SMITHSONIAN CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 The Native Polity of Ponape Saul H. Riesenberg SMITHSONIAN INSTITUTION PRESS City of Washington 1968 A Publication of the SMITHSONIAN INSTITUTION United States National Museum LIBRARY OF CONGRESS CARD 68-29128 UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON, 1968 For sale by the Superintendent of Documents, U.S. Govemment Printing Office Washington, D.C. 20402 - Price $4.25 Preface A large part of the material presented in this work was obtained in the course of fieldwork on Ponape from July 1947 to January 1948. This fieldwork was sponsored by the Pacific Science Board of the National Research Council, with support by the Office of Naval Research. It was part of the CIMA Project (Coordinated Investigation of Micronesian Anthropology), and the preliminary report I submitted is partially incorporated in the present work. Other materials obtained on the same field trip have been published elsewhere and still others remain to be published. In 1953-54, during a year of leave from teachirT- duties at the University of Hawaii and while in the position of Staff Anthropologist to the High Commissioner of the United States Trust Territory of the Pacific Islands, I was on Ponape in the course of other duties several times, for periods ranging from a few days to a month, and took the opportunity to add to my field materials. In January to March 1963, I was again at Ponape performing other fieldwork imder a grant from the National Science Foundation, and during that period was able to fill some additional lacunae in my notes. I have made use of the unpublished field notes of Dr. J. L. Fischer of Tulane University, and of the letters, journals, and other documents of the American Board of Commissioners for Foreign Missions in the Houghton Library at Harvard University, and have drawn on the rich knowledge of Father Hugh Costigan, S.J., a missionary stationed on Ponape since 1947. To these institutions and individuals and to my Ponapean informants and friends, too numerous to mention here by name, I wish to express my enormous debt of gratitude. I also wish to thank George Robert Lewis, Scientific Illustrator, Office of Anthropology, for preparing the hne drawings and maps from my pencil sketches and Dr. Clifford Evans, my colleague, for kindly allowing me to use photographs taken by him during our joint field trip to Ponape in 1963 for Plates 5b, 6a, b, 7b, 8a-d, 9a, b, 10b, 1 la, b, and 12a-d. Standard Ponapean orthography as developed by Drs. Paul Garvin, Isidore Dyen, and John L. Fischer and as taught nowadays with some modifications in Ponapean schools has been followed in this work. It is as follows: 1. Vowels are roughly as in Spanish except for the di^jraph oa, a low back vowel. 2. Consonants are roughly as in English except that there is generally no aspiration, and presence or absence of voicing is of no phonemic significance. The exceptions are: a. ^ is a retroflex alveolar stop. b. d approximates English t. c. n^ is a digraph for a velar nasal. d. 5 is about midway between English s and sh, and is sometimes heard as English ch, ts, and almost;. e. pvu and mw each represent single phonemes, pronounced approximately as written but with hp rounding beginning before closure. 3. h represents no phoneme and is used only a5 a sign of phonemic length of the vowel preceding it. 4. Those consonants that can occur doubled {m, mw, n, vy. I, r) are written by repeating them, except for double mw, which is written mmw, and double I, which is written nl when this is its etymological derivation (but there is no difference in pronunciation between nl and / / ) . Smithsonian Institution SHR Washington, D.C. Contents Page Introduction 1 The physical setting 1 Discovery and early history 2 Population 5 The Tribes 8 The tides 8 Clans and classes 10 Grades of chiefs 16 The title series in Net 19 Subtribes 21 The sections and other subdivisions 21 Sections, section chiefs, and clans 29 Functions of the section chief 32 Capital sections 33 Promotion and Succession 34 Theory and practice 34 Irregularities in political advancement 39 Promotion ceremonial 40 Other Tides 43 Priestly titles 43 Titles of address and titles of reference 44 Honorific forms 45 Feminine titles 47 Machinery of Government 49 The position of the Nahnmwarki and Nahnken 49 Privileged behavior of the B-line 50 Political councils, courts, and trials 51 Direct punishment by chiefs 53 Banishment 54 Atonement for offenses 55 Supernatural sanction of chiefly authority 58 Taboos 60 Wars 60 Prerogatives of Chiefs 63 Confiscatory powers 63 Attendants and servitors 65 Insignia, deference, and etiquette 66 Marriage and sexual privileges 72 vii ^"^ CONTENTS Page Presdge Competition 76 Competition for titles 76 First fruits 77 Lihli 80 Other first fruits 82 Ceremonial presentation of fish "2 Feasts for chiefs 83 The major feasts 83 Other feasts 86 Competitive feasts 90 The feast pattern ^^ The community house and seating arrangements 96 Feasting procedure and ceremonial 98 Division of food 100 Kava 102 Kava and feasting 103 Kava ritual 105 The role of kava in Ponapean culture 108 The Ponapean Polity 110 Literature Cited 112 Tables 1. Population of Ponape from Native Period to 1914 5 2. Population of Ponape, 1947 6 3. The A-lme of tides 12 4. The B-line of tides 14 5. Stages of kava ceremonial 107 Illustrations FIGURES 1. Map of Ponape showing the five independent tribes 9 2. The tribe of Kiti and its sections 22 3. Variations in slitting fish for ceremonial presentation 84 4. A Twisting Coir Twine feast in the community house of section 19, Kiti 94 PLATES 1. Abandoned house site and burial cairns. 2. Preparation of breadfruit for lihli. 3. Preparation of breadfruit for lihli. 4. Ceremonial food preparation. 5. Ceremonial food preparation. 6. Community houses. 7. Servitors in commimity house and chiefs belt. 8. The growing, transporting and cooking of yams. 9. Transporting yams. 10. Transporting sugar cane and kava. 11. Kava drinking ceremony. 12. Preparation of kava. The Native Polity of Ponape 298-818?68- Introduction T H E PHYSICAL SETTING The island of Ponape hes at 6?54' north latitude and 158? 14' east longitude, near the eastern end of the archipelago that comprises the Caroline Islands. Ponape and the nearby atolls of Pakin and Ant constitute the Senyavin group. Kusaie, the next major island to the east and last of the Caroline chain, is 307 nautical miles dis? tant; and Truk is 383 miles to the west. Ponape is about midway between Honolulu and Manila, 2,685 and 2,363 miles distant, respectively. The land area of Ponape as usually given is 334 square kilometers (129 square miles). Except for a few coastal plains and lower slopes, most of the island is ruggedly mountainous with several ranges and high peaks, the high? est rising to 791 meters (nearly 2,600 feet), the highest peak in the Carolines. The mountain tops are often cov? ered with cloud and mist. The island interior consists largely of basalt, with some andesite and other volcanic rock. The lower slopes and level areas are mosdy sand and gravel. Here and there, most spectacularly on Sokehs Island, are high cliffs of columnar basalt, with columnar talus at their bases. Streams and waterfalls abound. The streams are very active after every rainfall and deposit alluvium in great flats. The main island is roughly circular in outline. About 20 square miles of its area consists of coastal mangrove swamp; there are few beaches. Surrounding the island is an encircling reef, distant from the coast about 2 miles on the average and broken here and there by passages be? tween the lagoon and the open sea. Where the reef rises above sea level, some 15 coral islets are formed. The lagoon between the main island and the encircling reef is of varying depth and contzdns many heads of live coral that may rise close to the water surface. It occupies about 98 square miles of water and includes in it 23 small islands of the same volcanic materials as the main island. Also in the lagoon are a number of alluvial islands lying close to the shore and, in the east, off Temwen Island, some 90 artificial islets are grouped together that contain the well- known archeological ruins of Nan Madol. All of these smaller islands together occupy an area of about 5 square miles. Ponape, like the rest of the Carolines, is in the doldrum belt. This belt swings north May to July and south August to November, accompanied by stormy weather and heavy rain. The trade winds blow from the east December to April and move around more to the south the rest of the year. Typhoons are much less common than in western Micronesia. Relative humidity is high; the monthly averages are 79 to 91 percent and are lowest in March and April. Rainfall is very heavy and is rather uniform throughout the year, though somewhat less in January to March. The annual average precipitation is 178 inches. The monthly temperature means are from 78? to 82? F., the extremes are 68? and 92? F. The extensive mangrove swamps that line much of the low shore consist of Sonneratia, Rhizophora, Bruguiera, Lumnitzera, Xylocarpus, Heritiera, and also Nypa palm. Strand vegetation occurs largely on the reef islands and where the land of the main island starts to rise inland of the mangroves. Behind the strand and mangroves is a narrow strip of coastal plain vegetation, originally primary rain forest but now mainly occupied by single dwellings or small clusters of buildings and by cultivated areas. (Very few localities on Ponape can properly be called villages.) The cultivated areas contain coconut groves. 1 S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 banana, breadfruit, citrus, sweet potatoes, manioc, aroids [Cyrtosperma, Alocasia, and Colocasia), and a few other cultivated plants of less importance, as well as such trees as Ficus, Calophyllum, Terminalia, ivory nut palm, Pandanus, Macaranga, Morinda, and Hibiscus, with some grasses and undergrowth. The rain forest is very dense and rich. Ponape has the most extensive native forests in Micronesia. Uninter? rupted jungle covers most of the interior and reaches to the coast in a few places. The forest covers most of the steep slopes and summits and some lowlands. The lower primary forests contain large trees, pahns, climbers, ferns, orchids, and other epiphytes. On the slopes where rain forest has been destroyed there are mixed coconut and breadfruit groves, also bananas, Alocasia, and some shrub? bery. The montane rain forests consist of scrubby, mossy woods on the steep slopes and ridges, where the soil is thin. Tree ferns, Exorrhiza and Ponapea palms, Freycinetia, and many stunted broad-leaved trees bearing epiphytes predomuiate. On the high open crests are dwarfed shrubs, dwarfed Exorrhiza, tree ferns, dense growths of Pandanus, or open bogs. Grassland is rare. According to Baker (1951) some 39 species of birds occur on Ponape, mcluding sea birds, migratory shore birds, and land and fresh water birds. Among the species are a heron, a duck, seven of the snipe-sandpiper family, six gulls and terns, two doves and a pigeon, a lory, an owl, a kingfisher, two flycatchers, two starhngs, three white- eyes, and two weaver finches. Insects are numerous but comprise only a small number of genera. The only land reptiles are a few species of lizard. There were only three mammals, aboriginally: rats, bats, and dogs; pigs were introduced m very early postcontact tunes, and some are now feral, as are the deer brought in duruig German times. In contrast, the lagoon is rich in fishes, mollusks, Crustacea, and other marine fauna, including two kinds of turtles.^ DISCOVERY AND EARLY HISTORY Ponape's first European visitor was Pedro Fernandez de Quiros, commanding the San Geronimo, the last sea? worthy vessel of the four that had comprised the Men- daiia expedition. Alvaro de Mendaiia had died in the Santa Cruz Islands, and Quiros, the first pilot, took com? mand. On December 23, 1595, he sighted Ponape whUe en route to Guam and Manila. The description he gives is brief and he did not land; he only saw a few canoes and established no communication, other than by gestures, with the distant natives (Markham, 1904). But the name Quirosa, which the island thus received, became attached by mistaken cartographers to Truk instead of to Ponape. Two earher sightings of Ponape may possibly have oc? curred, by Loyasa in 1526 and Saavedra in 1528 (Navar- rete, 1827, pp. 99-100, 468). But the meager description appears to apply to low, coral islands, not to Ponape. Equally doubtful sightings are those of Musgrave in 1793, Ibargoita in 1801, and Dublon in 1814 (Hambruch, 1932 I, p. 78). But a Ponapean legend very circumstantially tells of a ship that came to Kiti, its men having skins of iron and one man in black wearing a cross. The visitors could be killed only by spearing them in the eyes. Evidently this is a memory of some unrecorded European voyage, probably Spanish. Another story tells of the removal by the Sou Kiti (a high chief's title) of a small brass cannon from some foreign ship that visited Kiti. The cannon was kept until 1839, when Captain Blake of HMS Lame carried it off to Hong Kong (Blake, 1924). It bore a Spanish coat of arms, according to the missionary L. H. Gulick (1862, p. 175); evidence of French or Chinese origin, according to Rosamel (Hambruch, 1932 I, p. 118). In the 1830s, whaler and trader captains, rummaging in the ruins of Nan Madol and elsewhere for supposed treasure, found various European objects, including two silver crucifixes, coins, and a pair of dividers (L. H. Gulick, 1862, p. 175; A. Gulick, 1932, p. 88; Maigret, 1837-38). The second visitor, after Quiros, for whom we have good documentation did not arrive at Ponape until Sep? tember 10, 1825. He was Capt. John Henry Rowe {Hobart Colonial Times, May 25, 1827; Sydney Gazette, June 15, 1827). He also did not land. His vessel was chased by five canoes; this was his only contact with the natives. He gave the name John Bull's Island to his dis? covery, after the name of his bark.^ It is not clear who the next visitors were. In a note in an AustraUan newspaper of February 1835 (Lhotsky, 1835), there is a reference to a Mr. Ong of New South 1 The foregoing is compiled from Bascom, 1965; Baker, 1951; Fischer, 1957; Gressitt, 1954; Hambruch, 1932; and Glassman, 1952. 2 I am much indebted to Mr. Harry Maude and to Mrs . Honore Forster of Australian Nationad University for providing me with this information concerning Capt. Rowe, for details of Australian history that affect the veracity of O'Gonnell's story (discussed later in this section), and for drawing my attention to several Australian newspaper accounts. INTRODUCTION Wales who "some years back" was at Ponape and re? mained several months. A brief description of the Nan Madol rums is given. The same article states that the island was discovered "very lately" by HM Sloop of War Raven; I have been unable to leam anything more of a voyage to Ponape by such a vessel, despite search of various archives. One can only guess what "some years back" might mean. Equally indefinite is the "discovery" by Captain Harper. According to Ahck Osborne, who arrived at Ponape on December 20, 1832, in the ship Planter, the island, which he calls Harper's Islands, "was discovered by Captain Harper, in the ship Ephemina, bound from Sydney to Canton about six years ago . . ." (Osborne, 1833, p. 34). This would make the date of Harper's visit about 1826. But the captain of Osborne's ship, L. Frazer (1834, p. 74), refers to the island as his own original discovery and names it William the Fourth Group. (The name WiUiam the Fourth is sometimes apphed to Ant atoll, as is also the name Frazer's. Ant is only 8 miles from Ponape and in the same Senyavin group, but Frazer's description is not of an atoll but a vol? canic island and must be of Ponape.) Ong uses the name Ascension for Ponape, as did Capt. J. H. Eagleston of the bark Peru (Eagleston, 1832-33) when he sighted the island on January 3, 1832; it was already on Eagleston's charts under that name. It is uncertain who first called it Ascension Island, but it quickly became established in the literature and was used by nearly aU later visitors. The first description of the island and natives of any length is by the Russian, Capt. F. Lutke (or Liitke), whose vessel, the Senyavin, gives its name to the group consisting of Ponape, Ant, and Pakin (Lutke, 1835-36). He was at Ponape only 5 days (January 14-19, 1828). His boats attempted to land but could not make the shore because of the throngs of canoes that pressed about them and the show of hostility. But natives came aboard, some trading occurred, a short vocabulary was compiled, and a map was made. F. H. von Kittlitz, a member of the expedition, wrote a supplementary account (KittHtz, 1858). The two reports together provide the first real knowledge of Ponape. That others visited Ponape about the same time as or before Ong, Rowe, Frazer, Osborne, Lutke, and Eattlitz is apparent from the article pubHshed in 1836 by a Dr. Campbell ^ who arrived on the cutter Lambton about the end of 1835. (The Lambton was the next year to par? ticipate in the infamous Falcon affair, which resulted in a war between two parties of Ponapeans, one of them ' Campbell's article was republished in The Polynesian, July 11, 1840, where Francisco Michelena y Rojas, the Venezuelan traveler who was at Ponape in 1841, evidently saw it. He proceeded to plagiarize from it, zJmost verbatim, in his own account of Ponape (Michelena y Rojas, 1843) as though he were the original author. The story is told in detail in Riesenberg, 1959. allied with the survivors of the Falcon and the crews of the Lambton and two other ships, and in the overthrow of the ruling Une of chiefs of one of the tribes. This incident is described later in this work.) Campbell says that the island "has been occasionally visited during the last nine years by the masters of ships engaged in the whale fishery, for the purpose of refreshment . . . " (Campbell, 1836), which would put the beginning of these visits in about 1826. James F. O'Connell, the allegedly shipwrecked Irish sailor who was rescued from Ponape in 1833, makes remarks that lead to similar conclusions; he refers to a song celebrating "the barking of a dog on board some vessel which had visited them," he asked the natives "why they stole from vessels," he refers to "bits of iron hoop, an officer's coat, and other articles" on nearby Pakin, to traditions of guns, to fowl descended from a pair presented to one of the chiefs by people with moustaches who ar? rived in "a big canoe with one stick" about 40 years before, etc. (O'Connell, 1836, pp. 175, 178, 181, 192, 201). This "big canoe" is interpreted by O'Connell to refer to a Spanish or Portuguese schooner; but Rosamel [in Hambruch, 1932 I, p. 117) refers to traditions of a Chinese junk that was wrecked at Ponape about 1810 and which brought the first fowls, and an identical remark is in one of L. H. Gulick's letters. When O'Connell arrived is not certain. On the basis of his remark that he left AustraHa "in or about the year 1826" botii Fischer (1957, p. 24) and Hambruch (1932 I, p. 77) conclude tiiat he was already on Ponape when Lutke arrived in January 1828. However, I am inclined to agree with the Reverend L. H. GuHck (1859, p. 131), the missionary to Ponape in the 1850s, when he says that "so much of the irreconcilably and egregiously incorrect is mingled with O'Connell's narrative . . . conceming everything connected with the whole island . . ." that very little of what he reports can be trusted. His story of his adventures in AustraHa before he came to Ponape and in the Philippines and China afterward is fuU of improba- biHties, exaggerations, anachronisms, and outright false? hoods. Even his name appears to be a pseudonym. Al? though he denies it, he was probably an escaped Sydney convict. By his own account he stayed at Ponape for some? thing over 5 years; since he was rescued in November 1833, this would put his arrival in 1828. He names the ship he sailed on as the bark John Bull, under Captain Barkus, which he says foundered after striking on a reef; after 4 days in an open boat he and five shipmates made Ponape. However, Barkus and the John Bull (the same vessel previously mentioned as under Captain Rowe when he discovered Ponape in 1825) did not sail on their final voyage from Sydney until May 13, 1830, and the bark was seen by another vessel off Japan in August of that year {Sydney Gazette, May 15 and November 13, 1830). Moreover, a newspaper report of 1837 states that the S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 bark and crew were the victims of some Europeans and "other savages" 7 or 8 years before at Pleasant Island, a good 800 nules from Ponape, a distance hardly likely to have been covered in 4 days in a boat {Sydney Herald, September 4, 7, and 28, 1837; The Friend, May 1853, p. 38). The missionary Sturges, writing in the early 1850s, flatiy states that O'ConneU ran away by swimming ashore from a ship that was lying off Kiti. O'Connell says he saw no ships or other white men, apart from his five companions, all the time he was at Ponape, imtil Captain Knight of the Salem brig Spy res? cued hun Ul November 1833. But Knight (1925, p. 200) reports that O'Connell told him that "a Botany Bay ship left the coast only ten days before, after obtaining upwards of seven hundred pounds of shell and in consequence I should find it much scarcer than usual," implying that a regular commerce in tortoise shell with trading vessels already existed; also, "a lot of whites came off with a quantity of shell which I purchased. These fellows had been put into the canoes of the natives by the Botany Bay whalers as they passed the island and undoubtedly were convicts. . . ." Knight (1834, letter) states that when he was at New Zealand, earlier in 1833, he inquired of various English wheding captains as to where was the best place to trade for shell; "one and aU were united in saying that Ascension was the best island for sheU in the Pacific." Evidently, then, Ponape was already well known. O'Con? neU is silent about the Sydney bark Nimrod, which lay m harbor with the Spy, and whose captain told Knight that he himself had left nine white men at Ponape, apparendy on an earlier voyage. (This unnamed captain had re? placed Captain McAuHffe, who had been killed a few days before at Pingelap. Sydney Gazette, April 8 and May 8, 1834; Sydney Morning Herald, May 8, 1834.) This same Nimrod, a whaler, then under a Captain White, had, according to Gulick (1858b, p. 34), anchored at Kiti in November 1832, on an earlier voyage, a month before Osborne and Frazer arrived in the Planter. In the same month another Sydney whaler, the Albion, Capt. John Evans, also stopped at Kiti. James (1835, p. 74), whose brief description of Ponape is dated December 1833, must also have been at Ponape before O'Connell's rescue, but O'Connell mentions none of these visitors. From this time on the whaling and trading vessels came in increasing numbers, and very soon a large colony of beachcombers, escaped convicts, and ship's deserters be? came established ashore. The log of the ship Eliza, Cap? tain Winn, on October 30, 1834, already refers to several white men in canoes coming off to the ship. Rosamel's 1840 report of the voyage of the French corvette, La Dandide (pubHshed in Hambmch), states that between 1834 and 1840 some 47 vessels had visited Ponape, nearly aU of them whalers: two were English, the rest American. There are no further statistics until the Protestant mission? ary period, beginning in 1852, but in that year Gulick reports the arrival of 29 vessels. In the next 4 years the tabulations are 33, 42, 23, and 19 vessels, respectively, 146 in all, of which all but 11 were whalers (Gulick, 1853, p. 19). CampbeU states that in 1835 25 foreigners were living ashore and that a short tune before there had been 40. Blake (1924, p. 668) gives a census of over 30 whites and Negroes m 1839; Michelena y Rojas (1843) men? tions 40-50 Europeans in 1841; Cheyne (1852, p. HO) reports upwards of 60 foreign residents ui 1846, "chiefly bad characters"; the report of the Swedish frigate Eugenie {in Hambruch, 1932 I, p. 148) gives a total of 30 in 1852. The missionary sources (letters and various issues of The Friend) count 150 in 1850, 60-80 in 1852, 60-70 m 1855, and 25-30 in 1857. By 1871, accordmg to Mahl- mann (1918, p. 56), after the decline of the whaling in? dustry and before the Spanish established their colony, the foreign population had faUen to two missionaries and 12 beachcombers. The effects of these acculturative influences were enor? mous. But they did not aU stem from foreigners. The brig Harmony {Nautical Magazine, 1838, p. 138), at Ponape in October 1835, signed on as a crewmember a Ponapean boy of 16. CampbeU, who was at Ponape the same year, reports that one native had already been to Hawaii and had retumed. The cutter Lambton in 1835 and 1836 took at least one Ponapean as a crewmember to Sydney and back to Ponape (Blake, 1924, p. 18). Soon Ponapean saUors were plying the Pacific and some were reaching New England ports, and many retumed to their home island. Most of the foreigners attached themselves to chiefs, as did O'ConneU, who married his protector's daughter. Blake (1924, p. 668) describes the situation m 1839 m these terms: the European seamen reside with Chiefs or petty Chiefs under their immediate protection, to whose tribe they are considered to belong and whose people become as it were their working attendants or slaves, pulling them in their Canoes, fishing for turde for them, collecting shell, etca., in short doing whatever may be required for them; the only compensation they receive being occasional small payments in pieces of tobacco. The chief perhaps receives nothing for a long period; but, on the arrival of a ship when trading is carried on, he is presented, in re tum for his protection and the services of the people of his tribe, with one or two muskets, Axes, adzes for making Canoes, powder or a portion of Tobacco, or whatever he may most desire; and this seems to be the sort of tenure by which the white men hold their setdement in the Island. When the Chiefs have once engaged to protect, they have in general shewn great fidelity to the White men . . . There was another hold on the foreigners, for among the letters of the missionary L. H. Gulick is one dated in 1853 that says it was a common habit to rob a white man of aU he had if he attempted to leave the tribe for another tribe. In another letter Gulick remarks that the foreigners INTRODUCTION 5 were quite at the mercy of the chiefs, so far as their prop- sionaries in 1852; they established themselves at Ponape erty was concemed; the chief expected at least half of aU permanentiy, and a wealth of information about the is- the earnings from a foreigner under his protection. Com- land is contained in their letters and journals deposited merce with the ships was almost entirely in their hands. in the Houghton Library at Harvard University. Mahhnann, who Wcis on Ponape from 1868 to 1871, says The later history of Ponape is admirably set forth in (1918, p. 58) that the chiefs "were always very eager to detaU by Hambmch (1932), Bascom (1965), and have a white man to do the bartering for them. They did Fischer (1957) and need not be recounted here. Though not give them any pay, but treated them like their own the Spanish had some claim to the CaroHne Islands, it sons; gave them houses to Hve in, and supplied them with was only in 1886 after a dispute over sovereignty with everything the island produced, including wives." Germany that they reaUy began to exert poHtical author- The first missionary arrived on December 13, 1837. ity. In consequence of the Spanish-American War, Ger- He was Father L. Maigret, later to become Bishop of many, in 1899, purchased the islands, together with aU Honolulu (A. Gulick, 1932, p. 89; Jore, 1953, p. 33; the Marianas except Guam. The Japanese occupied them Yzendoom, 1927, pp. 117-120; Maigret, 1837-38; in 1914 and the Americans in 1945. The natives of Ponape Maigret, 1839). With him had saUed another priest, often refer to events in their past as having occurred, for Father Bachelot, but he died aboard the ship and was example, "in German times," "before the Spanish," and buried upon arrival at Ponape. In their company were a so on. For this reason it wiU be convenient in this work number of Mangarevans and Tahitians, some of whom when precise dates of particular events are not determin- remained behind and left descendants on Ponape. Maigret able to refer, as Bascom does, to the foUowing historical sailed away again, after an unsuccessful 7 months, on periods: July 29, 1838. (I mention these details because in the Native period Before 1825 accoimts by Fischer (1957) and Hambmch (1932) there Pre-Spanish period 1825-86 are errors of dating and names.) The American Board of Spanish period 1886-99 V-, . . ? r^ ? -KIT- ? -n .. .. .. ? German period 1899-1914 Commissioners for Foreign Missions, a Protestant orgam- Japanese period 1914-4r5 zation with headquarters in Boston, sent a group of mis- American period Since 1945 POPULATION The early estimates of the number of natives of Ponape are highly unreHable, since they are based on brief en? counters with the people in a few locaHties. The Pona? peans often say that the population was once much greater than it is today, and they may be correct, for abandoned dwelling sites can be found in many places in the interior of the island, which is completely uninhabited today (see Plate l a ) . A list of estimates from various sources is given in Table 1. It would seem from the estimates that soon after European contact a population decline began, dramaticaUy accelerated in 1854 in consequence of the smaUpox epidemic of that year; the decline continued until about the tum of the century, when more accurate figures began to be compiled and a reversal of the trend commenced. I have not compiled later figures because the official German and Japanese tabulations for the period up to World War II do not usuaUy distinguish between Pona? peans and natives of other islands. The out-islanders began to arrive about 1906 and were considerably aug? mented in 1912. Most of them, especiaUy from the Mort- lock Islands, were resettled on Ponape by the administra? tion in consequence of destructive typhoons in their home islands. The 1911-12 figure of 3,190, given in Table 1, is composed only of Ponapeans, the census showing in addition 585 Central Carolinians and 279 Melanesian soldiers, but later figures are unreHable in this matter. The official Japanese figures compiled by Bascom (1965, p. 7) show a fairly steady increase from 4,165 "natives" in 1920 to a maximum of 5,905 in 1939, with some decrease thereafter untU 1946, but they are affected by the forced labor drafts that involved movement of Pona? peans and natives of other islands both to and from Ponape. My own census (Table 2) , conducted in November and December of 1947, with native assistance, shows a total population of 5,628, of which 4,451 were Ponapean and 1,177 were natives of other Pacific localities. (In 1963 the population was nearly 10,000.) Nearly 700 of the non-Ponapeans were natives or descendents of natives of the Mortlock Islands; most of them live in the six villages on Sokehs Islzmd. The only other community, inhabited by both Ponapeans and out-islanders, is the town of Kolonia, where the successive foreign admin- S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 TABLE 1.?Population of Ponape from Native Period to 1914 TABLE 2.?Population of Ponape, 1947 Source Sturges, 1855 letter Lutk6, 1835-36 James, 1835 CampbeU, 1836 Hale, 1846 (from Punchard, verbal) Hale, 1846 Rosamel {in Hambruch, 1932 I) Michelena y Rojas, 1843 Cheyne, 1852 Hernsheim, 1884 Gulick, 1852 letter The Friend, June 1853 Doane, 1855 letter Sturges, 1855 letter Sturges, 1854 letter Sturges, 1855 letter Sturges, 1855 letter Gulick, 1855 letter Doane, 1855 letter Doane, 1855 letter Gulick, 1858a Mahlmann, 1918 Novara voyage (in Hambruch, 1932 I) Mahlmann, 1918 Hernsheim, 1884 Finsch, 1880 Miguel, 1887 Gomez, 1885 Moss, 1889 Yanaihara, 1940 Pereiro, 1895 Christian, 1899 Bennigsen, 1900 Reichstag, 1902 Reichstag, 1904 Reichstag, 1905 Kolonialamt, 1911 Fritz, 1912 Kolonialamt, 1913 Bascom, 1965 Date of estimate Native period 1828 1833 1835 1835 1838-42 1840 1841 1844 1845? 1852 1853 1854 (before smallpox) 1854 (before smallpox) 1854 (after smallpox) 1855 1855 1855 1855 1855 1857 1858 1858 1868-71 1875? 1880 1885 1885? 1887 1891 1895 1896 1899 1900-01 1902-03 1903-04 1909-10 1910 1911-12 1914 Population estimate 20 or 30, 000 2,000 1,000 5-6, 000 15,000 7,500 20, 000; more likely 15, 000 7-8,000 7-8,000 15,000 no less than 10, 000 10,000 10,000 10,000 less than 5,000 less than 5, 000 a little more than 4,000 about 5, 000 6-8, 000 5,000 5,000 2,000 2,000 2,300 less than 2,000 2,000 5-6, 000 6,000 2,500 1,705 nearly 5,000 5,000 4,000 3, 165 3,266 3,279 3, 500-4, 000 3,200 3, 190 4,401 istrations have had their centers. The rest of the population is scattered in single houses or in groups of a few dweUings along the shores and coastal plains of the main island and on some of the lagoon islands. The census figures of Table 2 are broken down into subtotals showing populations by clan (or by place of origin for the out-islanders) and tribe. The natives of Ponape, like most of the CaroHnians, aU belong to exogamous, matrilineal clans {dipw or sou) whose Ponapeans, by clan name: 1. Pwuton 2. Lasialap 3. Sounrohi 4. Dipwinmcn 5. Ledek 6. Dipwilap 7. Sounkawad 8. Dipwinluhk 9. Dipwinwai 10. Liarkatau 11. Lipitahn 12. Sounsamaki 13. Dipwinpahnmei 14. Sounmaraki 15. Nahniek 16. Sounpelienpil 17. Dipwinpehpe 18. Sounpwok 19. Unknown TOTAL Out-Islanders, by island of origin: Palau Yap Truk Mortlocks Nukuoro Kapingamarangi Ngatik Mokil Pingelap Kusaie Marshalls Gilberts Saipan Philippines Malaya New Guinea TOTAL GRAND TOTAL Tribe Uh 32 250 - 29 12 78 60 56 48 74 28 7 143 20 36 13 4 8 - 898 - - 8 - - - - - 4 4 - - 8 1 - 1 26 924 Madol- enihmw 1 56 - 99 31 27 124 100 134 1 12 - 229 48 7 2 24 110 8 1,013 - - 14 17 - 2 21 25 23 3 _ 4 12 - - - 121 1, 134 Kiti - 81 1 264 15 65 281 36 135 7 49 26 117 2 66 20 233 65 - 1,463 - - 21 12 - - - - - _ _ - - - _ - 33 1,496 Sokehs - 11 1 50 5 42 67 - 24 - - 13 4 - 4 1 17 3 - 242 - 21 3 605 1 1 28 105 85 _ 1 3 9 _ _ 2 864 1, 106 Net 11 50 - 35 12 184 251 22 75 11 32 25 61 - 36 - 18 12 - 835 2 5 27 30 5 35 2 2 14 1 4 3 1 1 1 - 133 968 Total 44 448 2 477 75 396 783 214 416 93 121 71 554 70 149 36 296 198 4,451 2 26 73 664 6 38 51 132 126 8 5 10 30 2 1 3 1,177 5,628 members Hve in aU five of the tribes. The 18 clans are nearly all subdivided uito sub-clans {kaimw or keinek), which may or may not correspond to the conventional definition of the Hneage; but these are not tabulated separately. The clans listed in Table 2 differ somewhat from those INTRODUCTION given by Hambmch (1932 II, pp. 25-69) and Bascom (1965, p. 19). Bascom's Soimiap, Dipwinpwehk, and Dipwinwehi are extinct. He also shows three divisions of the Dipwinmen clan, which here are lumped together. Hambruch's material on Ponapean clans, put together from his notes by Eilers after his death and in part evidentiy misinterpreted by her, lists as Ponapean two Mordock or Truk clans, Sapwinipik and Sorr. I have not attempted to Hst the sub-clans, for several reasons. For one thing, it is very seldom that any two informants wiU completely agree on the number, names, and relative seniority of the sub-clans withm a clan. Some? times a man wiU insist that a sub-clan whose name was obtained from another man does not exist, or is only a branch of another sub-clan, or is a synonym for it. Some of the names appear to apply to groups of descent units on a level between clan and sub-clan; two such inter? mediate levels may exist among the Dipwinmen. In many clans there is one named group that, it is said, can intermarry with the other groups, which by the usual rule of exogamy ought to cause it to be defined as a separate clan, yet is not so regarded by its members. And among the 18 clans I have listed are a few that are often said to be only subdivisions of others; for example, clan 14 is frequently given as part of clan 13, clans 3 and 16 as part of clan 4. AU of this is to be expected in an evolving, viable society whose unUinear kinship groups continue to ramify, and my Hst of 18 is to be regarded as necessarily a rather arbitrary one. In the distribution of non-Ponapeans to island of origin, native classificatory ideas were foUowed in compUing Table 2. This no doubt has resulted in a number of errors. For example, a person bom on Ponape of an out-islander mother is classed according to origin of the mother, even though the father is Ponapean; but if the mother was Ponapean and the father an out-islander, he is Hsted under his Ponapean clan. A number of clans are shared among Ponape, MokU, Ngatik, and Pingelap; and some clans in the Mortlocks, Truk, and elsewhere, though known by different names, are equated with Ponapean ones. It is thus very probable that there has been inconsistent treat? ment in classification to island of nativity. The fact that natives of the Gilbert Islands and Gilbertese descendants Hve on Pingelap and Ngatik is also a complicating factor; some of the people classed as Gilbert Islanders are probably natives of Pingelap and Ngatik, and vice versa. Also, descent in the two Polynesian outHers, Nukuoro and Kapingamarangi, and in the GUberts has a patrilineal bias and true clans are lacking, in contrast to the other Micronesian islands where matrilineal, exogamous clans exist; when intermarriages occur, clan affifiation of the chUdren is often in doubt. Reference to clan membership throughout this work wiU usuaUy be by use of the numbers preceding the clan names given in Table 2. The Tribes Ponape, as it is presentiy constituted politically, is di? vided into five independent states or tribes {wehi) :* Uh, Madolenihmw, Kiti, Sokehs, and Net (Figure 1). Each of them is headed by two principal chiefs, known respec? tively as Nahnmwarki and Nahnken. The tribes are organized on a feudal basis and subdivided into a number of sections {kousapw) whose heads {kaun or soumas) are appointed by and formerly held their fiefs as vassals under the pruicipal tribal chiefs. The sections are further sub? divided into farmsteads {peliensapw), occupied by sepa? rate households whose relation to the section heads was lUiewise a feudal one. In theory aU the land formerly be? longed ultimately to the Nahnmwarki and Nahnken, who received regular tribute and whose rule was absolute. T H E TITLES Each of the two principal chiefs, the Nahnmwarki and the Nahnken, in each of the five tribes is the first of a series of ranked titieholders. In theory the two series of titles are the same in each tribe; actually they vary, for reasons to be discussed. The first 12 titles of each series are considered the most important, but the lists of titles obtained from informants often contain many more than 12; one Net man recited a series of 210 such titles. FoUow? ing Bascom, the series of chiefs headed by the Nahn? mwarki (the pali soupeidi) wiU be referred to as the A-Hne; the series headed by the Nahnken (the pali nahn? ken) as the B-line. If the Nahnmwarki is considered Al , the other titles then foUow, A2, A3, A4, etc., and for con? venience wiU usuaUy be referred to thus rather than by title. SimUarly, the Nahnken is Bl, and is foUowed by B2, B3, B4, etc. The A- and B-lines may also, for convenience in reference, be called respectively the royalty and the nobUity but without implying the European connotations of these terms. The lists of A and B tides for each tribe are given in Tables 3 and 4. < CommorJy called districts by the present Trust Territory administration. Most of the tides given in Tables 3 and 4, as weU as the many lower ones not listed, are purely honorific. Their translations are often suggestive of special functions (e.g.. Watchman of the Mountain, Gatherer of the Banana Fields, etc.), sometimes reHgious in nature (e.g., Lord of Thinking of Idehd, Idehd being the artificial island among the cluster known as Nan Madol where according to legend a sacred eel was once kept and fed; Sapawas and other place names that occur in the titles are also locali? ties at Nan Madol). But such functions have long lapsed; the few titles that stUl confer on their bearers particular duties or privUeges are discussed later. In these titles I translate the honorific words Nahn as Lord and Isoh as Honored One. These are arbitrary translations, since they have no specific meanings, and any English honorific would serve as weU instead. Other terms, as Watchman, Steward, etc., are fairly specific. Sou is given as Master; it has the implication of agency, simi? lar to the -er ending in EngHsh. Madau I have translated in some titles as ocean, in others as thinking. Both are correct, the choice in each case was that made by informants. I have given Sed as sea to distinguish it from Madau; Sed probably means T H E TRIBES P O N A P E 2 3 4 5 Km 6? 46' 158? 15' I : ^ p FIGURE 1.?Map of Ponape showing the five independent tribes. more exactly the water inside the reef. Ririn I have con? sistently translated as ladder; Fischer gives it as gate, but my informants insist it refers to the ladder that stood at the entranceway to the artificial island known as Pahn Kedira; it is not the ordinary word for ladder and is not used in any other case I know of except in the legend of a canoe descending from heaven, containing the god Luhk, which was suspended in the au* and was reached by such a ladder. The translation of Lempwei as successful side refers to a side in competition against another side. It could also be translated as a successful person who is outside of one's hearing. The Pei in Pohnpei (Ponape: "on" or "above the al? tar") and in titles is translated as altar for convenience. It actuaUy means any ancient stone stmcture, including burial chambers, house foundations, and mins, usuaUy with reHgious significance but not necessarily. 10 S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 TABLE 3.?A-line of titles (numbered according to ranking as given by informants and published sources) General Ch H a l Lu H a l l Uh A2U A2Uw BIU SoA* BaU Madolenihmw A l l M * BIMI Wa* BaM Hah Nahnmwarki (Lord of controlling titles) Wasai Dauk Noahs (Spray of the waves?) Nahnawa (Lord of endearment; Dear lord) Nahnipei (Lord of the altar) Nahn Kiroun Pohn Dake (Lord of the husbanders above the reserve) Nahlik Lapalap (Great lord of the exterior) Nahnid Lapalap (Great lord of the eel) Lempwei Lapalap (Great successful side) Soudel (Master of the ??????) Oundolen Ririn (Watchman of the mountain of the ladder) Nahntu (Lord of ) Isohlap (Great honored one) Kulap (Great ) Mwarekehtik (Little Nahnmwcirki [Al]) Kanikihn Sapawas (Steward of Sapawas) Nahnawa Iso (Honored dear lord) Soupwan (Master of ) Nahnsou Wehi (Lord of the masters of the tribe) Nahnsou Sed (Lord of the masters of the sea) Sapwetan Nahmadoun Oare (Lord of thinking of Oare [Kiti Section No. 15]) Kiroun (Husbander) Kaniki (Steward) Lepen Madau (High one of the ocean) Souwel en Wasai (Master of the forest of the Wasai [A2]) Nahn Pohnpei (Lord of Ponape) Oun Pohnpei (Watchman of Ponape) Oundol (Watchman of the mountain) Sou Madau (Master of the ocean) Nahn Kirou (Lord of the husbanders) Nahnkei (Lord of ) Luhennos (Remainder of the Noahs [A4]) Lepen (High one) Ou (Watchman) Nahniau (Lord of the mouth) Oaron Pwutak (Gatherer of the boys [boys=A2's title of address]) Kaniki Ririn (Stewzu^d of the ladder) Lepen Ririn (High one of the ladder) Sou Wene (Master of Wene) 10 9 13 12 14 15 11? 17 18 16 11 10 12 13 8 9 10 11 15 13 12 14 16 17 18 19 - 20 - 21 9 10 11 14 13 12 8 9 10 11 12 1 1 1 1 2 2 2 2 3 3 3 3 4 4 4 4 10 8 9 10 12 9 10 11 12 13 7 11 8 9 10 11 12 13 14 9 11 10 13 14 12 9 10 11 12 13 14 11 10 12 10 7 11 9 14 17 13 12 15 16 18 19 ?Numbers in these columns zxt used in the text for reference purposes. T H E TRIBES TABLE 3.?A-line of titles (numbered according to ranking as given by informants and published sources)?Continued 11 Nahnmwarki (Lord of controlling titles) Wasai Dauk Noahs (Spray of the waves?) Nahnawa (Lord of endearment; Dear lord) Nahnipei (Lord of the altar) Nahn Kiroun Pohn Dake (Lord of the husbanders above the reserve) Nahlik Lapalap (Great lord of the exterior) Nahnid Lapalap (Great lord of the eel) Lempwei Lapalap (Great successful side) LjUULlCi. ^IVidoLCr OI I l l C * J Oundolen Ririn (Watchman of the mountain of the ladder) TSTnVintn fT r\rtA r\f \ Ila-XUILLI ^^ijOrCl OI J Isohlap (Great honored one) V^^^^\T\ (C^rrit "1 IVUidp ^O^rcdL J Mwarekehtik (Little Nahnmwarki [Al]) Kanikihn Sapawas (Steward of Sapawas) Nahnawa Iso (Honored dear lord) Soupwan (Master of ) Nahnsou Wehi (Lord of the masters of the tribe) Nahnsou Sed (Lord of the masters of the sea) Sapwetan Nahmadoun Oare (Lord of thinking of Oare [Kiti Section No. 15]) Kiroun (Husbander) Kaniki (Steward) Lepen Madau (High one of the ocean) Souwel en Wasai (Master of the forest of the Wasai [A2]) Nahn Pohnpei (Lord of Ponape) Oun Pohnpei (Watchman of Ponape) Oundol (Watchman of the mountain) Sou Madau (Master of the ocean) Nahn Kirou (Lord of the husbanders) AT T l ' / T j r \ iNannJcei ^ i^^ ora oi j Luhennos (Remainder of the Noahs [A4]) Lepen (High one) Ou (Watchman) Nahniau (Lord of the mouth) Oaron Pwutak (Gatherer of the boys [boys = A2's title of address]) Kaniki Ririn (Steward of the ladder) Lepen Ririn (High one of the ladder) Sou Wene (Master of Wene) Kiti A6K* 1 2 3 4 5 6 8 7 9 14 13 11 15 - - - - - - - - - - - - 10 12 - - - - - - - - - - - Le 1 2 3 4 5 6 7 8 9 11 10 - - - - - - - - - - - - - - - 12 - - - - - - - - - - BaK 1 2 3 4 5 6 8 7 9 - 10 11 - - - - - - - - - - - - - - 12 - - - - - - - - - - - Sokehs B2S 1 2 3 4 5 6 7 8 9 10 - - - - - - - - - - - - - - - - - - - - - - - - - - - - Alb 1 2 3 4 6 - - 8 9 10 - - - - - - - - - 5 7 - - - - - - - - - - - - - - - - - Kw 1 2 3 - - - - 4 - - - - - - - - - - - - 6 5 7 8 - - - - - - - - - - - - - - BaS* 1 2 3 4 5 - 6 7 8 9 - - - - - - - - - - - - 12 10 11 - - - - - - - - - - - - - HaS 2 3 6 - 8 - 10 - ~ - - - 9 - - - - - - 5 - - 7 - - - - - - - - - - 4 11 12 13 Net B2NI 1 2 3 4 5 6 7 8 9 10 11 - 12 13 14 15 - - - - - - - - - - - - - - - - - - - - - - - - BIN* 1 2 3 4 5 6 7 8 10 9 11 - 12 13 14 15 - - - - - - - - - - - - - - - - - - - - - - - - AIN 1 2 3 4 5 6 7 8 9 10 11 15 14 12 13 16 - - - - - - - - - - - - - - - - - - - - - - - - BaN 1 2 3 4 5 6 7 8 10 9 11 ~ - 12 - - - - - - - - - - - - - - - - - - - - - - - *Nimibers in these columns are used in the text for reference purposes. 12 S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 TABLE 4.?B-line of titles (numbered according to ranking as given by informants and published sources) Nahnken (Lord of? Nahlaimw (Lord of ? Nahnsau Ririn (Lord of the masters of the ladder) Nahnapas (Lord of ) Nahmadaun Idehd (Lord of thinking of Idehd) Souwel Lapalap (Great master of the forest) Lepen Ririn (High one of the ladder) Ou Ririn (Watchman of the ladder) Nahn Pohnpei (Lord of Ponape) Oun Pohnpei (Watchman of Ponape) Kaniki Ririn (Steward of the ladder) Nahnku (Lord of ) Kaniki (Stewzud) Kirouliki2ik (Husbander of ) Oun Sapawas (Watchman of Sapawas) Isohlap (Great honored one) Soulikin Soledi (Master of the exterior of ) Lempwei Lapalap (Great successful side) Lepen (High one) Nahnipei Ririn (Lord of the ciltar of the ladder) Sou Madau (Master of the ocean) Ou (Watchman) Kulap (Great ) Nahn Kirou (Lord of the husbanders) Nahnkei (Lord of ) Nahmadaun Pehleng (Lord of thinking of "Under Heaven" [Kiti Section No. 35]) Soupwan (Master of ) Oundol (Watchman of the mountain) Nahnsou (Lord of the masters) Soulik (Master of the exterior) Kiroun (Husbander) Soulikin Dol (Master of the exterior of the mountain) Nahntu (Lord of ) Mwarekehtik (Little Nahnmwarki [Al]) Soulikin Sapawas (Master of the exterior of Sapawas) Lempwei Ririn (Successful side of the ladder) Nahnsohmw en Ririn (Lord of of the ladder) Nahnsaumw en Wehi (Lord of ? of the tribe) Kanikihn Sapawas (Steward of Sapawas) Nahn Kirou Ririn (Lord of the husbanders of the ladder) Nahnid Lapalap (Great lord of the eel) Oundolen Ririn (Watchman of the mountain of the ladder) Oaron Maka (Gatherer of the banzma fields) Kiroun Dolehtik (Husbander of "Little Mountain" [Sokehs Section No. 18]) Sou Maka (Master of the banana fields) General Hal Lu H a l l 1 2 3 4 5 7 6 9 10 11 8 12 13 14 1 2 4 3 5 7 6 8 11 9 13 12 10 14 Uh SoA* BaU 1 2 3 4 6 5 9 8 10 12 11 1 2 3 4 6 5 7 8 9 12 1 2 3 4 7 6 5 9 13 12 10 Madolenihmw A l l M BIMI* BIMII Wa BaM Hah 1 2 3 4 5 6 7 8 10 9 11 13 1 2 3 4 6 7 5 10 9 11 13 10 11 - - - - 11 14 - - 12 14 - - 12 14 10 11 12 1 2 3 5 4 8 7 9 13 11 12 10 14 *Numbers in these columns are used in the text for reference ptirposes. THE TRIBES TABLE 4.?B-line of titles (numbered according to ranking as given by informants and published sources)?Continued 13 i.^aUliJi.CXl ^JLJUJ-U. U I / X>i d l U a i l l i W ^ ^ X J O I Q OI ) Nahnsau Ririn (Lord of the masters of the ladder) ' M ^ ' U ? . n . ? ? n /^T -. J - . f N iNd-imapas ^j^oro oi ) Nahmadaun Idehd (Lord of thinking of Idehd) Souwel Lapalap (Great master of the forest) Lepen Ririn (High one of the ladder) Ou Ririn (Watchman of the ladder) Nahn Pohnpei (Lord of Ponape) Chin Pohnpei (Watchman of Ponape) Kaniki Ririn (Steward of the ladder) Nahnku (Lord of ) Kaniki (Steward) f ;..r^ . Mlj-i^h- fXJ ?U AC \ ?iviroTiIIKinK ^xxusDanoer or ' ' ' ) Oun Sapawas (Watchman of Sapawas) Isohlap (Great honored one) c ? . . K U ^ ? o ? i ? j ; /A>r??*rv-. ^c 4.1. -.4. "^ c \ oouxutin ooieoi ^ivLasier 01 tne exterior 01 ' ' ' ) Lempwei Lapalap (Great successful side) Lepen (High one) Nahnipei Ririn (Lord of the altar of the ladder) Sou Madau (Master of the ocean) Ou (Watchman) Kulap (Great ) Nahn Kirou (Lord of the husbanders) AT t. 1 - ? / T J r ^ i>ja.nnitei -^L*orQ 01 ) Nahmadaun Pehleng (Lord of thinking of "Under Heaven" [Kiti Section No. 35]) 0 /x X *. e \ ooupwan ^jviasier 01 ) Oundol (Watchman of the mountain) Nahnsou (Lord of the masters) Soulik (Master of the exterior) Kiroun (Husbander) Soulikin Dol (Master of the exterior of the mountain) AT 1 * - - / T J r \ IN annul ^L o^rQ 01 j Mwarekehtik (Little Nahnmwarki [Al]) Soulikin Sapawas (Master of the exterior of Sapawas) Lempwei Ririn (Successful side of the ladder) Nahnsohmw en Ririn (Lord of of the ladder) XT-^U - * TAT U * ft J r ^C *.\ *_.C1 \ iNannsaumw en weni _^L x j i It IS apparently mcluded as part of section 10. 3. Mwudok Soulik en Mwudok ^r? i i T i ? f r xn ? i 4. Sounkroun Soulik en Paleidi ^^ wUl be noted tiiat I have not given a hst of Y l tities 5. Olepel Lepenmadau en Olepel for the Sections of Kiti, and for those of Sokehs and Net 6. Pohrasapw Ounwene en Pohrasapw that foUow. My information for these tribes is in this 7. Ononmwakot Nahnawahn Ononmwakot matter contradictory and fragmentary, and by the tune . o em a n ro^n o sem j ^ ^ working in these three areas I began to have doubts 9. Kepmne Soulik en Kepinne , . , ,. , r i- r ? i ? 10. Nanpahlap Soulik en Pahlap whether tiie ideaHzed pa t tem of two Hues of tities m each 11. Pahnais Nahmadau en Kepinne Section, as expressed by informants, had much vaHdity, 12. Pehs Oun Rolong as I wiU discuss later. Sometimes an informant seemed to 13. Mwakot Soulik en Mwakot ^e searching for a titie to give me m order to fit it artifi- , -? ^?"