Palolo worm.

(14.) One of the main differences between thinking of wind directions as areas, rather than vectors (which is how most Western analysts tend to approach spatial terms of reference relating to wind directions and cardinal points), is that vectors are often representative of an Occidental, precision-based cardinal system of orientation that imagines a spherical horizon which is divided into 360 angular degrees, while areas evoke broad regions of natural (and often supernatural) influence and movement. As such, 'area' seems a more nuanced term with which to approach Melanesian models of orientation, which are not necessarily concerned with horizon-based measurement of the cardinal points.

(15.) On an everyday basis, Torres people employ two other systems of orientation, one relative (ego-centred) and the other deictic (speaker-centred), for expressing local movement. See Hyslop (1999:28), and Senft (1997), for comparative discussion of absolute and relative systems of orientation in Austronesian languages.

(16.) Nevertheless, a sagely informant from Raga was able to produce names for these (south and southwest) winds in the Raga language, probably reflecting the fact that, by contrast with Atchin island and most of the settlements of the Torres, which lie to windward or facing the north, Raga also lies open to the southwestern (leeward) horizon and the winds that originate there.

(17). The name for the langrevot was incorrectly cited as votven by Durrad (1940c: 188), who seems to have confused the root for 'north-west' (L: vot) with the term for a north-west wind that is in movement (L: vot ven).

(18.) On average, the skies around the Torres islands are characterised by high-altitude (stratus) cloud formations, known as paga in Loh, and small, scattered (and, during the S.E. trades, rapidly moving) cumulus clouds at lower altitudes, with relatively few overcast days (L: teroro, 'when the sun is especially strong', i.e. in the absence of clouds). It is the low cumulus clouds, meligu, which are most frequent throughout the year, and Tones people generally refer to them when attempting to manipulate the elements through magic. Another type of cloud which is an important marker of a man's ability to influence the weather are the thick red formations (L: merevelal) which gather over the horizon at dusk or dawn and are associated, respectively, with (human-induced) calm weather and rainstorms.

(19.) The faculty of powerful vision (tulung), conjoined with the term for efficacious weather magic (her or har) and with the term wutol, which stands for both physical strength and 'capacity' (literally, 'like he who can cut an enormous quantity of fresh wood in little time'), is neatly encapsulated in the name of a powerful ancestor, called Tulungharwutol, who is one of the few long-dead persons that is still widely remembered because of his capacity to control the weather.

(20). This is in contrast to other places, such as West Ambae, where people are said to measure the position of the moon with respect to its distance from the hump of the large volcano that towers above the local eastern horizon (I have relied on this information as it was passed on to me, in great detail and following local linguistic categories, by an Englishwoman who resided in West Ambae for several years as a voluntary teacher, but whose name I unfortunately did not retain after our meeting in the town of Luganville).

(21). The lack of storied narratives regarding the nature of the moon and other heavenly bodies has long been noticed by other observers of the Banks and Torres areas. For example, in the 19th century Codrington reported that 'in the Banks Islands the Sun and Moon are thought to be rocks or islands' and were sometimes associated with mythical persons, but stated that these stories were 'told without any serious belief' (1891:348). Nevertheless, in Atchin the moon is presently still held to be an agent of female fertility, and when related to the consumption and exchange of garden produce and seafood it becomes an important idiom for the reproductive nature of inter-island marriage ties. In addition, I have recorded a substantial body of ethnoastronomical knowledge regarding Torres islanders' perceptions of the night sky and the celestial sphere, all of which contain strong mythical components. The information about Atchin was provided by Bernard Georgie Rowsie, of Atchin, and Kirk Huffman, pets. comm. See also Biersack 1982 and Wagner 1987:37-38 for relationships between female sexuality and horticultural abundance in relation to the moon in other Melanesian societies.

(22.) Once a year, over the course of two or three nights, these creatures appear by the thousands in the evening, and are attracted by the light of kerosene lamps and hearth-fires. Their emergence marks the moment when these ants undergo a specific metamorphosis during which they sprout wings and leave their nests in order to mate, but in the absence of specialist descriptions on this creature I was unable to identify either the type of ant or the reason for its sudden appearance. The wingless phase of this insect is known as nevuho; the name for the flying ant, vuhorow, means 'voho that goes down,' from the directional row, 'to go down'; the meaning of this name might possibly be related to the fact that the flying ants 'come down' towards the coast from their hilltop nests. The vuhoro is employed as a marker for the end of the period when the posts for yam vines should have been erected, and effectively signals the end of the planting cycle within the new gardens (see Durrad 1940c).

