This short essay is a partial introduction to the canvas of Melanesian music by means of reviewing a baker's dozen of CDs. It is meant to be appreciative, rather than critical in the comparative-evaluation sense, but has to be selective for two reasons : first, that there are not many CDs in the catalogues, yet; second, that the region is rich, complex and varied, both linguistically and culturally. It is estimated that Papua New Guinea (PNG) alone has over 750 languages (a count for West New Guinea is not with me), the Solomons 80, Vanuatu over 100, New Caledonia about 30‹we can say maybe 1000, perhaps 20% of all the world's recorded tongues, although the degrees of linguistic difference in Melanesia are not as striking within each of its two main families, usually called Papuan and Austronesian, as they are between, say, Chinese and Russian. There is also some homogeneity in the natural environments, all tropical, with a prevalence of «island effect». By population size, by proximities, by variety, Melanesia contrasts with Polynesia (Austronesian Melanesians being in effect the ancestors of the supposedly more homogeneous Polynesians). Everywhere, it is hard to predict (but not impossible) which area will be polyphonic, or which will employ coordinated ensemble slit-drum music, or which scales may be found where, et hoc genus omne. Despite its astonishing variety, there seems to be a Melanesian traditional music «sound», directly recognisable to anyone who has heard enough, but eluding description : it is something sensed, and it would be good to know how and why that is so. The presence and persistence of traditional musical culture in Melanesia today offers a patchwork scene : in some areas it has sunk from view, but happily not in all.

In common with artworks and material culture, Melanesian music and dance was greatly affected by exterior contacts impinging on the cultural ecologies that had inspired their variety of songs and ceremonies. We are reminded of ideology being disease. It was not merely the measles that reduced many populations : not merely as to their numbers, but also as to the cultural richness they had created, animated.

Today, one may see beautiful-object displays in every important ethnographic museum : here the Malanggan carvings, there Ambrym slit-drums ; here the Kanak masks, there basketry from Papua ; here the shell-inlaid carvings of the Solomons, there pottery‹the feathers, shells, bones and weaving‹all resplendent, and mostly gathered long ago. Such things were and remain obviously collectable, valued as tangible objects upon which prices could be set, counted, accounted, assimilated as riches for display like trophies of war or commerce‹much of it because museums demanded filling with collections. Then, contact severely altered ways of material culture production. It is heartening that museologists these days concern themselves with the social actions of creating and making material art-works (also with the cultural ecology), and have also begun sending back some surplus holdings.

As for the intangible arts, it is a different story. How to count what has been lost and is now gone forever? Music and dance could never be readily collected, crated and consigned‹not until recently, not until recorders and films could gather cultural expressions. However, it is as if these techniques borrow facsimiles.

To imagine being in the place of an ethnomusicologist, those who work in museums would need to consider going to the field merely to collect 3-D holograms instead of solid objects. For the general public this could result in walking around a museum of apparently empty vitrines which, one imagines, would have permanent labels, eg «Operating a loom, Santa Cruz island», or «Techniques of pig-butchery, Mota». The visitor would drop coins in the slot (or, more likely, slide a credit card through a magnetic field), whereupon the vitrine would self-activate, switching on lasers from the four points of an equilateral pyramid onto four fields, so that a 3-D image would appear, inspectable on all surfaces from any angle, but not tangible. If the vitrine were meant to arrest idle or curious fingers, that would be a concession to taboos of times past ; if the hologram were «free», one might try and walk through it, without seeming to touch anything. Such an image, ghost or facsimile of an object could hardly be weighed, cut, or otherwise physically tested. How could it obtain a C14 radio-carbon date?.