^ f^ ^? ^^ emwei ciaUy into the scheme, and I am certain that many of the 15. Empempowe Soulik en Semwei t, , -.ri . , x i ? i i i ? i i 16. Enipeinpah Kroulikiak alleged Yl tities I obtamed are only the tities possessed at 17. Pwohk Souwen en Pwohk the time by men who occurred to my informants as being 18. Kipar Kanikin Kipar prominent in their sections. 19. Rohnkiti Krou en Wein In 11 of the 37 sections Bascom's X I tities differ from 20. Pweipwei Ritmg en Pweipwex ^ ^ ^ ^ ^^^^^^ ^ ^ ^ McGarry's " Ul five. In two 21. Salapwuk Saum . , ' , _. J T - I H T / - . 22 Mahnd Kasa of these instances Bascom and r a t h e r McLrarry agree, 23. Wenik Lepen Weruk and I have therefore deferred to them in the foregoing 24. Lauatik Sihpwen Lauatik Jigt; these are in section 14, where my own information 25. Oare Soulik en kepinne ^^^ ^^^^ -^^^ Semwei was X I , and section 3, where I zi' ?^^ t^^ -if ^^\^,^u^,^ had Nahnsoused en Mwudok; in the last case. Father 27. Poatoapoat Kaniki en Akahk . . , ^ . , . . . , . . , 28 Alauso Nahnsahu en lieu McGarry wntes that a imnonty of his informants agreed 29. Sewihso Nahnsahu en Sewihso with me. He also says that untU quite recentiy sections 4 30. Seinwar Soulik en Seinwar and 5 were one, with Soulik en Paleidi as X I , and that 31. Paliapailong Nahmadaun PaUapailong j ^ ^ jggg ^j^^ Nahnmwarki, bemg unable to discover what 32. Marahu Soulik en Marahu f *? c u u u J 33. Tamworoahlong Soulik en Lohd the proper X I titie of section 6 should be, made one up, 34. Pwudoi Luwen en Pwudoi SouHk en Porasapw. I t is probably for reasons of this 35. Pehleng Nahmadau en Pehleng kjnd that information about such tities is often 36. Dien Nahnsahused en Pehleng contradictory. 37. Nanpaies, or Paies Soukoahng en Paies * . , , , , . , T.- i j - ? ? i .i. ^L As m Madolenihmw, pohtical divisions larger than the Some informants list Sahpwtakai as a section; others sections once existed. As given to me by informants they include it in section 27. Here is the former seat of the a re : Nahnmwarki of Kiti. Today it is inhabited by only one o ? , ,o Wene Sections 1-13 family. Lukoap Sections 14-19 .. By 1966, according to Father Costigan, ?.h the resetdement K ^ P ^ - J (Kiti proper) Sejons 20-34 of people from Pingelap in addition to those from Losap atoll ^ already mentioned, there were five new sections, and these are . . T^ reckoned now as a seventh pzvihn. ? Correspondence with Father McGarry, missionary at Ponape 26 SMITHSONIAN CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 Wene and Kiti are terms stiU used for the two major halves of modern Kiti, and people describe themselves as belonging to one or the other, but the other names are of only historical interest. Luelen's manuscript gives several simUar lists of pofitical divisions, which apparentiy apply to different periods of history. One of these Hsts adds "Pasau" to the foregoing four areas and substitutes Pehleng (the modern name of section 35) for Lehnpwel. Another list consists of Wene, Kepihleng, Lehnpwel, and Ant; the last is the atoU 8 mUes off Kiti that nowadays is reckoned as part of sec? tion 19. A third Hst is the same as the last except for the substitution of the name Kiti for Kepihleng, its equivalent, but subsumes aU four areas under Onohnleng, which is almost always regarded by modern informants as the same as Wene only. A fourth Hst gives Wene, "Puasa," Ant, and Deleur?the last is a name otherwise appHed to the ancient seat of the Saudeleurs at Temwen Island in Madolenihmw but also nowadays to a locaHty in Wene. StiU a fifth Hst gives only three areas: Wene, ICiti, and Pehleng. What is now the tribe of Kiti apparently came into being in precontact times, in consequence of a series of wars between Wene, Kiti proper (Kepihleng), and Peh? leng. The date of these events is suggested from the in? formation, given by several informants and by Luelen's manuscript, that there have been 22 Nahnmwarkis since the time of the wars. (Other informants say there have been only 19.) The last six of these men, beginning with the present incumbent, were Benito, Sigismundo, Paul, Mikel (or MensUe), Hezekiah, and Luhkenmweimwahu (or Penena). It was either Luhkenmweimwahu or his predecessor who is referred to by L. H. GuHck as having died shortiy after his arrival in Ponape from Hawaii in 1852. This would work out roughly as an average of a 15- or 20-year reign for the last six Nahnmwarkis, and application of this formula to the earlier 16 incumbents would make the total span of time about 330 to 440 years. However, this seems far too long a time. Another line of evidence suggested by remarks of Gulick concerning a war which occurred a generation or two before 1852 would place it about 1800, if it is the same war. The Sou? lik en Semwei, who was 63 years old in 1947, informed me that when he was a small boy he had seen the son of a certain weU-known participant in the final battle (one Mahsohr, referred to elsewhere), this man then being quite old; the war, therefore, could hardly have occurred very long before 1800. He also said that the war occurred in the Hfetime of his great-grandmother, who had already Hved through the reigns of 12 successive rulers; if one assumes she was about 80 years old, the average reign of a precontact chief was about 7 years, and the last 16 Nahnmwarkis before 1852 occupied a period of 112 years, putting the war in about 1740 (or about 1760 if one figures 19 instead of 22 Nahnmwarkis since the war). In the tune before these wars Wene was ruled by a kind of priest-king with the title of Soukisehnleng ("Master of part of heaven"), commonly abbreviated to Soukise. He was regarded as the highest priest of all Ponape and was the leader of a cult and of religious practices diflferent from those elsewhere on Ponape. It is not altogether clear whether he headed a single Hne of titles in Wene or whether two fines existed; the latter is more Hkely. Luelen's manuscript gives the four senior titles of Wene, in order, as Soukise, Sou Wene, Madau, and Sohmw, and indicates that they aU belonged to the ruling clan, hence would have been one line. My own informants give the first three of these tities in the same order. In another place, in teUing the story of the conquest of Kiti proper by Wene, Luelen describes how the holders of the same three tities, Soukise, Sou Wene, and Madau, after the victory took the new titles of Nahnmwarki, Wasai, and Dauk (Al, A2, and A3), whUe another titleholder, Soumadau en Ponta, be? came Nahnken ( B l ) , " which would suggest that two clans were involved and that two lines of titles already existed in Wene, their holders taking equivalent titles of the conquered land. Hambmch (1932 II, p. 133) also says that the Soumadau en Ponta was the equivalent of the Nahnken. He names the Kroun Tamw instead of the Sou Wene as A2 or Wasai, and elsewhere (1932 II, p. 38) he gives Soukise, Sou Wene, Madau, and Soumadau en Ponta, in that order, as the four highest tides of a single ruling clan, which would make the last equivalent of A4, not Bl, contrary to Luelen, my informants, and those of Father McGarry. The fact that the titie Souma? dau en Ponta was held at one time by a man of clan 17 and later (as now) by clan 11, whUe Soukise and his Hne were members of clan 4, mcreases the probabUity that two lines of titles did actuaUy exist m Wene. Besides these titles there was a group of ranked priests who did not belong to the foregoing series (see the section on priestly tities to foUow and the Hsts of tities in Ham? bmch, 1932 II, pp. 132-133). In the aforesaid war, Pehleng (or Lehnpwel) first at? tacked Kepihleng (or Kiti proper). At that time KepUi- leng was mled by the Soun Kiti, a sub-clan of clan 4, its senior member was the Soukiti, and the Soukiti was there? fore the highest chief of KepUileng. Pehleng's high chief, Nahmadau en Pehleng, a member of clan 17, overcame Kepihleng and apparently assumed the titie of Nahn? mwarki, a titie either not previously used in Kiti or else a secondary titie that had belonged to the Soukiti. Wene was under another sub-clan of clan 4, and its high chief, '7 Father William McGany has written me that his information is that Soukise, Sou Wene, Madau en Ne, and Kroun en Ne became Nahnmwarki, Wasai, Dauk, and Nohs (Al, A2. A3, A4) respectively while Soumadau en Ponta became Nahnken (Bl). SUBTRIBES 27 Soukise, was appealed to on the ground of clan soHdarity. He thereupon, m concert with a high chief of Net, the Nahnsoused, marched against Kepihleng, overthrew clan 17 in a battie at Sahpwtakai, and took the titie of Nahn? mwarki as already described; aU of Kiti now became united into one tribe and Pehleng was reduced to a section within Kiti, the situation that exists today. The Nahn? soused of Net was rewarded for his help in the war with the title of Soukiti and the fief of section 32, both of which were retumed to Kiti only much later. Just before Souldse's conquest of Kepihleng he sent messengers, according to Luelen, to sections 14-19 to invite them to join him in the battle, which they did. My informants refer to these sections as Lukoap ("middle place") and from the context it would appear that they were semi-independent at the time. The five different lists of geographical units given by Luelen, as mentioned earHer, suggest that pofitical aUiances varied in composi? tion from time to time, and Lukoap appears as an entity in only one of them. Some informants give only sections 17-19 as comprising Lukoap, and Luelen's account of the war, which uses antique names for all six sections, has "Pasaulap" for section 14 and "Pasautik" for section 15, probably the "Pasau" or "Puasa" of two of the five lists, set off against Lukoap, suggesting again a condition of fluid boundaries. In Kiti proper, after the war and until German times, aU of the sections adhered to the Nahnmwarki as his fiefs, except for sections 25 and 31, which were under the Nahnken. Some of these sections were directiy under other high chiefs who in tum were enfeoflfed under the Nahnmwarki; thus section 32 came under the chief A7, section 27 was under the A3, sections 28 and 29 belonged to the A2, section 21 to the Saum, and sections 30, 33, and 34 were provinces of the Soukiti, a title that con? tinued in use in Kiti as well as in Net though its holder was no longer Al . The other sections served the Nahnm? warki directly. (See page 30 for a discussion of change of aUegiance of some of these sections.) Wene was also divided between the Nahnmwarki and Nahnken, the former holding sections 1, 4-7, and 12, the latter sections 2, 8-11, and 13." To tiie high chief of section 35 (Pehleng) two other sections, 36 and 37, had owed al? legiance, as already noted; after the Kiti v/ar his title (Nahmadau en Pehleng) was given to chief B2 (or Nahlaimw), who held the three sections in fief to the Nahnmwarki; the B2 to this day holds both tities Nahlaimw and Nahmadau en Pehleng. (This is the reason for my earlier remark, p. 18, conceming the B2 title in Kiti.) In Lukoap, after incorporation into the tribe of Kiti, sections 14-16 came under the Nahnmwarki, the others under the Nahnken. SOKEHS Section 1. Nanipil 2. Tomwara 3. Palikir 4. lohl 5. Likie 6. Oumoar 7. Sekere 8. Pohninal, 9. Nankui 10. Peitie 11. Utui 12. Sekerelap 13. Mwalok Title of chief XI PALIKIR (the mainland of Sokehs) or Nanpohninal 14. Lupwur, or Lup 15. Naneir 16. Denpei 17. Tamworohi, or Nanimwinsap 18. Dolehtik 19. Roio 20. Ipwal 21. Soledi 22. Pakin SOKEHS en Epwel Soulik en Sokele Soulik en Tomwara Lepen Palikir Krou en Wehn Noahs Palikir Nahmadaun Oare Lepen Sekere Soulik en Kepin Kroun en Sokele ISLAND Soukoahng Peitie Nahmadaun Oare Soumadau en Sokehs Mwarekehtik en Sokehs Lempweilapalap en Sokehs Kroun en Dolehtik Nahnken en Sokehs Nahlaimw en Sokehs Oun Sokehs PAKIN ATOLL Souni 18 This is in contradiction to Bascom (1965, p. 66) who says that the Nahnmwarki owned all of Kiti proper, the Nahnken all of Wene. This list of sections and section chiefs is given as it existed before the Sokehs rebeUion against the Germans in 1911, foUowing which there was a wholesale expulsion to Palau and elsewhere; the exUes were not permitted to retum untU Japanese times, when they found their lands, especiaUy on Sokehs Island, occupied by out-islanders who had been settied in Sokehs by the Germans. The organization of Sokehs had meantime been considerably changed. Today section 1 is in Net. Sections 3-5 in 1947 constituted modem Palikir section (which some informants reckon as two sections, Palikirpah and Palikir- powe), but the 1954 Ponape District list shows section 5 separately. Section 6 is now part of section 7. Section 9 is now part of section 8. (Section 8 itself was once part of a section called Kepin, a name preserved in the titie of its XI as given in the foregoing list.) Sokehs Island (sections 10-21) is divided into six viUages of out- islanders. Sections 10-14 constitute the modem vUlage of Pingelap, which takes its name from the atoll where its inhabitants originated. SimUarly, sections 15-17 are the modem MokU vUlage, named after that atoU. From the atoU of Ngatik came the settlers of Ngatik viUage, made up of former section 18 and part of sections 19 and 20. Three more viUages are composed of islanders from 28 S M I T H S O N L \ N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 Lukunor atoU and from Te and Satawan in Satawan atoU, aU in the Mortiock group; Lukunor viUage com? prises part of sections 20 and 21, Te viUage part of sections 19 and 20, and Satawan viUage part of section 21. Bascom's Hst (1965, p. 25) of 28 Sokehs sections differs so markedly from mine that it is weU to give my inform? ants' interpretation of his section names. His sections 1, 2, 3, 4, 5, 8, 10, 11, 12, 14, 15, 17, 22, and 27 correspond respectively to my numbers 16, 15, 14, 13, 12, 21, 20, 19, 17, 8, 7, 6, 4, and 2. But according to my informants, 12 of his sections are only farmsteads. No. 6 is in my section 11, No. 9 in my 21, No. 16 ui my 7, Nos. 18-19 in my 5, Nos. 20-21 and 23 in my 4, Nos. 24-25 in my 3, and Nos. 26 and 28 in my 2. Hambmch (1932 I, p. 321) gives a list of 12 sections only for Sokehs Island; he omits my sections 11, 12, and 18 (which he regards as farmsteads) as weU as 19 and 20, but he Hsts five others. It wiU be noted that the Xl ' s of sections 13, 14, 17, and 19-21 are represented as having tribal tities (see Tables 2 and 3) , as shown by the inclusion of the word "Sokehs" in their titles. These could hardly be section tities. A man may have both a section titie and a tribal title, as wiU be seen later, but in this Hst of Sokehs section tities there is no doubt some confusion in the minds of informants; the events foUowing upon the rebelHon of 1910-11 have probably dimmed their memories and they apparently remember better the highest tribal chiefs Hving in a section than the possessors of the section XI titles. Since a man may have both a section title and a tribal title, it is probable that informants have remembered only the latter as of the time of the rebelHon. It is not clear how the tribal area was divided between the highest chiefs of the A and B fines. (I do not refer to Al , or Nahnmwarki, here because Sokehs for a long period of time was ruled by the Wasai, a titie which is A2 in the other tribes. The reason for this is a legendary one. Today there is a Nahnmwarki.) Clan 7 traditionally suppHes the holders of the higher titles of the A-line, clan 6 those of the B-line. The Xl ' s of sections 1, 6, 8, 9, 11, 15, and 22 belonged to clan 7, those of sections 2, 10, 12, 17, and 19 to clan 6; however, in view of the situa? tion in the other tribes, this is not necessarily a clue to the division of Sokehs between the two highest chiefs. As for the other sections, the XI of section 21 belonged to clan 4, sections 18 and 20 to clan 5, sections 3-5 to clan 9, section 13 to clan 13, and section 7 to clan 17; section 14 fell under either clan 4 or 12, it is not clear which. Section 16 was the A2's seat and had no hereditary XI . GuHck's Hst of place names in the "Jekoits tribe" men? tions, besides "Jekoits" Island, only "Tomora" (my sec? tion 2) and "PaHga" (my section 3) . The last of these, Palikir, in GuHck's day was much larger than today and at one time apparently included all of what is now the mainland part of Sokehs and was poHticaUy independent. NET Section 1. Parem 2. Lenger 3. Dolonier 4. Pohrakied 5. Ninseitamw 6. Mesenieng 7. Nanipil 8. Eirike 9. Meitik 10. Kahmar 11. Pahnimwinsapw 12. Paliais 13. Nanpohnsapw 14. Ninsoksok 15. Tamworohi 16. Leprohi 17. Sekerenkauki 18. Lukapoas 19. Areke 20. Pohnpeil 21. PeUe 22. Pohnmanga 23. Dolokei Title of chief XI Lepen Parempei Lepen Lenger Krou en Doaroapoap Nahnawa en Mesenieng Nahlaimw en Metipw Soulik en Sokele Soumadau en Sapwalap Lepen Paremkep Sou Kahmar Sou Net Soulik en Ais Soulik en Sapawas Soulik en Ninsoksok Madau en Rohi Nahmadaun Ided Nahnapas en Net Soukoahng Krou en Rohi Soulik en Net Nahnsahu en lieu Luwehrei Net Koaroahm en Dolokei Sections 1 and 2 are islands. Section 7 formerly be? longed to Sokehs. Most of section 6 is now the site of the town of Kolonia and is not usuaUy reckoned as under Net. A former farmstead in section 6, Komwonlaid, is today sometimes considered a section. Section 4 was also formerly part of section 6. Section 23 is sometimes con? sidered as two sections, Dolokeipowe and Dolokeipah. The Net sections as in other tribes were ancientiy grouped into larger divisions. Sections 7-10 are coUectively known as Sapwalap and were under the XI of section 7. Sections 11-13 were under the XI of section 12. Sections 15-17 were headed by the XI of section 15. Sections 18-21 and a section caUed Leptohmara, consisting of modem sections 22 and 17, were under the XI of section 19 (though overlordship of section 17 could not have been held sunultaneously with that of the XI of section 15). Informants distmguish Net proper, meaning the Net penmsula, from the rest of Net as includmg sections 12 to 23 (some say 14 to 23). Bascom's Hst omits my sections 5, 11, and 22. The Ponape District Ofl5ce list of 1954 omits my sections 4-5, 11-12, and 15-22, apparentiy lumping most of them under a section caUed PaH Pohn Net, or Net proper. GuHck gives for the "Nut tribe," besides the island sec? tions, only four on the mamland: Misinipali, Nut, Misi- niung, and Definiur. The last two, judging from his sketch map, apparentiy comprise everythmg west of the Dahu Sokele River and are identifiable under the names of my sections 6 and 3. "Nut," shown as the peninsula east of the river, corresponds fairly well with the native usage of SUBTRIBES 29 the tribal name m its meaning of Net proper. MisinipaH is unidentifiable, but is shown on the Uh border. In Net the sections nowadays have been reduced to mere geographical significance. The fimctioning units for administrative and ceremonial purposes are the pwihn, artificial units into which the sections have been grouped. These are as foUows: 1. Section 1 2. Section 23 3. Sections 15-22 4. Sections 11-14 5. Sections 7-9 6. Section 10 7. Sections 3-6 and Komwonlaid Section 2 is not included because it is unoccupied and is used by the civU administration. It should be noted that whUe the section numbers as given here for Net and other tribes are of the writer's devising for use in refer? ence, the pwihn numbers, as in Madolenihmw, are used by the natives. SECTIONS, SECTION CHIEFS, AND CLANS The boundaries of the sections are very precisely known, and some appear to be very ancient. Nevertheless a proc? ess of spfitting and of combination is apparent from the legendary and historical examples previously mentioned. The boundaries and names of many defunct poHtical units are remembered, but when such areas are deserted and revert to wUdemess they are no longer considered as sections. The various old sections that constitute modem Palikir (see the Sokehs list of sections) show the process of combination at work. A number of pairs of sections bear the same names, with suffixes powe (upper) and pah (lower), indicating spHtting of one section into two. Chiefs' tities in fuU are usuaUy composed of three parts, as wUl be seen upon inspection of the various Hsts: the title proper, foUowed by the word en (of), then a word which in section titles is most often a section name. The fact that in each Hst of section titles a number of titles appear in which the geographical names that form part of them are different from the names of the sections to which those titles today pertain suggests a continual his? tory of political reorganization and recombinations. For example, the Soulik en Semwei is XI not of Semwei (section 14 of Kiti) but of Enipeinpowe (section 15). The Xl 's of sections 11 and 25 in Kiti bear titles containing the name of section 9. In section 4 of Net the titie of the XI refers to section 6. The XI and Yl titles in section 15 and the Yl title of section 17, in Madolenihmw, con? tain the names of ancient sections that have become de? populated and reverted to forest and are no longer con? sidered as sections. And the XI of section 8 of Net has as part of his title the name of the ancient area that includes the present-day sections 7-10. Some present-day or former sections are known as kap, areas in the forest where everything reputedly grows weU and where breadfruit remains ripe longer than elsewhere. Such a section is said always to belong to the same clan, though other sections may change clan affUiation from time to time. The areas covered by the ancient sections of Senipehn and Kepin in Sokehs are by some reckoned as kap; in Kiti, sections 9, 21, and 35 are so classed; in Net, section 7; in Uh, the hinterland of Awak; and in Mado? lenihmw, those sections in the area caUed Sapwalap that are under the Lepen Moar. (AU of these are under clan 7, except Awak, which comes imder clan 2.) There are a number of differences between the poHtical organizations of the tribe and the section. The clan affifia? tion of the XI is less consistently the same from generation to generation than is that of the Al , and there is no at? tempt, at least nowadays, to reserve aU the section tities in one line for a particidar clan. The ideal is to have each clan and each famUy in the section represented in the series. For example, in the X-line of section 16 of Kiti only 5 of the first 12 titles (given on p. 21) are held by the clansmates of the XI . In the organization of section 15 of Kiti the titieholders by name are as foUows: X-line 1. Cosmos 2. Moses 3. David 4. Ignace 5. Ignace 6. Luis 7. Aleck 8. Margarita 9. Joseph Clan 16 12 15 4 16 15 12 16 15 Y-line 1. Emereno 2. Jacques 3. Pema 4. Pe trigo 5. Michael 6. Johannes 7. Lucius 8. Luciana 9. Senio Clan 15 8 4 16 8 8 8 8 15 Five, instead of only the ideal two, clans are represented in the above rank order. (The total population of this section is 37 and is divided among seven clans.) It is obvious that the paraUel with the tribal system has broken down, unless we suppose that it never was anything but a pale imitation. Since the issuance of individual deeds to land in German times, the XI no longer holds all the land in fief to the Nahnmwarki; his famUy therefore is in no more advantageous an economic position than any other, except for the first fmits he may continue to receive 30 S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 through force of tradition, and is less Hkely to monopoHze the higher tities than formerly. The population of a sec? tion is smaU enough that, in contrast to the tribe, every man and many women receive a title of some sort; for no unUineal kinship group can ordinarily fiU even the most important tities in one line and members of other clans are necessarily drawn upon. Thus in Uh in 1947 the total population of one section (section 4) was only 7; the most populous section (section 2) had 138 men, women, and chUdren. Moreover, since exclusive inter? marriage of two clans does not exist on this political level, kinship bonds and the obligations to reward kinsmen with tities become dispersed over a number of clans. In the smaU population of a section nearly everyone is related to everyone else; so in section 15 of Kiti, X6, X9, Y5, and Y9 in the foregoing list of names are Xl 's grandsons; Y8 is his daughter; X8 and Y4 are chUdren of his sister; and Y6 is a brother-in-law. As a result, there is less feeling that a Hne of titles belongs to a single clan than there is on the tribal level. Also, there is no objection to free movement between the two lines during a man's career of poHtical promotions. Thus the present XI of section 15 was formerly Yl . Nevertheless, the tradition persists in section 15 that clan 15 has some proprietary right here; only five members of this clan are now residing in the section, among a total population of 37, but aU of them are in? cluded among the highest 18 titles as shown above. Uh seems to be the most conservative tribe in keeping XI tities in the same clan. The XI tities of aU but five sections in Uh are supposed to belong to the same sub-clan of clan 2 as do the A-fine tities of the tribe. The exceptions are section 1, which belongs to clan 6; section 11, which has been under clan 11 but is now under clan 7; sec? tion 13, which should be under an XI of clan 13 but has also had Xl 's of clans 8 and 9; and sections 2 and 5, which are under two different sub-clans of clan 2. In recent times section 8 has faUen under clan 5, section 9 under clan 10, section 12 under clans 13 and 15. In the Y series of tities of Uh sections there is less regularity of inheritance. The Yl titles of the 14 sections belong to eight different clans at present, no clan having more than three sections, and in only one section is it possible to find a Yl title persisting as much as three generations in one clan. Clan affiliation of successive Xl 's in Kiti is apparentiy less stable than in Uh, In the following Hst of Kiti sections the term "former," with reference to clan membership of the XI , appHes to some unspecified period of time in the past, as remembered by one informant: Present Clan Affiliation of XI {by clan number) 4 18 !8 4 Section 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. Former Clan Affiliation of XI {by clan number) 4 4 11 11 11 4 11 17 4 4 15 15 13 4 13 16 8 7 7 7 4 13 4 7 4 4 7 4 9 2 17 9 Section 1. 2. 3. 4. Former Clan Affiliation of Xl {by clan number) 4 11 4 4 Present Clan Affiliation of XI {by clan number) 4 7 18 13 18 11 18 17 17 4 16 4 13 4 18 18 18 7 17 2 6 17 4 17 17 4 16 6 7 2 9 7 It was previously remarked that various sections in Kiti were formerly held by the Nahnmwarki and Nahnken respectively, and each section was given in fief to an XI , directly or through subordinate chiefs. That the rule, or at least tendency, was for the Nahnmwarki and Nahnken to choose Xl 's of their own clans, namely clans 4 and 11, can be seen from the foregoing fist. Thus sections 1, 4-6, 14, 27, 29, 30, and 32, which came under the Nahn? mwarki aU formerly had Xl 's of clan 4; sections 1, 4, 5, 14, 27, and 30 stUl do. SimUarly sections 8, 9, 11, owned by the Nahnken, had Xl ' s of clan 11. In 25 out of the 36 Kiti sections fisted above, the XI now belongs to a different clan than formerly. Here, as in Uh and elsewhere, the causes are various, including dying out of the proper line, interference by the Japanese administration, and the issuance of private deeds of own? ership and the institution of patrifineal inheritance of land by the German administration, which had the effect of releasing land ownership from direct connection with either clan membership or the political stmcture. Many of the Kiti sections shown as now under clan 18 are legally owned by Oliver Nanpei, the A6 of Kiti, the wealtiiiest man in Ponape and a member of that clan. The clan affiliation of the Xl ' s of the various sections of Sokehs has already been discussed. Most mformants agree that the XI title should be inherited lUce that of Al , going first to younger brothers and then to eldest sister's sons in order of age; they at- SUBTRIBES 31 tribute the deviations from this rule to the effect of the German administrative measures just mentioned. With the title went the land; chieftainship was inseparable from land ownership, as noted elsewhere. Several instances were coUected of patrilineal inheritance in pre-German times, but it should be remembered that the influence of the missionaries was also directed toward this end. Some time before 1865 the then Nahnken of Kiti, Nahnku, who held about 10 sections of Kiti, left his land to his son, Henry Nanpei, instead of to the successor to his title, thereby foimding the fortune of the Nanpei fanuly. The OHver Nanpei mentioned previously was Henry's son and heir. The Nahnken had married the haU-Ponapean daughter of an EngHshman, James Headley, who helped him to compose a written document formalizing the legacy, which was later accepted as valid by foreign ad? ministrations. That it was not chaUenged by later Nahnkens is a tribute to the strength of Nahnku who was in other respects as weU an extraordinary personaHty, but it is also an indication of the weakening of matriHneal in? stitutions. The chieftainship of section 26 of Kiti has been inherited patrilineaUy for at least seven generations, go? ing from a man of clan 16 to one of clan 9, then in turn to a member of clan 4, 15, 7, 13, and now 17. On the other hand a number of sections have retained matrUineal inheritance of the XI title, regardless of how the land was inherited; thus the title of XI of section 30 of Kiti has gone to a man of clan 4 as far back as informants' mem? ories served, and in section 15 of Uh it has remained in clan 2 for at least eight generations. Perhaps related to the introduction of private deeds to lands by the Germans is the fact that a man may hold more than one section title simultaneously, having one "every place he works," i.e., wherever he holds lands or works lands for someone else. Thus the present X3 of section 16 of Kiti is also Yl of section 15. Whether this might have occurred in pre-German times could not be determined. The section organization is also different from the tribal poHtical organization in that the division into two lines of tities is less clearly marked. The fact that two sub-clans do not tend to monopoHze the two fines as they do in the tribe has already been pointed out. When clansmates find themselves in opposite lines the distinction between the lines becomes blurred. For a number of sections it was impossible to obtain lists of tities ranged in two series, and in a few cases the existence of a Yl was denied; either the series have been forgotten or the dual system was never in effect in some areas. The minor role of the Yl in many sections in contrast with that of the Bl in the tribe indi? cates that the division into two fines is not very significant poHticaUy in the section. UsuaUy the section organization has the appearance of an imitation, but a weak one, of the tribal poHtical system. 298-818?68 4 Yet exceptions are to be noted to these generafizations. For example, in section 2 of Uh the Yl by far over? shadows the XI in importance and dominates in aU situ? ations. This is of course a function of the personalities of the two men involved and is parallel to the situation that sometimes occurs in a tribe involving the Al and Bl. It should also be pointed out that in the older sections the XI title can sometimes be very important. In the section the XI may or may not belong to the royal or noble clans. In Madolenihmw, according to native theory, if a single man of clan 13 fives in a section that is tmder (and before the German division of the lands belonged to) the Nahnmwarki, himself of clan 13, that man wUl ordinarily be XI . An XI of the same clan as the Nahnmwarki or Nahnken is in an advantageous position, for he has a freer hand in his section than an XI who does not belong to one of the ruling clans. Before German times a section under an XI of commoner dan affifiation reverted, upon death of the XI , to the Nahnm? warki to be reissued on feudal principles. UsuaUy the as? signment of the section and the awarding of the XI titie that went with it foUowed matrilineal principles, but these could be disregarded at the pleasure of the Nahnmwarki. If the XI was of the Nahnmwarki's clan, however, such re? version did not occur, and inheritance proceeded along strictly matrilineal fines. In native thought, statements like "XI A owns section 5" and "clan B owns section 5" are equivalent, with reference to pre-German times, if XI A belongs to clan B; and "the Nahnmwarki owns section 5" means that ultimate titie to the land resides in the Nahnmwarki, who grants it in fief to the XI . The Nahnmwarki would hardly disturb the normal rules of succession and inheritance among his own clansmates. The principle that the XI should be of the clan of the Nahnmwarki if the section concemed feU under the Nalinmwarki, and of the clan of the Nahnken if the sec? tion was in the part of the tribe belonging to the Nahnken, though sometimes observed more in theory than in prac? tice, sometimes meant that a high-ranking tribal chief was not the XI of the section in which he fived. At one time the A2, A3, and B2 of Net all fived in section 1 of Net, a section faUing under the Nahnken, but a man of much lower tribal title, Oaron Mwar, was XI , and simul? taneously held the section XI titie, Lepen Parem. This was because the section belonged to clan 9, the clan of the Nahnken, and aU of the three tribal chiefs were mem? bers of other clans, whUe Oaron Mwar was the senior member of clan 9 fiving in section 1. Of aU these chiefs oiUy he had the right to banish a man from the section. Nevertheless, In such situations the XI often defers to the opinions of the coresident tribal chiefs and frequently consults them. Most informants use the terms kaun and soumas inter? changeably to designate the XI of a section. But a Net 32 SMITHSONL\N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 informant says that the highest tribal titleholder residing in the section is the soumas, whUe the holder of the highest section title, the XI , is the kaun. A Kiti man, however, says exactiy the opposite. A Sokehs man states that kaun is the proper term for the XI , but when the XI is on a visit elsewhere or is Ul and designates someone to take his place temporarUy, the substitute is caUed soumas.^^ Some of the section heads were virtually independent of the tribe. Sometimes this state of affairs resulted from the personal quafities of a particular XI and persisted orUy as long as he held office; for example, when PauHno, later Nahnken of Net, was the XI of section 2 of Net he had a free hand to receive offerings like a Nahnmwarki or a Nahnken; his status was like that of an independent baron in feudal times in Europe. In other sections there was a traditional semiautonomy; this was true of Awak in Uh and of PaUkir hi Sokehs. Some of the XI tities are considered to be very high, even when held by com? moners; the XI of section 15 of Kiti, for example, belongs to a nonroyal clan but nevertheless is said to take prece? dence over chiefs below A2 and B2 in such matters as food distribution at feasts. The A4 of Kiti, who is the son of the Al 's mother's sister's daughter and therefore ranks high in his sub-clan as weU as in titie, is also XI of section 16, and informants say that it is his XI title, Kroufikiak, that is the more important and entities him to more honor. The present XI of section 27 of Mado? lenihmw possesses simultaneously a tribal title but never uses it, regarding his section title as more important. The same is tme of the Xl ' s of sections 18, 24, and 25 of Madolenihmw; the first of these three men was recently offered a fairly high tribal title but refused it.^? FUNCTIONS OF T H E SECTION CHIEF The XI must keep an eye on the productive abifities of the various farmsteads of his section. He attends the Cookhouse Counting feasts (q.v.), since he must know how the crops are progressing. He counts aU the food articles brought and may instruct the people of any farm? stead to do better. One farmstead may be occupied by a family of only five people, another by 30, yet the first wUl try to do as weU as the second; so the XI may attempt to arrange with the more populated farmsteads to make larger offerings to the high chiefs at feasts, in order that demands on the smaUer farmsteads wUl be less. If a man with a tribal title lives in the section, the XI comes to hun, reports on the difficulties of the smaU farmsteads, and secures advice. The XI was formerly charged with the keeping of those tribal properties {kepwe en wehi) not retained by the Nahnmwarki. Formerly each section annually made a tribal canoe {wahr en wehi). All of these canoes were defivered simultaneously, by the fourth month of the year, to the Nahnmwarki. Such of the canoes as were not retained by the Nahnmwarki were retumed to the custody of the individual sections, but were caUed upon when needed for the tribe, as on the occasion of a visit in force to another tribe. If a section did not have its canoe ready by the designated time, it had to be destroyed and the people of the section did not attend the ceremony that was held then; an XI might be demoted for such negligence. Other objects made for the tribe but retained in the section included pit-breadfruit, sennit, sleeping mats, spears, ?md various other valuables. Nowadays only large yams and kava are placed in the category of tribal prop? erty. They are actually grown on the individuaUy held land, but the concept persists that these goods belong to the Nahnmwarki, even though the land on which they are grown no longer belongs to him but has been issued under private deeds, and they are caUed for on the occa? sion of a visit by a Nahnmwarki of another tribe. The XI would also be caUed upon to defiver baUs of sennit and mats, kept for his Nahnmwarki, especially for such visits, to be distributed as gifts; the visitmg Nahnmwarki who received the gifts would redistribute them to his own section heads to keep for him, and they might be retumed to the original donor, often with interest, on a retum visit. UsuaUy as many spears were kept in the section as there were male residents, and they were given out before a battie. The Nahnken and other chiefs would count the spears after the fight and thus would know who had fled the field; such flight was a great shame, and a coward might be put to death. The spears were kept in the house of the XI and m the houses of any tribal chiefs who resided there. "?Hahl (1904, p. 10) defines kaun as "Obrigkeit, Herrschaft, Herrscher, Haupt" and soumas as "Hauptling (=erster in Stamm)." 20 Father Costigan tells me (1966) that when the present Al of Madolenihmw was A2 (he was A4 during my 1947 field trip) he was simultaneously XI of section 8, where he had the title Soulik en Metipw, and it was the latter titie that was called out in the food distribution at the section feasts, not his tribal tide of Wasai (A2). SUBTRIBES 33 Chiefs' fiber kUts, differing from ordinary kUts in that each fiber was crimped along its length, and the loom- woven banana-fiber belts wom especiaUy by men of high rank were made by the section women for the tribe; the Nahnmwarki would keep some for hunself and retain others for gifts. Pit-breadfmit might be tribal or section property; each section made a pit near the house of the XI and the contents were used on the occasion of a visit by the Nahnmwarki to the section or at a section festival; a sunUar pit was made for the Nahnmwarki from the breadfruit brought to him as offerings on various occa? sions. The people of each section made communally the three types of seines used in Ponape and deposited them at the home of the X I ; the same was done with the com? munally buUt section canoe. Anyone might use these articles as he needed them. Today there are no section nets, but some section canoes are kept. The change is due to the encroachment of money economy on the native subsistence pattems; people prefer now to seU their catch, hence they use their own fishing apparatus, whereas formerly they used section property and divided the catch with the XI . Just as the section heads were tenants {kohwa) of the tribal chiefs who owned the sections, or of other chiefs on intermediate levels who in tum held fiefs under higher chiefs, so the commoners were tenants of the XI . (A patch of ground in pre-German days might have several "owners," each holding it under another.) Whether a farmstead reverted to the XI on its owner's death is not clear; there are many statements to the effect that farm? steads reverted directiy to the Nahnmwarki; possibly the reversion to the XI occurred only if he was a member of the same clan as the Nahnmwarki (or Nahnken, if it was the Nahnken who owned the section). In any case, matrilineal rules of inheritance were usuaUy foUowed, with only occasional interference by the X I ; even if a family died out, the land was not redistributed to other residents of the section but relatives were often brought in from elsewhere to occupy it. If a stranger came from elsewhere to take up residence, the XI would consult with lesser titieholders hi the section and they would decide upon a piece of land to give the man in tenancy. There was usuaUy plenty of vacant land in the interior for this purpose, for Ponape at its maximum population in pre? contact times seems never to have been overpopulated. The stranger would then get a section titie from the XI and would offer first fmits to him. CAPITAL SECTIONS Included in the Hsts of sections are some designated as nanwei; properly these are not sections at aU and do not have hereditary Xl 's , for they are the seats of residence of the Nahnmwarki. There was no such permanent place of residence for the Nahnken, who might Hve anywhere. The nanwei might justifiably be referred to as the capital of the tribe. In JCiti the Nahnmwarki always Hved in Sahpwtakai; the term nanwei here is used by some in? formants to include also sections 24 to 34, which constitute most of Kiti proper. In Wene, before its union with Kiti proper, the seat of the highest chief was Eleniong, a farm? stead in section 7. The capital of Sokehs was anciently section 16, though some later rulers Hved in section 20 as weU as in a number of other places. In Net it was section 21. In Uh six farmsteads, all in section 7, comprised the capital: Nanpei, Lukoapoas peiei, Lukoapoas peUong, Nandeke, Tipwenkepwei, and Pohnweinpwel. The first five of these are now privately owned, but the sixth is the site of the modem-day tribal govemment offices in Uh; it is thus StiU the seat of govemment, although the Nahnmwarki no longer lives there. In Madolenihmw, however, a portion of the ancient capital in section 23 is StUl the residence of the Nahnmwarki. Most Nahnmwarki of Madolenihmw originate in sections 4 or 8, where, until German times, the land of their sub-clan was inherited matrilineaUy and where they stUl five until they reach the title of A2; but even today when they become Al they move to section 23. The Nahnmwarki and Nahnken are said to ride or sit astride {dakedake) the sections where they Hve. Nowa? days they hold land under deeds, like other people; formerly they might have farmsteads of their own before they took office, but upon accession of the titles of Al or Bl they would give their land in tenancy to near relatives since the first fruits they received would be enough to support them. Inheritance of the capital does not seem clear. Most informants say that it always went from one Nahnmwarki to the next. Some assert that it proceeded from father to son, but it is difficult to see how the Nahnmwarki could have continued to five on it generation after generation if it passed to another Hne, as it would with patrilineal in? heritance combined with matrUineal clan exogamy. Possibly a part of it was inherited by sons; but it appears that the actual dweUing site of the Nahnmwarki itself (the pos) went always to the next Nahnmwarld. Promotion and Succession THEORY AND PRACTICE In native theory, as repeatedly expounded by inform? ants, when a title becomes vacant through death or other cause, all men of lower title move up one place in the line of titles affected. In theory, again, aU of the higher titles in the two lines are held by members of the two ruling sub-clans, and position in the order of titles foUows pre? cisely seniority withui the sub-clans. The proper succession to a vacant title is thus supposedly a matter of common knowledge; the Nahnmwarki is the senior man of the royal sub-clan, the successor to a deceased Nahnmwarld should be the man of the same sub-clan of next highest seniority, and this man should previously have held the next lower title, A2. Likewise with the Nahnken and the noble sub-clan and his successor the B2. The actual state of affairs is, however, quite different. We find that instead of automatic succession there is often much defiberation as to who wUl actuaUy succeed. For ail tribal titles except their own the Nahnmwarki and Nahnken make the decision when vacancies arise; in practice the Nahnken often has the decisive voice, but he usually seeks advice.^^ If the office of Nahnmwarki itself becomes vacant, the first two or three titieholders in the B-Hne, and sometimes also A2, meet and choose a succes? sor; similarly, if the title of Nahnken is to be disposed of, Al , A2, and perhaps A3 and B2 decide on the new occu? pant of the office. The criterion of seniority within the sub-clan is one affected by several modifying considerations. Thus it is not Joseph, the present Nahnken of Net, who is the senior 21 In Uh the exclusive right of the Bl to choose a new Al has now become a matter of law, the modern-day council of that tribe having passed an ordinance to that effect following a dispute over succession that is recounted elsewhere. (Information from Father Costigan in 1966.) 34 man in his sub-clan, but the present Lepen Net. This title was once the highest in Net, being the equivalent of A] , but when Net adopted the title system of the other four tribes, as previously described, it and the other titles in the older series continued in use but sank to a lower posi? tion. It has been given to its present holder, even though he belongs to the B-clan, in a kind of sentimental gesture of harking back to older days. Though the senior man of the noble sub-clan, he was inefigible to become Nahnken because his father was not a member of clan 7. In Net the titles in the A-line are supposed to be held by the men of the senior sub-clan, which bears no name, of clan 7, and those of the B-Hne by the men of the senior sub-clan, Luke? lapalap, of clan 6. Supposedly, in each tribe the royal and noble sub-clans formerly intermarried exclusively, so that because of matrUineal descent aU the holders of the tities in the A-Hne were either fathers or sons of those of the B-Hne, and vice versa. The present Lepen Net's mother failed to contract such a marriage, whereas the present Nahnken of Net is of proper descent on both sides. Hence, though the Lepen Net is considered as chief of the noble sub-clan, Joseph outranks him in the titie series and is Bl. A sub-clan chief is caUed meseni en keinek, a contrac? tion of mese, face; en, of; ngi, tooth; en, of; keinek, sub- clan. This refers to the smUing, benevolent countenance of the chief, accordmg to some; to his fearsome, wrathful ex? pression, according to others, Ngi also can mean a certain hard, refractory wood. All of these altemative transla? tions are suggestive of the attitude of the members of the sub-clan toward their chief and of theu- expectations of him. But meseni is also translatable as first-bom. In cases where the Nahnmwarki or Nahnken (or other high-titied man) is not the senior member of his sub-clan he is not usually caUed meseni. It is the above-mentioned Lepen PROMOTION AND SUCCESSION 35 Net who is regarded as the meseni of the Lukelapalap of clan 6, not the Bl of Net, who is also a member; it was to the Lepen Net that the A6 of Net went hi 1947 to ask permission for his daughter (who is also of that sub-clan) to marry, as is the custom, not to the Bl . Agam, the meseni of the sub-clan to which the present Al of Mado? lenihmw belongs is not the Al but the Kulap, a title that always adheres to that meseni, whatever other titie he may hold. Relative age is also a modifying factor in matters of seniority. Johannes, the present B4 of Net, also belongs to the same sub-clan as Joseph, and unHke the Lepen Net both of his parents belong to the proper sub-clans. More? over, he too is senior to Joseph, for his mother's mother was the elder sister of Joseph's mother; in the Ponapean kinship system Johannes is "great maternal nephew" {wahwahlap), Joseph is "littie maternal uncle" {uhlap- tik) ,^ However, Johannes was allegedly too young at the time of the occurrence of the last vacancy in the title of Bl to be considered eHgible, The present Bl was already B5 when Johaimes, at the age of 20, got his first tide, B12. Supposedly he is next in line for Bl.^^ Although it is generaUy clear who is most efigible for the two highest offices, occasionaUy a promotion comes as a surprise to the populace and to the new titieholder him? self. In one case the AlO of Sokehs was made Nahnmwarki in a single jump; he had thought he was going to become A2. The reason for this promotion has been forgotten, but usuaUy when promotions come unexpectedly it is because the person actually entitied to the vacancy is considered too young, as Johaimes, the B4 of Net, was aUeged to be. When Paid, a Nahnmwarki of Kiti, was about 12 years old, he was A2. The old Nahnmwarki died and Paul was scheduled to take the vacant place. A boy may succeed to a title at any age, but Paul's mother refused to allow it, considering that he was too young and would therefore be too vulnerable to sorcery directed against him by jeal? ous persons; instead she advised the Nahnken to appoint another man, Mikel, the father of the present Nahnken, Benito. This man belonged to the proper sub-clan but was well down the seniority list, only A14, and had never expected to become Nahnmwarki. In the community house, where the promotion ceremonies are held, the Nahnken raised the traditional bowl of kava and caUed the A14 to come up to the main platform from the central area where he was sitting. The latter was greatiy surprised and demurred on the ground that he was unworthy. The 22 Actually, though it was Joseph himself who told me that Johannes was his senior, many Ponapeans would dispute this formulation that a man is senior to his mother's mother's younger sister's son. 23 The source of this information was the Bl , whose own ambi? tions must have been involved. Since there are a number of holders of high rides even younger than 20 years, other factors possibly entered into the decision not to promote Johannes to B l . Nahnken grew angry, came and seized him by the hair of his head, and dragged hkn up to the main platform, where he presented him with the bowl of kava, put the royal wreath on his head, and proclaimed him Nahn? mwarki. Paul remained A2 but received first fruits from Kiti proper; the new Nahnmwarld got first fmits only from Wene. This was the second time Paul had passed up the title, the first opportunity having come when he was only a baby. He was past 30 when the third opportunity came and was accepted. Sturges, the missionary, describes the making of a new Al of Kiti in 1852, the old one having died on Septem? ber 30 of that year. On the third day of a feast prepared for the occasion, in the presence of the assembled people, the Nahnken, holding a cup of kava in his hand, put the question to the A2 and the A3, "Which of you wiU be the Nahnmwarki?" The question was put three times, without reply; whereupon the Nahnken rose up and threw them both bodUy out of the community house. After a few moments of thought he then sent the cup to the A3, thus constituting him the new Nahnmwarki. According to Sturges he did not promote the A2 in order to sUence the talk of certain busybodies who were saying, "Of course he wiU make the A2 the new king, for he is his uncle." (This description by Sturges raises some questions that are not answerable after this lapse of time. But I would guess that the Nahnken's question was purely a rhetorical one; it is almost impossible for a Ponapean, in pubfic, to be so immodest as to put himseU forward or to state an ambition openly. The Nahnken had evidently already made up his mind anyway. The display of temper is en? tirely in accord with Ponapean expectations of chiefly behavior; the Nahnken here concemed was Nahnku, also caUed Solomon, a man of unusually strong personality and quafities, to whom we have occasion to refer several times elsewhere. The reference to an uncle-nephew relationship must have to do with a patrUineal one, for mother's brother and sister's son would have been in the same clan and the same fine.) Additional factors are to be considered in choosing a new titleholder, some of them due to foreign influence. The current A2 of Madolenihmw is debarred from becoming Nahnmwarld because he is blind. (Hambruch mentions also ringworm as sufficient cause to prevent ac? cession to the highest offices, but modern informants deny this.) Also, he belongs to the third-ranking sub-clan of the A-clan (clan 13) of Madolenihmw. Moses, the Al, be? longs to a branch of the highest sub-clan, the Upwutenmei, as did his predecessor Alexander, his mother's brother, who called Moses back from Kiti, where he had been living, to become A2 and then to succeed to Al . The A3 belongs to the senior branch of the Upwutenmei, which should theoretically mle Madolenihmw, but nearly every- 36 S M I T H S O N L \ N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 one says neither A2 nor A3 wiU succeed the present Al , but A4, Samuel, the Al 's brother, wUl.''