(23). At his request, I presented jif (B: chief) Peter Wotekwo of Loh island with a copy of this calendrical scheme. I do not know whether jif Peter will manage to re-learn and transmit it to future generations as we reconstructed it, but his and other islanders' interest in recovering this knowledge in the format which I produced it were genuine and deeply felt.

(24). Damon has advanced similar arguments regarding calendars in the island of Muyuw and the broader Kula region (Papua New Guinea): 'No person or place has the function of linking celestial bodies for the purpose of keeping time. What is at issue here is not knowledge about celestial matters but constructing what is known into a conceptual system which speaks to something more important than the moon, sun, and stars, and perhaps even yams'; the point being that Kula societies 'use celestial phenomena [in order] to organize production.' (1982:227 italicised emphasis and brackets added).

(25). Compare my argument regarding the relationship between temporality, capacity and the evocation of social values with Roy Wagner's observations of the Usen Barok, in New Ireland: 'Because food production plays a central role in gender relationships, in defining and realizing the moral complementarity of men and women. however, the awat [traditional Barok calendar] is something more than mere speculative analogy. As a means of correlating concerted human activity through the observation of external events, and of focusing human concern on an image of productivity in the process, it becomes in fact a means of eliciting, or indirectly bringing about, the social and moral relationships of gardening' (1987:44, brackets added).

(26). While this approach appears to be at odds with Alfred Gell's rejection of 'the trend of thought which distinguishes different species and varieties of time on the basis of different types of processes happening in time' (1996:315), I would argue that it is not so much different species of time as different experiences of temporality (such as may derive from experiencing the growth of generations of people, sequences of exchange transactions, forest successions, etc.), and the attendant values and concepts that guide or derive from these experiences--which is not the same as 'time' understood in the narrow sense of measure, duration or process--that is at issue here.

REFERENCES

ADAM. B. 1994. Perceptions of time. In T. Ingold (ed.), Companion Encyclopedia of Anthropology. London: Routledge.

ALLEN, B.J. and R.M. BOURKE. 1997. Report of an assessment of the impact of frost and drought in Papua New Guinea. Australian Agency for International Development: http//www.ausald.gov.au/publications/general/other/pngdroughtrep.pdf

AMMARRELL, G. 1988. Sky calendars of the Indo-Malay archipelago: Regional diversity/local knowledge. Indonesia 45:85-104.

ATTALI, J. 1982. Histoires du Temps. Paris: Librairie Artheme Fayard.

AUSTEN, L. 1939. The seasonal calendar of Kiriwina, Trobriand Islands. Oceania 9:201-219.

1950. A note on Dr Leach's 'Primitive calendars'. Oceania 20:333-335.

AVENI, A. 1989. Empires of Time: Calendars, clocks and cultures. New York: Basic Books.

BALLARD, C. 1998. Melanesia in review: Issues and events, 1997. Irian Jaya. Contemporary Pacific 10:433-41.

BARNES, J. 1971. Time flies like an arrow. Man (N.S.) 6:531-552.

BIERSACK, A. 1982. Ginger gardens for the ginger woman: rites and passages in a Melanesian society. Man (N.S.) 17:239-258.

1990. Histories in the making: Paiela and historical anthropology. History and Anthropology 5:63-85.

BOLTON, L. 1999a. Introduction. In L. Bolton (ed.) Fieldwork and Fieldworkers in Vanuatu. Oceania Special Issue 70:1-8.

1999b. Women, place and practice in Vanuatu: a view from Ambae. In L. Bolton (ed.) Fieldwork and Fieldworkers in Vanuatu. Oceania Special Issue 70:43-55.

BONNEMAISON, J. 1994. The Tree and the Canoe. Honolulu: Hawaii University Press.

1997[1993]. Les lieux de l'identite: vision du passe et identite culturelle dans les lies du sud et du centre de Vanuatu (Melanesie). In E. Bernus, J. Polet and G. Quechon (eds.) Autrepart: Empreintes du passe. Paris: Editions de I'Aube/ORSTOM (Cahiers des sciences humaines, n.s. 4).

BROOKFIELD, H.C. and D, HART. 1971. Melanesia: A geographical interpretation of an island world. London: Methuen.

BOURDIEU, P. 1968. The attitude of the Algerian peasant toward time. In J. Pitt-Rivers (ed.). Mediterranean Countrymen: Essays on the social anthropology of the Mediterranean. Paris: Mouton.