If museologists were to consider, with that example, the dilemma of the ethnomusicologist! The only way to take real songs and dances out of their natural environment remains to import-export the performers. Then you have the people, but not the environment : the correct leaves do not grow in Europe or USA. Because they go away and die, the performers are neither fully countable nor accountable except by dealing in shadows‹skiamachies over images, not real objects, only with «good counterfeits». The surfaces of such art and dance works, intangibly «solid», will seem to have been captured, may be heard and may permit some inspection, some apparently diligent scientific classification. Indeed, recent digital-technology developments have brought all this very close. The portable CD sound-recorder or a digital videodisc-camcorder may be had or hired, powered by rechargable solar cells.

Although we have here a tidy bunch of recent CDs of Melanesians making music we don't‹mark you!‹have the people, nor their leaves, in our drawing-rooms. The smells are not there. Are these (or any) CDs not somehow reminiscent of trophies, like stuffed stag-heads above the mantlepiece? Urgent ethnomusicology makes every effort to quell doubts like that. The accompanying booklets for each of these discs do the best possible within the constraints of the physical, packaged form of the CD to provide the all-important contexts. For printed texts, the 30 cm LP was rather big for comfort (and photocopying!), while CD booklets, miniature at 12 cm square, have to squeeze 400 words on a page so that they don't grow too thick to fit inside the plastic boxes‹designed for mass-market popular music needs, which hardly ever include learned (or even simply informative) essays, not to mention diagrams, maps and photographs. In the first of the items being dealt with here, the booklet, 180pp thick, shows all the problems of making this bastard format reader-friendly. (The mass-market is not concerned with people who can, or who indulge in, things like‹well, hell!‹reading!) It is a brave attempt, this Western New Guinea six-pack booklet ; one only wishes Berlin had taken more care over idiomatic English translation.

Let Artur Simon set the tone. He led a museum-based research expedition team to the so-called Irian Jaya, 1975-76. Did he consult missionary pilots first about which were the still pristine valleys? One can understand a scientist's desire to visit «virgin» populations, in a spirit of urgent ethnology. He had fantastic luck ; his contacted people did not. He recounts this with eloquent concern:

Dite lelalamak‹they sing songs‹, mote selamak‹they dance the mot dance, they, the Eipo from the central mountainous region of Western New Guinea. When we recorded their songs in the middle of the 70s, we could not know that this was taking place virtually a few moments before the end of a traditional culture of New Guinea or, in a larger sense, Oceania. What we discovered was a society as good as undisturbed by outside influences, an extremely sensitively structured, traditional society and culture with simple technology and a simple economy. In contrast to this simplicity stood a highly developed language and intellectual culture, the reconstruction and comprehension of which belong to the most difficult things that can confront a cultural researcher.

On December 12, 1975, I entered the Eipomek Valley ; on April 28, 1976, I left it without imagining that on June 26, Munggona, the chief village of the valley, and also my place of residence, would be completely destroyed by a devastating earthquake registering 7.6 on the Richter scale. A great mudslide had buried the houses and huts of the village under it‹including mine, which had stood on the edge of the village. On October 29 there followed yet another massive earthquake. This also meant physically a disruption of serious consequence to the Eipo Society. Their garden fields suffered widespread destruction. From outside, not least of all through our project, food aid was organized and flown in by the Indonesian military. Further research permission was, however, refused by the Indonesian authorities. Soon thereupon American fundamentalists began with their missionary work. The very self-assured society of the Eipo fell under this external influence. «At Christmas 1980, the Eipo burned their holy relics under the influence of the Unevangelized Fields Mission and in 1983 there followed the first baptisms.» (Heeschen 1990:9). The time of our recordings, here published as historical documents, was before these events. These recordings represent a musical culture that, as such, no longer exists. As irretrievable evidence, they justify the unusual amount of documentation in hand.