* In Uh, the A5 is sister's son to the present Nahnmwarki and next in line, if strict seniority were considered, since there are no younger brothers. His clansmate, the A2, is senior to him in titie but junior in blood, being mother's sister's daughter's son to the Nahnmwarki, The A5 should therefore go ahead of the A2 in the succession to the title of Nahnmwarki, But the people befieve that the A2 knows more about modem govemment practices and the A5 may defer to hun on this account.^^ OccasionaUy a Nahnmwarki wiU express his wishes re? garding his successor on his deathbed, though this is re? garded as unusual. In Net, when the last Al , Saturlino, was an old man, the A2 was Benito of his own sub-clan. ActuaUy Benito was senior to Saturlino, his mother having been Saturfino's mother's older sister, but Benito had de? ferred to his cousin because of his relative youth at the time the previous Al , Eduardo, died and the office be? came vacant. Benito expected to succeed SaturHno as Al . When Saturlino approached death he took the titie of A2 away from Benito and gave it to Max who was A4, Informants' accounts differ as to why he did this; some say he was angry at Benito (and also at the A3) for not helping him during the hard times of World War I I ; others say Benito agreed to this arrangement amicably, while waitmg to become Al , and that Saturlino had dic? tated a written statement that specified Benito was to fol? low him. As Saturlino in his dotage lay dying, in 1946, there was much maneuvering for position, with Joseph, the B1, intriguing on behalf of his son Max, who belonged only to the eighth-ranking sub-clan of the royal clan. The Net people were rent in two by the struggle. The A6, 2? Information from Father Hugh Costigan in 1966 is that when the A2 died in 1960 Samuel, the A4, received the A2 tide, and that when Moses in t u m died, Samuel became A l , 2? According to Father Costigan, the A2 did succeed when the Al died, and the A5 is now the A2. As in the case previously mentioned in Net, the Al of Uh, Edmund, decreed on his deathbed that he would be succeeded by the A3, Bruno. There had been a dispute over the titie of B2, described elsewhere. The A2, Johnny, had not joined the dissident faction but was known to be sympa? thetic toward it, his own sister was married to the B6, who was a member of the faction; and Edmund continued to retain some resentment against the dissidents after the matter was settled. This was in spite of the fact that both Edmund and Johnny were members of the same sub-clan of clan 2, the Sounpasedo, while Bruno belonged to a different sub-clan, the Sounpeinkon, which once ruled Uh but was long ago deposed. When Edmund died Bruno immediately assumed the role of A l , but after the funeral the Bl formally awarded Johnny the titie. A stalemate developed, and both factions went around Ponape visiting the Nahnmwarkis of the other tribes to get their support. It was finally decided to employ the modern-day institution of a popular vote, and Johnny won the election. It was after this that the Uh council passed the ordinance previously referred to giving the Bl sole right to choose a new Al in Uh . Arnold (now the A4), actually went to Benito and anointed him as Al , a highly irregular and improper maneuver. The U,S, Navy had just taken over from the Japanese, and Joseph put Max forward to a naive Ameri? can administrator as the best candidate and induced him to go to Saturlino and ask him to designate his successor. Although the Bl normaUy selects the new Al , Joseph evi? dently felt that he could not bring off such a coup by himself. There was much badgering of the old man, par? ticularly by his sister's daughter, untU Saturfino finaUy acceded and named Max, in the presence of the Ameri? can. Benito took the usual measure that Ponapeans em? ploy in such a situation; he went into self-exUe to Mado? lenihmw. Foreign influence has sometimes affected succession in other ways; the present Bl of Uh passed by his mother's brother, the former B2, because of interference by the local Japanese poficeman. In Madolenihmw there was even more direct foreign interference with the succession in consequence of the incident of the Falcon in 1836, This vessel, a whaleship, was burned, its contents looted, and its captain and some of the crew kiUed by the A5 of Madolenihmw and his adherents. In revenge, a party of white men in league with the A2 and four or five hundred of his men made war on the A5 and finaUy captured and hanged him. The Al , whose burial name was Luhkenkasik, was the A5's brother; though less di? rectiy impficated in the affair he was shot. The two brothers belonged to the senior sub-clan of clan 13, caUed Upwutenmei. The man who then became Al was ap? parentiy the A2 who had fought alongside the whites, and he was placed in power with the help of those whites. This man, known as Luhkenkidu (see p. 16), seems to have belonged to a more junior branch of the Upwutenmei, to which aU succeeding Nahnmwarkis of Madolenihmw have belonged. (Some people refer to this junior branch as Pahnmei and regard it as a separate sub-clan; as with the other ramifying unilinear groups on Ponape, different people have different opmions as to whether their groups are subdivisions of others or have become independent.) At that tune it was also decided by the chiefs that no one of Luhkenkasik's branch could ever rise again above the position of A5.^* 26 Moses, who was Al in 1947, however, had broken this tradi? tion by elevating a member of Luhkenkasik's branch to the position of A3. Father Costigan says (1966) that when Moses was dying the chiefs agreed that all rides would move u p one place; Samuel, the brother of Moses, was then A2, so he became A l , and the A3 became A2. The Madolenihmw chiefs have agreed that in the future there will be no skipping of places in the hierarchy, so it is possible for the new A2 some day to become Al and for his branch of Upwutenmei to be restored to the power they held before 1836. But since both A2 and A3 are older than Samuel, the present A l , and will probably predecease him, it is likely that the A4, who is of the same sub-clan as Samuel and his predecessor Moses, will succeed and that the second branch will remain in power. PROMOTION AND SUCCESSION 37 In Kiti the B4 should succeed to Nahnken; Kiti is unique among the five tribes in that the B2 never belongs to the noble clan but is always a member of clan 9 and hence is not eHgible to become Bl ; and the B3 titie is held at present by a commoner. The B4 is therefore next in luie, being next senior in the Bl's sub-clan. But the present B7 also belongs to the proper sub-clan, is the Secretary of the tribe under the civU administration, and is famUiar with office procedure, and many people think he wiU succeed. In MadolenUimw there are several possible contenders for the position of Nahnken, The present B2 belongs to one of the proper clans (clan 2) but not to the proper sub-clan (Sounlehdau). The man considered best versed in office work is the A14, who is "acting Nahnken" under the American administration; though he holds an A title he belongs properly in the B-line, since he is a member of the second clan (clan 9) of the two clans which in this tribe are noble. He belongs to the same sub-clan as the Nahnken, being a son of the Nahnken's sister's daugh? ter, hence is efigible.^ ^ The Nahnmwarki, it is said, how? ever, wiU probably choose as next Nahnken his own son, who is a member of clan 18 and is presentiy B6; the Nahnmwarki has not foUowed the traditional pattern of marrying a woman only of the opposite fine but hcis married instead a commoner. If the son becomes Nahnken, he wUl be the first holder of that titie in this tribe who was in neither clan 2 or clan 9,^ ^ In Uh the situation is more complex. Here the Nahnmwarki traditionally puts his own son up as next in line for Nahnken, whatever his clan may be. Informants elsewhere in Ponape often say that the incumbent Nahnken is actuaUy demoted when a new Nahnmwarki of Uh steps in, bemg replaced by the latter's son. Study of past successions does not bear this out; ^^ it may be that the reference is to an ancient practice that has not been foUowed in recent years; but it is true that the son of any past or present Nahnmwarki is raised high in the B-line, The last three Nahnkens have successively been members of clans 9, 8, and 5. The present Nahnken ob? tained his office, according to informants, by currying favor with the Japanese and by false accusations against the previous Nahnken; he is a member of clan 5, which never held the office before, and he is not the son of a Nahnmwarki. His younger brother should follow him if the principle of seniority were foUowed, smce he is next 27 This man, who was A14 in 1947, subsequently fell out of favor, Father Costigan says. He is now Bl 1. 28 The foregoing, given in present tense, is from 1947 notes. In 1963 I learned that this prediction, upon death of the old Nahnken, had become fact. Also the new Nahnmwarki, Samuel, is married to a woman of clan 4, so it is possible that that clan may some day provide the Nahnken. 29 Gulick in his vocabulary does say that the Bl is "usually" the son of the Al. Ul line in the clan; but informants say that the B2 wiU probably succeed, since his father was a Nahnmwarki ^? and since the Nahnken's brother holds only a smaU titie of Uh.^^ Previously (p, 16) I referred to the fact that in Net aU but one of the first fifteen titieholders of the A-fine belonged to the royal clan, while only nine of the first 15 B-titled men belonged to one of the two clans regarded as noble. In Uh the contrast is even more pronounced; of the first 12 A-tities, 11 are held by clan 2, the royal clan here. The one exception, A6, is held by Ofiver Nanpei, a wealthy landholder, who is also A6 of Kiti, and who has influence far beyond what his title suggests. Of these 11 tities, eight are held by members of one sub- clan of clan 2, the other three by men of three other sub- clans. But in the B-line, the first 12 titieholders belong to nine different clans. It is especiaUy in the foregoing description of affairs in Uh that it becomes clear that statements by natives that such-and-such a sub-clan constitutes the only legitimate royal or noble class are to be discounted as ideafizations, based perhaps on knowledge of a limited period of history or a nostalgic harking back to things as they are supposed once to have been, and sometimes on personal interests. This is particularly appHcable to the B-line, where shift of power seems to have occurred most often. The earfier remark (on page 18) that a commoner in Kiti can advance only as high as B3 now requires qualification. In Kiti clan 11 has been the B-clan for many generations, having replaced clan 17 more than 100 years ago; the A and B clans having preserved the custom of intermarriage, there was always available to become Bl a clan 11 man with the proper clan 4 (A-clan) father. But the present Al's wife is of clan 7, which in Kiti is a commoner clan; a chUd bom to her now would be Bom Upon the Ditch (see page 18) and nearly everyone (except members of clan 11) says that he would by custom become Bl ahead of any brothers, real or classificatory, of clan 11; thus clan 7 would become noble. The Al and his wife are not likely to have more chUdren, but the A3 who is a potential successor and who is sister's daughter's son to the Al (hence of the same sub-clan, Lipohnroahlong, as the A l ) is young and his wife is stiU having chUdren; the wife is of clan 17, and any son bom after A3 became Al would be Bom Upon the Ditch and hence would outrank other candidates for the titie of Bl, member of clan 17 though he would be, again displacing clan 11 as the noble clan. Therefore the people of clan 11 are pressing hard for A2, who has a clan 11 wife, to succeed 301 also have the conflicting information that his father was a younger twin brother of a Nahnmwarki named Eluit and died when he was A12; but the Nahnmwarki would have been regarded and addressed as father. " The B2 actually did succeed, about 1955. 38 SMITHSONIAN CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 Al , even though A2 belongs to a low ranking sub-clan of clan 4, the Sounkiti, which once ruled Kiti proper but has not ranked high since the Kiti-Pehleng war, when the last Sounkiti Al was kiUed. Succession to the title of Nahnmwarki or Nahnken by a member of an "improper" clan is not a new phenom? enon, even though it causes grave headshaking by the old men. A number of instances are recorded from pre- Spanish times. The A-line of Net has traditionaUy belonged to clan 7 for many generations, but the eighth Al (then bearing the title Lepen Net) before the present one designated his own son, of clan 6, to succeed him.^^ The B-Hne of Sokehs has traditionaUy been in hands of clan 6 since the precontact conquest of Net and Sokehs by clan 7, which became the A-Hne; but the succession of clan 6 members to the title of Bl has been interrupted by pre-Spanish Nahnkens who belong to clans 5 and 13 and who gained office because their fathers were particu? larly powerful Nahnmwarki. Martial exploits in the past often would advance a man out of tum. During the wars with the Spanish the Nahnken Alexander of Madolenihmw fled to Kiti to take refuge. Nicholas, a man of an eHgible sub-clan but of low station because his father was a member of the commoner clan 8, fought bravely to protect the Nahn? mwarki, Paul, and was rewarded with the vacant tide of Nahnken. When Alexander retumed, he was made B2 and had to wait for Nicholas to die before he could become Bl again. There are a number of instances of accession to office by violent means and of the overthrowing of a whole line of titieholders by another Hne. In the semihistorical legend of the conquest of Ponape by invaders from Kusaie, the pre-Nahnmwarki Hne of kings known as Saudeleur was replaced by the new Nahnmwarki Hne; but this was essen? tiaUy the overthrow of the then ruling clan in Madol? enihmw, clan 6, by clan 13. In Uh one sub-clan of clan 2, the Sounpeinkon, formerly furnished the chiefs of the A-line. One day all the men of this Hne went fishing and during their absence another sub-clan of the same clan, the Sounpasedo, held a war feast and laid their strategy. When fishermen retum from fishing they carry their catch to the community house to be distributed and receive in exchange land produce from the other people who have assembled to await their coming. When the Sounpeinkon arrived at the community house they found that the Sounpasedo had taken aU the places of honor on the main platform and had crowned themselves with royal wreaths. The Sounpeinkon were few in number and retired with? out a fight, givhig up aU their tities to their kinsmen, who StUl constitute the A-line of Uh.^^ Another coup d'etat occurred ui Kiti; here the now ex? tinct Inanweies or Liesenpal sub-clan of clan 4 displaced the Liesenpahlap sub-clan of the same clan as the A-Hne m a bloodless war that found the latter fleemg at the sound of the triton tmmpets of the approaching enemy. The aggrandizement of a junior branch of the senior sub- clan of the A clan (clan 13) of Madolenihmw, at the expense of a more senior branch, has already been alluded to, A saying, "The younger stands in front" {mehtik uhmas), refers to the fact that junior sub-clans are some? times in political status superior to that of senior sub-clans of the same clan, and also to cases where a man holds a higher titie than his older brother. War as a means of accession to power is nowadays a thing of the past; it has been replaced by the law courts of the successive foreign administrations and in American times by elections, in certain instances. Other means are sometimes also employed when disputes arise. In 1953 Edmund, the Al of Uh, attempted to resurrect an old title, Soumaka en Mesentakai, and to award that title to the Oun Souna, XI of section 6. The titie is one of his? torical significance and power, and it was Edmund's in? tention to place the Oun Souna in charge of a subdivision of Uh containing about six sections. Various people de? murred on the ground that the title Soumaka en Mesen? takai was one that belonged to or was the equivalent of Nahnmwarki, and that they would thus be serving two Nahnmwarkis. The B2 in particular objected very strenu? ously. At a feast he spoke his mind, was told by the Al to hold his tongue, and stalked out in a temper. On the way to his canoe he told his people to take a particular yam from the pUe of yams outside the community house, claim? ing it as his share, although the food had not yet been divided. The Oun Souna followed indignantly, an alter? cation arose at the shore, and the Oun Souna was stabbed and died. Whereupon the Al deposed the B2 and awarded his title to the Oun Souna's sister's son, who had been B5. A number of high-titled men, especially men of sections 10 and 11, immediately resigned their tities; these in? cluded the B3, B6, A5, and the XI of section 11. The dis? pute reached the law courts. But before a decision was given the people agreed to caU in Moses, the Al of Madol? enihmw, the senior tribe. Since the tune of Isohkelekel, who was the semilegendary first Al of Madolenihmw, and of his son, Lepenien, who was the first Bl of Madol? enUimw and the later the first Al of Uh, the Nahnmwarki of Madolenihmw has had a kind of arbitration role in the 32 The fifth Al before the present one ruled about 1875, which should place the eighth well into pre-Spanish or even precontact rimes. ^ Another version of this affair omits the fishing incident; the Sounpasedo considered the Sounpeinkon too weak to look after state affairs properly, hence seized power. PROMOTION AND SUCCESSION 39 other tribes, but especiaUy in Uh, and can assume author? ity when it becomes obvious that the local Nahnmwarki cannot solve his problems, Moses sounded out pubfic opin? ion in Uh and decided in favor of Edmund and of his appointment of the B2, and the tities of the other dis? sidents were retumed to them. IRREGULARITIES IN POLITICAL ADVANCEMENT Thus, in spite of native theory, methodical progression up the ladder of titles is a principle of great flexibiHty, Clan seniority principles conflict with it. A boy might skip several places simply because he was the last of his sub- clan, and therefore outranked men of junior sub-clans who had achieved higher tities heretofore because they were older. Industry and obedience to the Nahnmwarki and Nahnken might produce differential rates of promo? tion, a sluggard being passed over to give preference to a more active man, MUitary feats formerly were also impor? tant factors. Sometimes a title awarded to a member of a clan that is neither royal nor noble remains in that clan through force of tradition. Thus the B2 of Kiti, as already de? scribed, is always a member of clan 9, instead of clan 11 as are the rest of the holders of B titles, because of certain historical reasons; he is succeeded always by a member of his clan. The present B2 of Net, also mentioned earlier, belongs to clan 2, instead of to the noble clans 6 or 9, He holds the position because of his admirable personal qualities, and the Bl says that he wUl probably be suc? ceeded by his sister's son (who of course is also in clan 2) . Another example is that of the title of A7 of Kiti which, untU its present incumbent received it, had been held for a number of generations by members of clan 11, the B-line in Kiti, The reason for this state of affairs goes back to the series of wars in Kiti already described, when the ml? ing lines were overthrown by dissidents from the section of Pehleng, and Wene, hitherto uidependent of Kiti proper but mled by the same clans, joined forces with her deposed brethren and reconquered Kiti, installing Wene's priest- king, the Soukise, as the Nahnmwarki of the now com? bined tribe. A man of Pehleng who had taken the titie of A7 was kiUed during the battie by Mahsohr, a man of clan 11, For this feat Mahsohr was granted this very title of A7, and it remained in the clan for several generations. There is today some resentment against the present Nahn? ken for having retumed the title to the royal clan, which holds the other A titles, A clan in this position of holding a title properly belonging to another clan comes to think it has proprietary rights to it, especiaUy if the first member to achieve it did so through warlike deeds, and strenuously objects if the title is given to someone else. Analysis of individual careers shows that progress in the hierarchy of tities is far more often a series of skips than a step-by-step rise, the factors previously described being the reasons, Francisco, a former A2 of Sokehs, be? came A2 in one jump from A5. (In Sokehs in those days the A2 was the ruler, and for legendary reasons there was no Nahnmwarki for several generations untU Spanish times,) This was because the A4 belonged to a nonroyal clan, and the A3, though of the proper clan (clan 7), was juruor by blood to Francisco. For simUar reasons a Nahn? ken of Sokehs, Oundoleririn,^* jumped from B7 to Bl. The present Nahnmwarki of Kiti first held the Enipeinpah section titie of Nahlik en Enipein (X4 of that section), then Souan en Enipein (X2), then successively the tribal titles of A9, A3, and Al . The present Nahnken of Kiti has held only three tities in his lifetime, becoming B12 as a smaU chUd, then B4, and finaUy Bl. The present Nahnken of Madolenihmw at the age of ten acquired the titie Souwel, then Nahnsahu, and then Nahnkrou, aU tities of Mesihsou (section 6 of Madolenihmw); as a young man he acquired his first tribal title, B12 of Madolenihmw, after which he was B2, and now for many years Bl. In succession to the Al and Bl tities the ideal, if often not actual, order of preference is: first, the brothers of the previous holder of the title in order of birth; second, the oldest sister's sons in order of birth; third, sons of younger sisters in order of the sisters' births; fourth, male paraUel cousins of the same sub-clan in order of their mothers' births. The succession thus foUows clan seniority. Each candidate should have a father of the proper sub-clan, that sub-clan which furnishes the titieholders of the oppo? site fine. Only after men with proper fathers are exhausted does a member of the sub-clan with a father of commoner clan membership come into consideration, as has been the case of the last four Al's of Madolenihmw. When the sub- clan is exhausted, the next senior sub-clan of the same clan inherits. If the logical successor fives in another tribe, he may be called back to accept the vacancy. During the last years 3< This is properly a titie but, as is common in Ponape, is used as a given name. Sometimes a man wUl be referred to by a title he once held in lieu of using his actual but secret name. A title is frequently given as a name when a child acquired the tide at birth; for example, in 1947 the B8 of Madolenihmw, a boy, had both name and titie of Ouririn. 40 SMITHSONL\ .N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 of Pwas, who was a Nahnken of Sokehs, his sister's son, Soukep, who Hved in Net and previously had had no Sokehs title, came to Sokehs to wait for Pwas to die so that he might take his place, Soukep was given the title of B9 in the interim. But when Pwas died, the Nahnmwarld insisted on another man, SoiUik en Soledi,^' taking his tum first. Soukep then accepted the titie of A4 and be? came Bl only after Soulik had died, Soukep's career iUustrates a number of points. In Heu of a brother, Pwas' sister's son was Pwas' logical successor, since in clan seniority he outranked SouHk en Soledi. But Soulik en Soledi was an older man, he was a Nahnm? warki's son, and he had married a Nahnken's daughter. These factors outweighed mere seniority in the opinion of the then Nahnmwarki, and Soukep was wUling there? fore to defer his promotion. Such factors are stiU more important when members of more than one sub-clan (or more than one clan, as in the B-line of Madolenihmw) are equaUy eHgible to provide a successor. They are subtie factors, which are weighed one against another and make prediction difficult for an outsider, even though he may have recorded aU pertinent genealogies. Deference to a junior is not infrequent. When a number of brothers are Hving, the younger ones may sometimes aUow their sisters' chUdren to go ahead of them; or when two sisters have sons, some of those of the elder may step aside to give their cousins a chance to advance. In a hypothetical case expounded by a Kiti native, if the senior sister had three sons the first would be Al , the second A3, the third perhaps A7, leaving the intervals to be fiUed by sons of the second or third sisters; but in the promotion to Al , A3 would normally succeed his brother. The temporary acceptance of an A-line title by Soukep also reveals how freely future candidates for the position of Al or Bl migrate from one Hne to the other during their early careers. Such lack of restriction to one line is ordinarily characteristic only of sons of a Nahnmwarki or Nahnken, Factors involved in promotion include also judicious choice of a wife. Thus the previous A2 of Kiti owed his position to the fact that he married the sister of the Nahnken, who was his mother's brother's son; he belonged to the A clan (clan 4) and, in fact, to a senior sub-clan thereof, Liesenpahlap, the one that ruled Wene before the conquest of Kiti proper; but this sub-clan was deposed long ago and has sunk in rank, and he would not have risen so high in the hierarchy except for his marriage. The present A2 of Kiti belongs to stiU another sub-clan (Soim- kiti) of clan 4 and advanced this high only because he married the daughter of the earfier A2 and the Nahnken's sister. (See also the section "Feminine Titles," p. 47.) Promotion through martial exploits and through service for and tribute to the chiefs is discussed elsewhere. PROMOTION CEREMONIAL The duty of making promotions belongs to the Nahn? ken, who, however, is supposed to consult with the Nahn? mwarki. The Nahnken also does the actual conferring of a tribal title {ta mwar or pwuk mwar, to raise up to a titie). First he makes a short speech recounting the good quafities of the man about to be promoted. Then he per? forms the Kava Holding ritual {sapwsakau), that is, he takes a cup of kava in both hands and raises it high before him, and he says loudly, "This is the coconut sheU vessel of ?" {Met ngarangar en ) . The title rep? resented by the dash in this formula is conceived to be in the vessel. Then the man being promoted comes up to the main platform of the community house where the Nahnken is standing; if he is a member of the A-Hne, he wUl have been in the central ground level area of the commimity house, as most members of this line should be. With him comes as his sponsor the senior member of his line, next below the Nahnmwarki or Nahnken, that is, the A2 or B2; but if these are not present, the next in seniority takes his place; or it may be his sub-clan chief or his mother's eldest brother. The sponsor, if the man to be promoted is not a member of the royal or noble clans, is most Hkely to be his sub-clan chief. Sometimes he comes up alone and takes the cup, then he calls the sub-clan chief to come and drink it. The function of the sponsor is to make a lapweiepwei; ^^ that is, he takes the vessel of kava from the Nahnlcen, turns toward the assembled people so that the Nahnmwarki and Nahnken are behind him, says "This is our cup," and takes a draught; then the newly promoted man and his sponsor retum to theu- proper places. Frequentiy, before he drinks, he wUl caU out the names of all the distinguished members of the sub-clan who are present, OccasionaUy the candidate for the title himself drmks. Rarely the Nahnmwarki functions instead of the Nahnken, on a special occasion such as when he is elevatuig his own son. 35 This again is a title used as a name. '* Apparentiy some religious ceremonial was once connected with this performance, since the term lapweiepwei is also applied to a prayer made to a spirit by a soothsayer. PROMOTION AND SUCCESSION 41 At the coronation of a Nahnmwarki, the Nahnken places a wreath of Ixora carolinensis (an elin katieu) on the new ruler's head. (This royal wreath is also caUed nihn or ninnin. A wreath wom by a commoner is caUed mware.) SimUarly, the Nahnmwarki crowns the Nahnken when the latter assumes office. The A2 is supposed to crovm the B2 and vice versa; but their crown is elin seir, made of the blossoms of the tree of this name {Fagraea sair; caUed pwur in Kiti). The crowning is foUowed by the drinking of the vessel of kava. Men promoted to titles lower than A2 and B2 are not supposed to be crowned but only to drink the kava, but nowadays it is done for men of much lower titie. No one except the Al and Bl is sup? posed to wear a wreath of Ixora, unless the blossoms are mixed with other flowers, ^^ The Nahnmwarki of Madolenihmw was formerly crowned in secret at night at a place caUed Peipwel, in section 26 of Madolenihmw, Lepen Wapar, the XI , was custodian of the place. The crowning foUowed immedi? ately upon the death of the old Nahnmwarki, which was likewise kept secret for fear of anarchy and civU war. Peipwel was a stone structure where the Nahnken and seven other men of title awaited the Nahnmwarki-to-be, who was led thither by the Nahntu hi a torchfight proces? sion. (By another account the Nahnken and his party, who came by canoe, were met at the shore of section 26 by Nahntu, who vrith the Kirounaip led the procession to Peipwel. The Kirounaip ["Steward of the dmm"] had particular duties connected with the drums that were part of the arrangements.) After the wreath of flowers was placed on the Nahnmwarki's head (or after he was struck by the Nahnken with a turmeric plant, according to Luelen) the Nahnken took in his hands a stone (a short basalt prism, says Luelen) upon which two sedei had been laid. (A sedei is a coconut pinna whose ends are folded together in a pecuHar manner to form a crown and which is wom on the head for various ritual purposes. See Plate 5.) He then thmst the stone toward the Nahnmwarki's face, as though he were about to strike him, whUe he ut? tered a phrase meaning roughly, "Useful and long reign." If the Nahnmwarki fluiched, it was a sign that he would soon die. Then all went to a place where a certain stone stands; aU the people were called to assemble and prepare a feast in the community house there, and when the new Nahnmwarki took his place on the main platform, they would know for the first time that the old Nahnmwarki was dead and see who had been chosen to be their new ruler. The secrecy involved and the dmm beating (at least two sizes of drums hi Kiti) applied to aU the tribes. The foregoing description was obtauied largely from modem informants. The missionary L, H, GuHck, writing " Hambruch speaks of the Nahnmwarki giving his crown to someone else as a method of conferring honor upon him, but my informants had never heard of this practice. in 1855, describes simUar ceremonies that, how? ever, are different in some details. The Nahrmiwarki of Madolenihmw, Luhkenkidu, who in 1836 succeeded LuhkenkasUi of Falcon fame, died in September 1854 during a smaUpox epidemic and was in tum succeeded by Luhkenmweiu. Then on June 5, 1855, Luhkenmweiu died. He was buried immediately, as was the custom, enveloped in mats and together with all his belongings. Then guns were fired to armounce the death to the people. The A2 now became Al , but for some reason the usual immediate instaUation did not take place. On June 7 the new Al with six of the highest chiefs and four priests went in canoes to a place in the artificial islands of Nan Madol (which, from the description given, is evidently Pahnkedira). Here a shed had been erected and in it two priests pounded kava and brought out cups to the chiefs who sat in a line in order of rank. Prayers were said over the chiefs, sips of kava were drunk, miniature cups were stuck on reeds before a certain stone, and circles and fines were drawn on the Al's hand. GuHck terms this the reHgious inauguration. Then the whole party retumed to the community house at the capital and all the chiefs were confirmed in their new offices by further cups of kava over which the different titles had been pronounced. In referring to this last as the political inauguration, Gufick says that more often it came ahead of the reHgious one. Then on June 9 the party visited another place that is evidentiy the aforementioned Peipwel. Here the new Al sat before an altar consisting of a pUe of stones, with pecufiarly folded coconut leaves around his neck and arms and in his ear lobes, and the other chiefs sat some distance from him. The Bl, serving as both chief and priest, said incantations over a cup of kava, gave it to the Al , who drank, then threw the cup under a stone of the altar. Nowadays, the ceremonial is much simplified, and the ritual frequently consists only of handing the recipient of the new titie a bowl of kava or a piece of sugar cane. On one occasion that the writer witnessed, which was at a funeral feast, the Nahnmwarki of Kiti only delegated a man, apparently arbitrarily chosen for this one occasion, to stand, announce the successor to the dead man's titie, and counsel everyone to respect him as such; the new appointee, though present, made no acknowledgment. AU tribal titles except those of Nahnmwarki and Nahn? ken must be confirmed by a titie-payment feast {irara- mwar or kapasmwar) held in the community house and given for the Nahnmwarki. At 1947 values this was esti? mated by natives to cost, for nearly aU tities, something like $200 worth of pigs, yams, kava, etc. Not all the cost is borne by the man who is being promoted, since his famUy contributes a large part to it. Vacancies in section tities are fiUed by the XI of the section, but the XI hunself, though often chosen by the people of the section, must be confirmed in office by the 42 S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 Nahnmwarki or Nahnken, depending on which is over? lord of the section. Succession is not automatic; the Nahnmwarki or Nahnken usuaUy agrees with the wishes of the people, but sometimes he may consider that the candidate is unsuitable and reject him, then a second choice is necessary. He might then choose another mem? ber of the same sub-clan, or the dead man's son, or some? one else. The title-payment feast is then made by the chosen man to the Nahnmwarki or Nahnken. After the feast is made, the title carmot be taken away without cause, but if the promotion is not confirmed by a feast, the title is revoked. Lower titieholders of the section make their title-pay? ment feast to the XI , who alone makes the appointments, and the tribal chiefs do not attend. SimUarly the XI performs the office of instaUation for the head of a farm? stead; this practice continues today, even though the actual inheritance is according to the terms of the German deeds and the confirmation by the XI has therefore become purely formal. Elsewhere than Net each man who is being promoted makes a separate feast, and there are no standards for his lavishness. In Net the present Nahnmwarki and Nahnken, with their customary acumen, have systematized promo? tion feasts by having them aU held in December and January of each year, so that a number of titles are simultaneously paid for; the payments also have been standardized, so that each man is required to pay a pig, a large kava plant, and one or two 2-man yams. The sime pair of chiefs has simUarly consoHdated the honor feasts (q.v.) given by the various sections into a single feast. On the same day that an XI dies the appropriate chief, Nahnmwarki or Nahnken, is supposed to come and bury him and award the title and, untU German times, the land to his successor. The section, which was held in fief under the Nahnmwarki or Nahnken or indirectiy under them through intermediary chiefs, theoretically reverted upon the death of the XI to be reissued to his successor. In? formants almost always say it reverted to the Nahnmwarki; when pressed for details they usuaUy give, instead, the title of some chief who held the land under the Nahn? mwarki or Nahnken, The chief of any area is stUl expected to attend the burial of his tenants; on one occasion the writer was forced to postpone a trip with A6 of Puti because one of the A6's tenants had died and, as he put it, his people would have become "discouraged" had he not gone to the funeral and funeral feast. In former days all work would cease as soon as it was known that a section chief or a Nahnmwarki had died, and a state of lawlessness would prevaU. To avoid anarchy and outbreaks of violence the dead chief was immediately buried in secret and, ordinarUy, the people would not leam of what had happened until his successor was in? ducted. When the Al of Madolenihmw died in 1854, Gufick reports his immediate train continued to talk as though he yet lived. Gulick, in describing the funeral of another high chief m 1853, states that from the moment of death untU the grave was closed?during which time the corpse was rubbed with oil, wrapped in mats, dressed in his finery, the grave dug, and the chief's woven belts, bead ornaments, adzes and other belongings placed in the grave with hun?a total of 25 minutes elapsed. Hundreds of people joined for several hours in the howling, shrieking, and contortions that foUowed, and formal waifing in set phrases by relatives continued for weeks, the mourners aU with shaved heads. (The standardized mourning phrases are stUl heard nowadays, but most of the other customs have vanished, including the hastiness of burial, which has been superseded by Christian practices.) The mis? sionary Sturges, in 1856, describes in simUar terms the funeral of a high-ranking woman, the aunt of the Bl of Kiti, and states that feasting aU day and dancing all night foUowed for a week at Ronkiti, where the Bl Hved, and then for another week at Wene, where the Al Hved. AU of these early reports, including others by the missionary Doane, refer to lawlessness and to general destruction of coconut trees, yams, and dogs at the place where the dead chief Hved and to a lesser degree elsewhere. His lands and other property would be taken and divided up among the other chiefs, sometimes in an orderly way, but often a msh was made by aU to seize what they could. This was tme after a commoner died too; all his accumulated baUs of twine, his mats, sometimes even his house, would be ap? propriated, and the widow turned out. My own informants give simUar descriptions and add that for a tune after? ward many people would feign madness, run about wUdly besmearing themselves with mud and throwing dung at each other, lose their normal inhibitions in dress and speech, and sometimes engage in sexual Hcense. It was to avoid these excesses that the attempt at secrecy was made and that the new Al was immediately installed in oflfice; ideally the general populace was not to know any? thing until they were summoned to the feast in the com? munity house and saw the new incumbent in the place of honor on the main platform. Other Titles PRIESTLY TITLES Although Hambruch states that the now extinct priests {samworou) were Royal Children, this seems to have been tme only of the higher ones. The principal priest was the Nahlaimw who today is the B2 chief. In Madol? enihmw he could be a member of either clan 2 or clan 9, the tvv^ o clans that constitute the Royal ChUdren in that tribe; but whichever of these two clans the Nahnken belonged to, the Nahlaimw had to belong to the other, as is also tme today. The second highest priest, the Nahnapas (B4 chief today), however, had to be of the same clan as the Nahnken. The third priest, Soulik en Madolenihmw, could likewise belong to either of the two clans, but the lesser priests might belong to any clan; in Madoleruhmw they were Oaron Maka, Serihnei en Patele, Seioar en WUek, Ninakap en Temwen, SihpwU Lihken, Kroun en Mwalok, Lepen Souleng. One addi? tional lesser priest, Soused en Roti, a title that plays a significant role in the Isohkelekel story and which persists today, however, has always been a member of clan 2. There are two simUar Hsts of priests, presumably appli? cable to ICiti, from natively written manuscripts. One by Luelen gives the Nahlaimw (the present-day B2) and the Nahnapas (B4) as the first two priests, as in Madol? enihmw. SouHk en Kiti is given as the fourth priest, whereas in the previous paragraph I have given SouHk en Madolenihmw as third. Soulik in the various columns of Table 4 is today B8 to B14. The third priest was Nahnkei (B6-B11 in Table 4 ) . Then foUowed Nahnsaumw (BIO- 12), Soulikin Sapawas (B15), Nahnawa Iso (A18), and Nahnsou Wehi (A20). The other manuscript, by a man named Silten, probably written about 1935, lists Ounsona third, Nahnkei fourth, then follow SouHk, Oun (possibly my Ou, B8-13), Nahnsaumw, and Soulikin Sapawas. The history of these priestly titles is not clear. Several informants say they formed a separate line in former times, a statement the two native manuscripts would support, though also known as Royal ChUdren; they say the priestly tities were fitted into the present-day B-Hne by Paul, Nahnmwarki of Madolenihmw in Spanish times. Accord? ing to these informants the other tribes then copied Paul's scheme. Presumably this was because priestly duties had become defunct after Christianization. Whether or not this is true, it is certain that the Nahlaimw succeeded the Nahnken in office, so whether there was one line or two lines of Royal ChUdren, they functioned much as the B- line does today. A number of tities apparentiy antedate the titles of the present-day A and B series. Most of these do not bear the prefix "nahn," in contrast with the majority of modern titles. The native explanation most commonly encoun? tered is that the Nahnmwarki-Nahnken series began only with the semihistorical conquest of the island by Isohkel? ekel; before that, Ponape was unified under a single, tyrannical ruler whose title was Saudeleur. If a ranked order of titles existed under the Saudeleur it is no longer recaUed, but many of the titles of that era persist outside the present ranked A and B series and are accorded high honors. Such chiefs as Lepen Moar, Lepen Deleur, Lepen Palikir, Soulik en Awak, and Nahmadau en Pehleng, for example, receive the third or fourth share of food at a feast. Some of them nowadays are only section chiefs, such as Lepen Palikir and Soulik en Awak, but it is apparent that their influence was once much greater; Palikir, for example, was always a semiautonomous area, only nomi? nally part of the tribe of Sokehs, and its ruler, the Lepen Palikir, frequently acted entirely independently. The Sou? Hk en Awak was likewise largely independent of the rest of Uh, once joining with his CathoHc co-refigionists in 43 44 SMITHSONL\N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 Net and Sokehs in a war agamst the Protestant area of the remauider of Uh, Madolenihmw, and Kiti. This event was in Spanish times but is indicative of a pre-Spanish cleavage. The Nahmadau en Pehleng, now only a section head, formerly had under him two sections besides his own; the story of how a holder of that title overthrew the Nahnmwarki of Kiti and for a period ruled in his stead, until Wene reconquered Kiti, has already been related. Some of these earfier titles seem to have combined many priestly offices with their chiefly duties; some of them untU relatively recently seem to have been held by priests who had temporal powers as weU. Such was the Saum en Long, a high priestly titie in Kiti, who mled section 21. The ruler of Wene was Soukise, who was similarly a priest- king. These priests were apparentiy of a different category from those in the later "nahn" series such as B2 and B4, who seem to have had no temporal functions untU quite recentiy. Several refigious cults devoted to worship of different gods seem to have existed in various parts of Ponape, and Wene was apparently the center of this development. Hambmch (1932 II, pp. 130 f.) discusses these cults and theu: priests rather briefly; already in his time memory of them was fragmentary. Each cult seems to have had its separate body of priests, with their own sets of tities, apart from those already discussed. Also a major division into two levels of priesthood cut across the cults, the upper level known by the general term for priest, samworou, the lesser priests caUed laiap or leiap. But native information about them is today so scanty as to make it unprofitable to discuss further their connection with the pofitical organization. TITLES OF ADDRESS AND TITLES OF REFERENCE The holders of most of the older tities, i.e., the ones antedating the "nahn" series, have a special title of address, pwoud or pwoudo. These include Lepen Net, Lepen Pafikir, Lepen Enimwahn, Lepen Moar, Saum en Long, Kroun en Lehdau, Soulik en Awak, Lepen Wenik, Nahmadau en Pehleng, Kirou Meir, Lepen Deleur en Kiti, and Kelahk en Takaiu. GuHck (1872, p, 39) gives the term pwoudo (which he speUs pauto) as another titie of the Nahnken, but it is not used as such today. It is the same word as the word for spouse, in reference to the close bond between chief and commoner. The Nahnmwcuki also has special titles used in address. The Nahnmwarki of Madolenihmw is addressed as Isipau ?^ and has the additional titles of reference Soumaka en Ololap, Lepen Seipel, an a third, secret one. The Nahnmwarki of Kiti has Rohsa, Soukise, and Soumaka en Long as additional tities; the Rohsa title has been awarded to the present A6, and Soumaka en Long to the former A12, but they are properly the Nahnmwarki's; Soukise was the ruler of Wene, and when one of the holders of this title ousted the Nahmadau en Pehleng from his brief occupation of Kiti proper and united all of modem Kiti under his rule, he retained his old titie while assuming that of Nahnmwarki as well. In Uh, the Nahnmwarki bears the additional title of address of Sangiro,^^ In Net, the present Nahnmwarki 38 In Gulick's vocabulary this title seems to be applied to the Nahnmwarkis of all tribes, not just Madolenihmw; he defines it as "the highest OTowja/) title in a tribe (equivalent to narzamara/:? , . .) ," not making a distinction between address and reference. It is sometimes used nowadays for the Al of Uh too. It may be derived from the name of the god Isopau. 3? The word may be related to the name of the god Sahngoro. (who has replaced the former ruler Lepen Net) has three additional titles.*" The highest chief of Sokehs was for a long period the Wasai, who is chief A2 in the other tribes, and his title of address was Nahnpwutak, "Lord Boy," which is that of every Wasai.*^ He also seems to have been addressed as Isoeni, "Sir Ghost," but a particular ruler of Sokehs who Hved about 1870 is usuaUy meant when that term is used. AU Nahnmwarki are likewise known as Wasa Lapalap,*^ "Important Place," though Luelen and some informants apply the term only to the Nahnmwarki of Madolenihmw; each Nahnmwarki is also known as Mwohnsapwaka, a form of "First of the Land," an expression already discussed. Each of these terms has its special use. In Madol? enihmw, Wasa Lapalap is held by some to be preferable to Isipau as a form of address for the Nahnmwarki. In Uh, Sangiro is preferred. Isoeni was used in place of Nahnpwutak during a period of mouming for the dead A2 of Sokehs. In direct address to a Nahnmwarki, Wasa Lapalap is used to attract his attention, Mwohnsapwaka *<> One of these tities, Soulik en Daun, listed on p. 19 as having been that of the A2 of Net until German rimes when the Nahn? mwarki-Nahnken series of titles was adopted, has more recentiy been given by Max, the Al, to Benito, who was Max's rival for the titie of Al and who had lost a lawsuit that he brought in the American courts in an attempt to gain it. In 1947 the title Soulik en Daun belonged to the Bl. " T h e story goes that Sokehs once had a Nahnmwarki, like every other tribe, but a series of misfortunes to these rulers occurred and finally the Wasai refused to accept the final promotion due him but ruled in the name of the second highest A title. *2 Gulick gives it as Oj lapalap, which would mean great or important "growth," not "place." OTHER TITLES 45 during speech with him. Soukise is used ui referring to the Nahnmwarki of Kiti during the distribution of food at a feast, but ui the part of Kiti known as Wene where the Soukise titie had its origin, it is used on other occasions as weU. In referring to the belongmgs of the Nahnmwarki of Kiti, as when one observes his canoes passing by or asks where his house is or refers to his wife, the proper term is Rohsa; this is also used in address when his B-Hne relatives are present. Corresponduig to the honorific titie of address Nahnpwutak for A2, the address forms for A3 and A4 are Nahniau and Nahimo respectively. In Net, the form Iso is used in addressuig A5 and A6, according to one informant; A5-A10, according to another. Madolenihmw informants apply this term to A5-A14." The A l l and the Soumadau (a low A-title) are according to different informants addressed as Sahu in Net, as is the Souruko in Madolenihmw. B1-B4 in most tribes are caUed Iso, but one Net informant gives B3 as Sahu and B5 as Luhk. There are apparentiy no such special forms of address for women. The titles of address are supposed to be courteous and deferential forms, the proper tities (Tables 3 and 4) are theoreticaUy reserved for reference only. But whatever may have been the practice in the past, nowadays they are frequently used interchangeably in direct address. HONORIFIC FORMS A large vocabulary of honorific forms serves to emphasize the differences in rank between the various classes on Ponape. Most of the forms are used to distin? guish between commoners on the one hand and the royalty and nobiHty on the other, or between the two classes of chiefs, or between the higher and lower chiefs in the two lines. The subject of honorifics requires a Hnguistic treatment for which this is not the place.