BURMAN, R. 198 I. Time and socioeconomic change on Simbo, Solomon Islands. Man 16:252-268.

CIPOLLA, C. 1981. Le Macchine del Tempo. Bologna: Societa Editrice il Mulino.

CODRINGTON, Rev. R.H. 1891. The Melanesians: Studies in their anthropology and folfklore. Oxford: Clarendon Press.

COOMBE, F.E. AND H.J. BRAUNHOLTZ. 1934. Magical stones from Mota, Banks Islands. Man 34:34-35.

CUISINIER, J. 1956. Un calendrical de Savu. Journal Asiatique 144:111-119.

CURRY, L. and R.W. ARMSTRONG 1959. Atmospheric circulation of the tropical Pacific Ocean. Geografiska Annaler 41:245-255.

DAMON, F, 1981. Calendars and calendrical rites on the northern side of the Kula Ring. Oceania. 52:221-239. 1990. From Muyuw to the Trobriands: Transformations Along the Northern Side of the Kula Ring. Tucson: University of Arizona Press.

DEACON, B. 1934. Malekula: A vanishing people in the New Hebrides. C.H. Wedgewood (ed.), London: Routledge & Sons.

DURRAD, W.J. 1940a. Notes on the Tortes Islands. Oceania 10:389-403.

1940b. Notes on the Tones Islands. Oceania 11:75-109.

1940c. Notes on the Tortes Islands. Oceania 11:186-201.

DWYER, P. and M. MINNEGAL. 2000. El Nino, Y2K and the 'short, fat lady': Drought and agency in a Lowland Papua New Guinean community. The Journal of the Polynesian Society 109:251-272.

ECKLUND, J. 1977. Marriage, Seaworms, and Songs: Ritualized responses to cultural change in Sasak life. PhD Thesis. Cornell University.

EVANS-PRITCHARD, E. 1939. Nuer time-reckoning. Africa 12:189-216.

FINNEY, B. 1998. The Pacific basin: an introduction. In The History of Cartography Vol. 2: Cartography of traditional African, American, Arctic, Australian an Pacific Societies. Chicago: University of Chicago Press.

FOX, J. 1979. The ceremonial system of Savu. In A.L. Becker and A.A. Yengoyan (eds.) The Imagination of Reality: Essays in Southeast Asian coherence systems. Norwood, N J: Ablex.

FORTH, G.L. 1982. 'Time' and the expression of temporality in Eastern Sumba. Ethnos 47:232-248. 1983. Time and temporal classification in Rindi, Eastern Sumba. Bijdragen tot de taal-, land -en volkenkunde 139:46-80.

GALIPAUD, J.-C. 1996. Recherches archeologiques aux iles Torres: Premiers resultats de la mission realisee du 9 au 23 Novembre 1996 dans les ties de Metoma, Tegua et Toga. Port Vila: ORSTOM (Rapport de Terrain No. 5).

GELL, A. 1996. The Anthropology of Time: Cultural construction of mental maps and images. Oxford: Berg. HALEY, N. 1996. Revisioning the past, remembering the future: Duna accounts of the world's end. Oceania 66:278-285.

HOSKINS, J. 1993. The Play of Time: Kodi perspectives on calendars, history and exchange. Berkeley: University of California Press.

HYSLOP, C. 1999. The linguistics of inhabiting space: Spatial reference in the North-East Ambae language. In L. Bolton (ed.) Fieldwork and Fieldworkers in Vanuatu. Oceania Special Issue 70:25-41.

JOHANNES, R.E. 1981. Words of the Lagoon: Fishing and marine lore in the Palau district of Micronesia. Berkeley: University of California Press.

JOLLY, M. 1982. Birds and banyans of South Pentecost: Kastom in anti-colonial struggle. In R.M. Keesing and R. Tonkinson (eds) Reinventing Traditional Culture: the politics of kastom in Island Melanesia. Mankind Special Issue 13:338-356.

1994. Women of the Place: Kastom, colonialism and gender in Vanuatu. Amsterdam: Harwood Academic Publishers.

1999. Another time, another place, Oceania 69:282-299.

KAHN, M. 1986. Always Hungry, Never Greedy: Food and the expression of gender in a Melanesian society. Cambridge: Cambridge University Press.

1990. Stone-faced ancestors: The spatial anchoring of myth in Wamira, Papua New Guinea. Ethnology 29:51-66.

KEESING, R. 1984. Rethinking mana. Journal of Anthropological Research 40:137-156.