All the same, some Melanesian groups since culture-contact have shown a reassuring record of resilience and adaptability, and even if the art-works embodying the past have now frequently gone with that past, what any optimist for cultural integrity can point out is that recent art expressions tend to support a variant of the Sapir-Whorf hypothesis : that is, that the nature of a language presages its cultural expressions, for local culture is never completely destroyed, it is only 'submerged', and exists awaiting recall. The trouble with this is that some vocabularies, when not in use and renewal through practice, also have a decay-rate (cf. Crowe 1993:223). Will the Eipo ever be able to dance and sing the mot again, as they did when Simon was there? This seems depressingly unlikely.

Meanwhile, we may take some wry satisfaction from the fact that we now know better, thanks to the Berlin expedition, the ethnomusicological picture of the vast mountain chain that runs across New Guinea. May I humbly ask if it will do us any good to know this? Yes, we shall be better informed, but not they, not yet anyway. And being better informed, what are we going to do about it? Heal ourselves, by means of borrowed musical potions, à la Lévi-Strauss in Tristes Tropiques? Would that the short-term benefit be that some of us may learn to let well alone, that the earth may thrive again, by means of songs and forests! How sanguine is this hope? Meanwhile, the Eipo are being drained by foreign gods : the mission, the political authorities, the cankers of modernity. The Eipo have been submerged in more than mud. What has become of the «highly developed language and intellectual culture» that Simon admired? This is the point where some of the politics of the fourth-world may apply.

Despite such gloom, taking the Berlin CDs in hand, and wrestling with its doughty booklet, one can vicariously experience the extraordinary richness these isolated and intact societies used to be able to forge. The Eipo (in 1976) were not numerous, and they worked hard at gardening, playing and fighting. The booklet says they «worshipped» the ancestors (I regard that as a cliché and prefer to think they «dialogued» with such spirits). There was no stocking of surplus garden or forest produce (except that pig-breeding is a kind of accumulation). The songs are disarmingly frank (small kids sing about fucking) and the metaphorical meanings are cleverly explained by one of the Berlin team. When Eipo lament, it is penetrating, absolutely terrible. When they dance, it is at turns funny (sophisticated irony) and magnificent. They know how to turn the voice to social and personal account, for it is a virtue of this CD six-pack that you will hear both formal and informal performances (with all the usual, natural, local noises). There are almost no instruments‹a jew's harp, imported hourglass drums‹it is the body that matters. I sketch these few details to try and show how much vivacity it had.

Now the dead-hand of the Unevangelized [sic] Field Mission has caused the Eipo to droop and bow their bodies when gathered together «to worship», it has taken the sting from their texts, it has robbed them of gamey sex, it has imposed a false and useless historicity on their relations with the environment (did Jesus really come to save the Eipo?). How did these fundamentalists gain permission to be there and stay? I fear it is the old political story of «pacification». We hear on CD #1 how beautiful and self-assured the singing was ; by CD #6 there are down to the languid vocalisations of the newly colonised. That was a rapid change, non?

One may progress, geographically and intellectually, through an evocation of the former Eipo society on CDs #1-4. On CD #5 we hear their neighbours and can understand differences and similarities. On CD #6 we listen to the degradations that are caused by cultural logging. The trees have been cut, but there is no real product ‹save statistics as furniture for some generals in Jakarta, or some bosses of bullshit in Utah. Voilà. Musique d'ameublement is well on the way.

It makes an instructive journey. On the CDs one encounters unique ways of singing together, enough to pose problems for all previous definitions of polyphony (even those of Arom). The scales or note-rows used need thought about why they seem so comparatively «simple». The Eipo have redoubtable listening/memorising capacities, shown by improvised canons. The Berlin notes don't mention rhythms (but some colleagues have been equally scotched by the Corsicans) although first listening reveals there are poles of rapid syllabification (c.400 per minute) versus sustained (echoic) tones (for up to 10 seconds) resplendent with the sustained control of beautiful timbres. In between, various cellular rhythmic bits may appear, to be used for a while, not seeming very stable, and so whither permutations? Have the mountains (very often over 3.000m) given an imprint to this music? I should like someone to explain why all this is like that. It is hard to believe it is arbitrary.