** A number of terms and usages connected with ceremony, however, deserve some description here. When the Nahnmwarki is concemed food and aU appurtenances connected with food are spoken of as koanoat. In the food distribution at feasts the Nahn? mwarki's share is formaUy announced as koanoat. The equivalent term for the Nahnken is sahk. These two terms are appHed to the persons themselves as weU as to the objects involved; that is to say, the Nahnmwarki himself is said to be koanoat, the Nahnken sahk. Everything else belonging or pertaining to both Nahnmwarki and Nahn? ken?clothes, houses, canoes, etc.? i^s sapwellem. When the Nahnmwarki is present at a feast in the com? munity house, the food of all other chiefs is formaUy an? nounced as kepin koanoat, except that of the Nahnken, which remains sahk. (An equivalent term to kepin koan? oat is tungoal, and the titled men who receive kepin koan? oat shares are referred to as being tungoal, although not announced as such.) But if the Nahnmwarki is absent and the A2 has taken his place, the latter is no longer kepin koanoat but instead becomes sahk, Hke the Nahnken.*^ <3 Father Costigan teUs me he has heard the term Iso applied only to A5, A7, and A8 in Madolenihmw. ** For a fuller treatment see Garvin and Riesenberg, 1952. " H a m b r u c h , following Hahl, is in error in making the A2 koanoat; he is sahk. SunUarly, if the A4 is the senior chief of the A-line present at the feast, he likewise becomes sahk instead of kepin koanoat. But if the A3 is the senior chief present, he is koanoat, Hke the Nahnmwarki. No satisfactory explana? tion could be obtained as to why only Al and A3 in the two ranked series of tities should be koanoat. A5 and holders of lower tities in the A-line remain kepin koanoat and do not become sahk Hke A2 and A4, regard? less of whether they are the senior A-luie chiefs present. If the Nahnmwarki is absent but the Nahnken is pres? ent, aU other men of titie become kepin sahk, except for the A3, who is kepin koanoat. If the Nahnken is absent, the B2 takes his place and is sahk and lower tities in both lines are kepin sahk; but if the B3 or the holder of a lower B-titie is the senior B chief, he remains kepin sahk. A few informants, however, extend the use of the term sahk as far as A12 and B12. Wives of men who are koa? noat are called pweniu, but wives of men who are sahk are themselves sahk. When titieholders below A4 and B2 are at home and no titieholders senior to them are present, or at a section meeting under the same conditions, they may be caUed sahk. Between commoners, in the absence of chiefs and ceremony, the term kahng is used for food. Besides the Nahnmwarki and the A3, certain titie? holders outside of the two series are also koanoat. Most of these tities are ancient priestiy ones, though other ancient titles are sahk. In Net these include, among others, Sou? ruko, SouHk en Daun, Sou Kiti, SouHk en Popat, Souedi, and Lepen Net; *^ ui Sokehs they are the Lepen Palikir, Soulik en Pafikir,*^ SouHk en Soledi, and Sapadan; in Uh, *" A Kiti informant says Lepen Net is sahk. *'' A Kiti informant and a Sokehs informant state that Lepen Palikir (XI of Palikir) is koanoat only because he simultaneously holds the title Soulik en Palikir, otherwise he is sahk. 46 SMITHSONLVN CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 Souruko, Souedi, and Sonhk en Awak; in MadolenUimw, Lepen Moar, Souedi, and Kroun en Ledau; ** in Kiti, Aufik, Saum, Soumaka, Soumko, Sau Wene, Nahmadau en Pehleng,*^ Soukise, Sou Kiti, and Lepen Deleur. Un? Hke the A3, many of these are koanoat regardless of the presence or absence of the Nahnmwarki, Many of them are entitled to places of honor in the community house and to other manifestations of deference. Some of them, as Soulik en Daun and Lepen Moar, have the privUege of sitting between the Al and the Bl on the maui plat? form, facing the people, in contrast to others who must sit to the left of the Bl ; they are third ui food distribution, after the Al and Bl, and they are entitied to be addressed in the highest forms of honorific language. The Lepen Moar, who for certain legendary reasons is "the crown of the Nahnmwarki," may lean his back against the Those Who Face Downwards post (the rearmost central post of the main platform of the community house), he may leave the feast without making his excuses as ordinary men must do, the kava pounding rhythm called sokemwahu (see the section on kava ritual) is different when done for him, and he has the privilege of conferring certain tities that pertain to the old area of Senipehn without reference to the Al or Bl, and the further privUege of receiving the titie-payment feasts for them. The Nahnmwarki is in aU five tribes considered koanoat in food distribution, but some informants say this is only because of the additional, older titles he possesses; thus the Nahnmwarki title in Kiti is said to be sahk in itself, but its holder is koanoat because of the additional titles of Soukise and Rohsa that belong to him; and since the present chief A6 of Kiti has been given the title Rohsa, he too is considered to be koanoat and is third in food distri? bution at feasts, foUowmg Al and Bl. SimUarly, in Madol? enihmw the Nahnmwarki is said by some to be koanoat " According to the Bl of Madolenihmw, the Kroun en Ledau is only sahk. Also, the same informant said that Moses, the t h e n A l , had only recently elevated two more titles to koanoat, Lepen Sehd and Soueda, and that in Madolenihmw only A l , A3, and Lepen Moar are properly koanoat. Hahl (1901, p. 7) gives also Saun and Aulik as koanoat in Madolenihmw. " The Nahmadau en Pehleng, who is XI of section 35 of Kiti, is simultaneously B2 of Kiti. This is cilways true, whoever is the occupant of these offices, and is traceable to an event in the time of the Kiti wars described elsewhere. The B2 title is sahk every? where, but in Kiti the XI title of Nahmadau en Pehleng that the B2 holds makes him koanoat. Similarly with the wife of this man; as wife of the Nahmadau en Pehleng her ritle is Nahnkedin en Pehleng and she is pwenieu, as are wives of all men who are koanoat, although as wife of the B2 her title is simultaneously Nahnkulei and she would ordinarily be sahk, as the wife of the B2 is elsewhere than in Kiti. only because of his older title of Isipau; in Uh because of his title of Sangiro. Men entitled to koanoat used to receive large shares in the food distribution at feasts. As an indication of the position to which some of the old titles not in the two ranked series have sunk, the present Soukoahng of Net, whose titie gives him the right of koanoat, refused, at the time the titie was offered him, to accept it unless he was considered as only sahk; he felt that he would be subject to ridicule if, when his smaU share of food was announced Ul pubfic assemblage, the presentation was accompanied by the pretentious term koanoat prefixed to his title. The titles are simUarly ranged according to the honor? ifics used for the verb "to come." In Net, the tities Al to A7 and Bl to B5 are katido; chiefs from A8 and from B6 down to an indeterminate level are apehdo; the com? moners are kohdo. In Madolenihmw, Al to A8 and Bl to B8 are said to be katido and lower tities are apehdo. The references do not change when the higher chiefs are absent, as do the words for food.^? The ancient priestly titles just mentioned, which are koanoat and which do not become kepin koanoat or sahk under the circumstances described, are also katido. A recent practice that began in the early 1960s in Madol? enihmw and is being taken up in the other tribes is to give the senior man of each clan and of some sub-clans a koanoat title. Sometimes these are ancient titles that have been regarded as belonging to those clans and now are dignified by being referred to as koanoat at feasts. These titles often do not include the word Madolenihmw in them but rather the ancient name of that tribe, Sounah- leng. The rationale is that the clan has given much service to the tribe over the years and the Nahnmwarki wants to demonstrate his pleasure and gratitude by this means. These men are also regarded as First of the Land (see p. 17). If they are older men they are sometimes led up to sit on the main platform facing the people. But they are koanoat only if the Nahnmwarki is not present, otherwise they are kepin koanoat; and they are not katido, nor is the highest honorific speech used with them. " H a h l (1901, pp. 6-7), in discussing Madolenihmw, shows A1-A3 and B1-B2 as katido, A4-A11 and B3-B14 as apehdo. (My reference to titles by letter and number in this instance follows Hahl 's list, as given in Tables 3 and 4, column Hah) . Also Saun and Avilik are shown as katido, and a whole group of other titles as apehdo, some of them belonging in the A and B series as given by my informants. Hahl adds: "Den Konigsgeschlechtern gegen- iiber heisst 'kommen' kotito, dem Adel gegeniiber apeto, beim gemeinen Volke koto. Die Mitglieder der koniglichen Familie werden in der Mehrzahl und mit 'Hoheit ' (koten) angeredet. Die Eintheilung der Wiirden geschieht geradezu nach diesen Sprachregeln, indem man von einem Manne sagt, er is koten, kotito, apeto u.s.f." OTHER TITLES FEMININE TITLES 47 Each title in the chiefly series has a feminine counter? part, which the wife of a titleholder automatically as? sumes. When her husband is promoted to a new titie, she abandons her old title and takes the title appropriate to her husband's new position, though without the ceremony or promotion feast incumbent upon him. In the days of polygamy a man held a title for each wife he took and each wife had the feminine counterpart of it; indeed, a few informants questioned whether new titles were taken for new wives or whether a man took as many wives as he already held tities for. The feminine series in Net that parallels the masculine series of Table 3, column B2NI and Table 4, column BIN is as foUows: 1. Nahnalek 2. Nahimep 3. Nahnte 4. Nahnado 5. Nahleo 6. Nahnpweipei 7. Nahlikirou 8. Nahlikiei 9. Nahnihdpei 10. Pweipei Lapalap 11. Kedindel 12. Nahntupei 13. Likendlap 14. Kedinkapw 15. Nahnkar 1. Nahnkeniei 2. Nahnkiilei 3. Nahlisau 4. Nahnapasepei 5. Nahnkedin Idehd 6. Eminalau Lapalap 7. Lempein Ririn 8. Li ou Ririn 9. Kanep 10. Kadipwan 11. Li Oimpei 12. Li Oundol 13. Li Ou 14. Kedpwan 15. Nahnkupei Titles elsewhere vary sHghtiy, apart from dialect differ? ence: in Kiti, for example, Bl 1 becomes Li Oun Pohnpei and B9 is Kanepein Ririn. In Uh and Madolenihmw, Al is Likend, In addition to the title a woman receives because of her marriage, there are in three tribes two tities outside the foregoing series that are reserved for daughters and sisters of a Nahnmwarki or Nahnken. These are Nahn- kakas and Idingel in Kiti and Uh, Isohlap and Idingel in Madolenihmw, The last Nahnmwarki of Madolen? Uimw, Alexander, was brother to an Isohlap and son of an Idingel; the same Isohlap ^^ is mother of the present Nahnmwarki and her daughter is now Idingel; another daughter wUl take the Isohlap title when the mother dies. In Madolenihmw the rule as stated by informants is that only sisters, not daughters, of a Nahnmwarki can have these titles, and that a sister or daughter of a Nahnken can have one only after marriage. But in Uh it is said that only one daughter of the Nahnmwarki and no sister *i According to Father Costigan, this woman was given the title Kedindeleur when she approached death. This title was buried with her and may not be used again. The reason was that the Madolenihmw people did not want to bury so important a title as Isohlap, which was given to one of her daughters. may have one of these titles, and only the eldest daughter or eldest sister of the Nahnken may have the other. Ac? tuaUy, the Nahnkakas of Uh today is the eldest daughter of the Nahnken, but the Idingel title belongs to the daugh? ter of the B2, himself a son of a former Nahnmwarki. The previous Nahnmwarki issued the Nahnkakas title to his adopted daughter. These titles are given to women of proper descent, not because of marriage. In addition, a number of women have masculine tities of the A and B series. This is said to be irregular and is looked upon with disfavor, but it does not seem to be a practice very recent in origin. Daughters of a Nahnmwarki or Nahnken may acquire such titles in infancy. In Madolenihmw these tities may be, among others, B9, BIO, and Ou. (Column BIMI of Table 4, which we have chosen to use for purposes of reference for Madolenihmw titles, does not list the title Ou, but Column BIMII Hsts all three tities immediately adjacent to one another.) Another example is provided by the present A5 of Kiti, who is daughter to the present Bl. If the female holder of such a title marries a man of lower title, he takes her title and she takes the feminine counterpart thereof; if she dies, he reverts to his former titie, urUess it has been reissued, when he receives another of about the same level. If she marries a man of higher titie, she gives up her old title and takes the female counter? part of his; then if he dies the next man in line gets his title and she goes back to her old title, ^Vhen, in this manner, the husband tcikes his wife's title, or when she has not held any title but he is awarded a high title upon marriage, he is said to hold the title "for" her. Thus, the present A13 of Net belongs to neither of the clans that form the A or B fines but is a member of clan 12, and his proper title is a section one; but he is married to the daughter of a former Nahnmwarki, and he holds his tribal title because of her. If she died he woiUd lose his title.^^ The wife herself did not hold the title before her marriage. The "sister" of the Nahnken of Uh is mar? ried to the B5, who is a member of clan 15 (a commoner clan); in this case she held the title before she married him. (In Uh this title B5 is subject to a pecuHar provision: it is said to "go back" on the death of its holder to the Nahnmwarki and Nahnken to be reissued as they please without reference to seniority or clan membership, in con? trast to the other titles.) The present B8 of Uh is the un? married young daughter of the Nahnmwarki; when she grows up she must have a higher title (informants empha? size the word "must"), which her husband wiU hold for her, because of her father's rank. A previous Nahn- Uh informants deny that such loss of title is customary. 48 SMITHSONIAN CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 mwarki's daughter's husband was first B l l of Uh, later A8; he got the first titie from his wife upon marriage to her; the second he earned, but by native theory it was she who reaUy got the promotion, giving it then to her husband. A man who elevate himself in this way by marriage to a high-ranking woman is called a "Big Bones" (tihlap), because he must work harder than other men and make a better showing at feasts; formerly he had to be in the forefront in battie. Other influences are sometimes involved in achieve? ment of tities by women. The BIO of Net today is a wom? an and is said to have been given the titie because her former paramour, the Bl , wanted her to be in a position to have a good share in the distribution of food at feasts. Machinery of Government T H E POSITION OF T H E NAHNMWARKI AND N A H N K E N The existence of two supreme chiefs in each tribe pro? vides a mechanism of govemment that is always in a rather delicate state of balance. The Nahnmwarki, in virtue of his semisacred person, was formerly considerably removed from the general popidace, who came more closely in contact with the Nahnken. The Nahnken made most of the practical decisions and the promotions, though theoreticaUy in consultation with the Nahnmwarki. In many respects, as Bascom points out, his position com? pares with that of the talking chief in Samoa. He was looked upon as a sort of champion of the people and inter? vened for them in the face of the capricious temper of the Nahnmwarki. Consequently, he was in a position to be? come the real master, and some incumbents of the office seem actuaUy to have done so; thus Nahnku, the Nahnken of Kiti from about 1850 to 1864, was the real mler of Kiti and completely ecHpsed the Nahnmwarki, But Nahnku was an exceptionally powerful individual, as many anecdotes about his career show; the missionary Gulick writes in admiring terms about his impressive fig? ure and strong personafity. In theory, at least, the Nahn? mwarki had and in practice nowadays has more power, and in any difference of opinion the decision was supposed always to go his way. Sometimes the two chiefs reverse their traditional roles and it is the Nahnmwarki who ap? proaches the Nahnken to intercede with him on behalf of the people. That open discord seldom arose attests to the stabUity of the state of balance between the two chiefs. Occasional quarrels might arise; at one tune a dispute between Paul? ino, Bl of Net, and the then Nahnmwarki nearly spHt the tribe asunder. But Paulino after 3 days made a feast of apology. He was considered extremely rash to have waited so long, for after 3 days without an apology a war between the A and B clans is supposed to ensue. Another time a ruler of Sokehs, who is remembered only by the honorific form Isoeni, in a coup d'etat seized power in Net and became head of the A-line in both tribes; the Nahnsoused (at that time the titie of the Bl of Net) sulked for a period, but his ultimate reconcUiation to Isoerd at? tests to the pressures exerted against a falling out between the two fines. The present Al's of both Kiti and Madol? enihmw have had violent quarrels with their Bl's but these were quickly made up. Although numerous in? stances were recorded of clan wars and of wars between sub-clans of the same clan, no wars were reported between the A-line and B-line clans of any tribe. If a quarrel arose between the Al and Bl, a reconcUiation by means of a feast of propitiation must soon be effected, otherwise the tribe would be considered to be falling apart. The devices used to force such reconcifiation, particularly the power ascribed to kava to bring about a change in attitude, are arbitrarUy recognized mechanisms in which aU concemed concur. The social pressures for poHtical stabiHty are numerous. The A and B lines are supposed to intermzury, and the members of the B-Hne stand in the fictitious relationship of chUdren to the A-line, as their designations Royal Men and Royal ChUdren respectively show. Sometimes a Nahnmwarki and Nahnken wUl be actual father and son or son and father, as is true in Net today. The mutual re? spect or affection immanent in this relationship, real or ascribed, is mdicated by the faUure of overt jealousy to develop when a section switches aUegiance from one to the other. Each of the two leading chiefs finds hunseU sur- 49 50 S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 rounded by attendants of the opposite line, since members of one's own Hne may not take the famUiarities that neces- sarUy develop in that situation. A variety of attachments may thus spring up. An extremely strong bond in the kin? ship organization is that between a man and his sister's husband or between a man and his wife's brother; brothers-in-law {mwah) are bound to go to one another's assistance in any difficulties, to respect and honor one another, if need be die for each other; the saying goes, "My mwah is my Nahnmwarki," Since the two fines in? termarry, every man of A-title must have a number of men in the B-line in the actual or classificatory brother- in-law relationship, and vice versa; thus the kinship bond serves to reinforce the political bond. Formerly when the Nahnmwarki and Nahnken could not come to an understanding the priests might intervene. The B2 as principal priest was particularly charged with this duty and, though he is no longer a priest, assumes the same function nowadays. Certahi other powerful chiefs might likewise mediate; m Madolenihmw the Lepen Moar is one of these. In addition the personal attendants of the Nahmwarki and Nahnken, particularly the Oun Mwarki and Lepen Pun, who were simidtaneously priests, might come between the two mlers. PRIVILEGED BEHAVIOR OF T H E B-LINE The Royal ChUdren or B-line titieholders in their posi? tion of fictitious chUdren of the A titieholders have a number of privUeges pecufiar to them. When the Nahn? ken, the highest member of the B-Hne, is not present in the community house during feasts, the Royal Children may stand or sit anywhere, in contrast to the commoners and the members of the A-line. High Royal Children need not bow when they pass before the Nahnmwarki. Tradition? ally, they may violate aU sorts of standards of behavior to which others must conform. A Royal Child can often be recognized by his loud talking and his free demeanor, par? ticularly in the community house but also outside it; "they go about yeUing and respect nobody." The head of the B-Hne, the Nahnken, who is considered to be the eldest son of the Nahnmwarki, can take famUiarities with his ascribed father that are permitted to no one else; at a feast of propitiation {tohmw), if the Nahnmwarki proves too obdurate in forgiving those who are seeking to atone for some offense, the Nahnken might go so far as to vio? late the sacredness of the Nahnmwarki's head by seizing him and forcing him to drink the proffered cup of kava, which is the sign of forgiveness. He may take an imperti? nent tone in conversation with the Nahnmwarki, and there have been incidents when the Nahnken on being struck by the Nahnmwarki has repHed in kind. Traditionally such privUeged behavior dates back to the time of Isohkelekel, the conqueror of Ponape and the first Nahnmwarki, whose son, Lepenien, became the first Nahnken. In the Isohkelekel legend the son does aU man? ner of forbidden things: he climbs up on his father's canoe from the outrigger side, he hands his father fish on the end of his spear instead of strung together, he stands in the canoe in his father's presence as it nears the bank, he alights from the middle of the canoe, he enters the com? munity house from the front end of a side platform in? stead of through the central area, he walks down the in? side edge of tiie side platform uistead of down its center, and he steps to the main platform inside the comer post where the platforms join instead of going around behind it. From these legendary times on, unconventional and indecorous behavior was to be expected of members of the B-line. The Isohkelekel legend referred to several times in this work is in a sense a kind of poHtical charter for Ponape. The theory beliind these ascribed behavior pattems is that the Nahnmwarki is an indulgent "father" and per? mits his "sons" to take fiberties that no one else dares. Not so with members of the A-line; such men, though potential heirs to the Nahnmwarki's position, must be meek and humble in demeanor, for they have no fiction of being favored chUdren to support any untoward behavior. Whether the fiction has any fact behind it other than the legend of Isohkelekel is problematical. Some informants state that formerly, when a Nahmnwarki whose son was Nahnken died, the A2 would take his place, depose the Nahnken, and instaU his own son instead; Kiti informants charge other tribes with stUl foUowing this practice, and everywhere Uh is held up as the horrible example. It is true that upon accession of a new Nahnmwarki to office his chUdren, who are of course in the B-Hne, are in a favorable position to receive promotions and often are soon elevated. A saying that the five tribes apply to each other goes "Tribe of afighting, embarking" {Wehi en keredi, kerada); it refers disparagingly to this shufifing of position to benefit Royal ChUdren who are literally such. But examination of the history of promotions of titie? holders in Uh and elsewhere does not corroborate the charge insofar as the Nahnken himself is concemed. A number of cases in various tribes in which the Al had been father to the Bl were revealed, but only rarely did MACHINERY OF GOVERNMENT 51 the son assume office whUe the father stiU lived; and the converse situation, when the Bl was father to the Al , also occurred. A possible explanation for the conventionafized behavior pattem, other than the legendary one, is that such deposition actually occurred formerly. If this is so, the Nahnmwarki must have played a more active poHtical role in earHer days than he seems to have done in early postcontact times, and the paraUel between the position of the Nahnken and that of the talking chiefs of Samoa loses some of its force. There may be some connection also between the ficti? tious parent-chUd relationship and the Crow system of cross-cousin terminology which exists in Ponape, but orUy if we assume that the institution of exclusive intermarriage between A and B clans of each tribe that prevaUed untU recently was in more ancient times asymmetrical. With symmetrical marriage, an A-clan man marries a woman of the B-clan, and vice versa, so that in this matrilineal society a man of either clan should always find his father and sons in the opposite line of titles to his (except for those few cases where a man takes a title in the wrong line, as discussed elsewhere; that is, in the case where A- clan is not equivalent to A-line of titles, and B-clan to B- line). In the Crow system, maternal cross-cousins are caUed by filial terms, paternal cross-cousins by parental terms, and these terms are extended downward several generations.^^ Thus, if A and B clans intermarry symmetri- caUy and exclusively, a man of one line wiU find not only his father and sons in the other line, but also his mother's brother's son, caUed "son" by him, and his father's sis? ter's son, called "father" by him. But if orUy men, not women, of the A-clan married into the B-clan, there would be "sons," actual and classificatory, but no "fathers" in the B-Hne, and the fiction that Royal Men are "fathers" but not "sons" of Royal Children would become reafity; and conversely, men in the B-Hne would have "fathers" but no "sons" in the A-line. Presumably women of the A-line and men of the B-Hne would have to marry com? moners. However, we have no evidence to support a hy? pothesis that such a state of aff aus did once exist. We might also speculate that there was once no B-clan, and that there was one ruling clan which took marriage partners from any of the other clans. The chUdren of the royal women would stiU belong to the royal clan, but the chUdren of the men would be commoners. They might receive special titles and deference would be paid to them by virtue of their fathers' positions, but then their chUdren, if they in tum married commoners, would sink to lower position. The temptation would become strong to require the children of the men of the mling clan to marry back into their fathers' clan, so that the grandchUdren would become royal again. This practice might then become institutionalized and the tities of the Royal ChUdren would become a formal series, like that of the existing A-Hne, The behavioral expectations and the kinship ter? minology involved would then be extended from the actual to the classificatory relationships, and the B-line that thus came into being would consist of fictitious as well as real Royal ChUdren, as it does today. POLITICAL COUNCILS, COURTS, AND TRIALS None of the Nahnmwarkis ever approached the abso? lute, despotic authority of their semihistorical predecessors, the Saudeleur Hne of kings, which ended with the conquest by Isohkelekel. These kings, it is said, made such continual and arbitrary demands upon their subjects that, inform? ants report only in half-jest, a man could not so much as find a louse on his head without having to deliver it to the Saudeleur, But different Nahnmwarkis varied con? siderably according to their individual characters and those of theh" opposite numbers, the Nahnkens; some were tyrannical, some were permissive or weak; and from time to time a section chief or a strong warrior might chaUenge their authority. That the cluefs, at different levels, did not have every? thmg theu- own way is evident from the occasional caUing together of the people in an assembly to discuss various projects, such as the buUding of a big community house. This might be a meeting of a sub-clan, of the people of a section, or of all the people of a tribe, under their respec? tive heads. They were apparently caUed when the majority of the people seemed to be opposed to a course of action by the chief, in order to persuade them to come around to his point of view. A chief who acted too unUat- erally might come to grief, A Nahnmwarki in Madol? enihmw in precontact times is said to have made numerous decisions without consulting his subjects, who rose and marched against him. He called upon his own clansmates to help, but he had afienated them too and they did not respond to his plea; finaUy, he was kiUed. ^ It might be noted at this point that with exclusive inter? marriage between A and B clans all cross-cousins would be in the opposite line. In spite of the use of filial and parental terms for cross-cousins in the kinship terminology, cross-cousin marriage is preferred and is urged on children by A and B clan parents. 52 S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 Nowadays, of course, poHtical councils are thuigs of the foreign administrations and have a markedly westem complexion. Meetings of the highest chiefs of a tribe, of a clan, or of any poHtical or social grouping are known as kopwung. Hambruch describes them as much more formal affairs than they actuaUy appear to have been. He refers to them as lawsuits, whereas actuaUy they seem to have been meetings of chiefs for consultations conceming war, work projects, or the awarduig of tities, and only occasion? aUy involved law. In its legal aspects the kopwung resembled a criminal trial more than a lawsiut. In addition to the tribal chiefs the head of the section where the accused fived would attend. GeneraUy the Nahnken would preside, less often the Nahnmwarki, who was usuaUy con? sidered to be above such mundane affairs and would not know his subjects weU enough to function effectively in such matters. Today, however, the Nahnmwarki is occa? sionaUy the judge under the civU administration. The section head would previously have investigated and have brought the accused to trial, and the only testknony taken was from him. The Nahnken would pronounce judgment on any commoner and on members of his own clan; if a member of the royal clan was on trial, the Nahnmwarki or A2 would give judgment. For seduction of the wife or widow of a Nahnmwarki or Nahnken, and sometimes wives or widows of other men of high rank, the penalty was death. Gazing upon a married woman of high rank at her bath was the equiv? alent of seduction in native eyes, and the same punishment was required. Sometimes, however, banishment was employed instead. These crimes might be forgiven if the whole clan of the guUty man quickly made a feast of atonement, but even then the culprit was restricted for a time to his section. The woman was not ordinarily punished. Rape was considered the same as seduction. But if the woman was unmarried rape was no different from any sexual adventure, to which no penalty was attached, except that there was some element of disgrace when force was involved; but, in any case, rape seems to have been very rare. Sexual perversion was simUarly regarded, and ridicule was the only social sanction employed; that it was effective is attested by a number of suicides that resulted. Murder of a member of the royal or noble clans was punished by execution; simUarly with incest involving members of these clans. SteaHng from a chief resulted in a beating, but most thieves were punished only by ridicule as lazy fellows. It seems obvious that for the most part the tribal court dealt with crimes against the mling clans; shnUar crimes against commoners were most often directiy avenged. It is difficult to draw the line between personal venge? ance and legal punishment. When a commoner exercised blood revenge for the murder of his clansmate, there was Httle difference, essentiaUy, between his act and that of a judge, the Nahnken, in delegatuig someone to execute the murderer of one of his own clansmates. An offense against a commoner seldom was brought before the tribal court; instead there might be a meeting of the clansmen of the uijured person and his relatives in other clans, and from among them were chosen the instmments of revenge. In the court the assembled judges would be? long to one or both of the two chiefly clans, and the crime being tried would have ordinarily been committed against one of their clansmates; since the members of these two clans were considered to be in the actual or fictitious relationship of parent and chUd, and smce they were further bound by ties of marriage, the circum? stances were not very different from a meeting of a group of related commoners, and the crime involved virtuaUy the same factors as did a crime which among commoners was punished by blood revenge. Informants in speaking of these matters make fittie distinction be? tween clan and tribe. For example, they say that in Net the punishment for incest was not death, as it was else? where in Ponape; but it develops that they are speaking of clan 7, the A-fine in Net, whose members were different from members of other clans in that they are supposed once to have intermarried among themselves. Some informants refer to the court as a clan, not a tribal, meeting. Here, as in other respects, the state has the aspect of a clan, but a clan paramount among other clans and able, because of its coercive powers, to lend some show of formality and legal trappings to an insti? tution it otherwise shared with the clans it ruled. Under such circumstances, then, the legal code coiUd not be a very formal one. Descriptions of various trials are vague and inconsistent, as one might expect with such a loosely organized institution. For the same crime a trial might be held on one occasion and direct vengeance taken on another. There was also inconsistent treatment of quarrels between commoner clans. Individual chiefs varied greatly in this respect; usuaUy they tried to settie matters by arranging a trial court, but often they did not interfere with blood revenge; at other times they might become angry when a clan war started and send messengers to stop the fighting or kiU the man who began it. UsuaUy a crime was expiated when one act of retribu? tion had taken place, and the quarrel was not pursued further unless the first act led to a war. If the actual cul? prit was not caught, anyone else in his clan would do as a substitute. If a clan war began and one clan was driven away by the victorious side, or by a Nahnmwarki who had decided to punish one side, the exUed party would sometimes seek to take vengeance on clansmates of their enemy in their new home. MACHINERY OF GOVERNMENT 53 When personal property was damaged or destroyed, the usual thing, if the victim felt strong enough, was to reimburse himself directly from the culprit's property or to destroy it. When an injured man did not feel capable of taking up the quarrel himself, he could lodge a com? plaint with the Nahnmwarki or Nahnken, and the chief would try to redress the harm. He could beat the wrong? doer, or drive him from his land, or assess him a fine in such valuables as sennit, mats, and canoes, which he would pay over to the v^onged party. But such com? plaints seem to have begun orUy in later times; people were usuaUy ashamed of them, for they constituted a con? fession of weakness, and to some extent this feefing of shame persists in the modem courts. Informants find it difficult to understand Hambmch's statement that the Nahnmwarki punished "Schuldigen Oder Streitenden" (1932 II, p, 150). Such matters were largely up to the clans concemed. It was considered most reprehensible for a clan not to attempt to take revenge for an insult or injury to one of its members, and such craven conduct earned the ridicule of aU. Nevertheless, sometimes an unfaithful woman and her lover might be taken before a chief to be scolded or beaten, or in Ger? man times to be given manual labor to do. Possibly such action was taken when blood vengeance was deemed fike- ly to array relatives on opposite sides, since in a war broth- ers-m-law were expected to help, yet being necessarily in different clans because of clan exogamy they would be subject to opposing pressures. If capital punishment was imposed, certain men were delegated to carry out the sentence. About the tune of first white contact the Nahnmwarki himseK or his broth? er seems occasionaUy to have been the executioner, but in earHer times certain titled men were appointed to this duty. Torture, it seems, was rarely practiced. The doomed man had his hands tied behind his back and he was then clubbed or speared; a brave man would sit with his hands unbound. In Spanish times there was a case of bummg alive. For first offenses within a sub-clan there was ordinarUy no punishment, even for murder. But if a man made a reputation as a troublemaker, he would be disposed of by his feUows under the direction of the head of the sub-clan. A man of clan 2, Reisip by name, fiving in section 3 m Madolenihmw, was an obstreperous person, always bul? lying and striking his clansmates and acting proud and in? dependent, FmaUy three men of the same sub-clan (Soun Tamworohi), after consultation with theh" sub-clan chief, knifed hun to death. The sole survivor of the trio. Sun- rait, reports that the Spanish jaUed him for 4 months and the Nahmwarki, Paul, who had been converted to Christianity, gave him 4 months of labor on the canals; probably a non-Christian ruler would not have concemed himself. Incest withm a kuiship group was always punished by death. But nowadays a number of mcestuous unions, though matters of common gossip, are tolerated. DIRECT PUNISHMENT BY CHIEFS The causes of offense to a high chief were many, the consequence of which was condign retribution. Among them were faUure to pay suitable deference, to observe proper etiquette, to respond to a call to give service, and especiaUy to offer the customary tribute and first fruits. Chiefly retafiation took several forms, but strippmg a cul? prit of his land and titie and banishment were perhaps the most common. Tenure by a commoner of a piece of land was always precarious, the concept bemg that aU land belonged to the Nahnmwarki and was issued and revoked at his pleasure; however, suice the institution of private ownership by the Germans the chiefs have m this regard lost theu* coercive power. But taJdng a man's titie from him is stiU practiced, and ui fact some say more frequentiy than m pre-German times, perhaps m com? pensation for the loss of power by the chiefs m other respects. A chief might formerly have the house of someone at whom he was angry bumed down; this act is isimw. Only the Nahmwarki, Nahnken, and a few other high chiefs possessed this privUege, Henry Nanpei, who was chief A6 of Kiti, is said to have exercised it several times, but his position was far greater than his titie would indicate, owing to his father havmg been a Nahnken when Henry was born and especiaUy because of his great wealth. On one occasion a relative of his, married to a man of clan 2, eloped with a man of clan 17; Henry went and brought her to his own home, but the man came that night and carried her off agam; whereupon Henry grew very wroth, gathered his people together, and went and burned a number of houses of people of clan 17 in sections 25 and 29 of Kiti. UsuaUy the Al or Bl did not need to go and do tiie burning himself, or even order it to be done. The A2 54 S M I T H S O N L \ N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 would know that the Al was angry at someone and would take it upon himself to inflict the punishment; similarly the B2 might do it for the Bl. If housebuming was the punishment imposed no banishment ordinarily foUowed. Besides his house, a man's canoes, fishing nets, and other belongings might be bumed; or his canoes might be broken with stones and the outrigger lashings cut. Other punishments included shooting of pigs and dig? ging up of bananas, yams, young breadfruit, young co? conut trees, etc. Such destruction at the order of a Nahn? mwarki or Nahnken is known as oudek, the same term that is appfied when a Nahnmwarki or Nahnken Hes Ul and the people of all the sections must bring him quan- ties of food to eat; it is of interest that the people view these heavy obligations, at least so far as vocabulary goes, in the same Hght as punishment for an offense. Giving of blows was common enough in a moment of pique, but when the Germans first used flogging as pun? ishment the natives were, to use their own words, "greatiy astonished," Worse than blows by a Nahnmwarki or Nahnken are considered his scolding and curses. Stone-throwing was a common practice of chiefs. It is mentioned in the early 1830s by O'ConneU. It was done not only out of anger, however; often it was intended simply to impress upon the people their low status. Or it might be done simply because it was expected of a chief. As soon as a man was promoted to Nahnmwarki he would often throw stones at the assembled people, and they would flee and retum bearing kava in suppfication. At a feast of atonement it was considered proper conduct for the person being appeased to throw stones, A Nahnken of Kiti, Nahnawa en Mwudok by name, is described as carrying a stone in each hand every time he entered the community house and flinging them at random among the people; he considered this as a species of elegance, incum? bent upon a man in his position. This was about 1870, but the custom persisted untU very recentiy; Sigismundo, Nahnmwarki of Kiti in Japanese times, is said to have practiced it frequently. The Nahnken even more often than the Nahnmwarki foUowed the practice, and occas- sionaUy would throw stones for the Nahnmwarki if the latter was vexed for some reason. Banishment The Nahnmwarki, the Nahnken, and the head of a section had the power to banish. A lesser tribal chief could banish someone only if he were simultaneously a section head (but one informant states that chiefs A2 and B2 also had this power in their own right). The Nahnmwarki and Nahnken seldom carried out an act of banishment or destruction by themselves, but usuaUy delegated chiefs A2 or B2 to do it, (ActuaUy, cases of banishment as recorded often show men of much lower rank carrying out the execution of the order,) The section chief derived his power to banish from the fact that he was, in many instances, of the same clan as the Nahnken or the Nahnmwarki, depending on which of them owned the section. Banishment of an offender is known as kalipehda. The term is sometimes extended to cover aU acts of destruc? tion by a high chief, including destruction of a man's crops, oudek. Banishment of a whole sub-clan or clan {pokousala) also occurred. Persons banished had to leave at once, UsuaUy they took refuge with their sub-clan chief, even if the latter fived in the same tribe; one man, banished by the Nahnmwarki from his home in section 23 of Net, went to live in section 2 of the same tribe, at the home of the chief of his sub-clan. But the Nahnmwarki might order the sub-clan chief not to shelter him, in which case he would go to live with clansmates in another tribe or, if these were lacking, with members of his father's clan. As a last resort he could go to a friend who was an out? standing chief and become his servant. Banishment was seldom imposed for a first offense. Most often an apology feast sufficed to atone in such a C?ise. Usually banishment was for Hfe, though when the chief who had banished a man died the offending party might sometimes retum. Sometimes he chose to remain in his new home if circumstances were better there. Reasons for banishment included not presenting first fmits and other food offerings, eating foods forbidden to commoners, adultery with the wife of a high chief, or faUure to obey any order. For example, in buUding a community house or a chief's house each section would be assigned as its share of the constmction the width between two upright studs in the waU (a dinak, one arm- span Ul width); if one section failed to complete its share its chief might be banished. About 1840 a man of clan 2 named Tulal, livmg in section 16 of Kiti, was caught gazuig upon the wife of the chief of that section when she was at her bath. He could have been executed for this offense, known as mwoanipil, for it is equivalent to adultery. But the chief, since he was a member of clan 4, hence clansmate to the Nahn? mwarki, had the power to banish hun, Tulal took his famUy and went to Uh to live. The last banishment m Net occurred about 1900. The Soumadau (X6) of section 1 went fishing and caught a merer fish, a species reserved for the highest chiefs. In section 1 the senior tribal chief was the Nahnsoused (the titie which then was Bl in Net, corresponduig to Nahnken in the four other tribes) and the fish should have been presented to hun. The misdeed came to the ears of this chief, who summoned the culprit and gave him orders to go fishmg again and bring hun the whole catch. The Soumadau did not obey. Then the Nahnsoused instmcted the SouHk (X8) to go and bum the house of Soumadau, MACHINERY OF GOVERNMENT 55 kUl all his pigs, dig up his yams, and drive him and his famUy away. With all the food the Soulik then made a feast for the Nahnsoused, The refugees went to Uh and lived there untU the Nahnsoused died, when the section head called them back. The land vacated by a banished man was awarded to someone else or taken by the chief who banished him. In a number of cases the high chief of Sokehs called upon the Lepen Net (the then Al of Net) to carry out the banish? ment for him and the two chiefs would then make a feast together, using the exUed man's belongings. The confis? cated land, though in Sokehs, would then belong to the Lepen Net, This was possible only between Sokehs and Net, not between any other two tribes, because their re? spective rulers were clansmates. OccasionaUy a man was in a strong enough position to defy the Nahnmwarki or Nahnken, Such was the case of a member of clan 5, the head of section 18, Madol? enihmw, in Spanish times; this man repeatedly refused to obey the Nahnmwarki of the tribe but suffered no penalties in consequence. His independence stemmed from his fame as a brave warrior against the Spanish. A man may also seek to be banished. He may become convinced that there is no chance in his present residence of getting a higher title because elder clansmates and brothers precede him in seniority. But there is no way for him to state his ambitions frankly, since under the Ponapean pattem of personal modesty it is a shameful thing operUy to seek advancement, and he cannot move elsewhere without his brothers questioning him or insist? ing that he remain. Therefore he wUl provoke banish? ment by his behavior. For example, the present A4 of Net formerly lived in section 23 and despaired of ever be? coming the section chief, since his elder brother took pre? cedence over him. He therefore began to flout Kalisto, the then Nahnmwarki, He kUled pigs to seU and offered none to Kalisto; he stopped offering first fmits; he pre? tended to be sick when caUed upon to join a work group. Suice this was in Japanese times the Nahnmwarki no longer had the power of banishment, but Kalisto acted as nearly in the old pattem as possible; he and the elder brother advised the younger brother to move to section 10, where his relatives gave him a piece of land. The head of this section was an old man and the present A4 ingra? tiated himself into his favor, taking over his work and ultimately becoming head when the old man died. Now he was able to offer first fruits direct to the Nahnmwarki (at this time Eduardo) instead of induectiy through the section head. He also insinuated himself into the good graces of Joseph, the present Bl (then the B2), and his wife, without directly asserting his ambitions, although it was obvious what he wished. Joseph helped him become Soukeperoa, next A7, then A4. The elder brother was only A12; he became ashamed and left Net to live in self- exile in Madolenihmw. For a simUar reason the present A 7 of Sokehs, who was formerly a Madolenihmw man, moved to Sokehs. Ostensibly this was in order to marry a girl who Hved there, but really it v/as because he had littie hope of pro? motion at home. He knew it was easier to get a titie in Sokehs; Sokehs is populated largely by out-islanders who have not been thoroughly assimUated to Ponapean title- seeking ways, and competition is therefore less severe. Self-exUe because of humUiation, usuaUy caused by being passed over for a high title, is not uncommon. The case of Benito, who failed to become Al of Net and went to Hve in Madolenihmw, has already been related. A sim? ilar case is that of Leon, who was B2 of Sokehs and ex? pected to succeed Tionis, the Bl, when the latter died a few years ago. Kalio, the Al , however, promoted the B3, Joharmes, instead and Leon in anger went to live in Kiti. Agaui, when the B2 of Net, Vicente, was dying, the B4, Mainrat, fuUy expecting to succeed him, brought pigs and prepared lavishly for his anticipated titie-payment feast. But Vicente was offended at Mainrat because he had failed to visit him whUe on his sickbed or to come and wel? come the high chiefs who did come (both Vicente and Mainrat lived on Param Island, and Mainrat was chief of that section, being the Lepen Param as weU as B4 of Net; he was therefore remiss in his duties). So Vicente advised the Bl, who was then Augustine, to appoint Joseph (the present Bl) to foUow him, which Augustine did, and the present B2, Johannes (a different Johannes from the man mentioned previously), became B3. Mainrat in shame and frustration exUed himself to Uh, where he is presently Souruko, ATONEMENT FOR OFFENSES A feast given to beg forgiveness of an offended person is known as tohmw. It is rather similar to the feast caUed aluh, which is given when an iUness afflicting a person is diagnosed as being supematural punishment for an affront to a chief or other person of authority; but with 298-818?&8 5 tohmw there is no propitiation of a god or ghost, as there is with aluh.^* An intermediary is involved. To beg par? don of a man of the A-Hne a man of the B-Hne should ordinarily serve as the intermediary, and vice versa. If the " For a fuller account see Riesenberg, 1948. 