KIRCH, P. and R. GREEN 2001. Hawaiki, Ancestral Polynesia: An essay in Historical Anthropology. Cambridge: Cambridge University Press.

LAYARD, J. 1942. Stone men of Malekula: Vao. London: Chatto & Windus.

LEACH, E. 1950. Primitive calendars. Oceania 20:245-262. 1954. Primitive time reckoning. In C. Singer, E. Holmyard, and A. Hall (eds), History of Technology Vol. 1, Oxford: Clarendon Press

LEENHARDT, M. 1979. Do Kamo: Person and myth in the Melanesian world. (trans.) B. Miller Gulati. Chicago: Chicago University Press.

MALINOWSKI, B. 1927. Lunar and seasonal calendar in the Trobriands. The Journal of the Royal Anthropological Institute of Great Britain and Ireland 57:203-215.

MEGGIT, M.J. 1958. Mac Enga time-reckoning and calendar, New Guinea. Man 58:74-77.

MONDRAGON, C. 2003. Las Ples, Las Aelan, Las Tingting: Living respect and knowledge in the Torres Islands, Vanuatu. PhD Thesis. University of Cambridge.

n.d. Yam gardening, substance and 'recognition' in the Torres Islands, Vanuatu. Ms.in preparation for publication.

MUNN, N. 1977. The spatiotemporal transformations of Gawa canoes. Journal de la Societe des Oceanistes 33:39-53.

1986. The Fame of Gawa: A symbolic study of value transformation in a Massim (Papua New Guinea) society. Cambridge: Cambridge University Press.

1992. The cultural anthropology of time: A critical essay. Annual Review of Anthropology 21:93-123.

PALMER, J.D. and J.E. GOODENOUGH 1978. Mysterious monthly rhythms. Natural History 87:64.

RODMAN, M. 1987. Moving houses: Residential mobility and the mobility of residences in Longana, Vanuatu. American Anthropologist 87:56-72.

RUBINSTEIN, R. 1981. Knowledge and political process on Malo. In M. Allen (ed.), Vanuatu: Politics, economics and ritual in Island Melanesia. Sydney: Academic Press.

SCHROTER, S. 2000. Creating time and society: the annual cycle of the people of Langa, in Eastern Indonesia. Anthropos 95:463-483.

SENFT, G. (ed.) 1997. Referring to Space: Studies in Austronesian and Papuan languages. Oxford: Oxford University Press.

SHORE, B. 1989. 'Maria' and 'Tapu', In A, Howard and R. Borofsky, eds. Developments in Polynesian Ethnology, pp.137-73. Honolulu; University of Hawaii Press.

SILLITOE, P. 1994. Whether rain or shine: weather regimes from a New Guinea perspective. Oceania 64:246-270.

SORUM, A. 1984. Growth and decay: Bedamini notions of sexuality. In G.H. Herdt (ed.) Ritualized Homosexuality in Melanesia. Berkeley: University of California Press.

STURZENHOFECKER, G. 1994. Visions of a landscape: Duna premeditations on ecological change. Canberra Anthropology 17:27-47.

SUGGS, R. 1997. Le calendrier lunaire marquisien. Bulletin de la Societe de Etudes Oceaniennes (Polynesie Orientale) 273:105-121.

TAMBIAH, S.J. 1973. Form and meaning of magical arts: A point of view. In R. Horton and R. Finnegan (eds), Modes of Thought, London: Faber and Faber.

TANNENBAUM, N 1988. Shan calendrical systems: The everyday use of esoteric knowledge. Mankind 18:14-25.

THOMPSON, E.P. 1967. Time, work-discipline and industrial capitalism. Past and Present. 38.

WAGNER, R. 1987. Asiwinarong: Ethos, image and social power among the Usen Barok of New Ireland. Princeton: Princeton University Press.

WEIGHTMAN, B. 1989. Agriculture in Vanuatu: A historical review. Cheam, Surrey: The British Friends of Vanuatu.

WHITROW, G.J. 1980. The Natural Philosophy of Time. Oxford: Clarendon Press. 1988. Time in History: The evolution of our general awareness of time and temporal perspective. Oxford: Clarendon Press.

Article Details

Printer friendly Cite/link Email Feedback
Author:	Mondragon, Carlos
Publication:	Oceania
Geographic Code:	8VANU
Date:	Jun 1, 2004
Words:	13253
Previous Article:	Western arrernte Pmere Kwetethe spirits.
Next Article:	A critique of Alfred Gell on Art and Agency.
Topics:	Climate Research