The notes Feld supplies to his Kaluli rainforest-soundscape CD hint at some in the interrelationships that connect natural environment and human music-making. This disc is a labour of love for an endangered people, who will surely have their rainforest severely damaged in the search for oil, minerals and timber, all for the cash economy to line politicians' and businessmens' and bankers' filthy pockets, in the name of «progress». He says the rot has already set in, young people no longer know how to intone their own natural environment. Feld seems almost ready to unravel the famous Zen question over «what is the sound of one hand clapping?», by using the difficult art of literature to convey what he feels. Bravo! Part of the proceeds of the sale of this disc are for rainforest protection (Rainforest Action Network and Cultural Survival are two USA groups cited in the booklet, with addresses).

There is, in Feld's annotations, judicious use of translated Kaluli terms, such as «lift-up-over sounding», indicating something like emotional transcendence as brought about by musical means. The native terms for times of day and night, and both morning and evening crépuscules, are highly evocative. This disc is not, at first sight, straight ethnomusicology, far from it. There is not very much straight music on it : some work-singing, a jew's harp, some drumming and a portion of a song ceremony. The timbres of the speaking and singing voices sound cousined to those of the Eipo, and the environments would not be dissimilar, nor their remote Papuan ancestry. There is a world of difference in the fieldwork approaches of the Berlin team and that of Feld, who also worked with colleagues (see Feld 1982, 1991 and Schulte-Tenckhoff 1988), but one presumes for much of the time solo ; I do not want to say I prefer either approach : I should like to see them both employed in the practice of ethnomusicology, but it is true that some people are not temperamentally adapted to it. The great value of Feld's CD is this : that you can, as a listener, make the vicarious voyage where he has been, and I believe it would make an interesting weekend listening-course, for even the most hard-bitten among us, to hear Feld, take a teabreak, and then start on the Berlin set, so that by tea-time a group would be ready to discuss musical, and indeed pertinent moral matters. The next day, the imaginative voyage could take the tour through Island Melanesia, but starting at the Trans-Fly, in the order of discussion followed below. (Who will join me? Write to the editor of this publication! ).

Feld's soundscape inspires me to coin a neologism. We have monophony, heterophony (the Russians distinguishing at least four varieties), polyphony (with countless types), and by the evidence of Feld's «soundscapes», we should add 'ecophony'‹or, «sounding with the environment», which appears to be sometimes mimetic, but sometimes quasi-contrapuntal. Ecophony in this sense is reminiscent of Strawinsky's use of the word «symphony», to mean simply «sounding together» rather than a classical European orchestral form, as in his memorial piece for Claude Debussy, his Symphonies of Wind Instruments.

If the Eipo and Kaluli are Papuans, inheritors of up to 40.000 years of New Guinea occupation (where agriculture was independently invented some 9.000 years ago, according to Golson 1977), those Island Melanesians who are Austronesian of origin can only claim a brief 4.000 years. If they did not go beyond neolithic technology, their ancestors used their brains and brave bodies to conquer the Pacific (Irwin 1992). So let me go back a bit on this review-voyage, this curving around and down the Melanesian crescent, to mention Laade's CD of the coast of the Western Province of PNG, on the equatorial, northern side of the Torres Straits.

The peoples of the Trans-Fly region were first recorded by the Australian photographer Frank Hurley in 1921, on cylinders (which this reviewer wants to relocate). A photograph of Hurley's shows him «recording a concert» in which the physical details of traditional village life look little affected by culture contact (see photo). In Laade's CD, recorded 40 years later (note : the CD is 70 years later), the aural evidence now seems «untidy». That is, the singing seems ragged by contrast with the «look» of Hurley's situation : one has the impression that self-confidence in performance has slumped over the years of culture contact, but of this one cannot be sure. Laade worked on the coast of an extensive region, where considerable contact with the longer-acculturated Torres Strait islanders (missionaries, pearling industry) took place. The booklet gives useful indications of the disruptions the area has since suffered, including efforts to stamp out head-hunting as from the end of the 19th century. Laade's fieldwork represents a kind of salvage ethnomusicology, which is, and was then in the 60s, just as important as the urgent variety (let us say, fieldwork undertaken when culture-rupture or change is thought to be imminent). Laade is also alert to the probable transience of popular styles, and has intelligently included some modern «island dances» of the so-called guitar, or «stringband» kinds, known in parts of PNG as «lokal musik» (cf. Webb 1993).