56 S M I T H S O N L \ N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 1 0 Al is the person whose forgiveness is to be sought, the go- between should be the Bl, B2, or the Al's mother's brother's son or daughter. These persons are in the B-Hne, but sometimes the Al's mother's brother serves as go-be? tween. If the Bl is the person offended, the Al , A2, A3, or the Bl's equivalent relatives should come. If the Al is angry at the Bl or the B2, or the Bl at the Al or the A2, the offending parties must present themselves without in? termediary. Kava is indispensable at any atonement. For a Hght of? fense the kava alone wiU suffice, although usuaUy sugar cane and coconuts are also offered. For a serious trans? gression, a regular feast with a stone oven and all the cus? tomary trappings is usual. If the affair is between Nahn? mwarki and Nahnken, the stone oven with baked pig or dog and yams is mandatory; such a feast is caUed Burn? ing Unhappiness {isikala insensued). The foregoing is native theory as informants expound it. In practice, detaUs vary considerably. In one ob? served case the Lepenkin, whose title is a low one in Net, had slandered the Nahnken, and the fact had come to the Nahnken's ears. The Lepenkin requested the senior tribal chief in his section, the A6, to go and beg pardon of the Nahnken. But the Nahnken felt too ag? grieved to accept a man of such comparatively low title as mediator; he scolded and raUed at the A6 and refused to accept the proffered apology, insisting that a man of higher A-title be sent. The Lepenkin hovered about the door of the house where this scene took place, but the Nahnken ignored his presence. I did not witness the further development of this episode but was told that a higher chief was later employed as intermediary, and that the apology was accepted and reconcifiation effected. The intermediary usually brings with him as a token of suppfication kava, sugar cane, a basket containing five young coconuts with the bottoms opened for drink? ing, or nowadays cigarettes. (This information is largely from Net people; a Kiti man says one or two drinking coconuts are brought.) Only the kava and sugar cane are ordinarUy brought to a Nahnmwarki or Nahnken; the other objects are suitable for presentation to lower chiefs or between commoners who are becoming reconcUed after a quarrel. In other circumstances, if sugar cane is brought in whole stalks and laid on the floor before a chief, it al? ways signifies an invitation to a feast or other affair; but if the sugar cane is prepared for eating, it can mean only pardon-begging. In the episode involving the Lepenkin the intermediary presented a length of sugar cane of five sections, with the terminal section peeled back so that eight strips hung from the node; the peeled section he ringed with three shaUow grooves to facUitate the break? ing off of pieces for chewing. Then he knelt on the floor, holding the sugar cane stalk upright with the lower end on the floor. The Nahnken expressed his displeasure by refusing to accept the stalk; in the course of his scolding he suddenly lashed out and knocked off the uppermost section. Another apology I observed again involved the Nahnken of Net, The Nahnken and I, sitting m the Nahnken's house, saw the A2 approachuig with a kava plant on his shoulder. The Nahnken immediately ceased talk, sat upright, and assumed an air of calm re? pose. The A2 entered bearing the kava with root foremost, as is the usual practice, but instead of laying the plant down with the root toward the Nahnken, as is always done with an ordinary offering, he flipped it over so that the branches and leaves lay on the head and shoulders of the Nahnken, who con? tinued to sit immoving. This act at once showed that it was neither an offering nor an invitation to a feast, but an apology. The A2 proceeded to hack off the roots and then, speaking for the first time, gave a greeting; this again was a clue that it was an apology. Then he took the roots away to the nearby community house to pound them. On an occasion Hke this the kava must be pounded with as rapid a beat as possible. The A2 said nothing until he brought back a cup of kava, then he offered it to the Nahnken and explained his mission. The daughter of the Nahnken had left her husband the week before, and the husband, who has only a small titie, was now askmg pardon of the Nahnken for offendmg the woman, usmg the A2 as intermediary. The Nahnken accepted the apology. Had he shown anger or cut at the kava plant with a knife, the A2 would have had to leave without saying anything more and would have retumed with more kava; this he would do up to three tunes, then the Nahnken would have had to accept, for as natives say it is uiconceivable that he could have continued to refuse atonement involving kava. On stiU another occasion witnessed by me, the offended party was a man whose younger brotiier had not helped him to make the feast known as the honor feast due the Nahnmwarki of Uh, but instead had jomed with other people to make a separate feast. When the younger man heard of his brother's displeasure, he made an apology feast to him. Again, an apology feast was given by a chief of low position who had been summoned by the Nahnmwarki of Net to record songs for an American Hnguist, He refused to come because he beHeved that the Nahnmwarki was being paid and tiiat he hunself would get notiimg; whereupon tiie Nahnmwarki caUed him to account before a large assembly of people, rated him severely, and required hun to make the apology, with his section head acting as mtermediary. The missionary Doane, writing m 1870, describes a rather dUferent kind of formal atonement which he wit? nessed. A man had indirectiy caused another's death. The culprit assembled a large number of articles (mats, twine, MACHINERY OF GOVERNMENT 57 pigs, dogs) and brought them to the commimity house, where he gave a feast; he led a procession of his people into the buUding, each carrying some of the goods. AU were presented to the offended party, a high chief. Then he came, trembling, to that chief and gave him a Hghted pipe, which was accepted. The chief in turn gave him a piece of sugar cane, and other members of his clan did likewise. There foUowed discussion, after which the Nahnmwarki announced that the sin was absolved. This peaceful settlement Doane ascribes to the effect of the visit of the U.S,S, Jamestown to Ponape in 1870. But in 1871 a chief of the offended side gathered his forces and attacked a house where the offending man and his party were sleepmg. We are not told why the settiement was not enduring. Occasionally the procedure for atonement is more com- pficated. This usuaUy is foUowed when it is the Nahn? mwarki who is the offended party and a commoner who is the culprit. If the commoner feels that his own status is too low to approach a high B-line chief, he may go to the senior member {meseni en keinek) of his sub-clan to secure his services as go-between. The sub-clan head, if he is also a commoner, may not approach the Nahn? mwarki directly but first makes the apology to the Bl or B2. Then the two go together to the Nahnmwarki and repeat the feast to him. If it is the Nahnken who has been offended, the Al or A2 may be asked to intercede. If a member of one commoner sub-clan has insulted or injured a member of another and a war threatens, the procedure is simUar; the head of the first sub-clan approaches the Nahnken, who intercedes with the Nahnmwarki, who then summons the head of the second sub-clan. If this sub-clan head and his party are very angry there is fittie talk at first; a simple apology is made with a kava plant. The Nahnken gives the head of the second sub-clan a cup of kava and begs him to be ap? peased, and when he has drunk, the Nahnken hands the first sub-clan head another cup, which he in tum hands to the second sub-clan head; then the Nahnken gives the head of the offended sub-clan a third cup, which is passed on to the offending sub-clam head, who drinks and passes it back for the other to drink. Thereupon the offense is erased. Apologies between sub-clans should be made as soon as possible; formerly the offended party would wait at most three days before beginning hostUities. A war of this sort occurred in German times in Kiti in a quarrel over a seduced woman. It is the Nahnken who in such cases is most often ap? proached to act as go-between because he is considered the protector of the people against the wrath of the Nahnmwarki, who is traditionaUy remote from the com? moners. When the Nahnken intercedes with the Nahnmwarki, he takes the cup of kava and offers it to him; if he refuses to drink, the Nahnken wUl argue and scold, and has been known even to seize the Nahnmwarki about the head with his left arm and pour the draught down his throat. A Nahnmwarki who sees the Nahnken coming with a kava bush will sometimes flee from his house. If the intermediary meets the Nahnmwarki on the path, he may take two smaU pieces of kava root, roll them in a kava leaf, and offer them as a symboHc cup of kava. Lesser chiefs than a Nahnken, acting as inter? mediary with a Nahnmwarki, may not raU or force the drinking of the kava; but the Nahnmwarki's father or father's brother, who of course would be in the B-Hne, may scold. Sometimes the argument lasts for hours, the Nahnken speaking for the people, until the Nahmnwarki at last accedes. Less often the two high chiefs reverse their roles; the Nahnmwarki then acts as intermediary and may force the other to drink. At such times the suppHcating chief quotes old sayings, e.g., "chiefs make naught of things" {menin kasohr sou? peidi) and "commoners destroy things" {menin kau ara? mas) ; the purport of this being that though commoners, being base in nature, do not overlook an offense, it shoidd be disregarded by a high chief. A chief is supposed to have the attitude that aU difficulties and unpleasant events are things that pass and should be ignored; a Nahnmwarki shoiUd not scold or lose his calm but should remain digni? fied under any stress. Therefore he must aUow himself to be placated. If the Nahnmwarki in anger begins to throw stones (the patterned outlet for chiefly spleen), the Nahn? ken might unmediately come with a kava bush and flip it over on the Nahnmwarki's head; his anger would sup? posedly be immediately stiUed, A high chief is ideaUy responsive to the needs and re? quests of his subjects. The proverb "a chief is a hibiscus in the wind" {keleunieng soupeidi) expresses this chiefly at? tribute ; just as the wUd hibiscus tree bends readUy before the wind, so should a Nahnmwarki accede to his people's wants, Kava plays an extraordinary role in these circumstances or, for that matter, in any emotion-charged situation. To kava is ascribed the greatest inducement in making atone? ment. A Nahnmwarki or Nahnken is not supposed to be able to scorn an offer of kava; if he did, he would be held in contempt by the people. The Nahnken of Net regards Hambmch's remarks about severity of punishments and the use of the death penalty as exaggerated, for a high chief ought sooner or later to be moUified by repeated of? fers of kava. Often he only plays a role expected of him, his reluctance to forgive is feigned, and should the inter? mediary leave without a settlement being made, "he would feel bad," On one occasion, previously referred to, Pau? Hno, a Nahnken of Net, quarreled with Eduardo, his con? temporary Nahnmwarki; Paulino went to his home and sulked for three days, but when he returned and made 58 S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 two apology feasts to Eduardo, one in his home and one in the community house. Paulino was very rash to wait this long, for it might have led to a war between his clan and that of the Nahnmwarki. If either a Nahnmwarki or a Nahnken faUed to accept a proffered cup of kava, it would be a sign that the tribe was falling apart. Moreover, only two refusals of kava are permisible. The same Paulino was once angry at some people who f aUed to appear when he summoned them to help buUd a house. The offenders brought him a kava bush but he slashed it ui two with a knife. They went away and re? turned with more kava, pounded the root, and gave him a cup, but he dashed it from him. But the third tune they came to make the apology he felt it necessary to accept the cup and they were reconcUed, (Some informants say that more than two refusals may occur, but when a cup is finaUy presented as "sakau en peidi," fiterally "kava facing down," its refusal would mean war,) We have untU now been discussing the kind of apology and apology feast caUed tohmw. In spite of native theory to the contrary, there have been instances when the apology was not accepted and severe punishment was meted out or war foUowed. But there is another kind of propitiation and atonement more drastic than tohmw. It is called sekenpwoud {sak, to eat; or possibly from sakara. to petition; pwoud, spouse; referring to a woman's giving herself to a man to use as he wishes); in this case no go- between is used, but the culprit defivers himself com? pletely uito the hands of the chief, body and Hfe and pos? sessions, throwing himself abjectiy upon the chief's mercy. Supposedly a plea of this nature is impossible to reject. About 1898 two men of section 1, Net, went to section 23 to fetch a parem tree to make a canoe. It was the custom for the canoe makers, after the huU was shaped, to lay theu- adzes before the highest chief of the section and beg permission to fmish the job. But the two men here con? cerned faUed to do this. The section head heard that they were making a canoe; he took an axe and came to the place where they were working; as he approached they took to theu- heels, but he pelted them with stones and laid them both senseless; then he chopped up their canoe. The two men, when they had recovered, made a seken? pwoud feast to the chief; one of them brought a large boar with long tusks (a ngihpwar, "teeth emerged"), the other a large sow past productive age (a iepwou, "barren"), both of these being exceedingly valuable offerings; each also brought five 10-man yams and one 10-man kava plant; and they put themselves completely at his mercy. Thereupon they received forgiveness. SUPERNATURAL SANCTION OF CHIEFLY AUTHORITY The authority of a chief is supported by his protective spirit (ani or eni). Each chief has one or more such spirits, who may be an ancestral ghost {eni mel or eni aramas) or his clan deity {enihwos). When a Nahnmwarld or other chief is angry at a man who has neglected to perform prop? er acts of fealty, the ghost also becomes angry, because the chief and his protective ghost are always in the same, har? monious mood; or even if the chief is not aware of the af? front, the ghost wiU know about it and become irked; in either case, the result is a sickness of a special kind {riahla) that befalls the offender or, more often, his chUd. Such a disease comes if the Nahnmwarki or Nahnken is not in? vited to a new-house ceremony to dedicate the buUding or to any other feast where his presence is required; if the owner sleeps in the new house only one night, he wUl faU sick and die. Likewise, if a man faUs to present first fruits or any of the fishes and turtles which are the due of a chief, the disease wUl come. What is required when the riahla disease strikes is the giving of an apology feast of the kind caUed aluh, which is intended to propitiate a spiritual being and which differs from the hitherto discussed temporal apology feast called tohmw in certain detaUs described below. It is not possible to make a clear distinction between the kinds of offenses that result in the disease and those that do not, nor is it easy to distinguish between the symptoms of the riahla disease, which vary widely, and those of ordinary diseases. The distinction is made by having recourse to a diviner, and it is thus determined which kind of apology feast is neces? sary. The imminence of supematural punishment was in former times enough to ensure complete control by a high chief over his subjects. Nahnku, who was an exceptionaUy strong Bl of Kiti in the middle of the last century, had only to intimate that the spirit of his father, who had been Al , required a certain feast to be made, and the people would drive themselves day and night to execute his wishes. Some recorded accounts of ghostly punishment shed light on the ideas surrounding these beliefs, A sick chUd was brought by its parents to a female shaman to have the cause of the disease diagnosed. The shaman entered a trance and the god Sahngoro spoke through her fips; he was angry because the father of the chUd had pounded MACHINERY OF GOVERNMENT 59 a large kava root and dmnk with his friends in private, not offering any to the local high chief, the Lepen Palikir. Sahngoro is the deity of the Dipwinwai clan, to which the local high chief belongs, hence it was he who had avenged the affront to his human partner. The father of the chUd now made a spiritual apology feast to the chief and the chUd recovered at once. Another chUd died from supematuraUy induced dis? ease because its parents had caught and eaten royal fish and had not brought any to the Nahnmwarki of Madol? enihmw, whose subjects they were; that this Wcis the cause of the Ulness was seen from the fishlike actions of the chUd on its deathbed. In a thud case a fittie girl was sick and a female doctor, in preparing the medicine, saw signs that the disease was caused by a ghost, the chief of the section where the parents Hved being angry over their neglect to bring him first fmits, fish, and kava and their failure to bring pigs to feasts. The parents performed a spiritual apology feast and the chUd immediately re? covered. Again, a woman who had adopted the chUd of the A6 of Net refused, when ordered to do so, to contribute a pig to a feast the A6 was making to some visitors to section 1, where he is the senior tribal chief; the A6 was vexed and the ghost of his dead father, as determined later by divination, caused the child to become iU; it recovered orUy when the woman made a spiritual apology feast to the A6, The woman herself did not become sick because the A6's father's ghost had no power over her, even though they were of the same clan; but the child, though of a different clan, was his lineal descendant. An ancestral ghost can cause only his own descendants to fallUl. The concept of supematural punishment for an af? front to a chief is identical with ideas about the effects of disrespect shown to the head of a sub-clan or the head of a family; the ancestral ghost or deity associated with a kin group helps to enforce the authority of the senior mem? ber of that group over its junior members in exactly the same way that the authority of a chief over his subjects is supported. Ghostly retribution is exacted even from younger siblings when they faU to grant the eldest his pre? rogatives. Thus, a woman's newly purchased chickens died and her largest pig disappeared; the pig was found orUy when she came to beg forgiveness of her elder brother for not having given him any of the fowl. Somewhat simUar is the experience of the sons of the A6 of Net, who had new canoes made but faUed to invite theu* father to the dedication feast when they were completed; as a result, the protective spirit caused the canoe hulls to split. Taking things without permission from an older relative, speaking disrespectfuUy to him, or faUure to make presents cause the anger of the offended relative to be transmit? ted somehow to the god or spirit of the clan, sub-clan, or famUy, who then exacts supematural punishment. The famUy ghost acts even to enforce domestic authority; in one instance an unruly boy's mother's dead brother visit? ed him at night and caused a recent wound to throb unceasingly, then explained why he had done so; the in? formant could hear him plainly, he insists. The spiritual apology feast, held in consequence of spirituaUy induced disease and requiring propitiation of a ghost, differs in some respects from the feast offered to someone as an apology and in appeasement of anger for a temporal offense or, as in the incident of the canoes pre? viously related, when supernatural punishment not in? volving disease is concemed. Three objects must be brought by the parents or other relatives of the sick per? son ; kava, a dog or pig, and yams to be baked. The kava bush is laid before the Nahnmwarki or other offended per? son and the already slaughtered animal is placed among its branches, (This resembles a first fruits presentation ex? cept that in the latter the animal, presented afive, is never laid on the kava branches.) After the stone oven is start? ed, the Nahnmwarki makes a special prayer {sakarkihda) to his protective spirit for the offerers; nowadays many chiefs pray instead to the Christian God, If the offended ghost is a famUy spirit, as determined by a diviner, it is sometimes propitiated directly, with an old man or wom? an famifiar with the proper procedure making the prayer and the whole famUy assembled by the sick person before whom they lay the food being offered. Formerly a shaman might function in the community house in place of a chief. The foregoing descriptions throw some Hght on the relationships between political units, kinship units, deities, and ghosts. Most gods {enihwos) are said to have existed always, and on this ground are distinguished from ghosts, which are the disembodied spirits of dead persons. But gods are also linked genealogically to clans. Thus Isoh- kaniki is a clan 11 god, Likendkahnpein is a clan 13 goddess, Luhkensed is a clan 17 god. In the tribe of Net the gods of the Nahnmwarki are Liesenkomwmwad, who is the god of clan 7, to which the Nahnmwarki (and of course his mother) belong, and Lumwohdeleng, the god of clan 6, to which the Nahnmwarki's father belongs; at the same time the Nahnmwarki has, as an additional spirit protector, the ghost of one of his more immediate ancestors, who is fikely to have been a high chief, Legend- arily, the deities are in the position of mother {ihn) or mother's brother {uhlap) to members of the clan. It therefore appears that the supematural being which be? comes angered at disregard of political prerogatives and dues of men senior in title and the being who likewise is vexed at neglect of kinship obfigations owing senior men of kin groups are always the actual or legendary ancestors, direct or coUateral, of the persons they protect, A com? moner's protective spirit, who is ordinarUy a recently dead ancestor, is called a human ghost {eni aramas), but the 60 SMITHSONLVN CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 equivalent ancestral spu-it of a Nahnmwarki or other high chief is a great ghost {eni lapalap) and serves as a bridge in the probable apotheosis of gods {enihwos) from ghosts; the various types of supematural beings are prayed to indiscriminately at the feast of spirit propitiation, with no difference in attitude or objective. A principal difference between these beings, however, is that a god can make a man who is not his descendant faU Ul, whereas a ghost can visit disease only upon his own descendants, A Nahnmwarki, who is both head of his clan and chief over all the clans of his tribe, is thus supported in his position by supematural sanctions applying to aU his subjects, whUe an ordinary senior man of a kinship unit is sup? ported in his authority over only the members of that unit. We may therefore speculate that an evolution of gods from ghosts on Ponape was paraUeled and accom? panied by the development of state from clan and para? mount ruler from clan head. Taboos A taboo {inepwi) may be imposed by a Nahnmwarki during times of scarcity of any product. Yams, kava, pigs may be placed under a ban and during the period in? volved no one may make a feast and aU must eat sparingly. Yams particularly were tabooed formerly, since they were fewer in number and variety than nowadays. Breadfmit was never tabooed, since it grows without cultivation and is plentiful. The eating of some types of fish is also pro? hibited periodically; kioak, which are scarce in February, are often tabooed for the month and sometimes, if the scarcity continues, during March as weU. The eating of some species is prohibited during their spawning period. Such prohibitions have no spiritual concepts attached to them and merely serve practical purposes. Nowadays a simUar edict by the Trust Territory administration is caUed by the same term, inepwi. No supematural sanc? tion Is involved in support of them. The same word, how? ever, is also used for the act of a man who protects a valuable tree or other object by tying a coconut leaf around it, at the same time saying a charm; or when he wishes to prevent trespass on his land and lashes a coco? nut leaf between two posts. In such a case spiritual pun? ishment would come to the transgressor in the form of affliction with ulcers, whether the taboo were violated wittingly or not. WARS Fights between two tribes consisted of single raiding expeditions of short duration, urUess a formal invitation to battle was involved. Usually the victors retumed home the same day. No one caught, regardless of age or sex, would be spared. The invaders would destroy banana, kava, and yam plantations, as weU as the smaller bread? fruit trees; the large trees were too difficult a job to tackle in a short time with sheU adzes and knives. The houses would be burned after being looted. Such prisoners as might be taken woidd be kUled, either on the way back or after arrival home, A retum raid by the surprised party usuaUy foUowed, and sometimes intermittent hostilities of this nature might continue for years. Formal battles often Involved an invasion with a fleet of canoes that would be met by the opposing fleet some distance off shore. First sHngstones were used, then at closer quarters spears and clubs came into play, O'ConneU describes also bows and arrows, but all informants insist that they were never weapons of war. If the defending fleet was forced back to shore, the battie continued on land. O'ConneU gives an account of one such encounter in which he took part, when some 300 lost their fives, but living informants who participated in wars after the intro? duction of rifles recall at most twenty deaths In a single engagement except against the Spanish. The missionaries, writing m the 1850s, after the m- troduction of guns, speak deprecatmgly of these engage? ments. They describe them as exchanges of voUeys at safe distances untU one side exhausted its ammunition, then the other side would attack and carry away or de? stroy property, A surprise attack or an ambush might per? haps accomplish the death of one old woman. But by Spanish times war had become considerably bloodier, and the numbers of Spaniards who feU in single engage? ments against the Ponapeans occasionaUy numbered In the hundreds; Finsch (1893, p. 235) mentions one battle alone m 1890 when over 300 casualties occurred. The history of the earfiest documented wars between Kiti and Madolenihmw, though fragmentary and largely compUed from missionary letters, is revealmg as to the character of aboriginal warfare, even tiiough It was carried on with guns. In 1850, 2 years before the Protestant mis? sionaries arrived, there was a cuhnmating battie between the two tribes, won by Kiti, For 15 years previously, ac? cordmg to Gulick, the high chiefs of the two tribes had not met except in battie, and Madolenihmw had con? sistently dominated Kiti, This 1850 warfare Is apparentiy MACHINERY OF GOVERNMENT 61 the same as that described ui a native text ui Hambmch (1932 II, pp, 356-357). Accordmg to the latter, the Bl of Kiti invaded section 18 of Madolenihmw and klUed a man. In retum Madolenihmw sent a party to section 3 of Kiti and kiUed three people. Whereupon Kiti forces descended on Madolenihmw sections 27-28 and left sev? eral people dead, then climbed the mountains across to section 14 and kUled one more. Section 14 now sent Its men to KLiti section 20 and took a life, and MadolenUimw laid waste several small Kiti islands. The Bl of Kiti was now wrought up and sent a message to the Al of Madolenihmw, appomting a formal battle ground off Madolenihmw sections 27-28. He assembled an army from among his own foUowers of Kiti sections 9-11, 17-20, 22, and 25, Many were slaui, and the battie ended in a decisive defeat for MadolenUimw, and "since then no more fighting has occurred" between the two tribes. A kind of treaty was effected, and the Bl of Kiti became overlord of section 18 of Madolenihmw, placing the head of Kiti section 9 in charge of It. Nevertheless, In September to November of 1852, ac? cording to the missionary letters, eleven people were kUled In three skirmishes between the two tribes. Because of a series of thefts, war resumed in October 1854, and con? tinued untU May 1855, with littie parties robbing and murdering almost nightly. A typical report is one of De? cember 1854 when a Kiti fleet of 28 canoes and 160 men attacked, joining an equal number of Kiti men who had gone by land; the net accomplishment was the death of two Madolenihmw women. In the same period there was cIvU war In Uh, fighting between Sokehs and Uh, and some skirmishing between Uh and Madolenihmw, with three "bloody battles" In 1853, Through 1856-59 Uh, or at least that part of Uh known as Awak, was at war with both Sokehs and the northern part of Madolenihmw; the reports typically mention "a woman and a girl klUed," "five people kUled," "two boys kiUed," "sbc men kiUed," etc. "Tmly wonderful" breastworks of stone and watch- towers are described. The outbreaks of fighting in both 1856 and 1858 are attributed to elopement or theft of women married to high chiefs. In 1868 the Lepen Pafikir (Palikir being part of Sokehs), the Bl of Sokehs, and the Bl of Uh formed an aUiance with the Al of Madolenihmw and fighting occurred aU year against the A2 of Madol? enihmw and his party, with what results we are not told. It is of interest that In 1855, 5 years after ICiti had over? thrown Madolenihmw, under whose domination it had previously been, it was stiU rendering tribute to the latter tribe even in the midst of hostUities; and at the annual major religious ceremonies, held In the fourth month, when new canoes that had been built during the year were launched, the two Al's exchanged new canoes, sus? pending warfare when necessary to do so. O'ConneU records that in battie there was strict segre? gation by class, chiefs fighting only against chiefs, com? moners against commoners. Modem mformants know nothing of this but do remember that the Hfe of the enemy Nahnmwarki or Nahnken would always be spared; nevertheless, in the conquest of Kiti proper by Pehleng (described elsewhere), before Wene intervened and united aU Kiti, it is related that the Soukiti, whose position corresponded to that of Nahnmwarki, was kiUed. The usual plan of formal battle was to send a challenge to the opposing side, appointing a time and place for the engagement. But mUitary strategems were not want? ing. There were numerous cases of spying and of ambush. The clan 7 conquerors of Net and Sokehs, according to legend, spied upon the enemy by hiding under taro leaves. In the Kiti wars the Pehleng people were defeated when they massed against an apparent attack by a fleet of canoes only to discover that the sea attack was a feint and the canoes were fiUed with dummy warriors made of coconut leaves, the main Wene forces having come from behind them by land. The legendary account of the over? throw of the Saudeleur dynasty by the hero Isohkelekel, who became the first Nahnmwarki, describes an order of attack Involving groups of men caUed fiteraUy "living water," which informants interpret as reserve units. In Spanish times a clan 13 man klUed the brother of the Nahnmwarki of Madolenihmw, also of clan 13 but of a different sub-clan. He took refuge In a mountain fort. The Nahnmwarki and the Lepen Moar, after planning their strategy, pretended to have a falfing-out with one another and staged a mock battie, during which a few men on each side received fight wounds. The Lepen Moar then retired with his people to the area he governed and caUed upon the murderer to come and help him. When he arrived with his forces, they set out together, ostensibly to attack the Nahnmwarki, but en route through the woods the men of the Lepen Moar seized the murderer, bound him and defivered him to the Nahnmwarki, who had him bumed afive.^ ^ Land was seldom taken by a winning army. When the object of a war was to replace one ruHng clan by another there woiUd be a wholesale shift In the holders of tities, but the only immediate economic effect was that the new Nahnmwarki or Nahnken and such section chiefs as were replaced received offerings instead of the old ones; the greater number of famUies remained on their farmsteads. Occasional shifts of a section from one tribe to another " Another version I recorded attributes the fight to a quarrel between two sisters; the son of one, Oun Sapawas, in revenge killed the A5, son of the other, and it was the A5's brother, the A2, who concerted with the Lepen Moar to take vengeance in this manner. Christian (1899, p. 116) gives an account of what seems to be the same incident but apparently puts it in an earlier period. 62 S M I T H S O N L \ N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 are recorded, but these were more often through peaceful means than by conquest; thus section 32 of Kiti was for a period transferred to Net because of the gratitude of the Al of Wene to the Bl of Net for his help in the war against Pehleng, The simUar transfer for a time of sec? tion 18 of Madolenihmw to Kiti In consequence of the war between those two tribes in the 1850s has already been described. Wars between tribes, according to native theory, re? sulted not from economic causes but from vainglory and pride. A Nahnmwarki "would feel proud when he saw how many people he had and that they were ready to go to war"; he might himself pick a quarrel on sHght pretext. Net and Uh frequently warred for no apparent immediate cause; the members of clan 2, the A-line of Uh, remained bitter after they were ousted by clan 7 from their simUar position in Net and would seize every opportunity to take revenge. The wars between the two tribes are referred to as games of Uh {kousor en Wenik; kousor being a certain violent game; Wenik, the ancient name for Uh, but by some definitions also Including Sokehs, Net, and the part of Madolenihmw called Enimwahn); they were thought of orUy as manly sport. After such a war each side was supposed to be satisfied for the time being; the members of the two tribes would visit one another with large fleets of canoes to exchange the formal apology feast expected after cessation of hostifities, at which times they would behave fike good friends. Sometimes, when a man was given a titie. Instead of making the usual title-payment feast he would prefer to gather his kinsmen and wage war for his overlord in token of his fealty. Paying for a titie in this fashion was con? sidered to be of higher merit than giving a feast. The Nahnmwarki might order the would-be warrior not to set forth, but usually he consented to the war; if so, he might choose the enemy against whom the man would exhibit his valor. OccasionaUy a clan war resulted from a quarrel over titles. It might happen that a title was given to a great war? rior in return for his deeds in battie, after which his clan might come to think of the titie as belonging to it. In Spanish times a clan 7 man of Madolenihmw distinguished himself in fighting against the Spanish; in retum he re? ceived the titie of Karuki of Madolenihmw. When he died, however, the Nahnmwarki Solomon gave the title to his own adopted son, a man of clan 6. The members of clan 7 were incensed. They raided the property of the head of clan 6 at Temwen and destroyed his houses, canoes, and yams; then they fortified themselves under their principal chief, the Lepen Moar, in their own sections. The clan 6 people ranged themselves around the Nahnmwarki at Temwen, and the Nahnken and aU the men of high title came to stay with them. Neither side could approach the other. Finally a number of chiefs, friendly or related to both sides, mediated and some of the leading men of clan 7 were induced to come and make an apology feast to the Nahnmwarki. Though aU was forgiven, Ponape was by this tune under German mle and each side had to send a number of Its partisans to penal servitude in Rabaul for a year. The title of Kaniki was given back to clan 7. Wars between sub-clans of the same clan were not uncommon. One such occurred in Madolenihmw, at some tune after first contact but before 1860, between two sub-clans (the Pahnmei and the InanpaUeng) of clan 13. A woman of clan 9 was married to a man of the second sub-clan but had a lover who belonged to the first; they were surprised together at a place of assignation. He was speared as he fled, but she escaped without punishment. A war then broke out between the two sub-clans and the people of the first sub-clan were driven away to Uh and Net. Prerogatives of Chiefs CONFISCATORY POWERS Aggrandizement and avarice of chiefs seem to have had few checks. A chief who possessed one section but wanted another as weU might simply go to the second section, command the people to prepare a stone oven for him, and proclaim that henceforth the section belonged to him. Such an act was known as Knotted Basket {kiam pwuk; translatable also as Pleading Basket). This term seems to be appficable primarily to cases of dispute in land inheritance, when each applicant sent a basket of coconuts to the Nahnmwarki In token of his claim. The term is also applied when a section voluntarily, or even against the wiU of its chief, transfers its allegiance to an? other chief. But here It is appHed to a case of outright confiscation. The chief of the section supposedly could do nothing to prevent it if the Interloper had a title higher in rank than his. A confiscation of this kind might also be made by the head of a section immediately after he learned that the head of another section had died; haste was necessary, be? cause there might be competition In the seizure. The ex? pression Knotted Basket as used here appHed ordy to a change in ownership between clans, since members of the same clan were not supposed to aggrandize at one another's expense, UsuaUy the change was between the royal and noble clans, and might Involve even the Nahnmwarki and Nahnken themselves. Thus, in Kiti, sec? tions 2 and 13 at one tune belonged to the Nahnmwarki but were taken from him by the Nahnken (see p, 27), Some of the sections that form the part of Kiti known as Lukoap were simUarly taken from their independent heads by the Nahnken, Sections 23, 28, 29, and 32 were 298-818?68 6 once under the Nahnken of Kiti, but whUe (about 1870) a man named Nahnawa en Mwudok was Nahnken he buUt a community home that required a large amount of cordage, then buUt a chief's canoe {warasapw) whose outrigger also took much of this material. The people of these four sections despaired, since they were called upon to defiver the required fiber which Is arduous and tedious to make. They decided to switch their aUegiance to the Nahnmwarki, In this transfer, referred to as a case of Knotted Basket, as in those of sections 2 and 13, there were no difficulties; It Is reported that the two chiefs concemed merely laughed the matter off, since a Nahn? ken is always in the position, actual or fictional, of son to the Nahnmwarki, and there is supposed to be generosity and harmony between them. In Net and Sokehs, which were mled by the same clan, the Al of one tribe might ask the Al of the other for a piece of land, and when he went to take up the claim the second Al would give him a feast, which was also caUed Knotted Basket. But such a thing was possible only be? tween these two tribes, since the others were under dif? ferent clans. Or if a high chief drove someone from a piece of land In order to possess It himself, the tenants of the land involved would make a feast called by this name to him. Knotted Basket Is a term also appHed to confiscations of land in punishment for a misdeed. The banishment of the X6 of section 1 of Net, referred to previously, was an act of Knotted Basket insofar as It Involved reversion of his land to the B1 of Net. A high chief might confiscate any article he wished, simply out of greed, not necessarily as an act of punish- 63 64 S M I T H S O N L \ N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 ment. Such an act was known as kuhl. (In Kiti the term is also used for acts of destmction by a chief,) A chief or his messenger would go into a house and point out things he wanted or go to the community house In a section and demand that various things be brought him. People with fine objects would hide them In the bush or surrender a poor substitute. If they could get away with it, when they heard that such demands were being made. A Nahn? mwarki or Nahnken (but no lesser chief) might send a messenger from house to house with a tiny sponge hang? ing from a strip of hibiscus bast or a smaU shuttie wound with a piece of twine; the messenger would show these symboHc articles to the householder, saying only "com? mand," and the people would have to defiver to the messenger whatever sponges or balls of twine they had. Such a demand for twine Is pohi kepei mwahu {pohn, upon; kepei, thigh; mwahu, good; referring to manufac? ture of twine by roUing coir on the thigh). The Nahnmwarki might also demand other articles, for example, he might require that all the women make sleep? ing mats. He would notify the section heads, who would pass the word on to the women in their sections; the women would begin to gather pandanus leaves In Jan? uary or February, the period of less rain, and present the finished mats to the Nahnmwarki at the end of the year; he would keep the best specimens, distribute others to his higher chiefs, and retum the poorest ones. The Insatiability of the chiefs in their demands on the people in former times is well expressed by the Ponapean proverbs "The chiefs are hoUows of no retum" {Peden seupwur soupeidi) and "In the bottom of the channel there is water of the chiefs" {Kepindau pilen soupeidi; meaning, no matter how low the tide, even if the channels through which canoes must pass be dry, tribute in food and goods must nonetheless constantly flow to the chiefs). Not only the highest chiefs felt free to caU on their subjects to deliver up their possessions to them. The Soukoahng of Net (A3 In the old Net series) of three generations ago used to caU fishermen ashore from where they were fishing in the lagoon and pick out the best canoes, which he would keep for himself. These forms of confiscation were practiced frequently, though the form caUed kuhl was somewhat rarer and was considered more drastic. They are said to have been com? moner in Kiti than elsewhere and persisted into late Spanish or early German times. More recently the term kuhl has been used to mean legal permission given by a Japanese judge to a creditor to seize property In satisfac? tion of a debt. But the arrogance described above was not always tamely brooked. About 60 years ago the Bl of Net, who fived in section 21, came to sections 8 and 9. In each section he went to the community house, seated himself, and demanded that kava be prepared. He drank, then caUed for various objects to be brought hun and took them away with hun, the form of confiscation caUed kuhl. He returned a second tune and repeated his performance. The third tune he came to section 9 he seized the whole section. This strauied the patience of the people beyond the Hmits of sufferance; despite the high status of the Bl, a number of them, led by two men of clan 6, seized his canoe as he sat in it at the water's edge, capsized it, and held him under water untU he was nearly drowned. Though the Bl had four men with tum, they were out? numbered and stood helplessly by. He was aUowed to leave and never retumed to these sections nor exacted any vengeance. The function of the pofitical council as a check on chiefly arrogance has also been mentioned in previous pages. Another view of the power of the chiefs is pro? vided by the Protestant missionaries, writing In 1855-60, after smaUpox had decimated the population, reducing It to perhaps one-half or one-third of Its former size. The chiefs had become disproportionately numerous, since it had been necessary to elevate commoners to the vacated titles. Sturges states that the commoners, hitherto always subject to a chief's caU, never certain of retaining their few possessions and greatly oppressed, had now become Inso? lent and the high chiefs were finding It hard to get men to do any work. Mere boys gave their opinions in pubHc with as much authority as the highest chief. Chiefs be? came careful not to give commands or even to make requests not In conformity with ancient customs or with the active desires of the people. A commoner, if piqued, would escape to another master, where he was made welcome; this losing of men was regarded as disgraceful and kept the rulers almost in bondage to their subjects. Young men, who had been given titles and were anxious to prove themselves, would in fieu of a title-payment feast issue a chaUenge to battle to neighbor tribes, Independ? ently of the wishes of the high chiefs. Subtribes and sections had become virtually independent of the tribe. This was of course an unusual and temporary state of affairs brought about by the epidemic, but Sturges says that the process had been In operation on a more limited scale over many years before?presumably since European contact. And the chiefs never again regained their erstwhUe position of near-absolute power. PREROGATIVES OF CHIEFS ATTENDANTS AND SERVITORS 65 Chiefs Al to A5, A7, Bl and B2 have a number of personal attendants {erir) who receive special titles. These are as follows :^ ^ Al A2 A3 A4 A5 A7 Bl B2 Madolenihmw 1, 2. 3. 4. 5. 1. 2. 3. 1. 2. 3. L Lempwei en Isipau Souwel en Isipau Oaron Mwar Luhen Mwar Oun Mwarki Souwel en Wasai Lempwei en Wasai Oaron Pwutak Oaron Ahu Oaron No Oaron Awa Oaron Kirou Oaron Kin Oun Kin Lepen Kin Oaron Laimw 1. 2. 3. 4. 1. 2. 3. 4. 1. 2. 3. L 2. Mt Mwarekehtik Oaron Mwar Oun Mwarki Luhen Mwar Lempwei en Wasai Souwel en Wasai Oaron Wasai Oaron Pwutak Oaron Ahu Oaron No Oaron Awa Oun Kin Lepen Kin Oaron Kin Oaron Laimw Wahlaimw 1. 2. 3. L 2. 3. 1. 2. 3. 1. 2. Kiti Lempwei en Isipau Mwarekehtik Oaron Mwarki Lempwei en Wasai Oaron Pwutak Oaron Wasai Oaron Ahu Oaron No Oaron Kin Oun Kin Lepen Kin Oaron Laimw Wahlaimw In the other two tribes, Uh and Sokehs, the titles and number of attendants are said to be simUar. Some of these titles are of fairly high rank, sufficientiy high to have been listed in Tables 3 and 4, in the A and B lines. In former days the successor to one of the titles which carried with it the right to attendants would in? herit the same attendants his predecessor had. The main duties of these titled attendants are to function as body? guard, messenger, and poHce, (A Kiti Informant refers to them also as cooks.) As a member of the guard, an at? tendant. If he sees someone standing on the main plat? form of the community house with his head higher than the heads of the highest chiefs, tells him to desist; for? merly he would have thrust him through with a spear, which he always had by him. (But attendants were not the official executioners; these officers were junior clansmates of the Nahnmwarki, whereas his attendants belong to any *" The three lists here given are from informants; the Kiti informant admits that his list is incomplete. The manuscript by Luelen, who was a native of Kiti, lists six attendants for the A l , presumably the Al of Kiti. The first four are the same as those I have listed above under Net (with numbers three and four in reversed order and number three spelled Aun Mwar) ; the fifth is the same as my first as given above under Madolenihmw and Kiti; the sixth is my second under Madolenihmw. The manuscript gives as attendants of the A2 the same four tities I have listed under Net, but in the order 3, 1, 2, 4. For A3 and A4 it gives the same tities as mine. clan but his.^ ^) If one of the workers walks about or idles In the central area of the community house or drinks kava without permission (a kava worker is supposed to drink only at the caU of "dipenkeleu" by the master of the kava ritual), he is seized by the attendants and forced to perform an apology ceremony {tohmw) immediately; the kava already pounded on his stone is thrown away and he is made to supply another kava plant and pound It. This is tme whether it is the Nahnmwarki or the Nahnken to whom disrespect has been shown, but offenses of such nature are looked upon with more severity by the Nahn? mwarki's attendants. The attendants also see to It that no one uses commoner speech before the highest chiefs. Their function as mediators in a quarrel between Al and Bl has already been described. The attendants formerly did no work beyond their various specified duties. Some of them were part of the domestic menage of the chiefs they served. They always accompanied their master when he traveled. Today, some of the titles listed above are purely honorific titles and are given to men who do not perform any of the functions theoreticaUy associated with them; aU of their holders nowadays maintain Independent households. The attendants cannot belong to the clan of the man they serve, since clansmates of a chief are prohibited from undertaking the various famUiarities with him that at? tendants must perform In the course of their duties.^' For example, a clansmate of the Nahnmwarki is not sup? posed to enter his house (although today this prohi? bition Is no longer stringently observed), an act an at? tendant was often required to do. In Madolenihmw, where the Nahnken may be of either clan 2 or clan 9, an attendant of the Nahnken can belong to neither clan, regardless of the clan affUiation of any particular Nahnken, The word used for attendant, erir. Is often extended to those persons who sit before the high chiefs on the main platform of the community house and serve them kava or pass them other objects. Although the titied attendants sometimes fimctlon in this fashion, far more commonly *' In Madolenihmw the royal executioners were three of the Nzihnmwarki's clansmates: The Kroun en Lehdau, the Kulap, and the X I of section 26. 