Moving southeastwards, after PNG we come to the Solomon Islands, now known to have been originally occupied by proto-Papuan peoples as from c.30.000 BP, of whom there are pockets surviving as far to the east as the Santa Cruz group. Most of the present populations are descendants of the more recent Austronesian migrations. Solomon Is. is a jewel in the diadem of Oceanic polyphony, with miriad facets and colours. One finds areas of rich instrumental variety and development, within the limits of the natural resources and neolithic technology, while in other areas the music may be almost exclusively vocal. This is shown in the CDs made by Hugo Zemp, as considered here. The music of Savo is almost entirely vocal, in a Papuan language, and perhaps because of proximities and diffusion, now seems to be musically structured in an Austronesian way.

The political boundary between PNG and the Solomons Is. obfuscates the existence of a suite of archipelagoes from the Bismarks (eg Manus), through the Admiralties (eg New Britain, New Ireland), down through Buka and Bougainville, all of which are parts of PNG, and then on through what one sees on the maps as the Solomon Islands. Using the term archipelagoes in the plural may be misleading, depending on artificially chosen distances of separation ; it is perfectly valid to think of all these islands as one great mainly-Austronesian archipelago, or as the northern region of Island Melanesia (the southern region being Vanuatu and New Caledonia), as distinct from quasi-continental PNG. (As for Fiji, many reference works say it is «physically» Melanesian but culturally Polynesian, so it does not fit into this review ; yet Fiji was indeed a traditional-culture pivot, see below.) In this region the matter of island-to-island proximities is and was important for cultural diffusion, yet this is complicated by the presence on these high islands of two types of peoples : those of the «salt-water» (coastal) and «man-bush» (interior). The well-kown kula trading-ring (as in Malinowsky's classic, Argonauts of the Western Pacific), possibly has echoes in the phenomenon of «musical tours» mentioned in the Zemp CDs.

Despite the opportunties for music diffusion available to salt-water peoples, all of whom had large ocean-going canoes at some point in their (pre-)histories, the particularity of each region's or island's musical styles (or systems) seems to have its integrity, but we know little about cross-influence, about borrowings, about musical change and retention in the past, which resist reconstructions‹except perhaps via song-texts, using techniques such as lexico-statistics (but complicated by the use of special song-languages, and numerous other factors), when there are song-texts. Guadalcanal women's songs on the Guadalcanal/Savo CD (tracks 8-14) have few words, mostly vowel-vocalisations ; Zemp suggests this may be in imitation of panpipe tones. In general, Oceanian songs are text-heavy, but here is an exception, and also one can remark the comparative lack of singing among the 'Are'are (cf. Zemp's repertoire-catalogue film Musique 'Aré'aré, reviewed in Crowe 1987).

Laboratory measurements of music are often more readily made, and then interpreted, of instrumental music than vocal music. At the Musée d'Homme in Paris, Zemp (with the collaboration of Jean Schwarz) made careful measurements of the scales employed by panpipe ensembles of Guadalcanal and Malaita. Perhaps the most unusual result was the determination that the 'Are'are consistently employ an equiheptatonic scale in much of their panpipe music. How did these people hear such a scale, or how did they find a way to obtain it regularly? Is this a chicken-or-egg question, in that by following divisions of body measurements (based on the cubit in this case) the seven equal-step scale was the inevitable result? When Zemp arrived on Malaita, he found the last man, the late 'Irisipau, who still knew how to do the traditional «sizing» procedures. All other panpipe makers simply copied the extant models, but apparently faithfully, else interchanges between ensembles would have had difficulties over tuning each to the other, or even within the one ensemble, if the manufacturing had not been standardised somehow. With Zemp's meticulous work, the traditional fine-knowledge is restored and made available to everybody.