58 The Mwarekehtik, who is A14 or 15, is an apparent exception, but though in the same line as the Al he is considered as the latter's son and has a number of peculiar privileges, such as sole right to remove the kava plant placed in the beams above the Al ' s head in the community house. His tide means literally "little Nahnmwarki." His position in the A-line is probably con? nected with the privilege, previously discussed, of sons of the Al and Bl taking a titie in either line during their early political 66 SMITHSONIAN CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 they do not, and to distinguish the attendants from the kava servers I shaU caU the latter servitors. The servitor who waits on the Nahnmwarki is very often the head of the section where the feast is being held, though he might also be someone else, picked just for the occasion. Lesser chiefs, when they are the highest men of title present, may sit In the place of honor on the main platform and have servitors to sit before them and pass them kava; fikewise with a visitor and whoever else is entitled to sit facing forward on the main platform. When a servitor functions on ceremonial occasions, he sits not In the usual crosslegged manner of men but side? ways. In front of the chief. In the position normally taken by women, that is, with both legs bent at the knees and to one side, and leaning with one shoulder toward the chief, the elbow on the leaning side resting on the knee on that side; the head is averted from the chief and bowed. If a chief has two servitors, they sit facing each other in front of him, one leaning toward him with the right shoulder, the other with the left. The servitors are the only persons not of the highest rank who may hand things directly to the Nahnmwarki or Nahnken, Among the chiefs, Bl to B4 and the sons of men of the Nahnmwarki's sub-clan may hand objects to the Nahnmwarki; and similarly chiefs Al to A4 and sons of men of the Nahnken's sub-clan may serve the Nahnken. In Uh the master of the kava ritual may pass a cup of kava directly to the Nahnmwarki or Nahnken Instead of to a servitor first. The servitor hands the kava or other object to the chief whom he attends with either hand, depending upon which of his shoiUders is leaning in the chief's direction. If his left side inclines toward the chief, as Is commonest when there Is only one servitor, he takes the cup with his left hand from the kava pounder or master of the kava ritual; then vrith his right hand he wipes it clean and passes it on to the chief, who also takes it with his right hand. The chief passes back the drained cup with the same hand with which he received it, and the servitor re? ceives It likewise with the same hand. If the servitor leans with his right side forward, he and the chief use their left hands In giving and receiving. In giving, the servitor supports the hand holding the object with the outer side of the crooked elbow of the other arm; this gesture is characteristic and striking, though often performed in a careless and perfunctory manner. When the chief has dmnk, he thrusts the cup back, without lookuig at the servitor and paying no heed to whether the servitor is prepared to receive It, for the latter Is supposed always to be ready. The chief's wrist Is caught on the crooked elbow of the servitor as he thrusts the cup or other object back, then the object is taken with the opposite hand. (See Plates 7a, 11a, and l i b for servuig position.) Besides attendants and servitors others formed a kind of retinue to the Nahnmwarki and other high chiefs; they were servants and retainers of aU kinds, some of them landless exiles, others tenants or hangers-on. There were enough of these so that the Nahnmwarki and Nahnken constituted a true leisure class, doing no work of any kind; some of the lesser chiefs, depending on personal power, likewise had their retauiers to do all the necessary work. Life In ancient days, by Informants' accounts, was easier and more leisurely for all. Such work as there was to do was mostiy gardening and fishing for the men, household tasks for the women; people today look back on the old days as a time of much feasting and Httle work. So many people were idle that, it is said, the chiefs would put them to work cleaning the forests and there was not a leaf on the ground. Many, for lack of occupation, went to Hve with their relatives of higher rank as retainers and servants. Titied men would also attract unrelated retain? ers from among those who had been banished from other tribes for various crimes. Thus, many lesser chiefs could afford to refuse to collect breadfmit or care for yams; many would not build canoes; no chief would ever paddle a canoe or make a stone oven, and some even today refuse to undertake such menial tasks. The wives of a Nahn? mwarki or Nahnken likewise had little to do; they occupied themselves chiefly with the making of valuable articles, such as sleeping mats or woven belts; the servants pre? pared the food, carried water, fetched wood, and cleaned the house, and the section chiefs brought in first fruits and other offerings daUy, so that all the necessities were pro? vided. But a man of lesser title often had only the work? ers of the section where he lived, and his wives might have more to do. INSIGNIA, DEFERENCE, AND ETIQUETTE Chiefs and rich men formerly wore necklaces {el en pwur, or el en pwul) made of discoid beads of a yeUow- ish or pinkish oyster {pwahke) strung on banana fiber. Fiber kUts {koal) are made nowadays of a variety of ma? terials, but aboriglnaUy they were of coconut leaves; these are first baked In a stone oven, then left ovemight in water to bleach, then dried and shredded with a special tool consisting of a slab of wood, the end of one face of which PREROGATIVES OF CHIEFS 67 is set with needles (formerly shark's teeth) and suspended from a waist cord. This is the dress of commoner men,^ ? but the special chiefs' kUts {koalihkos) were more elabo? rate; each strip of coconut leaf was finely crimped with a certain sea sheU {kommol or kopol). The crimping re? quired the laborious efforts of several women over a pe? riod of about a week. Such crimped kUts are not seen today. Above the kUt was wom a belt of loom-woven banana fiber called dohr (Plates 7b, 11a, l i b ) ; from it hung a number of pendants of the pinlc oyster sheU, shaped Into isosceles trapezoids. Headbands {nihn) were frequently loom-woven of banana fiber or made of tapa. Neither of these articles Is made any longer. (Royal head and neck ornaments, reconstmcted from archeological materials, are shown in Plates 11a and l i b ; their recon? struction was done by and according to the ideas of the B6 of Madolenihmw.) These omaments were probably only induectiy a badge of rank, more du-ectly reflecting the wealth accmed through the offerings that came to a man of titie and the retainers he gathered about him; It Is stated that com? moners could wear them "If they had enough relatives," that Is, became head of an extended family large enough so that there would be ample surplus time for women to spend making the articles. Priests also were eligible to wear them, and in the community house the attendants of the chiefs wore the woven belt. At a tribal assembly the Nahnmwarki was privUeged to wear a headband with four red "spurs," two pointing back and two forward; the chief priest (the B2) also wore one of these, but aU others, including the Nahnken, could have only two such "spurs," A Kiti Informant speaks of bamboo combs wom by chiefs and priests, who wore their hair pUed on top of the head; the chief's comb was red, the priest's comb yeUowish; commoners let their hair hang loose. But combs are denied by several Net Informants as having existed in Ponape; they say that people wore their hair loose or in "bundles," meaning, no doubt, the typical Carolines chignon.?" Drawings of Pona? peans in the Peabody Museum at Salem, made about 1850 or 1860, show both men and women with long hair loose down their backs, but other men with short hair; in this connection GuHck's remark that those with whom the spirits are supposed to hold Intercourse wore long hair is perhaps explanatory. Apparently no special distinctions in dress according to rank were made by women. The chief's house was caUed ihmw en nei. It was con? structed like a dweUing house, the ihmwalap, but Instead of being buUt of mangrove wood it was made of hibiscus wood, with the lower beams of breadfmit wood, and lavishly covered with coconut-twine lashings. In the house of the Nahnmwarki was a special sleeping-room for his use {nanweip). For visiting chiefs a special house was built, caUed ihmw en kinte. Only the ihmwalap persists, but much westernized in form. The houses of high nobles often receive special names. That of Nahnku, who was Nahnken of Kiti until 1864, was caUed "Ponape fears." Those not named are referred to as iehnpas, an honorific used In place of the common name for house ( ihmw); e.g., tehnpas en wasa lapalap for a Nahn? mwarki's house (from his title of address) or tehnpas en nihleng, the house of the present A4 of Kiti in the farm? stead of NUUeng. Certain relatives of the Nahnmwarki and clansmates with low title could not enter his house. The holders of the three highest tities below his in the same line (A2, A3, and A4) could enter his house but not his sleeping chamber. But the Nahnken and other titieholders In the B luie could enter both house and bedroom. And conversely, title- holders In the Nahnken's line, except for B2-B4, could not enter the Nahnken's house but those in the Nahn? mwarki's line might. These restrictions today are defunct. Near the community house and between It and the Nahnmwarki's dwelling house was buUt a Httie chief's house {ihmw en patok); this was shnUar in constmction to an ordinary house of four spans length {ihmwalap pahkis). Here the servants of the Nahnmwarki slept; sometimes It served as a cookhouse for the Nahnmwarki, and old women and widows who were relatives of the Nahnmwarki or his retainers kept his provisions here and prepared his meals. But Its main purpose seems to have been a place of concealment for the Nahnmwarki during preparations for meetings or feasts In the community house. When aU was made ready he would take his place on the main, front platform of the community house. Apparently there was no prohibition agamst the populace seeing the Nahnmwarki; ?^ his waiting ui the smaU house seems simply to have been a means of enhancing his dignity by his abstention from vulgar activities, and his appearance at the appropriate time was a kind of theat? rical demonstration. If the communal activities were In progress at night, the Nahnmwarld would sleep in the ?? However, Kubary (1874, p . 267) reports that in his day com? moners wore hibiscus-bast kilts, while chiefs had kilts made of young leaves of the coconut palm. 8" Hambruch does not mention any comb, but Gulick lists the term koko for it. Bamboo was introduced only in postcontact times, so the comb, if it existed, must have been of other material. " Yet this was not always strictly true. Father Costigan points out that in Madolenihmw, at least, the community house is usually built so that the open front faces east, whence the prevailing wind comes. The reason is said to be so that the smoke from the stone ovens will blow toward the main platform at the rear, where the Nahnmwarki sits, and will envelop him so that he will not be easily visible; damp wood is sometimes placed in the fire to en? hance the effect. See also the remarks later concerning the Nahnmwarki's concealment in the special cabiai on the royal canoe. 68 SMITHSONL\N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 Httie house afterwards, since commoners slept in the com? munity house. The Nahnken and a visiting Nahnmwarki coiUd also make use of this house, but the Nahnken felt no need to hide himself. Men with titles below A2 and B2 could not enter the house at all; A2 could enter if the Nahnlcen was there alone and B2 if the Nahnmwarki was alone, but neither could enter if both of the ranking chiefs were present. Men of lower titles (only as far down as A4 and B2 according to one Informant) could remain out? side and talk to the Nahnmwarki, but commoners might not. The house served also as a place for discussion between Nahnmwarki and Nahnken. Nowadays, the Nahnmwarki has been assigned various governmental functions by successive foreign administrations and his aloofness from the people has perforce been considerably diminished. The commuruty house of a section (the nahs en kou? sapw), which Is constmcted by the head of the section and his people, and that of a tribe (the nahs en wehi), buUt at the command of a Nahnmwarki, are actuaUy the same type of structure and are considered to be for the use of the whole tribe, despite their names. A tribal com? munity house, however, was not buUt in a section but in the tribal capital. Today community houses are usuaUy on privately owned land and men wUl often speak of owning this or that community house even though It may be called by one or another of the names appropriate for the public structures and is used for pubfic purposes for the most part. Under present conditions, with private ownership of land and a money economy, a not always clear distinction between private and pubfic community houses has come into being. In 1947 the Nahnmwarki of Net was buUding a private community house, with the aid of kinsmen, on the land of another man with whom he had made appropriate arrangements; suddenly he decided to convert the building into a tribal structure and summoned community labor from the whole tribe to do the job. There was some resentment until the people learned what the buUding was to be. In the front center of the main platform of the com? munity house was built a Httle stmcture with a wall 2 or 3 feet high; this was caUed kelepap isoh. Behind this the Nahnmwarki and Nahnken sat. The stmcture is not seen nowadays. Formerly the front right corner of the main platform was fenced about with a sImUar wall, made of cloth, within memory of informants, probably of a kind of cane, Saccharum spontaneum {ahlek), anciently. Into the privacy of this room {kepinpwalek) the principal wife of the Nahnmwarki and of the Nahnken (wives of A1-A4 and B1-B4 according to another informant), accompanied each by a female attendant, retired In order to be shielded from the gaze of commoners during men- stmal periods. The seclusion hut for menstruants, widely distributed In the Caroline Islands, Is, apart from this weak manifestation, absent In Ponape. The cane flooring of the budding's platforms continued over this comer too, but underneath it was a pit buUt into the stone foundation, just large enough for a woman to sit over, which was used for disposal of the menstrual tampon of cloth or, more anciently, sponge. Informants differ as to whether the woman remained in the room during meet? ings and feasts or only used it to change the tampon privately. No such provision was made for commoners. The little room continued in nearly every community house untU German times, when boards began to be used for flooring In place of cane and stone foundations were often replaced by piling. The Nahnmwarki's principal wife was not supposed to be seen by anyone but her husband, though this custom was subject to modification; in the community house, for example, she sat In her rightful place in full public view. To some extent the practice appfied also to the secondary wives, depending on such Individual matters as uxorlal jealousy and affection; sometimes one secondary wife would be regarded as highly and treated In the same man? ner by her husband as the principal wife. It was the clan- brethren of the Nahnmwarki who particidarly were not supposed to enter her presence; the death penalty might be inflicted for a violation of this custom. When aU the peo? ple were assembled in the community house and word came that the Nahnmwarki's wife was approaching the caU went out, "AU Sore Eyes {lekerwait ^ )^ go and hide," This term indicated the male members of the clan of the Nahnmwarki and was used only in these particular cir? cumstances. At once they left (one informant says they were chased out) and hid themselves. After she took her seat they might return. AU of this Is conceived of as a form of honor towards her. The missionary Doane, in comment? ing on these customs, ascribes them to the fact that the Nahnmwarki's clansmates were in some sense his equal and so might be tempted to take fiberties with her, SunUarly, no member of the Nahnmwarki's clan was supposed ever to see the Nalmmwarki's wife ui her house. But none of this appfied to the wife of a Nahnken or of any other high chief. Sometimes, on a visit by a Nahnmwarki to a community house in his tribe, he would send his principal wife du-ectiy to a dwelfing house nearby and let a secondary wife sit in 82 Gulick (1872, p. 25) gives a spelling close to this as a term of abuse. Modern informants describe it as a condition in which the lower eyelids droop, exposing the red membrane, and Hahl (1904, p. 27) defines it as an ectropion of the eyelid. Madolenihmw informants say that the expression was used in Kiti, and that in Madolenihmw, when the messenger came to the community house to announce the coming of the Nahnmwarki's wife, he called instead, "Are there any ghosts here? If so, get out," and this was a warning to men not only of the Nahnmwarki's clan but to those of his wife's too. PREROGATIVES OF CHIEFS 69 the pubfic budding m her place. But usuaUy the principal wife did not travel with her husband. The chief's canoe {warasapw) and the ordinary canoe {pahnta) differ princIpaUy in the ornamental lashings and painted and carved decorations used on the former. Both of these might have a platform {poang) of Saccharum spontaneum buUt out on the leeward side as an extension of the board {dinapw) that covers the central part of the huU from gunwale to gunwale. If this platform was added, the canoe would lack the vertical sheer strake {pedilik) that ordinarUy extends above the central portion of the leeward gunwale. Such canoes were caUed warapoang. On this lee platform was buUt a Httle cabui {katauk) where high chiefs hid from the vulgar gaze when travel? ing. The stmcture was shaped like a half-cyfinder, open at each end; the framework consisted of Ixora carolinensis {ketieu) withes bent into shape and thatched over with ivory nut {oahs) or parem palm leaves. When not in use it was stored in the community house, to which It was carried by two men by means of a pole thmst through a loop of fiber fastened to the top of the structure. The chief's canoe, the leeward platform, and the cabin are no longer seen, but middle-aged Informants remember them from their chUdhood, When the canoe in which a Nahnmwarki or Nahnken is traveling reaches the bank he alone may step ashore from the middle of the vessel; aU others must leave from positions closer to either end. When a canoe approaches the shore close to where a high chief is waiting it must not be under saU and aU persons aboard must remain seated until It touches land. A man who Is fishing when a Nahnmwarki approaches In a canoe must desist from his occupation and leave. Boatmen in a canoe meeting a chief's canoe make a wide detour and salute by slackening saU to aUow It to flutter, or, if he is a very high chief, they lower the saU entirely; if the canoe is not under sail but beuig poled, the boatmen sit down. The chief then teUs one of his men to wave the other canoe on. From my own experience such deference Is paid to chiefs as low as A8. A canoe passing a chief's house must likewise wait to be waved on. It is especiaUy the members of the clan of the Nahnmwarki who have to show respect to him; formerly, in passing his canoe or house, they had to jump out into the water, hold on to the gunwales of the canoe, and bow their heads; they stUl bow their heads but since about Spanish times they only pull in the saU and sit down, Clansmates of the Nahnken show simUar deference to him. The character of the respect shown to the Nahnmwarki is indicated by the fact that m his canoe in former times one attendant had the sole duty of sittmg, facmg him, in the same stylized attitude of respect that In the fishing ca? noe one of the fishermen is supposed to exhibit whUe sit? ting and facing the empty seat reserved for NahnuUap, the fishing deity.?^ On a canoe trip, when the usual methods of conceal? ment were not possible, the wives of a high chief had to cover their faces and heads with leaves of a variety of Alocasia macrorrhiza {sepwikin). The reason was not only so that they might maintain their Isolation and aloof? ness from commoners, but also so that young men might not take a fancy to them and seek to come to them by night. Canoes of Madolenihmw are supposed to tum their "small ends" {imwitik) foremost, canoes of Uh their "large ends" (imwalap), when passing a chief. The "smaU end" is considered to be that from the upper part of the tree from which the hull of the canoe was fashioned, the "large end" from the lower part. This custom, according to legend, dates from the time when the first Nahnken of Madolenihmw departed for Uh to become the first Nahn? mwarki of that tribe; his son, who later became Nahnmwarki of Madolenihmw, followed to persuade him to retum, and they took counsel at Pohn Nintok, the reef opposite section 4 of Madolenihmw; then, as they parted again and the father turned his canoe toward Uh, the son toward Madolenihmw, their canoes were oriented accord? ing to the custom which thereupon became established. Befief in the sacredness of the head persists today. Formerly it was taboo, on pain of death at the hands of the executioners, to touch the Nahnmwarki, especiaUy to touch his head or face or even his kUt. Though the pen? alty is nowadays void, the respect attitudes continue. (But at ceremonies in the community house a woman might smear oU over the Nahnmwarki's back and chest, as is the custom.) It was a form of respect not to awaken a high chief except by pulling the tuft of hair on his great toe, the part of his body farthest from his head. Few persons may stand so that their heads are higher than that of the Nahnmwarki; among those who may do so are the men whose duty Involves calling out the stages in kava prepara? tion and in division of food; but these officials must belong to the opposite (B) line of titles. His personal attendants, who are usuaUy of the opposite Hne, and the highest chiefs in that luie also have this privUege. The missionaries, in the 1850s, write complaints that though they would like the people to stand in church during certain prayers, they could not do so if a high chief was present. If a man wishes to cHmb a tall tree near the house of a man of high title, he must first obtain his permission; if a man is in a tree when a high chief comes near, he must climb down. In the 83 Gulick lists in his vocabulary the word keor, defined as "the respect paid the ijipau {isipau, a title of the Al) when he is on a canoe, by one person who does not paddle and who sits facing him; only the ijipau and the chief ani (ghost or god) are thus honored." The custom was thus evidently still followed in the 1850s. 70 SMITHSONLVN CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 community house, if It is necessary to fetch some object suspended in the rafters, when the Nahnmwarki is present, one of the highest B-line chiefs (Bl to B4) or a son of a member of the sub-clan to which the Nahnmwarki be? longs or the Mwarekehtik, who is considered as the Nahnmwarki's son, must cHmb up to get It; no one else may raise his head high enough. A man cannot pass the house of a Nahnmwarki or of a Nahnken but must sit outside in the path, some forty or fifty yards distant, and wait for permission from an attendant to go on, or eke take a detour through the woods; then, when he walks on, he must do so In a stooped posture. This behavior, sakarahl {sakara, to beg of a chief; ahl, path), is some? times exhibited toward men of lower title but It Is not con? sidered obfigatory toward them. In passing a seated chief, a commoner must bend low, and in his presence he is ex? pected to squat with his head at a lower level and in a bowed attitude. When leaving, he must crawl away back? wards for some distance before rising. This crawling and stooping behavior is exhibited also before Europeans. When a man met a chief's wife on the path, he was formerly obfiged to sit down untU she passed (but Infor? mants deny that he hid himself, as Hambmch asserts). As recently as 1925 commoners could not even talk to a Nahnmwarki, and to this day many of them do not ini? tiate conversation with a man of high title. Commoners take a special tone in talking to chiefs, the principal char? acteristic of which is a prolongation of vowel sounds. They should talk only In answer to questions and should make the answers as short as possible. This appfies also to men of high title when they are talking to someone in the same Hne with a stiU higher title; only members of the B-line may speak at length to the Nahnmwarki and only mem? bers of the A-fine to the Nahnken. If a titied man begms a conversation with the usual greeting, the commoner averts his gaze and answers with a prolonged "ehi," In? stead of responding with the same greetuig as to an equal. When the chief comes to a pause in what he Is saying, the commoner mterjects an "ah" in agreement. Children of four or five even today already know enough to snatch off their hats, stoop, and give a prolonged greeting when they meet a high chief on the path. It is said of Nahnku, a Nahnken of Kiti, that he was so powerful that he could go about mvisibly; hence people, when they were in sec? tion 19 of Kiti, where he lived, would bow continually, thinking he might be somewhere near by. A leaflet of "red" coconut whose rib had been removed was often inserted by a commoner under his kUt and when a Nahnmwarki or Nahnken was met or was about to be? gin a conversation the leaf was taken out, the end tom off, and the rest folded Into a sedei (a leaf folded in a pecuHar fashion; the length of the terminal fold foretells the future). At the same time a prayer would be mum? bled. This was less a form of respect than a method of divining, for so strong was the fear of the rulers that a simple encounter was sufficient to cause a commoner to seek what his fortune would be. But not aU people knew how to do this. A commoner may not eat with a Nahnmwarki or Nahnken; If he eats with any other man of superior rank, he must eat slowly, so as not to be sated first. Nor may he finish his drinking coconut first. Any water left In a Nahnmwarki's coconut may not be drunk by commoners or by the Nahnmwarki's clansmates, but only by men of high title in the opposite line down to Leperirin (B7 to B9 In different tribes); the same rule applies to eating rem? nants of his meals, SimUarly, no one may finish the Nahnken's leavings except men of the Nahnmwarki's Hne as far down as Nahnid lapalap (A8 or A9). A commoner could not touch a high chief's gourd or coconut-sheU water bottle; today the prohibition appHes to glass bottles. If a man of high rank stops at an Inferior's house, he must be given a fuU bottie to drink from; more recentiy, glasses or cups, once proscribed, have been aUowed but they must stUl be fidl. In the house of a host or In a com? munity house coconut twine must be tied around the neck of the bottle offered to a Nahnmwarki and twelve leaves of a sort of citms {peren) or of Campnosperma brevipetiolata {dohng) are lashed to the side; four leaves are used as a stopper. For a Nahnken ten mstead of twelve leaves are used. Commoners use all sorts of other leaves as stopper. The Nahnmwarki and Nahnken were carried on fitters when sick, lame, or numbed by overindulgence in kava. Holders of smaUer titles, section heads, and high-titied women were also sometimes carried,?* but this was not an institutionaHzed practice and varied according to individ? ual prestige. The two fitter-bearers had to be sons of men of the same clan as the man they were carrying but them? selves of the opposite chiefly Hne. Chiefs were sometimes carried to feasts even when ui good health. They were also carried astride the back of a servant. Not only prestige and power affected the degree to which a chief's subjects acquiesced in his demands for the observation of the extemal signs of his authority. A former head of section 1 of Net Hved on a high hiU but insisted that his people carry him and his family, seated in a canoe, from his community house on the hiU to the water's edge and back agam whenever he traveled somewhere by water. The men are said to have been so eager to act as porters that they would surround the canoe m throngs and many could barely lay a hand to it. But at least part of the reason for this enthusiastic fealty was that the chief had several fair and unmarried daughters whose favors were bestowed upon his most obedient subjects. 8* But informants deny Hambruch's remark concerning the carrying of the Nahnmwarki's wife to her bath. PREROGATIVES OF CHIEFS 71 People coidd not go to a high titleholder who was a medical practitioner and ask him to come to minister to a sick commoner but had to resort to someone of the commoner class. But a high chief might summon a com? moner to cure him if he wished. Such distinctions are no longer maintained. When a Nahnmwarki or Nahnken Hes gravely Ul, peo? ple of the tribe come In great numbers to pay their re? spects or sunply "to eat," as uiformants put it, and the commoners in the vicinity must furnish food for all of them. A Nahnmwarki who visits another Nahnmwarld is ac? companied by a large fleet of canoes. Kelmw Sapwasap, who ruled over the combined tribes of Net and Sokehs about 1875, used to get foUowers for such expeditions not oiUy from these two tribes but from Uh and aU of Enimwahn in Madolenihmw. When the Bl of Kiti, called Nahnawa en Mwudok, in 1870, went out to a man-of- war that was calling him to account for the burning of a church, some forty canoes foUowed him as a gesture of loyalty. In theory each section of a tribe sends three or four men to go along with the Nahnmwarki on a visit to another tribe, in the canoe which Is reserved in each section for tribal business. Even today, when a great chief goes on a visit, his attendants and tenants wUl shortly foUow him to "protect" their lord. On a visit to section 6 in Madolenihmw, made by the A6 of Kiti In company with the writer, by the second day some eight of his tenants had foUowed. This Is nowadays as much a func? tion of land ownership as of title-holding; the Nahn? mwarki of Kiti today has no large estate and no tenants, and there are fewer therefore to perform this service for him than is deemed proper, in spite of his superiority over lesser wealthier chiefs. When aU five Nahnmwarkis meet, the proper seating arrangement Is with the Nahnmwarki of Madolenihmw in the position of highest honor. In the center of the main platform of the community house, with the Nahnmwarki of Kiti and Uh sitting somewhat forward of him at 45- degree angles and to his left and right respectively. This arrangement Is said to have been decreed by the SouHk of Ant atoU at the time of the conquest of Ponape by Isohkelekel, the first Nahnmwarki, The Nahnmwarki of Net and Sokehs, whose positions date from much later, sit to one side. For commoners, travel between tribes was always dan? gerous, but if a man wished to settle in another tribe and could get his Nahnmwarki to send word ahead that he had permission to move, the Nahnmwarki in the other tribe might teU his own people to regard the newcomer peacefuUy. Some economic specialization between the five tribes existed, and there was traduig to some extent. Sokehs had the reputation of making the best mats, Madoleruhmw the best canoes, and Kiti the loom-woven belts. But such trade was virtuaUy a monopoly of the highest chiefs be? cause of the constant danger in travel. A commoner crossing a boundary was safe only if sent by a high chief. He v/ould carry as a sign of his mission a stalk of sugar cane or a kava plant to the Nahnmwarki of the tribe he was visiting. Men holding titles lower than A4 or B2 could trade only with their relatives In other tribes. Despite the risks Involved, visits between Nahn? mwarkis do not seem to have been rare. Finsch records the mutual avoidance between two Nahnmarkis when both visited his ship about 1880, but this might only have re? flected recent hostUities. Royal visits, however, were al? ways made in considerable force, owing to the precarious state of peace that prevaUed in the intervals between wars. On such occasions (known as seiloak) a number of un? married girls would be chosen by the Nahnken from both chiefly fines and made avaUable for the large company of chiefs and retainers accompanying the visiting Nahn? mwarki, and valuable articles would be presented. On the inevitable reciprocal visit a few days later equivalent gifts were returned. Such articles were sewed pandanus sleep? ing mats, decoratively wrapped baUs and cylinders of twine, canoes, baUers, sponges, loom-woven banana-fiber sashes, chief's kUts, besides the baskets of food always presented at feasts. The assumption of the frequency of visits is in part based on the fact that some balls of twine stUl In existence are said to have circulated around the Island in this way many times. The legend of the faU of the fine of monarchs which preceded the Nahnmwarkis attributes It, in part, to the sorcery of the XI of section 21 of ICiti, a powerful priest, who had visited the last of these rulers but had been In? sulted at receiving only some coconuts instead of the royal gifts he felt he had a right to expect. The majesty of the bearer of a high title could be affected by individual variabUIty in personality traits or by status not derived from descent, as a number of bio? graphical incidents Indicate. Solomon, who later became Nahnmwarki of Madolenihmw, was appointed A2 in Spanish times at the age of 13. At a feast given to him in section 6 he played with other lads, got into a fight and was soundly beaten, and retumed to his place on the main platform weeping bitterly. The spectacle of this small boy, the third chief of the land and the future su? preme mler, sitting In the place of honor and sniveling because of a thrashing, stUl tickles the risIblHtles of in? formants who describe the Incident today. The pHght of the Bl of Net, described previously, who was nearly drowned when his arrogance in demanding tribute and In confiscating land exhausted the patience of his sub? jects, is another Ccise in point. Nevertheless, deference to a chief was often carried to remarkable lengths. It Is told of the renowned Nahnawa 72 S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 en Mwudok, who became Nahnken of Kiti in 1864, that his runaway wife, who had eloped with a lover, was brought back to him whUe he lay In a drunken stupor; when he was awakened, he plunged a knife Into her breast. Her parents were his servants and were among the wit? nesses of the murder, yet did not dare to Interefere.?^ A high chief was formerly referred to after his death by a special burial name {edenpwel). Ponapean personal names, which are often derived from legend, are regarded as secret; it is considered as an Insult for a man to be addressed by It, and the proper form of address is by use of his titie, or by his special title of address, if he has one, even by his wife. Even a woman should be addressed by the feminine form of her husband's title. (Nowadays the personal name of legendary orlgui Is supplemented or often supplanted by baptismal names of nonnatlve origin, which are not held to be secret.) In the case of a high chief, the secrecy Involved amounted to a taboo. During his lifetime his name was not to be spoken; In those cases when it was also the name of a common object or, later, the introduced name of a foreign object, a new term had to be devised to refer to the article, for use in his tribe; the older term continued In the other tribes, which ac? counts for some of the dialectal variations that exist In Ponape, Names of many long-departed chiefs are not remembered, but these chiefs are spoken of today by their burial names or sometimes by a title they held at some tune, A Nahnken of Kiti who died in 1864 is not re? caUed by his native name even by his own grandson (though Luelen's manuscript refers to him as Solomon), but he is known as Nahnku, a titie he held ui his youth. His successor is also recaUed by a title he once held, Nahnawa en Mwudok. Usually, however, the burial name is used. This consists of the prefix Luhken (Luhk, the name of a god) foUowed by a descriptive suffix; for example: Luhkenlengsihr {leng, heaven; sihr, the dart-game known in Polynesia as teka, similar to North American snowsnake) Luhkensohpur {soh, negative; pur, return; i.e., died a hero s death in war) Luhkenmelmel {melmel, typhoon) Luhkensakau (sakau, kava; i.e., drank much kava) The name of a chief's widow takes the same suffix as that of her husband, but the prefix is Luhmo, A com? moner's burial name is prefixed by Nalang. The dead chUd of a Nahnmwarki or Nahnken, but of no other chief, is given a title before burial and the title is considered buried with him and may not be given out again untU the father has died. This applies even to a stiU birth. In November 1947 a dead chUd was born to the wife of Max, Nahnmwarki of Net; it was given the title of Nahnsahu of Net, which may not be reissued untU Max is dead. To announce the death of a Nahnmwarki tritons were blown In quick, short blasts, simUar to those made by a war party; In contrast to this the death of a lesser man was signaled by long blasts, like those sounded by fishermen re? turning from fishing with a new net. A slmUar class dis? tinction was made with the playing of the drum, though Its exact nature has been forgotten. Scattered over the island are a large number of stone burial chambers {lolong) of variable construction; this Is quite apart from the spectacular stmctures on the artifi? cial islands off Madolenihmw and elsewhere, A number of alternate mortuary usages seem to have prevaUed simul? taneously; thus suspension in canoes, earth burial wrapped in mats, and placing in stone famUy vaults are aU de? scribed. But it may be generaUzed that the stone struc? tures were used primarily for chiefs, whereas commoners were buried in earth graves. In Wene, at least, low, cairn- like structures (Plate lb) were in use for the priest-king, Soukise, in combination with secondary burial; burial was done at night in secrecy at Ninlepwel in Section 7; later the stone stmcture, regarded as sacred, was built and the bones placed therein. MARRIAGE AND SEXUAL PRIVILEGES In contrast to the prevaUing rule of matrUocal residence the sons of men of highest rank, perhaps as far down as A4 and B4, brought their wives home with them. If such a man married a woman whose father had a title higher than that of his father, however, residence remained ma? trUocal ; and If the two fathers were of approximately equal 85 Hambruch (1932 I I , pp. 74-75), in giving a somewhat differ? ent version of this, makes it appear that two similar incidents involving two different wronged husbands occurred. rank, the married chUdren lived for indefinite periods in either household, staying longest with the more prosperous one. If marriage was outside of the tribe it was usually patrUocal, regardless of the relative rank of the spouses, Hambruch states (1932 II, p. 148) that around the dwell? ing house stood the houses of a man's sons, but this would ordinarily be true only for the house of a high chief. WhUe a Nahnmwarki or Nahnken would often have 10 or more wives, fewer lesser tribal or section chiefs had PREROGATIVES OF CHIEFS 73 more than two. An informant past 90 years of age could not remember any polygynous commoners, and some in? formants deny that a commoner could have plural wives. Keimw Sapwasap, a Nahnmwarki of Sokehs, Is said to have had some 30 women in his harem. The first wife mar? ried, the inenmwohd, was the principal wife; aU the others were caUed pekehi. In addition, a chief would have sexual access to the female servants, the lidu.^^ Each wife of a chief received a title, (See p. 47,) The first wife of a Nahnmwarki was Nahnalek; the sec? ond Ked; the third Nahnte, The wives of a Nahnken were simUarly Nahnkeniei, Karekin, and Emekin, For any additional wives other tities were employed. Under present monogamous conditions the titles Nahnalek and Nahnkeniei persist as forms of address for the wives of a Nahnmwarki and a Nahnken respectively. The Lepen Net, who was Al of Net untU German times, when he was granted the titie of Nahnmwarki In order to conform to the practice in the other tribes, gave his principal wife the titie of Lempein, At least the later of the rulers who had the titie Lepen Net had the additional title Soulik en Daun, and to correspond to this title a second wife took the title Kedinlik en Daun; to his third title, Soumadau en Eirike, corresponded that of his third wife, Kedin- madau en Eirike, and for his fourth title, SouHk en Ais, his fourth wife had the title Kedinlik en Ais. These were his four principal titles; if he took more wives, additional titles were taken by him specificaUy for each marriage, and the wife received the feminine cognate form thereof. TheoreticaUy, at least. In Net each wife of the Lepen Net belonged to a different clan, the principal wife be? longing always to the clan of the B-line. This was in order to reinforce the pofitical authority of the Lepen Net, since several clans were thus Imked to him and were bound to support and revere him. He also enhanced his economic position, for members of those clans were ex? pected to bring him goods beyond normal requirements. The Lempem had to be of clan 9; the Kedinfik en Daun was of clan 6; these two clans have been the B-clans of Net. Clan aflSHatlon of the other wives was apparently not fixed. Elsewhere than Net a chief seems usually to have taken aU of his wives from the opposite chiefly Hne. Luhkenkidu (a burial name), who was a Nahnmwarki of Madol? enUimw, had ten wives, aU of them m clan 9 and aU closely related, A Lepen Pafikir had as wives two sisters, belonging to clan 4, Hezekiah, B2 of MadolenUimw in German times, who was one of the last polygynists, had three wives who were sisters; but they were of clan 18 instead of the A-clan. The husband lived In one house, the secondary wives in another close by. It is not clear where the principal 88 One informant denies such access unless the chief was un? married. wife fived; most Net Informants state that she Hved In the house with the husband whUe the secondary wives Hved apart with an old woman to guard them; but Madolenihmw informants agree that all the wives Hved together and apart from the husband. In at least one royal famUy (that of Keimw Sapwasap, a Nahnmwarki of Sokehs), the most recent acquisition In the harem was set up with her servants In a separate house of her own. In any event, the principal wife mled the harem. Her designation as inenmwohd, HteraUy "sittmg mother," re? veals her status. When her husband was away, she stayed at home and could not go about; but the secondary wives might travel with him. She did little work, whereas the secondary wives, unless there were enough servants, did such tasks as making kUts, sewing sleeping mats, and weaving belts, and sometimes they would cook, fetch wood and water, and clear grass away from the viciruty of the house. Though a high chief had considerable sexual freedom, many informants state that he had to obtain the permission of his principal wife to sleep with one of the other wives or, according to one Informant, to sum? mon any woman for an extramarital affair. If the first wife was a commoner, such permission does not seem to have been necessary and his sexual freedom was therefore greater. A man of high title had the right to demand any woman he desired (although informants in Net say that the right was restricted to the Nahnmwarki and Nahnken). This right is caUed klasopa; commoners, who did not have the right, were referred to reciprocaUy as klapata. It did not matter whether the girl was married or unmarried; her famUy or husband would be punished if she demurred. The husband could do nothing to pre? vent the affair; he might receive a present from the chief, but this was not obfigatory, A Nahnmwarki would simply send his attendants to fetch the woman. She would be kept at the pleasure of the chief and would be expected to massage and delouse him, pluck gray hairs from his head and otherwise attend him, as weU as to sleep with him. She was not anointed as in a regular marriage cere? mony, unless he Intended to add her to his harem, A Nahnmwarki, up to the time of Paul of Kdti, could even take a woman of his own clan, even a paraUel cousin, and none could say him nay, whereas incest meant death to a commoner. Moreover, there could be no joking at or ridicule of such behavior. The saying goes, "See but don't say . . . the chiefs" {Kilang seupwa?soupeidi), mean? ing essentially for a commoner, "No matter what you see or know about the chiefs, keep your mouth shut." An? other proverb is "Wickedness of the chiefs" {Sakanakan soupeidi); this is explained to mean that regardless of how reprehensible a chief's actions may be his high posi? tion excuses him. These royal prerogatives are ancient, for the Saudeleurs who preceded the Nahnmwarkis are said 74 SMITHSONL\N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 to have sent their chief fieutenant, the Soukampwul, to bring them wives of other men. The perpetrator of a rape would be chlded or punished by one of the highest chiefs, because It was a chief's pre? rogative alone to force a woman to fie with him against her wUl, A woman of somewhat higher rank than her legal hus? band had sexual privileges similar to those of the high chiefs and could summon any man she liked to come and be her lover. The sister of a Nahnmwarki who was mar? ried to B3 or lower, and the sister of a Nahnken married to A5 or lower, could avaU themselves of this right to take lovers Indiscriminately; but faithfulness was required of a woman whose husband was of equal or higher rank. The class or clan or marital status of the lover was of no consequence, and his tenure as paramour of the woman was entirely dependent on her whim. She would usuaUy notify the wife of the man who had taken her fancy to perfume and bedeck him and then to send him along to her. The degree of sexual freedom permitted to a woman thus seems to have varied roughly Inversely with the rank of her husband, A man of lower rank than his wife could apparently do little to halt her affairs, but he himself would be "thrown away" if he were unfaithful. The use of "higher rank" in this connection should not obscure the previously noted fact that a man of lower rank than his wife receives a title "for" her; he is called by the title she had before marriage, or by one that has been held In reserve for her future husband, but his acqui? sition of the title through marriage does not raise his rank to the level of hers. Her rank is thought of as being on the level of the titles of her brothers. Bascom states that royal and noble men and women married to commoners did not have to observe the forms of courtesy ordinarUy due to high chiefs, but the writer's Informants deny this. Only Royal Children and the few exceptions such as the Mwarekehtik already noted have the right to famiHar behavior, regardless of the status of their spouses. The adulterous wife of a Nahnmwarki or Nahnken was. In theory at least, put to death, along with her lover. The story of the Bl of Kiti who put his runaway wife to death has been related; her lover had already been slain by the wronged husband's attendants. Nevertheless, the fact that a chief's secondary wives fived ui a house separate from his tempted many a young man to take the risks involved, and It is said that few members of the harem remained chaste, A man of lower status who was cuckolded could kill his wife only if he thought he could get away with It, for he had to thuik about revenge by her relatives; most often he had to be satisfied with beating or otherwise punishing her; but he would usually make an attempt to klU her seducer. Ordinarily It was the Nahnwarki who had the unfaith? ful wife of a Nahnken killed, and the latter took upon him? self the reciprocal duty for the Nahnmwarki; they would delegate the actual execution to a lesser chief of their own lines. Hambruch (1932 II, p. 75) says that it was the clan of the woman who took revenge for a seduction, but pres? ent-day informants insist it was the clan of the husband; contrary to Hambmch, the seducer did not escape per? sonal retribution, but suffered along with his brothers and clan chief. A high chief could forbid a pretty young girl to marry, but might take her Into his house to be raised by his prin? cipal wife untU she was 13 or 14, then he would make a secondary wife of her. There were even cases of kidnaping. Widows and divorced wives of high chiefs are referred to as rohng ^'^ en soupeidi or karohng en soupeidi, by which term Is understood a prohibition to remarry. A widow of a man of high title is also known as rohng en (the dead man's titie). Hambmch, Finsch, and Kubary aU state that the prohibition was absolute. However, the pro? hibition seems actuaUy to have applied only to remarriage to someone outside the clan of her dead husband. That the clan of the dead husband retained rights to the widow is seen In the occasional use of the expression rohng en (the dead man's clan) to designate her. The brother or the sister's son or sometimes a more remote clansmate of a dead Nahnmwarki or other high chief could take such a woman as his vrife, the brother having first choice, regardless of his previous marital status; in fact, it was considered better to marry the widow than to require her to remain unmarried, lest she enter mto a fiaison with a commoner, (A commoner, however, could marry his dead brother's wife only If he did not already have a wife.) Usually only old widows were left to remain unmarried. If a member of a clan other than that of the dead husband took the widow, he was, withm the memory of Hvuig Informants, beaten up or cut with knives, then banished; more ancientiy, he might be kUled. The pro? hibition was rigidly enforced by tiie brothers, sister's sons, and successor In office of the dead man, for the widow was considered to be property of their sub-clan. Some informants Hmit the term rohng to widow of the three highest chiefs of each Hne, but extend It also to widows of men with lesser tities if they belong to the same sub-clan as the Nahnmwarki or Nahnken. Sometimes the prohibition was circumvented by a suitor of the widow by giving a feast to the clansmates of tiie dead man and by payment to them of such property as baUs of sennit, sleepmg mats, and other valuables. Some informants state that secondary wives of a dead man, unlUce the pruicipal wife, were ordinarily not con- ^ 87 The primary meaning of this word is "to hear" and apparently is used here figuratively, for no scandal was supposed to be heard about such a woman. PREROGATIVES OF CHIEFS 75 sidered as rohng; it was only a secondary wife who was dearly loved by her husband, one who was always taken about with him on voyages and to feasts, who would be held as rohng and prohibited from remarrying after his death. This view may be correct, smce I recorded genealo? gies which show several cases, three and four generations ago, of remarriage of a secondary wife to a man of a clan different from that of her first husband. However, the missionary Doane, writuig m 1870, sug? gests otherwise, at least for divorced women, if not widows, Ul his tune. The "kmg" of Sokehs had kidnaped a 12- year-old girl and added her to his harem. The captaui of the USS Jamestown, visituig Ponape in that year, at the behest of the missionaries forced her release and she re? turned to her famUy, But when the ship left there were attempts on her life because, as Doane puts it, it was a great sin for her to be free and mingle with the people, having once been claimed by the king. In the same year Doane writes of a high chief who had a principal wife and four concubines; he wanted to become a Christian and to have a church marriage to his principal wife, but Doane Insisted that he must first put aside his concubines. To which the chief repHed that if a woman once married or mistress to a chief were to be allowed to appear in pub? lic or to get remarried the people would