These CDs are a joy to listen to for the intrinsic interest of the musical items and the excellence of the recordings. The booklets are models of clarity and detail, beautifully illustrated by a man with a true eye (in speaking of Zemp as photographer and also, elsewhere, as his own film-cameraman), and by very intelligent diagrams. They also contain valuable (indeed, essential) bibliographic and audiovisual references for an amateur to pursue the subjects.

A reader of these notes brought up in the Western tradition of equal-tempered scales, or even those who are used to «perfect» (pythagorean) systems, might ask if the tunings found in the Solomons would sound weird. Not a bit of it. I have had the chance to play Zemp's recordings to many different audiences, and the music wins everyone's hearts at once. At a conservatorium of music, string-players (so aware of intonation problems, and possibilities) accepted the «strange» scales as if they were completely natural. I think this says more about the basic artificiality of the typical Western modes, as a canon, than anything else, for Westerners surely sing and play far more «in the cracks» than they are usually aware of.

The pieces tend to be quite short and have picturesque titles («The groaning of Pora'ahu», for example, or «What a mess, boys!») which, however, do not seem to Zemp to be onomatapıic. The capacity for polyphonic musical «thinking» is brought to a high art, implying that a constellation of choices in the aural world in the linear and vertical senses of pitch (aural space), not to mention the organisation of the progression of time, have been made. The mental acts involved are stupefying by their complexity (including their social conditioning), in the way they have made magical musical webs of the most intricate interlocking aural relationships conceivable, in the context of the natural environment and available technologies. That is to say, this music evokes the infinite potential extensibility of human creativity, made finite only by local circumstances.

By contrast with the Solomons, Vanuatu appears to have little polyphony today, although much is known to have been lost (Crowe 1981). Singing in canon reappears in New Caledonia, and Fijians sing in parts, so if one persists in thinking of polyphony as a stage of cultural evolution, all the evidence here is against it. Polyphony does not have to wait for a certain cultural «development» ; it appears and disappears along the routes of the Austronesian explorers, whose trajectories and voyaging calendars are now quite well-known, at least in sufficiently clear outline (Irwin 1992). Neither does polyphony require diffusion. It can be (re)invented whenever a group simply decides to exploit certain ideas of tonal simultaneity, an idea which might come from observing birds in chorus, or hearing humans chatter together from a distance, let us say, as much as through diffusion. There is a conceptual problem here.

Current ethnomusicological definitions of polyphony smack slighty, to me, of Eurocentrism. Arom (1985/1991) points out the need for rhythmic idiosyncracy of each of the parts in order to distinguish polyphony from, say, chordal singing (as in Christian hymns, which he sees as polyphonic in only a rudimentary sense) : for as Bach composed polyphonically, so sing the Aka Pygmies of Central Africa, and, it is evident above, so do many groups in the Solomons. It seems to me that exponents of polyphony need also consider coordination of events as much as results or artefacts, having isolatable, mutually interacting zones of definite pitch. It is a trifle bizarre that polyrhythmic music, as on drums, is set aside in another category because the degree of manipulation of pitches is considered, again, only rudimentary. To the extent in which polyphonists concentrate on tonal results they are avoiding basic implications of coordinated human acts, those which produce aural results not fitting neatly into a pre-determined polyphonic tonal-object category. I always thought a strength of ethnomusicology was to go back to the fundamentals of human choice which lead to «repeatable» aural artefacts.