not tolerate it; such an offense might be punished by death. Sexual activity with a rohng apart from remarriage had to be carried on very secretly; If it were discovered, the chief of her dead husband's sub-clan woiUd bum down the mean's house and chop dowTi his bananas, kava, and other plants. A commoner's widow was known as liohdi instead of rohng. She was free to marry as she pleased and was equaUy free to engage in amorous adventures. It was con? sidered best for a prospective husband to obtain permis? sion to marry her from the members of the sub-clan of the dead man, but it was not always done, and Is not done at aU today. Occasionally a commoner might let it be known that his brother's widow was to remain unmarried, but his wishes would be respected only if he were generaUy feared. The usages associated with rohng have to some extent persisted to present times. The widow of the Nahnmwarki Francisco of Sokehs remained unmarried untU her death; the widows of Luis, Nahnmwarki of Sokehs, and of Solomon, Nahnmwarki of Madolenihmw, remain unmar? ried today. The principal widow of the last Lepen Net married his clansmate and classificatory brother, Satur? fino, who later became Nahnmwarki. The widow of the last Nahnmwarki of Uh has married the present Nahn? mwarki, Edmundo. Informants say that probably no punishment would be Infficted today on a man who married a rohng but that a feast of propitiation would have to be made by the new husband to chiefs A1-A4 and Bl if she were a Nahn? mwarki's widow, to B1-B4 and Al if she were a Nahn? ken's widow. No actual remarriage of a Nahnmwarki's or Nahnken's widow out of her husband's clan is known, but the divorced wife of the present Nahnmwarki of Madolenihmw, who is rohng by definition, has so re? married without any payments or atonement being exacted. On the death of a chief the widow and chUdren formerly would take their belongings and steal off that same night to her f amUy, for fear that the common people would come and despoU them of everything they owned. Native in? formants in describing these behaviors appear to accept them as perfectiy natural; people, they say, woiUd obvi? ously be jealous of the power of a chiefs wife, and as soon as opportunity afforded, they would give vent to their spite. Apparently the dead man's brothers did not feel the same protective interest in her property that they did in her person. Prestige Competition COMPETITION FOR TITLES We have untU now spoken as though promotions and succession in the titie hierarchy were regulated mainly by the principle of clan seniority. We have discussed the ex? ceptions to this principle and described how a man might, through unusual merit, warfike deeds, or advantageous marriage, overcome the handicaps of birth and achieve a high position regardless of commoner origin. But a significant omission in this presentation remains, namely, the acquisition of tities through prestige competition. The royal and noble clans tend to monopolize the higher tribal tities, so much so that we have used the term A-clzm to indicate the royal clan which holds the highest A-tltles and the term B-clan for the noble clan which holds the highest B-tltles, Only an occasional commoner re? ceives a title in the first 12 of each series. But there are many more than 12 titles. In Net, which has a total male population of some 450, a list of 210 issued tribal titles was collected, and there are undoubtedly more. If we do not count boys and very young men, since few of them out? side the ruling clans would have tribal titles, it is clear that the majority of mature men must possess tribal titles and that among them must be a large number of commoners. Informants state that formerly there were fewer tribal titles, that new ones have been Invented and issued in or? der that the chiefs might profit by the title-payment feasts, and that most commoners In times past had only section titles. Nevertheless, a significant number of commoners has always achieved titles in the two tribal series; and in such achievement the principle of clan seniority cannot operate, for the commoner clans, though some are held to be worthier than others, have no title series considered to belong exclusively to them. Though a commoner nor? maUy has a higher title than another man who is his junior in blood in the same clan, he has no particular status in 76 relation to a man in another commoner clan, except with reference to the titie each of them holds. The status of a man of high title is for the most part ascribed, for with the exceptions already noted it comes to him In the main through birth. But the status of a holder of one of the lesser titles is largely achieved, for he gains his title through cer? tain types of activities, in competition for prestige against his feUow commoners. It is through such competition that merit for promotion is judged?the judges being primarily the Nahnken (for tribal titles) and the section head (for section tities). Promotions come about in part through bringing to feasts for presentation to chiefs larger and better and more frequent food offerings than other men, thus demon? strating industry, abUIty, loyalty, and affection toward the chiefs. But more important than presentations at feasts are the direct offerings of first fruits {nohpwei) and the occasional gifts of food between first fruits (kaiak or uhmw en kaiak). AU of these types of presentation are known as Service {uhpa, fiteraUy "to stand under? neath"). Perhaps even more Important than regular of? ferings is the bringing of a valuable article, such as the fermented contents of a large breadfruit pit, to the Nahnmwarki on a special occasion, as when an important visitor comes from another tribe and the Nahnmwarki wishes to make some display at a feast in the visitor's honor, AU of these acts are stored in the memories of the chiefs (in Net nowadays recorded In writing) and duly rewarded when vacant titles arise. A man who has a tribal title may perform Service direct to the Nahnmwarki and Nahnken and thus gain prestige. A man who has only a section titie, unless it Is that of the head of the section, cannot do this; but he can present articles to the head or to any man with a tribal PRESTIGE COMPETITION 77 titie who, he sees, lacks enough goods to offer to the Nahnmwarki or Nahnken at the tune of a feast; the recip? ient then presents the goods as his own but in the course of the feast praises his benefactor to the Nahnmwarki or Nahnken. The section head also reports to the Nahn? mwarki and Nahnken on the quality of performance at Service of his various subjects. Activities of aU sorts performed for and on behalf of the Nahnmwarki, mcludmg communal labor, proper expression of obedience, etiquette, deference and the vari? ous forms of Service, but not mcludmg warfare, are caUed Great Work (taulap). AU of these things are the due of the Nahnmwarki; In native theory promotions are eamed not so much for gifts to the Nahnmwarki as for looking after the property of the tribe; all valuables? large yams, large kava bushes, fine mats, etc.?belong to the Nahnmwarki, and offerings made on specified and other occasions are not tmly presents but merely defiver- Ing up, when called upon or when occasion requires, what is rightfuUy the Nahnmwarki's. Warfare on behalf of the Nahnmwarki constituted a second type of activity. It was caUed Littie Work {tautik). Both of these activities are considered in making decisions about promotions though warfare, of course, is now a thing of the past. Great Work is regarded as everyday, humdmm work and easy; Little Work though brief in duration as diffi? cult. Some Informants say that Great Work counted most In Kiti while Little Work was more Important In Ma? dolenihmw, but others assert that Little Work counted most everywhere. The hteral meanings of the terms make it difficult to evaluate these opinions; probably the suf? fixes, great and fittie respectively, refer not to the relative worth of the two types of service but to the period of their duration. Little Work, In any case, produced the more spectacular results. A man might be jumped from a low title to one as high as A3 ?* or B3 through valiant deeds, if he were a commoner; or up to A2 or B2 if he belonged to the proper clan. There were even instances when a successful war leader, upon retum from battie, dethroned the Nahnmwarki or Nahnken and took over the office. One informant includes under the term Littie Work the division by chiefs of the food offerings made to them at feasts and redistribution among the people, but several others deny the application of the term to this custom. (Such retum is more properly kepin koanoat. Koanoat, as discussed elsewhere, is the special honorific for food which Is appficable to the Nahnmwarki and some other chiefs, but it Is also extended to redistribution of feast goods as weU as to any acts of generosity by the Nahn? mwarki, such as the giving out of tities, land, or anything else, mcluding a daughter to a noble husband, in retum for Service.) The war pattern has already been discussed. Remain? ing to be discussed is Great Work, and more particularly Service, as expressed through first fruits, feasts, and other presentations. FIRST FRUITS First fmits offerings to chiefs, as distinguished from food offerings made at the regular feasts, are caUed nohpwei. Since some types of food are offered more than once, orUy the first of these is strictly speaking, first fruits, and this particular nohpwei is caUed mwohn dipwisou. After the first fruits offering for any plant food is made, the com? moners are free themselves to partake of that food and any variety of It. In general, first fruits are offered to the Nahnmwarki, Nahnken, and the section head, and unless otherwise indicated the term "chief" as used in the descriptions of first fruits to foUow applies to each of these three. But if a man of higher title than the section head lives In a particular section. It is he Instead of the head who has the right to receive the first fruits, and the section head Is sometimes Ignored in this regard. People who Hve In the same section as the Nahnmwarki or Nahnken need offer only two first fmits for a particular product, one to the Nahnmwarki and one to the Nahnken, and need not offer a third to the section head as other people should do. Many Informants state that formerly only the sections belonging to the Nahnmwarki gave first fruits to him and only those belonging to the Nahnken gave to him; other informants deny this. This difference of opinion is no doubt due to the changes effected by successive European Influences. The German land deeds which, in giving the lands to individual owners, authorized the giving of an honor feast annuaUy to the Nahnmwarki failed to men? tion the Nahnken, thereby affecting the pattern of food offering. In Net today, for example, both Nahnmwarki and Nahnken receive first fruits from those who hold to the old customs regardless of where they Hve, but many give only to one or the other, and some to neither. The concept is often verbalized that first fruits are a rental for the use of the land, and since the chiefs no longer own the land many people consider that no rent is due. A5, according to some. 78 SMITHSONIAN CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 The various tribes were formerly divided into pofitical subdivisions larger than the sections and some of the chiefs who headed these areas seem also to have been entitled to food offerings. Nowadays, large landholders, such as chief A6 of Kiti, receive first fmits regularly from their tenants, whUe the Al and Bl of Kiti, having only a smaU amount of land of their own, receive first fruits only in times of plenty. The formal presentation of offerings to the Nahn? mwarld and Nahnken Is not made directiy by household heads but through the section head, who receives such offerings and transmits them appropriately, retaining those that have been made to him directly. Anciently, failure to present first fruits resulted in loss of titie and land, as well as banishment, but today there is no compulsion, although titles are stiU taken away. But It was acceptable for a man who had nothing of a par? ticular food to offer to present another food as a symboHc substitute. Those who retain the old customs stUl offer some of the first fruits, particularly those described under the names kehmei, lihli, idihd, and kotekehp; but others have fallen into complete desuetude. It is undoubtedly correct to say that only a minority of Ponapeans nowa? days adhere to the old practices. One semiacculturated informant who used to offer first fruits but has given it up argues that his title is a low one, hence he gets little back from the chiefs when yams are redistributed at feasts and he therefore looks upon contributions to the chiefs as an economic loss to him; this attitude Is to be directly linked with the money economy that has been introduced Into Ponape since the missionary and whaHng days. Another man states that his mother bought the land he has now, he did not get it In fief from the chief In the manner practiced before the issuance of the deeds In German times, hence he feels no compulsion to pay In kind for the use of the land. The first fruits {nohpwei) are to be distinguished from the feasts {kamadipw) in that they are simple presenta? tions of food, whereas the latter are more formal occa? sions. Where cooking is involved, the objects offered as first fruits are usuaUy prepared by each man in his pri? vately ovmed, famUy cookhouse and presented to the chief at the chief's house; whereas feasts are held in the commumty house, where the food is communally cooked. A number of other differences are described later. The various first fruits occasions by name are as foUows: 1, Yams {kehp): a, kotekehp {kote, to cut; kehp, yam). This is the first of the first fruits of the yam season {isol). The vines are cut from the immature yam, which is then baked in the family cookhouse. The yam must come from a new garden. The whole yam must be brought, undivided, to the chief. Any variety of yam wIU do except "southern yams" {kehpineir). Commoners present what they can afford; a section head presents to the Nahn? mwarki and Nahnken a definite number of yams, which varies from one tribe to another; In Net It Is five. Supposed to occur in October. b. idihd (to grate). When the yams are mature, two or three are dug up, skinned, and grated on a piece of tin (formerly on a rough stone found In salt water), then mixed with coconut cream. Grated yam, when prepared for private consump? tion, is tied up into a leaf of Cyrtosperma {mwahng) to make a loaf and baked in the cook? house; the food is then called Bundle {koruk). But for presentation to the chief it is baked in a banana leaf, when it is called raisok. Both may also be caUed idihd, the name of the first fmit. The food is normally presented in a basket made of a section of coconut leaf (a kiam). A com? moner may present a smaU basket, but a section head presents one of three sizes, containing (In Net) five, eight, or 11 baked loaves; or sometimes. Instead, he presents a basket made of an entire coconut leaf (a pahini). Sometimes the dish Is made of Cyrtosperma, bananas (of any kind ex? cept utuniap or karat), arrowroot, or manioc; or bananas may be mixed with manioc or yams, when the dish is called repwrepw. This first fruit presentation comes In December. c. deulimau {deu, to fiU; limau, five). This also consists of grated yams baked In the cookhouse, but It Is wrapped in Cyrtosperma leaves and is always presented in five loaves, carried to the chief by two bearers in a long basket slung under a pole. This first fmit was never obligatory, but Wcis made in February by anyone who could afford it, 2. Breadfmit {mahiOT mei): The various first fmits Involving breadfmit are col? lectively caUed karihmei or mwohnmei (the latter is also the name of a particular one of these first fruits). a. karisimei {karis, to pluck; mei, breadfmit). Green breadfmit is picked with the stem and leaves attached, baked In the cookhouse, and taken to the chief. This has not been observed for about 15 years in Net, Occurs in early May,^^ b, kehmei (from ka, to bite or chew?). A type of breadfruit containing seeds {meikohl) is used. Baked ui the cookhouse. May, June, and July. ?? Hambruch (1932 II, p. 227) records this in April, but regards it as a feast, not a first fruit. PRESTIGE COMPETITION 79 c. mwohnmei {mwoh, first). Mature breadfruit baked and taken to the chief. Late May through July. d. lihli or mwohlihli {mwoh, first). This is the most important first fmit involving breadfmit. The people of an entire section bring great quantities of breadfmit to the community house, where the Nahnmwarki is in attendance, and the fmit is there baked and pounded, and coconut cream squeezed over It. This first fruit partakes also of the nature of a feast, since it involves pigs and kava as weU as the community house. Held in July and occasionally later, (See further remarks on lihli, p, 80), e. uhmw en pahini or dokapahini, names used in Net; probably the same as dokemei in ICiti (uhmw, stone oven; dok, to spear or punch a hole; pahini, the whole coconut leaf). A hole is punched in the eye of the breadfmit, water and sometimes a Macaranga kanehirae {ahpwid) leaf put in the hole, and the fruit left ovemight to ripen artlfi- ciaUy, Then it is baked in the cookhouse or oc? casionaUy in the community house, A long basket is made from the whole coconut leaf and some 30 to 50 baked breadfmit are carried In It to the chief at his house. Not done by a section but by every man who Is rich enough. Occurs In July and August. f. uhmw en inihn {uhmw, stone oven; inihn, to roast on a fire). UntU this first fruit is presented, breadfmit must always be baked in the stone oven, as they are on this occasion too, but after this they may be roasted over an open fire. The breadfruit are prepared in the community house in the presence of the chief. g. song en mar, name used In Net; same as pokolop- won In Kiti {song, to taste; mar, pit-breadfruit; poko, to roll into a ball; lopwon, a ball of baked pit-breadfruit). Done individuaUy; each man bakes some fresh breadfmit together with pit- breadfruit In his cookhouse and takes a small basketful to the chief, August and September. h. sakalap {sak, to eat; lap, much, big). Breadfmit artificially ripened, as in 2e, above, then baked in the cookhouse and carried to the chief in a basket made from the whole coconut leaf. September. I. uhmw en luhwen mei {uhmw, stone oven; luhwe, remainder; mei, breadfruit). This is made from the last breadfruit of the season; the people of a whole section bake the fruit In a stone oven In the community house with the chief in attendance and with pigs and kava simultaneously prepared (ac? cording to Net informants) or in the cookhouse (according to Kiti informants). These particulars In Net would cause this occasion to be classed as a feast rather than first fruits if one foUows the definitions informants give; nevertheless Net peo? ple call this first fruits. Held in September or October. A Madolenihmw informant gives also mwohn- delemei as a first fruit, when a food caUed delemei is offered. This is made from artificially ripened breadfmit, which is pounded, heated with hot stones, and added to coconut cream squeezed into a wooden vessel. Other Informants do not consider delemei to be a first fmit offering but agree that it is presented to chiefs. Some inform? ants also list kaiak as a first fruit, but others state that It is merely an occasional gift occurring be? tween the various first fmits and consists of the best portions from food of all sorts prepared at famUy ovens as part of regular famUy meals. ^ ? The information about sakalap (2h, above) comes from a single Kiti Informant, and was not checked with other Informants. 3. Kava {sakau): a. wisik pwehl {wisik, to carry; pwehl, earth). The first kava bush dug from a new kava garden Is brought to the chief. b. sahrpahn sakau {sahr, to remove; pah, under). After the kava has grown high, the lower parts are trimmed off and brought to the chief. 4. Bananas {uht). The first fruit is called mwohn uht. Offered to chiefs In breadfruit season. A Madol? enihmw informant says only mangat bananas may be offered; a Net Informant says utuniap and karat are offered; another Net informant says that all three of these, the only types of bananas which ought to be baked in a stone oven, are traditional offerings, but that nowadays the first of every sort of banana from a new garden should be offered.^ ^ When bananas are intended for use by commoners, the stalk is cut off close to the top of the bunch, but for a chief care Is taken to leave a long stalk with two yoimg leaves attached, 5. Pineapple. The first fmit, which had its origins ordy in German times. Is called mwohn painapal; pineapples are not native. It comes In the bread- '8 Gulick's vocabulary gives kaiak as "food prepared at the natives' homes and taken to a chief; better food than that usually prepared at a feast," thus agreeing with the second group of informants. '^ Offerable types of mangat are mangat proper, mangat en alohkapw, epohn, ihpali en pohnpei, ihpali en Saipan; types of utuniap offerable are utuniap and utumwas; karat types are karat proper, karat koldo, and karat en pahlil. Other than these three, which are considered native and are baked in the stone oven, the numerous imported varieties are prepared as "ainpot" (from English, iron pot), i.e., by boiling, a nonaboriginal method. 80 S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 fruit season. The stems and leaves must be left attached. The largest fruit are packed five to a basket and brought to the chief. 6. Mango {kehngid). The first fruit Is mwohn kehngid; it comes in breadfmit season. The fruit are laid along with separated leaves and branches In the long coconut-leaf basket {pahini) and car? ried by two men with a carrying pole to the chief; the basket holds several hundred fmit. Two vari? eties of mango may be used, kehngid en pohnpei and kehngid en salong, also the sports developed from each of these in recent times, kiewek en kehngid en pohnpei and kiewek en kehngid en salong. 7. Pandanus {kipar or deipw). The first fruit is mwohn deipw and comes in the breadfmit season. The fruit Is presented uncooked. As with bananas, a long stem and two leaves are left attached, but unlike bananas, pandanus is offered in a basket. AU types may be offered. Pandanus, however, is rarely offered except in Madolenihmw, 8. Sugar cane {sehu). The first fruit is mwohn sehu. Presented at any time whenever the cane Is of sufficient size. The leaves may not be removed; this is tme for all three purposes for which sugar cane is presented to a chief: first fmits {nohpwei), begging pardon for an offense {tohmw), and in? vitation to a feast {luk). All six varieties of sugar cane may be offered, as weU as sehu ahlek, which Is a kind of reed somewhat resembfing sugar cane, but no one of these may be offered by itself, 9. Polynesian chestnut, Inocarpus edulis {merepw). May be offered ordy under a basketful of pit-bread? fruit or of kehp palahi, a sweet yam. 10. Taro {sawa). Only the varieties pahmaru, pahnta and keiwetik may properly be offered to chiefs, though kuhwet has also been recorded, 11. Dioscorea bulbifera {palahi). A wild yam. The first fruit is mwohn palahi and Is made In Decem? ber. The yams are baked in a stone oven one night, then skinned with the fingemaUs and brought in a coconut-leaf basket to a brook and placed on a flat basketry tray (caUed a pwaht en palahi), 1 to 3 feet square, made of Saccharum spontaneum {ahlek) leaves; the sides are built up of a sort of fern {mahrek), which Is not woven Into the tray but worked up around Its edges, A supply of run- nuig water is fed into the tray from the brook by means of a flume constructed of overlapping half- cyfinders from the concentric layers of banana tree tmnks (Net informant) or bamboo (Madol? enihmw mformant). The yams are kneaded with the hands and, after the bitter quality leaches out, the fem Is removed and the tray placed with its contents (now caUed kedepw) into a coconut-leaf basket and presented to the chief as first fruits; in this condition it is described as Ice cold and is now? adays regarded as a rare and great delicacy. It may be baked again after leaching. On the occasion of this presentation, five places are laid on the main platform of the community house, each place being a sort of platter consisting of 10 breadfruit leaves and two leaves of Campnosperma brevipetiolata {dohng). Four of these places form the corners of a square, the fifth Is in the center of the square. Two spoons are made by slicing chords through the husk of a young coconut and the official in charge of food distribution at feasts (the soun ne) uses these to scoop out a portion of the food from the center of the tray, which he places on the central leaf platter. The second scoop is taken from the lower right comer of the tray and placed on the corresponding platter; then the lower left, upper right, and upper left comers are simUarly treated in that order. These five scoops are presented to the chief, who retains only the central one, the others are distributed to the highest men of title present. The remainder of the dish is scooped up according to no particular plan and distributed among the rest of the people, whose leaf platters contain no particular number of leaves. The ritual of division corresponds closely to that used for a dog or turtle. In the old days this dish was prepared by women on order of the high chiefs. Nowadays It Is made primarUy for Invafids and Is offered as first fmits only when there is enough left over after the Invafid has partaken of it. It Is considered difficult and tune-consummg to make; the tray alone requkes considerable labor. Formerly It was sometimes a substitute for idihd (yam first fmits l b ) . Lihli The making of lihli, mentioned above as breadfmit first fruits 2d, is sufficiently important and suflUcientiy as? sociated with ceremony and deferential customs to merit Its description at length. The dish consists usuaUy of baked and pounded breadfruit covered with coconut cream. It is offered as one of the pruicipal first fruits to the Nahnmwarki by the people of each section, Sometunes it is also made for the Nahnken, When these requirements are satisfied, it is then made for the head of each section, foUowmg which anyone can eat of it. It is made perhaps five or ten times yearly. The dish can also be made of banana {lihli uht) or taro {lihli sawa),^'"- and sometimes "Bascom (1965, p . 40) also includes Cyrtosperma {lihli mwahng) but the writer's informants say that only Pingelap, Ngatik, and Mokil people make this dish. PRESTIGE COMPETITION 81 bananas and breadfruit, or bananas and taro are mixed {lihli repwrepw). The foUovmig account is based in large part on a lihli preparation arranged for the observation of the writer by the A6 of Net and ceremonially held for the Nahnken of Net m December 1947: A wood fire Is made and stones heaped over it to make a roughly square oven with vertical walls Instead of the usual dome-shaped stone oven made for other occasions. Whole breadfmit are piled on top of the hot stones and no covering leaves added, whUe In the ordinary stone oven the coals are raked out, the hot stones buUt up again, and halved breadfmit pUed on the stones with leaves over them; the result is that breadfruit in an ordinary stone oven are baked In a steamy atmosphere, but for lihli they bake in the open, are turned over several times in the process, and the skin becomes charred (Plate 2a), This charred surface is peeled away by two men (designated as soun rar); they use a flat, oblong piece of hibiscus wood {mehn rar) to pry off the char, holduig the hot fmit with half a coconut sheU (Plate 2b). The hot breadfruit are then carried to the lihli maker {soun li),^^ who pounds them on a flat stone {peitehl, the same type of stone used for kava) with a wooden pounder {pein rar) of Morinda citrifolia {weipwul)''* (Plate 3a), The flat stone wUl have previously been cleaned by pounding a coconut husk on it and then washing with water. Close to the lihli maker Is a bucket or wooden vessel of water into which he dips his hand so that he can remove the hot core and seeds without burning himself; he also wets the striking surface of the pounder with his hand from time to time to prevent the breadfruit from sticking. He must work very fast, since the lihli is supposed to be stUl too hot to eat when it Is finished. A girl or boy sits near him and fans him, so that he does not sweat. He prepares four breadfruit at one time, and the resulting loaf-shaped mass is put on a banana leaf that had been previously seared in the fire to waterproof it; this is placed on top of another simUarly treated banana leaf (Plate 3b) and then laid in a certain type of coconut-leaf basket {Hail) which Is carried to the coconut-cream maker {soun piah).''^ The man who car? ries it holds the basket in both arms, with its stemward end (as determined by the direction of growth of the pinnae from which the basket was woven) and the banana-leaf stems to his left. A Madolenihmw informant says two or three baked breadfmit are used for each loaf, sometimes five for a '3 This person on Ponape is always a man, but Ponapean informants say that on Truk and the Mortlocks, where a similar dish is prepared, a woman may officiate. '* The wooden pounder was used on the occasion being de? scribed, but a pounder of white coral, takai mei, is said to be preferable. '8 From piahia, to squeeze out coconut cream; this term is not used for straining kava or any other plant. large one. A Kiti Informant says that four are used, but that for the Nahnmwarki and Nahnken only four halves are used to make one lihli loaf, the other halves are put aside for the commoners. As observed on this occasion In Net, four whole breadfrmt were used. The coconut-cream maker has meanwhUe been grating coconut meat and has a heap of It pUed in front of him on a banana leaf. He now pUes some of it on a mass of young coconut ''^ husk fiber, which he twists to squeeze out the coconut cream over the loaf of lihli held up before him. Both lihli maker and coconut-cream maker wear a headband of two coconut pirmae with the ends twisted together and turned up over the ears (a sedei; this is also used for divining and for what Hambruch calls "Bot- schaft" leaves) (Plate 5a) . The lihli maker must be bare to the waist. He sits on two banana leaves and holds another in his lap so that his kilt and the hair on his stomach do not show to the women present. Neither he nor the coconut-cream maker may speak, laugh, or cough during their work, lest a fleck of spittie faU on the food they are preparing. There is no calling out or notification that the lihli Is ready, since quiet must be maintained. Four lihli loaves are prepared, one apiece for the Nahnmwarki and Nahnken and their respective wives, before whom they are laid. The stems of the two banana leaves and the stemward end of the basket that holds each loaf are to the left of the recipient. In Madolenihmw six loaves are prepared, as described below. Additional loaves are prepared without ceremony for the men who prepare the oven, remove the char, pound the breadfmit, and make the cream to partake of after the chiefs have eaten. Before the coconuts can be grated the water from them is put into a kind of cup {putemei) made of an Alocasia macrorrhiza {sepwikin) leaf. The cup is made by sewing up the leaf around its edges with the rib of a coconut pinna (see Plate 5a) ; It is brought to the Nahnmwarki, on the main platform of the community house, in one of the special lihli baskets {Hail) previously mentioned,''^ To the Nahnmwarki's wife is also carried one of the special baskets, containing a number of objects: one baked and peeled, but not pounded, breadfmit; the two halves of a ripe coconut; an implement of wood {mehn rar) for coring the breadfruit like those used by the men who remove the char; and a piece of coconut husk, shaped somewhat like a club, used to tap and soften the breadfmit (a mehn pok). The whole (caUed a meirar) is presented, lUce the other baskets, with the stem end to the left. It is 76 On the occasion witnessed by the writer; but all informants questioned say that the straining apparatus should be strips of bast from the hibiscus tree. " The sketches in Hambruch (1932 II , Abb. 158, p. 372) are of an Hail basket and a leaf cup from Madolenihmw; both are made somewhat differentiy from those seen on this occasion. 82 SMITHSONL^N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 said also to be given to any of the other people entitled to receive lihli who do not like coconut cream and there? fore abstain from It. The special Hail baskets are made during the baking of the breadfmit by the high chiefs and the visitors on the main platform of the commuruty house, A whole coconut leaf is brought by a man of lower title to the chiefs and spHt in half dowTi the rachls; In Net 19 pinnae are counted from the stem end of the leaf, counting the first three as number one, then each succeeding pair as a unit, so that the actual count is only nine, and this section is then lopped off; then another 19 pinnae are counted and cut, and so on, to the distal end of the leaf. Each section is given to a titled man to weave into a basket. In Net there are said to be six types of these baskets; among them are pwat en kole (Plate 4b), oaralap, ulung en nahnpwatak, and pweiwer. They are made differently in each tribe and (traditionaUy) on Ant atoU. They are made only for lihli, and differ from the ordinary coconut-leaf general utUIty basket {kiam) in that the latter is made from the full width of the leaf whUe for these the leaf is spHt in two down the rachis. In Kiti, Informants give as the types of Hail baskets: 1. Hail en soupeidi, for the Nahnmwarki and Nahnken. Two pinnae remain inside the basket, unplalted; if the lihli Is to be carried home, these two pinnae are tied together to hold the basket tightly together and prevent the coconut cream from flowing away. For only this type of basket the two banana leaves in which the lihli Is put must be cut from the tree at the point where the stalk of the leaf joins the trunk; for the other types they are cut where the stalk merges Into the blade of the leaf. Also, the leaf blade is torn loose from the central rib and pushed back up the distance of one hand's breadth by mnning the hand up the stalk against the blade. 2. Hail en likend, for the wives of the Nahnmwarki and Nahnken. This Is made in the same way as the previous type, but the banana leaves have no stalk attached, nor is the blade torn back. 3. Hail en soun Uh, for the lihli maker. On the under? side of the basket are four unplalted pinnae, two on each side; each pair is tied In a knot, which prevents the basket from sHdIng when the lihli maker eats lihli, for he eats it with his hands, whereas the Nahnmwarki and Nahnken and their wives use spoons of coconut husk made by cutting a section through the outer husk on the plane of a chord. 4. Hail en soun piah, for the coconut-cream maker. Three pairs of pinnae are left unplalted on the upper side of the basket, one at each end and one in the middle. Each time the man whose duty It is to make the coconut cream wrings it out of the fibers he wipes his hands on one pinna, then tears It off and discards it. Six pinnae are required, since in Kiti (and also In Madolenihmw) six lihli are made for ceremonial presentation to high persons prior to general eating of lihli by the masses: one for the Nahnmwarki and Nahnken; one for their wives; one for the A2, A3, and B2; one for their wives; one for the other high chiefs present; and one for their wives. 5. Hail en soun rar, for the men who remove the char. One end of the basket is rounded, one end left pointed with the ends of the pinnae not tied together as with the other baskets; this end is used to clean off the char-peeling implement, which is then used as a spoon to eat the lihli. Other First Fruits Tobacco, beans, manioc, Cyrtosperma {mwahng), cucumbers, com, sweet potatoes, watermelons, and onions are occasionally presented as first fruits; onions were re? corded only once, In section 3 of Net, All of these except Cyrtosperma are European-introduced plant species. They are offered only from new gardens, as with the first of the yam first fruits. A number of animals are also offered to chiefs, and the offerings are caUed by the same name {nohpwei) as first fruits. That for a turtie is caUed mwohn par; both common species of turtle are offered, the hawksbUl ^^ {sapwake) and the green turtie {kalahp); a rare sea- creature, malipwur, which some informants identify as a turtle, possibly mythological only, is also considered to be a chiefly gift. A coconut-eating crab, omp, which occurs on the sandy islets off Madolenihmw, is similarly offered. Fish presented to chiefs, however, are given simply as their due and are not caUed nohpwei. Large specimens of Serranus {mwanger: several species of giant grouper), dep, moat, and Gymnothorax {lahpwid: several species of sea eel) are occasionaUy presented nowadays; aU merer caught are supposed to be offered; wrasse (kemeik) are presented to the chiefs at the third feast given on the occasion of the dedication of a new seine of the largest kind (see below), FaUure to offer the chiefs their due re? sults Ul supematuraUy caused disease, as described else? where. In the prestige competition for pofitical advance? ment these presentations, though neither first fruits nor feasts, coimt as Service too. Ceremoaial Presentation of Fish Certain fish that are offered to high chiefs to be eaten raw are presented with sfits made in them accorduig to '8 Some informants deny that the hawksbUl need be offered to chiefs. PRESTIGE COMPETITION 83 special patterns, which differ accorduig to the species of fish and from tribe to tribe. FaUure to prepare fish In this way would cause much offense to the chiefs who are the recipients. Cooked fish are not sHt except (at least In Madolenihmw) for very large ones; but when they are baked for presentation to a high chief the fins and taU are left out of the fire so that these portions wUl not fall off; otherwise they are not considered fit for presentation. In Net there are seven methods of sfitting fish, each appfied to a number of species of fish (see Figure 3 ) : 1. Four sfits are made transversely on each side, evenly spaced between the base of the tail and a point close to the gUls. Then the fish Is presented whole. Spe? cies: Pseudoscarus sp, {mwommei), pakas. (The identical pattem was seen on a pakas fish hanging outside the Bl's house in Madolenihmw too.) 2. The head Is cut off, and ten transverse cuts are made on each side, everUy spaced from the base of the pectoral fin to the anal fin; then a longitudinal sHt (a) is made above the gills, connecting with the first transverse slit at a point near Its upper end and curving upward and forward to meet the equivalent cut on the other side. For presentation to a high chief the three sHces between the third and sixth sfits are freed at their dorsal ends and laid back downward on each side (b) , the three sfices between the sixth and Tunth sfits are simUarly laid upward (c). The twelve sfices thus produced, counting both sides of the fish, are for the delectation of the high chiefs, the rest of the fish is for commoners. Species: Caranx sp. {arong) (Figure 3A), 3. After scaling, six transverse cuts are made on each side, the first at the base of the pectoral fin, the sixth about halfway to the taU. The three slices between the third and sixth cuts are removed entirely from each side, except for their most ventral portions, and presented to the high chiefs (these portions shown shaded in the sketch); the rest of the fish is for the commoners. Species: wrasse {kemeik) (Fig? ure 3B). 4. The fish is first scaled; then one transverse slit is made at the base of the pectoral fin; from the ven? tral end of this sfit another slit curves upward and posteriorly, then mns lateraUy along the flank and curves back down again to a pomt about midway along the length of the fish (a) . Also from the ven? tral end of the first sfit the beUy Is slit aU the way to the anal fin. Four more parallel slits are made from the base of the first dorsal fin and running down the flank, somewhat diagonaUy forward, untU they meet the sfit that runs lateraUy along the flank. These slits are duplicated on the other side, and the fish is presented whole to high chiefs. Species: mid- let {ah) (Figure 3C). 5. Eight parallel transverse cuts are made on both sides, from the base of the pectoral fin to a point near the precaudal constriction. Presented whole to high chiefs. Species: kioak, kereker (Figure 3D), 6. The two spines (a) on each side near the base of the tail are cut off, also the ventral and dorsal fins. Then the skin from the gUl sfits to the precaudal constriction Is laid back, inside out, on both sides, over the taU (b). Six paraUel transverse sfits are made, the first at the base of the pectoral fin, the last before the anal fin; the first, thud, and fifth of these are cut through to the bone, the second, fourth, and sixth only into the flesh. The whole goes to the chiefs. Species: pwulak (Figure 3E). 7. A portion of the back and stomach (marked a and b In the figure) and the head are given the chiefs; between the back and stomach portions four paraUel transverse sfits are made and the four sfices that result are cut out and given to the commoners, along with the rest of the fish. Species: merer {Figure 3F), A number of other fish that bear a resemblance to one or another of these seven types and which are eaten raw are treated In Identical manner. In other tribes than Net different methods of sfitting are used. The methods are not common knowledge, but are known particularly to the master fisherman {soused). There are other ceremonial details connected with presentation of fish, which again vary from tribe to tribe; for example, when toik fish are presented to the Al of Madolenihmw they must be wrapped in leaves In a certain manner and offered to him in special kinds of baskets. AU of the foregoing appfies orUy to fish caught In a new net, except for such fish as merer, which are always pre? sented to high chiefs, new net or old. FEASTS FOR CHIEFS The Major Feasts To an outsider, the multiplicity of feasts on Ponape and the development of the feasting complex seem almost hypertrophled in comparison with the rest of the culture. Yet we shall see how closely they are linked with the polit? ical organization, not to mention their importance In the economy. 84 S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 A. ARONG B. KEMEIK C. AH D. KIOAK E. PWULAK F. MERER FIGURE 3,?Variations in the slitting of fish for ceremonial presentation to high chiefs. PRESTIGE COMPETITION 85 Feasts of most kinds are known loosely as kamadipw, though the people sometimes distinguish the most impor? tant ones as kamadipw proper. The important feasts are always held In the community house in the presence of the chiefs, AU feasts involve the preparation and drinking of kava. Theoretically also a major feast always involves the use of yams, whUe first fruits are concerned with all types of food plants, particularly breadfruit, and need not in? clude yams. But these native definitions and distinctions are not always strictly sustained in practice. The feasts are by name as foUows, in the calendrical order In which they were formerly held: a. Start of Yam Season {ire ihsol). Two Kiti In? formants give the time it is held as January; a Net informant and one from Madolenihmw say March; another Net Informant says any time from October to May, depending upon the condition of the yams. It is supposed to foUow the first fruits caUed kotekehp. When the feasts were sharply curtaUed by the German regime, the deeds to land that were issued provided that an honor feast {kamadipw en wahu) would be provided once a year to the Nahnmwarki, and this feast is today reckoned as the old Start of Yam Season feast. Each tribe made the feast In sequence to the Nahnmwarki, Nahnken, A2 and B2, and each section made one for its head. At this feast yams of any variety may be brought for presentation, b. Countuig Cookhouses {wad wonuhmw). Given m Febmary (Kiti Informant) or AprU (Net m- formant). The Nahnmwarki and Nahnken each formerly went about from section to section, ac? companied by a wife and some attendants, to visit every section head and every important man and was feasted once yearly in this manner. The feast was made in the famUy cookhouse. Some informants state that each section gave one feast, others that every important famUy head gave one. One Kiti mformant also says that each farmstead would give the feast and Invite the other seven farmsteads that supposedly constituted a single section. (See p. 93 for the theoretical 8-fold divisions of sections.) Rarely given nowadays; occasionaUy accomparues the titie-payment feast {iraramwar). c. Twisting Coir Twine {dakadak dipenihd). This feast was given for the Nahrunwarkl and Nahnken separately (Madolenihmw Informant, any tune; Kiti Informant, m September) and was consid? ered the biggest feast of aU. It was held In the community house. The old men would sit and prepare twine for the chief by roUing coir on the thigh, whUe the yoimg people would make a stone oven and prepare the yams, pigs, and kava. The feast is no longer given (see Figure 4) , d. Kapei tehnrip {kapei, ?; tehnrip, the banana or breadfruit leaves used on this occasion). This is probably the same as the feast caUed In Madol? enihmw tehnkulop. The contents of a whole pit of fermented breadfmit were brought to the com? munity house and placed on banana or bread? fmit leaves. Each person present took a piece and kneaded it on a smooth stone, then made it Into a loaf about 3 feet long, 6 inches high, 8-12 inches wide, which he wrapped in banana leaves and put In the stone oven to bake. The feast was given in the community house, separately for the Nahnmwarki and Nahnken, e. Oven of Whole Yams {uhmw en kehp pwon). Yams baked whole, then divided. For the Nahn? mwarki and Nahnken separately, given by the whole tribe at one time; for lesser chiefs by the people of their sections. Formerly made In the community house, nowadays generaUy In the famUy cookhouse and brought to the Al and Bl. f. Oven of Half Yams {uhmw en pali en kehp). SimUar to feast e, above, but the yams are baked after breaking them up. The above six feasts are the most important. Supposedly these feasts involve yams, never breadfmit, despite the obvious discrepancies that are apparent in the descrip? tions, A number of less important feasts, however, are often included in the category of kamadipw and may in? volve breadfmit. Some informants include the feasts for new buUdings, new canoes, new fishing nets, etc., under this heading of kamadipw. A Madolenihmw informant adds to the fist above a feast caUed koummot, which is the same as feast b, above, except that the chief comes to visit without formal Invitation, A Kiti informant adds kaitihsol,''^ given in May at the end of the yam season and held in the commuruty house. Among the smaller feasts are tiepwel, given to a Nahnmwarki when he recovers from an iUness; uhmw en mwirilik or mwiririk, the death feast; iraramwar, a payment feast for a new titie; tohmw, a feast of apology or propitiation (already discussed in " The Luelen manuscript gives this term (meaning literally "no more scarcity," referring to the scarce period between breadfruit seasons when yams must substitute) as the last of four yam first fruits; the first two are my yam first fruits la and lb; the third one he writes "um en tanpuat" (which can be translated as Oven of Leaf Basket). 86 S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 detaU); famUy feasts; feasts given to visitors; and a num? ber of others described below.?? Other Feasts While Great Work is done and tities thereby earned principally through the forms of Service thus far discussed, a man's reputation is also affected by the degree of his generosity exhibited at other types of feasts, not classified under Service but at some of which chiefs receive shares of offerings also. These are therefore discussed as part of the prestige competition complex. For a new seine three feasts are given, called coUectively oulaid; most people do not refer to them as kamadipw. The first of these is Seme Commencement {simas uhk). The man who has ordered the net prepares food In a stone oven and pounds kava and brings these to the netmaker {sou uhk) at the beginning of his work. The net is made Ul two halves, and on the day that the netmaker joins the halves and has ready the floats another feast is made, Seine Joining {kohpene uhk) or Seine Binding Together {patpene uhk). The fishermen of the section meanwhile have for several days been assembled in the community house, from which women have been excluded, awaiting the proper tides; sexual activity Is forbidden during this time. The day after this feast the fishermen go fishing, taking with them the food they have received at the feast, and when they return, the third of the series of feasts, caUed laidkapw,^^ is made. The three feasts are made for all three of the types of seines used on Ponape, but only laidkapw is made for the four types of hand nets. The fishing undertaken on these occasions is always done for 4 days in succession. In Net, for a new smaU seine (uhketik), the fish caught on the first day are brought to a community house where the laidkapw feast is held for the 80 Hambruch (1932 I I , p . 227f.) first fruits and feasts. His list follows to my lists as given above): Breadfruit first fruits.? 