In the Vanuatu CD one may hear, track 19, the deliberate use of a counter-song. It is a song sung more or less at the same time to «cover» another song with a text on taboo matters of concern only to Qat Baruqu initiates. The combination here is at least reminiscent of a sort of quodlibet, and in Bach's use of that form (so named after the event, let us note) in the Goldberg Variations, everone says that it's polyphonic. Will a musicologist object that the two Qat Baruqu simultaneously performed songs were not composed with the intention of being sounded together? How can that be known? Is the conclusion taken (Eurocentrically, I should hazard) that it is an aleatoric combination? So what? Why would that be a disqualification? What about Stockhausen's Zeitmasse? Although these are not questions that can be quickly answered here, they instance some fundamental problems in the conduct of ethnomusicological analysis, especially regarding polyphony, that Melanesia inspires for reconsideration. Melanesia is under-valued as a laboratory.

Coordination of events, gestural and aural, are indeed shown on the Vanuatu CD by extracts from two major rituals : the pig-killing ceremonies of Ambae, and Maewo's Qat Baruqu. A major musical interest lies in the slit-drum ensemble music, here brought to a brilliant point of elaboration, and presaging the deployment of such drums in Polynesia (famous in the Cook and Society Islands, eg Tahiti). The CD also has a range of individual solo singing styles that seems much more varied than elsewhere in the repertoire we have available. One of the songs (track 17) has a text indicating an intention to set sail for «Mamalu», which one presumes was Fiji, but even if it were not, this is a rare and intriguing oral tradition relic of the sort of voyages that took place 3000 years before.

New Caledonia turns out to have remained musically richer than we thought was the case, after its dreadful colonial history, reappearing in public with the Melanesia 2000 festival of 1975. Discs published in the early 60s presented characteristic ways of singing «Christian choruses», but customary music was hidden‹or presumed lost. The Kanaks may have felt there was no appreciation of their identity through music to warrant performance before the overbearing colonists, and it were better kept private. A Kanak group came to the first South Pacific Festival of Arts at Suva in 1974, and performed various pilou, remarkable for the use of hissing, whistling, scraping sounds, but wordless and not evidently intoned as «singing».

Now, with this CD, we can hear that this is and was not all, not by any means. Beaudet presents a limited number of items, and one hopes there will be more to come for the general public. The most remarkable items are the trailing canons of a character reminiscent of Tibet, where the lead singer is paid homage by the echoing or following singer, endeavouring to keep pace, but carefully at a step behind, presumably out of respect. There is also an astonishing intoned, rapid genealogical discourse on track 1 which tests the boundaries of speech and song, not unlike the tau sequences in a formal Maori speech. One wonders if there is some kind of «latitude effect» at work, because from Erromanga in Vanuatu one has heard songs every bit like the oriori («lullabies» to teach the children of chiefs) of the Maori, and here in New Caledonia, the music suggests it works as a functional underpinning for the consolidation of hereditary hierarchies, for chiefly systems are often absent in the north of Island Melanesia. It seems well-established that New Caledonia was a SW terminus in Austronesian expansion, not proceeding to New Zealand (Norfolk Is., intermediate, remaining a puzzle). The route from Erromanga to New Zealand took Austronesians 2000 years or so to traverse, via Eastern Polynesia and a migratory «return» southwestwards. Wherefore these musical aspects so indicative one of the other, as if there had been diffusion, when the separations were of such orders? However, Jean Guiart has recently discussed (1993) some curious elements that suggest prehistoric contact with the Maori of New Zealand may have occured, but not yet supported by archaeology.

In the opening paragraph I mentioned «proximities» as being of importance to the cultural ecologies of the grand archipelago of Island Melanesia. Some writers have called the region the «Mediterranean» of the south : some 10.000 years ago you could have walked to most islands in the actual Mediterranean, the sea level being 150 metres lower than it is now. In the Pleistocene you could have walked from New Guinea to Tasmania. The Papuans ranged as far east as the Santa Cruz islands, by a procedure of island hopping, the next one always being in sight either from the island or from a safe mid-sea position (Irwin 1992). By the time the Austronesians arrived, the sea levels were raised to what they are now, and the craft of canoe-building became a matter of technological prowess, as did their elaborated, empirical sciences of navigation. Joël Bonnemaison's metaphor of «the tree and the canoe» (see Bonnemaison 1986 : 517-23 ; translated in Crowe, in press) is pertinent here. Bonnemaison sees the Austronesians as rooted in places as if trees, but in a network that depends on the maintenance of canoe-routes. Melanesians are simultaneously fixed and mobile in their island environments. This idea is winning, and by implication it undermines notions of rigidity in the construction of traditional Oceanic societies. What is even more winning, it must be said, is that it was the Melanesians themselves who explained it. On arrival at Tanna, Bonnemaison was told that he had not come to the end of his work (after fieldwork on many other islands), he was only at the beginning, and he took the lesson to heart to find out what it meant.

Writing a review-article like this is in itself a kind of migratory voyage. For an investment of about £150 a reader can buy all the baker's dozen of discs we have discussed. The hardest thing to do in music-writing is to give the reader an idea of what the stuff actually sounds like, and that applies to «exotic» music as much as to the première of a new work from the pen of Luciano Berio. I have deliberately tried to minimise technical jargon in the hope that enthusiasm on my part will, perhaps by some mysterious process of synæsthesia, inspire readers to immerse themselves in the truly extraordinary world of Melanesian musics.

(MS version as at 3/4/95).

BIBLIOGRAPHY

• AROM, Simha1985. Polyphonies et polyrhythmies instrumentales d'Afrique centrale. Paris : SELAF. 2 vols.
• AROM, Simha1991. African Polyphony and Polyrhythm. [English translation of Arom 1985.] Paris & Cambridge: Maison des Sciences de l'Homme & Cambridge University Press.
• BONNEMAISON, Joël1986. L'arbre et la pirogue [livre I ‹«Les fondements d'uneidentité. Territoire, histoire et société dans l'archipel de Vanuatu (Mélanésie)»]. Paris : ORSTOM.
• CROWE, Peter1981.«Polyphony in Vanuatu» in Ethnomusicology 25(3):419-32.
• CROWE, Peter1987. Reviews of Zemp's films, Musique 'Are'are & Tailler le bambou in Ethnomusicology 31(2):326-9.
• CROWE, Peter1993. Reviews of Love Samoan Variations & Wassmann The Song to the Flying Fox in Cahiers de musiques traditionnelles 6:218-26.
• CROWE, Peter in press. Translation of a passage of Bonnemaison 1986. J.of Pacific Arts.
• FELD, Steven1982. Sound and Sentiment. Birds, Weeping, Poetics, and Song in Kaluli Expression. Phildelphia : University of Philadelphia Press.
• FELD, Steven1991. «Lift-up-over sounding» in ICTM Yearbook 23 (with C60 cassette).
• GOLSON, Jack1977. «No room at the top : agricultural intensification in the New Guinea Highlands.» In Sunda and Sahul, prehistoric studies in Southeast Asia, Melanesia and Australia, ed. J. Allen, J. Golson and R. Jones, pp 601-38. London : Academic Press.
• GUIART, Jean1993. A propos de Simmons 1976. J. Société Océanistes 97(2) : 239-46.
• IRWIN, Geoffrey1992. The prehistoric exploration and colonisation of the Pacific. Cambridge & Oakleigh (Australia) : Cambridge University Press.
• SCHULTE-TENCKHOFF, Isabelle1988. Review of Feld 1982 in Cahiers de musiques traditionnnelles 1:214-21
• WEBB, Michael1993. Lokal Musik. Lingua franca song and identity in Papua New Guinea. [Port-Moresby] : Cultural Studies Centre, National Research Institute [of PNG].