1. puatseme'i 2. me'i aui 3. pai i ni (first fruits 2e) 4. lili (first fruits 2d) 5. tsakalap (first fruits 2h) 6. kamemem 7. tautau 8. um en lu en mei (first fruits 2i) gives a different sequence of (the parenthetical notes refer Tam first fruits: 1. puke men puel 2. puke lo pun (first finiits 2g) 3. kotse kep' (first fruits la) 4. itiz (first fruits lb) 5. um en peli en kep^ (feast f) 6. um en kep' uong (feast c) Feasts: 1. ire'isol (feast a) 2. takatak tipenit (feast c) 3. ke'itisol (see text above) 4. karisime'i (first fruits 2a) 3' This feast is also known as kapas and pasalaid to different informants. Strictly speaking, kapas refers to the farmers of the section who prepare the stone ovens for this feast, not to the fishermen. Nahnmwarki; the second day the same feast is given to the Nahnken, at the commimity house closest to his residence; the third day to the A2 and B2; the fourth day to the A3 and B3. For A2, B2, A3, and B3 the feast need not be held in a community house. In Net today the Nahnken and Nahmnwarki are father and son, and one feast is held for both. For the larger seines {uhkelap and sokesok) the laidkapw feast is held aU 4 days for the Nahnmwarki and always in the community house; the lower chiefs attend, of course. The laidkapw feasts for the three types of seines and the two larger types of hand nets {naikelap and naik en dokedok) are tribal feasts. Those for the two smaUer hand nets (luhkouk and naiketik) axe section feasts, may be held elsewhere than In the community house, and the head of the section in which the new net has been made may function In place of the tribal titieholders. At such feasts the fish that are distributed are con? sidered to be an exchange for the agricultural food brought and prepared in the stone oven by the farmers and are equally divided among them. AU the land produce, except that offered to the chiefs, is given to the fishermen, who retain none of the fish; the largest share goes to the master fisherman. This exchange is known as soawa. But if the fishing has been uiUucky, the sacred staff {ketia sarawi) used by the master fisherman for religious purposes is brought into the community house where the chiefs and the farmers are assembled waiting for the catch, the staff is taken to symbolize the wanting fish, and the food from the stone oven is divided among aU present. The assem? blage occurs when the signal of the returning fishermen is heard; this signal consists of long blasts on the conch tmmpets, simUar to those formerly given for a dead commoner. At the time of these feasts the section head and the tribal chiefs who dwell in the section must be present, and no fish can be sent to their houses; If they are not there they have to be summoned before the feast can proceed. Besides the fish consumed at the feast many are distributed to be carried home. The first fish caught {mwamwenieng) is always given to the highest chief present. Additional offer? ings depend on the size of the catch and whether any of the royal fishes have been caught. Large fishes are given singly to the chiefs; medium-sized ones are tied in bundles of three, small ones In bundles of five, and so presented, WhUe aU of the foregoing activities are centered about the occasion of the new net, there is ample opportunity for an ambitious man to exhibit his prowess as a farmer or fisherman, to demonstrate fealty to the chiefs by lavishness of offerings and by appropriate deference, and thus to further his political advancement. A considerable religious element formerly entered Into these feasts. How much of it survives It is difficult to say, though in the practices surrounding fishing more pagan customs continue than in any other phase of Ponapean PLATE 1 Abandoned house site and burial cairns, a. Stone platform of an abandoned house site on Sokehs Island, b, Burial cairns for the Soukise at Paler in Wene. PLATE 2 Preparation of breadfruit for lihli. a, The oven of stones with the breadfruit on top. b, Peeling the breadfruit after baking by holding it with half a coconut shell and prying off the charred exterior with a piece of hibiscus wood. PLATE 3 Preparation of breadfruit for lihli. a, Pounding the peeled, cooked breadfruit with a wooden pounder on a flat stone of the same type used to pound kava. b, The finished lihli of pounded breadfruit and coconut cream. PLATE 4 Ceremonial food preparation, a, A meirar (see p. 81, for explanation), b, One of the six types of coconut-leaf baskets used for lihli in Net?this type known as pivat en kole. PLATE 5 ?/'.:fe j i PLATE 6 Community houses, a. Side view of the community house at Tamworohi, Madolenihmw. showing various acculturative elements, h. Front view of the community house at Temwen, Madolenihmw. PLATE 7 Servitors in community house and chief's belt, a, The main platform of a community house in Net. Leaning against the wall is the Bl of Net; to his right, drinking a cup of kava, is a visiting chief from Sokehs; to his left, in the foreground, is the Bl's wife. Each has a servitor; the Bl 's servitor is seen in typical posture, b, Chief's belt, worn by the B6 of Madolenihmw. PLATE 8 The growing, transporting, and cooking of yams. a. A stone oven covered with leaves and with the edges weighted down, b, A yam planting protected against pigs by stones, c. Yams in baskets and litters of various sizes brought to a feast, d, A single, large yam on an individual litter. PLATE 9 Transporting yams, a, A pahini basket of yams, b, Four-man yam, decorated with leaves. PLATE 10 Transporting sugar cane and kava. a, Sugar-cane bearers in procession at dedication of a new community house in Kahmar , Net. b, Kava plants piled up before pounding. PLATE 1 Kava-drinking ceremony. ?, .Master of kava rcrcmony handiiiL( cup of ka\ H -n o 3} 0 0 NANRAS m o m > H -n o O X ? ? ? ? ? ? ? FIGURE 4.?A Twisting Coir Twine feast in the community house of section 19, Kiti. Numbers in squares represent the stone ovens of the eight farmsteads of section 19; farmstead No. 1 on one side competes and exchanges wdth fairmstead No. 1 of the other side, and so on. Numbers in circles represent the piles of yams belonging to each farmstead, left outside the com? munity house for display. " X " marks the heap of pigs brought by all the farmsteads, placed here before singeing off the hair on the hot stones from the ovens. The triangles on the main platform represent the location of the two principal kava stones, where kava is pounded untU the ovens are cleared away, when more kava stones are placed in two rows in the central area {nanras). One kava stone and a row of ovens are under charge of the Kroun Wein, XI of the section, the other stone and row under the Nahnawa en Rohnkiti, the Yl chief. Overseer of all is Oun Kiti, formerly highest tribal titieholder in the section. ing was as foUows (the six named areas are aU farm? steads of section 7 ) : Section 8 Section 9 Pohnkeimw Keimwin Nanseinpwel Sewihso vs. vs. vs. vs. vs. vs. Section 6 Section 3 Peien Uh Lukoapoas Section 5 Section 4 PRESTIGE COMPETITION 95 Informants say that this Start of Yam Season feast was then followed by another, sunUar one, which they call Taboo Lifting. It Is not at all certain that this was the same kind of feast previously referred to as Taboo Lift? ing; it is also possible that what the informants are de- scribmg is the same Start of Yam Season feast but some? how given a more competitive aspect. At any rate, the same pairing of the same 12 units occurred, and It seems to have been a regular affair, not depending on a chal? lenge; sometimes it ended In fightmg. In contrast to Kiti, the two sides did not compete simultaneously, the recipi? ent retumed the feast the next day. The other sections of Uh (1, 2, and 10-15) staged their Start of Yam Season feasts separately. In section 13 of Uh there were again 12 farmsteads, which were matched thus: lepilong Peidi Pohnpihr Sewihso Apweiakpeilong Apweiakpeiei Nanepwok vs. vs. vs. vs. vs. vs. lepiei Sep Soukiro Peilong en rohi Nindol Likomwpweilong LUoomwpweiei The expressed native theory that there were eight farm? steads in a section faUs thus to be borne out everywhere. The twelve farmsteads of section 13 just listed have been in existence since at least during informants' early years (as far back as 1880), as have the six of section 7 pre? viously mentioned. Elsewhere Irregularities in number may in part be due to the effects of the German issuance of deeds during the first decade of the 20th century. On such deeds the various holdings that were awarded as private property, when they were contiguous, were entered as a suigle piece of land and coUectively caUed by the native name of one of those holdings; thus references nowadays to named areas may become confused. Depopu? lation Ul some areas has also affected the number of ovens that can be made. But Net informants deny any general eight-oven or six-oven pattem and say that there may even be an odd number of farmsteads in a section. The B2 of Net, for example, recaUs that there were ten ovens at Start of Yam Season in section 1 some 30 years ago and there are 16 today at Its modem substitute, the honor feast. Each farmstead was occupied by one extended family, hence it is obvious that the subdivisions must have varied m number from time to tune. But where the num? ber eight was held to, as it seems to have been in Kiti, there must have been grouping of famUies together m order to achieve the ideal number of ovens, and a farm? stead could not always have corresponded to one oven. Or it may be that the word peliensapw is not always to be taken literally as meanmg a famUy farmstead but may indicate a poHtical subdivision with which farmsteads tended to, but did not always, coincide. In Kiti, at least, the ceremonial exchanges between the farmsteads occurred also at those first fruit ceremonies that Involved assemblage in the community house. The rivalry in attempting to outdo others in food presentation at such affairs persists today, but only in a minor degree. In Kiti there was the same ranging of the ovens on two sides, but the exchanges, at least in some places, were between any two farmsteads on opposite sides, not between members of designated pairs. In section 16 of Kiti the ovens were arranged as follows: Pokil Kepinuluhl Tonkiap s' Karki In section 15 of Kiti: Kepinsemwei Kepinsemwei Tonemwok Reiwei Kias In section 26 of Kiti: uhmw peiuhpa (i.e., four lower ovens) Pohnkulu Loahngenkiti Ta wenapa Longorik Pohnangieng Pohnangkiok Pahlap Soupir Kepilerohi Nanmangil Alamas Kepihle Kepinta uhmw peiuhpowe (i.e., four upper ovens) Dol en samaki Eipwa Imwindol Madol But section 29 of Kiti has only two farmsteads (i.e., two ovens): the northem half of the section is Dionpwo, the southern is Lehpwel; informants say that this was true jmclently also. Section 5 of Kiti likewise has two farm? steads, Sounsom and Ipal, each divided Into six mwokot (a term used rather variably in different areas, but here signifying a subdivision of a farmstead); each mwokot provided one oven, as foUows: Sounsom Nantipw Sapwet Nanpahlap Doleni Panor Pohniong Ipal Alapowei Pohntui Paler Pohsomwak Dopari Pahnkomwo It seems clear that hi this section the term farmstead is beuig used for the two halves of the section that exist elsewhere but are usuaUy unnamed, and that mwokot is here equivalent to farmstead as previously used. ' Given as Panios by another informant. 298-818?6S -10 96 S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 In Awak, the present-day sections 1 and 2 of Uh, the ancient organization was six ovens, three on a side: Under the XI (Soulik en Awak) PelienpU Sewihso riau Matar Paremitik Mesenpal Panol 87 FoUowing the German redistribution of lands the two sides in Awak became organized Into 12 ovens, four on one side, eight on the other, as foUows: Kopwun Peimore Wei Panpei Meikido Nanwou Dierekpwel Nankepiniak Alukeren Penietik Tipwenpohnore Panipa Just how this apparently unbalanced division functioned in food exchanges is not clear; another informant gives eight unnamed ovens instead of 12, Although these 2-fold divisions pitted geographical areas against one another, political and kinship ties were also Involved, For example, in section 29 of Kiti it was the XI of the section, the Nahnsahu en Sewihso, who led the people of Lehpwel farmstead in competition against the people of Dionpwo farmstead headed by the Yl , the Nahnkroim Sewihso, The foUowers of each of the two chiefs tended to be his clansmates within the section. In section 15 of Kiti the four ovens in KepUerohi were simUarly headed by the XI , the four in Kepinsemwei by the Yl. Each separate oven might be headed by a desig? nated section titleholder belonging to one or the other line of titles within the section. In Awak, Uh, this was organized as follows: Pelienpil Sewihso riau Matar under the Lepenien Awak under the Sidin Awak under the Soulik en Awak Under the Tl (Kroun en Awak) Paremitik under the Kroun en Awak Mesenpal under the Koaroahm en Awak Panol under the Soumadau en Awak Smce the three titles on each side (especiaUy ui Uh) be? long to the same Hne of chiefs and usuaUy to the same famUy, the competition within the section is essentially between two kinship units and is of a somewhat different cast from that between political units larger than sections and different also from the prestige competition of in? dividuals. When more than one section Is Involved, several kinship units may be In competition, and kinship ties may cut across political and residential bonds instead of rein? forcing them as they do within the section. There are, however, no data as to Intensity of competitive feeling at the different levels of competitive exchange, so we can make no generafizations conceming the reality to the native of this distinction. Again, In the presentation of food to the chiefs at ordinary feasts the individual strives to augment his prestige by surpassing the efforts of all other individuals; whereas in group competitions personal ambition must be temporarUy submerged In the com? munal cause; but the distinction Is not clear-cut, for in the individual offering a man is often helped by his rela? tives in his efforts to prove himself worthy of a promotion, and In the communal offering the individual's contribu? tion to the oven is noted by all persons present and his reputation is thereby affected. THE FEAST PATTERN The Community House and Seating Arrangements Feasts of most types are held In the nahs, which is vari? ously referred to in the published fiterature as community house, meeting house, canoe house, feast house and aU- men's house. Any of these terms is appHcable, since the buUding serves aU these purposes, except the last, which is an error based upon a false analogy with the bachelor's house of the central and westem Carolines. In this work the term community house is consistently employed. The buUding varies somewhat in architecture, but the most developed form is a structure containing platforms on " Mesenpal and Panol constitute modern section 1 of Uh, the other four ovens making up modem section 2. three sides, in the shape of a U with the front left open. The central, ground-level area (the nanras), enclosed on three sides by the platforms, is the place where kava is pounded and the stone ovens {uhmw) are made. The main, rear platform {lampahntamw) Is somewhat higher than the two side platforms {mongentik) and is some? times stepped (Plates 6a and 6b). On the main platform of the community house sit the soupeidi, or Those Who Face Downwards, an expression whose various meanings have already been discussed. The term soupeidi derives from this seating arrangement, since the people it refers to sit facuig the front of the buUding, whUe everyone else sits on a lower level and faces them. In Kiti the Nahnmwarki sits forward of the rear central post of the main platform (the post itself bears the name PRESTIGE COMPETITION 97 soupeidi), with his servitors in front of and facmg him. The B2 sits alongside hun in front of the next post to his right and the Nahnken m front of the next post to his left; to the left of the Nahnken the A2 may sit, though not always.^^ The pomt of this arrangement is that men of lower titie In the same Hne as the Nahnmwarki do not sit near hun, and juniors to the Nahnken In the B-Hne do not sit near that chief. Farther to the right of the B2 sits the Nahnmwarki's wife, and to her right sits the Nahnken's wife.?^ In front of each of these persons sits his servitor or servitors, but stUl on the main platform. Nowadays In Kiti, OHver, who is the A6 but who holds one of the tities properly belongmg to the Nahnmwarki, namely Rohsa, may sit In the B2's place; he is m the A-Hne but belongs to clan 18 ratiier than to clan 4 (tiie A-clan) or clan 11 (the B-clan). In Kiti the A2's position is not definite, smce his duties often take him down Into the central area where the other holders of A-tities are working. He often sits to the left of the Bl, but he can also sit elsewhere on the main platform, as long as he is not near the Nahnmwarki. In Net, accordmg to the Nahnken of this tribe, the Nahnmwarki and Nahnken should sit on the right of the mam platform, their wives on theu left; but the position of the Nahnmwarki relative to that of the Nahnken and the positions of their wives relative to each other do not matter (Plate 7a). As observed at feasts in Net the Nahnmwarki actuaUy sits in the center of the platform, as In Eliti, with the Nahnken and the visitors from other tribes to his right, his wife to his left and the Nahnken's wife to the left of her. The A2 in Net does not get on the main platform but should sit and work at the last kava stone in the right row of stones; he is not supposed to move about. The A3 has duties that keep him on the right plat? form. The A4 must, Hke the A2, stay in the central area and work, when the Nahnmwarki is present (although the present A4 Is one of the assistant food dividers). The B-Hne chiefs, being Royal ChUdren, are permitted to move about as they please. The present B2 in Net is the principal divider of the food and has duties that keep him standing on the main platform, so his proper place was not observed. The main assistants of the B2 of Net, the A4, A6 and B4, likewise have special duties, though these are not duties that adhere to these titles but to the particular men who at present possess them. On the left side platform, from the Nahnmwarki's point of view, sit the women; on the right platform sit the men who are not working in the central 88 Hambruch (1932 I I , p. 17) has a diagram of a feast in Wene (part of Kiti) that shows this arrangement of Nahnmwarki and Nahnken. The B2 is absent and the A2 is to the right of the Nahnmwarki, but the community house in which the feast was held belonged to A2, which may have affected the seating arrange? ments. 88 In Hambruch's diagram of the Wene seating arrangements the only high-ranking woman present is the wife of the A2, who was the host at this occasion. She is seated to the left of the Nahnken. area; but this division is not always enforced, and at some feasts, when the women do not fiU up their side, a number of men sit there too. In Madolenihmw the Nahnmwarki, instead of sitting in the center, sits toward the right side of the main platform, with the Nahnken more toward the center, to his left, contrary to the practice elsewhere. To the left of the Nahnken sits the Nahnmwarki's wife, and leftward of her the Nahnken's wife, AU of these sit facing forward. The A2 In this tribe sits toward the rear end of the left platform, facing Those Who Face Downwards; behind him, about midway down the platform, is the position of the A3; and behind the A3, toward the front end, is the A5, who must stand In order to perform his duty, the overseeing of the work in the central area. If the A5 is absent, the A4 may be positioned here. On the right side platform in Madol? enihmw the B2 sits in a position about one-third of the way to the front, and the B3 is at the front end of this plat? form; the B3, like the A5, must stand, for it is up to these two chiefs to see that no one else stands up straight so that his head wiU be higher than the heads of Those Who Face Downwards, that the people on the side platforms do not sit with their legs outstretched or apart, that their legs or skirts do not hang over the heavy beam forming the inner edge of the side platform, that everyone speaks quietly and does not use commoner speech, and that aU maintain the various forms of respect due the high chiefs.'" The right platform m Madolenihmw, as In Kiti but unfike Net, is the femuiuie side, although the B2 and B3 and sometimes other men sit here too." The B2 is In charge of the division of the food, as the bearer of this title seems formerly to have been in aU of Ponape in his capacity of chief priest. Whoever vrishes to go out to fetch breadfruit or firewood must first get his permission, and the breadfmit picker or the axe must first be shown to him. After the food is baked, it is brought to him In baskets and he pulls the sfip-knot of hibiscus fiber that fastens the baskets shut; then, in former days, he would give a prayer. After this, his assistants help him divide the food. Only then might food go to the Nahnmwarki or Nahnken, Much of the foregoing ritual Is rarely seen today, and the seating arrangements are not always taken seriously. ?" Hambruch's sketch of the Wene arrangements shows the B3 at the front end of the left instead of the right platform as in Madolenihmw, and the A3 at the front end of the right platform instead of midway down the left platform. These two chiefs are described as guards. The reversal of position with respect to the side platforms is no doubt connected with the fact that the position of the two sexes on the side platforms is also reversed as compared with Madolenihmw. ?i But the sketch in Hambruch of the feast in Wene shows the women on the left. It is of interest, also, that at a motion picture shown at Kolonia and attended by many Net people, the men, without any apparent direction, all took seats on the right, the women on the left. 98 SMITHSONIAN CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 The prayer by the B2 was his function as priest. The B4 is also said to have "blessed" food before the highest chief ate; he was the second priest. In addition to the high chiefs a number of other persons are entitled to sit on the main platform of the community house. Among these is the Lepen Moar In Madolenihmw, who Is not a member of either ruling clan but owes his special position to certain feats performed in war by an ancient holder of the title. In Net the Lepen Net and the Nahnsoused, whose titles were formerly those of the Al and Bl but are now bestowed on other men, stiU receive high honors along with the Nahnmwarki and Nahnken, whose titles have replaced theirs. Also visitors from other tribes take their places beside the high chiefs, regardless of the position of their titles, if they come in official status. Many of the XI titles rank very high and their holders may sit alongside the Nahnmwarki. There Is a tendency for the people who sit on the side platforms to arrange themselves by rank, those with higher titles sitting toward the rear, the commoners toward the front; but this is anything but a rigid ordering. Bascom (1965, p, 29) states that the Nahnmwarki and other chiefs in the A-line sit In order along one side of the buUd? ing, whUe the Nahnken and B-chlefs sit on the other, but neither Informants nor observation bore this out; in fact. It would be Impossible, for nearly aU men of the A-line must be working in the central area when the Nahnmwarki I^ present. Other than this requirement there are no assigned positions for the lower titieholders in either line except for those already specified. There Is only a faint resemblance to the kava circle of Westem Polynesia, When a titled man from one tribe comes to a feast in another tribe, his clansmates residing in the second tribe, who are normally commoners there, move up closer to the main platform. If the visitor Is a Nahnmwarki or Nahnken or comes as his representative he sits facing for? ward on the main platform alongside the local high chiefs; but if he has a lower title and is visiting for his own pur? poses, he faces up toward those chiefs. Besides the Royal ChUdren, the Nahnmwarki's sons-in- law, even if they belong to his own clan and thus have A-titles, may move about and may sit on the main plat? form (but not facing forward) if they wish. This is be? cause, like the Royal ChUdren, they are considered as his chUdren. As noted previously, the A2 takes the Nahnmwarki's place when the latter is absent, the A3 takes it if both of his superiors are absent, and so on; sunUarly, in the B- line, the order of the series is foUowed, At a section feast, unless a person with a high tribal title is present, the XI takes the place of honor. If the A2 takes the Nahn? mwarki's place, the other members of the A-clan may move up closer to the rear. Only the highest Royal Children, particularly the ac? tual chUdren of a Nahnmwarki or Nahnken, may climb up at the front ends (the karatak en seriiso) of the two side platforms. Sons use the masculine side, daughters the femlnuie side. This custom is attributed to the sImUar behavior of the first Nahnken of Madolenihmw in the legend of Isohkelekel. The two rear doors {wahnihmw en lampahntamw in Kiti, wahnmeimei in Net) and the first side door on each side opening onto the main platform are reserved for the use of Those Who Face Downwards. Holders of lower A-tities and male commoners must come through the large front entrance and cfimb onto the side platforms from the central area. Women of any rank may enter by the side doors on the feminine side platform, but on the masculine side the side doors may be used only by Royal ChUdren. Feasting Procedure and Ceremonial The preparation of the food at feasts is carried on in the central, ground-level arena of the community house, below the three platforms. It is entirely the work of men. A number of stone ovens are arranged in two paraUel rows along the length of that area. The oven is not the same as the earth oven of other parts of Oceania but Is a surface stmcture. A layer of basalt rocks is made on the ground or in a shaUow depression. On top of this is pUed firewood, then more rocks are heaped on to form a dome- shaped stmcture. When the rocks are hot, the oven is taken apart with wooden tongs, the coals and ashes are raked out, and the stones are buUt up again over the food. The food is wrapped in leaves, but halved or quartered animals are laid in place, and this Is foUowed by layers of hot stones. Then yams and any other vegetables being prepared are laid on the stones; over these are placed leaves, usuaUy of banana; then the pig or pigs or other animals are laid in place, and this is foUowed by layers of more leaves?banana, ieuieu, and coconut palm leaves are often used in successive layers, or sometimes old mats. The edges of the leaves around the oven may be weighted down to keep in the steam (Plate 8a) . For carrymg most smaU articles to feasts the carrying basket {kiam) made from a section of a coconut leaf is used. Articles may also be tied to each end of a carrying pole {ini) which is bome on one shoulder, or tied to its middle and carried on the shoulders of two men in fUe. A long basket {kiam ro) made of the whole coconut leaf {pahini, also used as the altemative name of the basket) is carried slung from a carrymg pole with one man at each end. For large quantities of food a litter of wood {peikini) with coconut leaves laid over it is carried by 6 to 10 men with two poles; such a contrivance can hold hundreds of breadfmit, 10 or 20 bunches of bananas, 20 PRESTIGE COMPETITION 99 or 30 smaU yams, or 5 or 10 pigs. In place of the fitter a "nest" {pahs) can be slung between two poles (Plates 8b, 8c, 8d, 9a, 9b). Articles carried to a feast are designated according to the number of men required to carry them. A 1-man yam is a kehptapan or, if very smaU, a kehptikitik. A kava bush smaU enough to be carried by one man is a sakau kepaik. A yam that requires two men is a kei, whUe a kava plant of this size is a ro. The term ro may also be used for yams, but kei Is not used with kava. In Net, kei is appHed to a yam that requires two to four men. For a yam or kava plant carried by four or more men (six or more in Net) the term appHed is pahs (the same as the carrying device). The number of men required for very heavy yams, as given by various Informants, may run over 30; one Informant gives the hardly credible figure of 50. On one occasion, in December 1947, at a funeral feast in Madolenihmw, a 5-year-old yam was carried in a "nest" by 30 men. Even 2-year-old yams are often large enough for this carrying device. The poles are heavy and there is also a tendency to add superfluous porters in order to magnify the impres? sion of opidence on the part of the donor of the yam; nev? ertheless such large yams must weigh very considerably to require so many bearers. Though the writer had no scales with which to verify his guess, he judged the aver? age weight of five 2-man yams at one feast to be about 50 or 60 pounds, which would be at least 25 pounds per porter. A yam that required 30 porters would, on this basis, weigh an Incredible amount, but I have no hesitation in saying that Ponapeans can grow yams weighing 200 pounds or more. For distribution at the feast among the people the large uncooked yams are broken up Into their component tubers {kutor). There are usually 2 to 10 tubers to a yam of the 2-man size, depending on Its kind. Pigs and dogs are brought to feasts slung under carrying poles, the legs tied over the poles with hibiscus fiber; or they may be brought on fitters if there are several of them. The pole is sHpped out and the animal Is kUled by stabbing in the heart with a machete; dogs may also be killed by having theu- throats cut. The hair is singed off by puUing the carcass back and forth over the hot stones scattered about when the stone oven is opened, or coconut-leaf brands may be appfied to the animal; then the skin Is scraped with a machete, A single ventral slit Is made and the entraUs drawn. The ceremonial preparation of dogs described and iUustrated by Hambmch (1932 II, pp. 246-252) was not seen in 1947-48, due to an epidemic that had virtuaUy wiped out the canine population. (See Plate 5b, a photograph taken In 1963 of a roasted dog in a coconut-leaf basket; its preparation was not witnessed.) UsuaUy, whUe the food is being prepared in the oven, one or two kava stones are pounded for the early kava, known variously as moming kava {sakau en menseng). quick kava {sakau en ahmwadang), and kava of the risuig sun {sakau en kapwarisou). Formerly these stones were on the main platform, and the kava produced from them was made only for the Nahnmwarki, Nahnken, and their attendants, but this is seldom seen nowadays; usually the stones are in the central area, just below the main plat? form. The hibiscus carrying pole, on which a kava bush is brought for this early pounding, is thrown out of one of the doors of the main platform. In the middle of the main platform, buUt into the stone foundations, there was formerly a pit about six feet square and two feet deep ( nanpehs or nanparas). Here a fire was kept burning, from which were fit the fires for the stone ovens. Sugar cane {sehu) is brought to a feast either in single stalks or in a whole clump {uhnsehu) joined at the roots, which is then divided into separate stalks. Single stalks are carried on a man's shoulder, the large clumps on fitters (Plate 10a). Like kava, sugar cane is carried root fore? most. If sugar cane is brought, it is carried by the women as weU as the men, for "sugar cane is women's worL" The fitters are placed with the root ends of the cane on the main platform, the other ends In the central area. One large fitter may be shoved up into the rafters. The women who bear the sugar cane are much less inhibited than the men who f oUow them with kava and they whoop and give the lUulatmg yeU {kadakadek) common through the Carolines; I have seen them occasionally breaking into solo dances. Some of the cane is cut up and distributed; in Net the persons of high titie present, male and female, get two pieces each, one piece of four sections and one of eight. The remainder of the cane is distributed whole, ac? cording to rank. After this distribution the ovens are opened and their baked contents are removed and distributed according to rules described below. The kava is then brought in, in a singing procession, after which the kava pounding stones are set up (Plate 1 Ob). If dances are held at the feast, the side platforms may be used or a special structure may be buUt out from them. But dances are seldom associated with the feasts nowadays. When the feast is over, the portions of food that have been distributed are carried home, the uncooked yam tubers to be replanted, the cooked foods often to be re- cooked. The Nahrunwarkl's share should be carried to his home by someone from the opposite line of titles, who takes commoners along with him to act as porters. The wife of a generous Nahnmwarki wIU give perhaps half of the food to the bearers, hence there are eager volunteers for this duty; a complaint against the present chiefs of Net is that they are too stingy to five up to this custom and theu* bearers go unrewarded. 100 S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 The baked pig carcasses are taken from the opened oven and brought up to the main platform to be cut up on coconut leaves. In Madolenihmw the pig is placed on its stomach before it is carved. First the head is taken off, cutting from the left side. Then two longitudinal gashes are made along the loin, from neck to mmp, first along the left, then the right side. After this the pig is tumed on its back and the left front leg, mcludmg the shoulder, is cut off; then the left rear leg; then a belly strip on the left side of the ventral sfit made before cooking for drawing the entraUs. Next the flank is sliced, from the ventral sfit to the loin cut, into a number of pieces; these include the ribs. Then the whole operation is repeated for the right side. The remaining part, the back between the two loin cuts. Is chopped Into several portions. Elsewhere than Madolenihmw the pig is placed on its back before it is cut. Butchering procedure in Kiti is sunUar to that in Madolenihmw, though the writer has seen the two belly strips removed first, even before the head, and the flank sliced up after the front leg is cut off and before the rear leg is removed; this m spite of the ex? pressly stated procedure. Occasionally, also, the butcher? ing begins with the right side. In Net the right side is the proper side to begin with, but if there are two men carving pigs at a feast, one begins with the right side, the other with the left side. In the preparation of a turtie the intestines are first removed without separating the carapaces and are baked in the oven without any container for some 15 or 20 min? utes. The baked intestines are considered a great deficacy, despite their overpowering odor. At the feast they are served on breadfmit leaves to the highest chiefs and their wives. The turtle Itself Is baked In the stone oven for about an hour; after it is removed from the oven the portion of the right front leg that protmdes from the shell is cut off, then the right rear, then the left front and left rear legs. (This is the procedure observed ui Net.) After this the lower carapace is removed and a certain portion behind the right foreleg is cut away, followed in order by the same portion from the left foreleg. A large portion of the carcass is then removed from between the hind legs, followed by the large front shoulder muscles, and then what is left of the legs and shoulders themselves in the same order of right front, right rear, left front, and left rear; with these portions comes most of the rest of the body of the turtle. At a feast a turtle should be cut up before the pigs, so that it can be served hot; they are baked together In one oven. A large group of eager observers usually gathers around to watch the butchering, since it is considered a difficiUt job and there Is much less opportimity to witness its execution than there is with pigs or even dogs. Formerly, a taro leaf fuU of water had to be nearby for washing the hands and the knife; nowadays, any con- tauier serves. The pahn leaves on which the butchering takes place are supposed to be thrown out of the doors leaduig into the back of the main platform; if they are bloody, the side doors may be used. In Madolenihmw the breadfmit-leaf platters upon which the portions of meat are placed have their stems attached, but elsewhere the stems are supposed to be removed. The men who cut up the pigs and distribute the parts are requu-ed not to wear shirts. Division of Food The official divider of the food (the soun nehne or soun ne) is appointed to his task by the Nahnmwarki and Nahnken. One man from each line of tities is appointed because only If he is a member of the opposite line may he stand with his head higher than that of a high chief. At one feast observed in Kiti, when the Nahnmwarki was present, there was no man in the B-line present who was considered skiUed enough to undertake the job of divider, so the A3 functioned. But being in the same fine as the Nahnmwarki he could not stand in the divider's usual place on the main platform, for the Nahnmwarki was seated there. The central area was too fuU of people and kava pounding stones to permit him to stand there and function effectively. So he stood outside of the buUding, In the rain, and caUed out the titles of the recipients of the various portions from that position. Also two women, one from each line, divide the wom? en's share of the food. In a section feast there are like? wise two functionaries, although their assumption of duties is more informal. In Net, at a large feast, the B2 and A4 function; at smaUer feasts the B4 and A6; the Oundol en Ririn is also a qualified food divider. The duties of the divider are not incumbent on the holders of these particular titles, even though the B2, as chief priest, was anciently the chief functionary and tends stUl to be so in each tribe. UntU a few years ago the present Lependeleur of Net was a food divider of that tribe, but he proved unsatisfactory and was replaced by the Oundol en Ririn. It is considered a difficult and delicate job, for it is easy to offend a titleholder by not giving him a portion commensurate with his position. In the division of the pigs each man's portion is raised up by the divider whUe he calls out the recipient's titie. Thus, at a feast where there are a number of pigs, the largest pig is held up whUe the official loudly calls "koa? noat Nahnmwarki"; the second largest pig is assigned to the Nahnken by a caU of "sahk Nahnken." (See p. 45 for explanation of these honorific terms for food.) This pro? cedure is caUed pwekpwek. The pig or portion of pig is not handed over to the person designated but is taken by commoners to the front of the community house and hung there in baskets from the main stringer. For men of lower PRESTIGE COMPETITION 101 tities portions are hung elsewhere in the buUding. The third portion, which goes to the A2 (the Wasai), is an? nounced as "Wasai kepui koanoat." If the Nahnmwarki is not present the Nahnken takes the first share and that of the A2 is caUed "sahk Wasai," It does not matter which di? vider apportions the food, but if he belongs to the A-lme and the Nahnmwarki is present he cannot shout out the proper words, the divider of the Nalmken Ime must do it instead; and likewise, the official of the opposite line must function in this manner for the Nahnken, The female functionaries caU "pwenieu Nahnalek" zmd "sahk Nahn? keniei" for the wife (Nahnalek) of the Nahnmwarki and the wife (Nahnkeniei) of the Nahnken respectively, and their baskets are also suspended. In Kiti if there are enough pigs the order of division is as foUows: Nahnmwarki, Nahnken, Nahnmwarki's wife, Nahnken's wife, chiefs A2, B2, A3, B3, A4, and B4, alter? nating thus between the A and B lines. Beyond this whole pigs are not distributed, ordy portions. Some of the sec? tion heads and some of the holders of priestly tities that rank high may, however, in the distribution of pigs, come ahead of the tribal chiefs just listed, as explained earfier, except for Al and Bl . Also nowadays the A6 of Kiti, OHver Nanpei, comes just after the Nahnken, because of his special position as the wealthiest and most powerful man In Ponape, and because of his extra titie of Rohsa, which is normaUy another titie of the Nahnmwarki of Kiti but has been awarded to OHver, When OHver attends a feast in Uh (where he is also A6) or Madolenihmw, he is said to receive first honors, ahead of even the Nahnmwarki or Nahnken, In Net the order of division is supposed to be Nahn? mwarki, Nahnken, Lepen Net, Nahnsoused, A2, Souruko, SouHk en Daun, B2, This tribe has apparentiy interleaved the old Net titles (see p. 19) with the new ones that have been copied after the other tribes. But In quite recent times there has been dismption of this ideal order; thus SouHk en Daun is a title that is supposed to belong to the A-Hne but is now held by the Nahnken himself as a sec? ondary titie; the title Lepen Net, which was formerly that of the Al instead of the title Nahnmwarki and belonged to clan 7, Is nowadays held by a man who is a member of clan 6, hence in the B-lIne, and he foUows the A2 in food distribution. The order of division is also mtermpted in aU tribes by the presence of visitors {tohn kapar) from other tribes, who share very generously In the food distribution. If there are only two pigs and only one visitor from another tribe is present, he is supposed to get one of the pigs. If there is only one large pig. It goes entu-ely to the Nahnmwarki, except for the legs and ribs. In Net the right foreleg goes to the Nahnken, the left foreleg to the Nahnm? warki's wife, the right hind leg to the Nahnken's wife; the other leg and ribs go to holders of lesser tities. But if there is only one pig a benevolent-minded Nahnmwarki should teU the divider to divide his portion among the people. In Kiti the beUy strip from the left side is laid on a leaf and put before the Nahnmwarki; he may redistribute it if he wishes. Another portion from the left side is simUarly given the Nahnken. In Net the equivalent portions for the Nahnmwarki and Nahnken are ribs, from either side. These portions are in addition to those received in the distribution procedure caUed pwekpwek (see p. 100), and they are intended to be eaten at the feast; most of the rest of the meat Is taken home and recooked, since there is actually little eating at a feast. Eating is considered im? proper when kava drinking, the principal activity at a feast, is going on. In Kiti aU of the left forelegs remain hanging In baskets during the division; after the division Is made they are taken down and the Nahnmwarki disposes of them as he pleases. He usuaUy gives some to each of the holders of the first 12 tities in the two lines and divides what is left with the Nahnken, taking his own share home. Besides the left forelegs, the head belongs to the Nahnmwarki and the portion of the back immediately behind the neck to the Nahnken. Actual procedure varies widely. Records were kept of three feasts in Net; at one, of three large and seven smaU pigs, the Nahnmwarki received the head, back, and right foreleg and shoulder of the largest, and his wife the two hind legs; the Nahnken, who Is the Nahnmwarki's father, received one whole pig and another left foreleg, and his wife received two hind legs. This left only the remaining portions of one large pig and the seven smaU pigs for divi? sion among more than 80 people. At a second feast, where some 80 or 90 people were present, the two largest pigs went to the Nahnmwarki and Nahnken, and a leg apiece from other pigs to their wives; three pigs were divided among the visitors from the other four tribes; the remaln- mg six pigs were left for the other people. At the third feast, one whole pig was given afive to the Nahnmwarki as weU as the head and back of the largest pig; one leg apiece from this pig went to the wives of the Nahnmwarki and Nahnken; the second largest pig was given to the Nahnken; four pigs were given to the visitors from the other four tribes; three pigs were eaten; and the remaining five pigs and the remaining portions of the largest pig were divided among the rest of the people. The food divider is in theory autonomous as far as the performance of his duties is concemed, just as Is the kava distributor {soun dei sakau). His job Is to apportion the meat and vegetable food baked In the stone ovens and such food as sugar cane, which does not have a high pres? tige value. The B2 of Net, who is the principal divider (and also kava distributor) of that tribe, asserts that he would dispute with the Nahnmwarki or Nahnken if they attempted to interfere with his proper functions. To be 102 SMITHSONLVN CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 sure, at smaU gatherings the writer has seen chiefs of vari? ous ranks issue sotto voce instructions to the man, ap? pointed only for the occasion, who did the dividing, or even countermand an apportionment that displeased them. But over the uncooked yams, which constitute by far the largest part of the yams brought to a feast, the divider has no authority except to present one such yam to the visitors from each of the four other tribes. The un? cooked yams aU belong to the Nahnmwarki and Nahnken, for they have been presented to them; they retain or dis? tribute them as they please, the food divider foUows their instructions with regard to them as weU as with their right? ful portions of the cooked articles. Hence there is consider? able opportuiuty for generosity or niggardfiness to make themselves evident, as there is in connection with pigs. The five present Nahnmwarkis are ranked according to pubUc opinion in the foUowing descending scale of gen? erosity; Kiti, Uh, Madolenihmw, Net, Sokehs; and the Nahnkens as foUows: Madolenihmw, Kiti, Uh, Net, Sokehs, The Nahnken of Madolenihmw is considered particularly generous, and is said to keep practicaUy noth? ing for himself. At the other extreme, the Nahnmwarki and Nahnken of Net are noted for their greediness, which is held to stem from their acquaintance vrith westem con? cepts of money economy. In Sokehs most of the chiefs are out-Islanders and do not foUow Ponapean feasting patterns. The divider makes two primary divisions of all the un? cooked yams: one heap is the Nahnmwarki's, the other the Nahnken's, Then he further divides them according to the wishes of these two chiefs, the best and largest yams being usuaUy retained by them. Men of lower titles in the A-Hne receive their shares from the Nahnmwarki's heap, those In the B-line from the Nahnken's. Commoners re? ceive what may be left over after the division. However, commoners have a device for augmenting their meager portions, known as yam seizing {doar or doarekehp). During the division of food the young people wiU occasionaUy make a rush for the yams and seize what they can and carry them off. The divider wUl scold at them but usuaUy only half-heartedly, and the Nahnmwarki and Nalmken "cannot scold because the yams wiU be planted and brought to another feast when they have grown," Most people are ashamed to do this sort of thing, but the elders wiU often whisper to their youngsters at the propi? tious moment to dash up and grab their share. The prac? tice is generally frowned upon, but the high chiefs In their ascribed role of generous and paternalistic despots are sup? posed to overlook it. A grievance against the Net chiefs is that they have forbidden the custom. It applies only to yams, but a similar practice, known as simw, occurs at the death feasts when the mourners, especiaUy the women, sudderUy swarm over aU the feast goods and carry off both cooked and uncooked yams, kava, pigs, etc. Cooked yams for consumption are distributed also ac? cording to rank. Visitors do not fare so weU here as they do with imcooked, plantmg yams and with pork; at a feast m Kiti attended by the writer a basket of cooked yams apiece was presented to the Nahmnwarki, the A2, the B2, and to the A6 (Oliver Nanpei); the Nahnken was absent; but the visitor from Sokehs, the B3 of that tribe, received only a single cooked yam, like holders of other, lesser titles, Kava There are several versions of the legendary origin of kava. One of them attributes the discovery to a rat, which nibbled at the root and whose actions were observed whUe under the Influence of the plant; thus the effect of con? suming the plant was leamed. This story Is found also in Westem Polynesia, Fiji, and on Pentecost hi the New Hebrides. Another story gives credit to Wuhtanengar, a native of section 10 in Uh, who was taken when an old man by the god Luhk to the mythical land in the south, Air, where he became young again and shed the skin off his f ootsole; the skin grew into a kava plant and a bit of its root, pounded in heaven, feU to earth in section 10; there it took root and grew. Another version combuies both themes. Hambmch (1932 II, p, 103) gives a simUar story, substituting the island of Kusaie for the land of Air, and he and Christian record stiU other accounts. Kava does not grow everywhere, but it is easy to tend; it is necessary to clean around it only once a month. The plant is grown from cutting, usually from the young branches of an old bush brought to a feast. The cuttings are made two joints long if the branch is more than an inch in diameter, four joints if less; they are severed diagonally, between the nodes. They are planted about one yard apart in cleared ground prepared first with a digging stick; later they are thinned out to two or more yards; but single plants may be seen sometimes growing in a thicket of other species of plants. The cutting is stuck into the ground somewhat diagonaUy to bury one node. Usually two cuttings are planted in the same spot to pro? duce a large plant. If they cannot be planted the same day they are cut they are bound into bundles and soaked In water by day, left in the dewy grass at night. A kava garden is Ccdled a kemenseng. A smaU one contains a hundred plants, a large one five or six hundred, A large garden is a great source of pride but is usually kept very secret, for fear of witchcraft which wUl cause the plants to dry up. Those men who are regular kava drinkers eat nothing before partaking. The high chiefs also are not supposed to eat before drinking kava, a rule not appHcable to com? moners. At most feasts a littie food {kepsakau) is served before the kava Is prepared; this is eaten by the nonin- PRESTIGE COMPETITION 103 dulgers^^ and the moderate drinkers. After each drink something is taken to paUiate the unpleasant aftertaste; this may be a sip of water, a morsel of pineapple, raw fish, sugar cane, a puff at a cigarette, or any of a large number of other things. Whether the drink Induces safivation or because of the disagreeable taste, many persons find it necessary to spit frequentiy between draughts; cracks be? tween floor boards are used for this purpose. The slime from the fresh hibiscus bast used as a strainer also pro? duces spitting. The immediate effect of the drink is a numbing of the fips and tongue; speech becomes thick, though the head remains clear. EventuaUy control is lost of the lower Hmbs, and men who have overindulged must be carried home. After retuming from kava drinking to one's house, food is eaten in ample quantities but It must be eaten slowly to experience the fuU effect of the kava. The result Is a deep and sound sleep. The next moming, whUe some persons seem to remain unaffected, others complain of a hang? over, have Httie appetite, and sHght noises seem to be intolerably magnified in volume. The writer observed none of the effects on the eyes and skin attributed to intemperate use of kava in parts of Polynesia, although Hambmch (1917, p. 113) describes overlndulgers whose skins looked as though they were sprinkled with a yeUow- green powder. Possibly the Ponapean variety of the plant produces a chemically different drink. Natives befieve that kava is beneficial to the health, and that heavy drinkers do not contract gonorrhea; whether this Is an aboriginal befief was not verified, and It is possible that they acquired the idea from the Germans, who listed kava in their pharmacopoeia, or the Japanese, who are said to have manufactured pills from kava as a specific against gon? orrhea. The Japanese are supposed to have bought sev? eral thousand yen worth of kava annually for shipment to Osaka to be reduced to pills; one native chief states that he sold 540 yen worth in one sale. Some of the pUls foimd their way back to Ponape for sale to the Japanese there. Dr. GIrschner, the physician on Ponape in German times, is said to have drunk kava every time the ship arrived, which it did semiannuaUy, In the befief that it prevented the cough the ship would bring. Hambmch (1932 II, p. 64) states that kava was taken as an aborti- faclent, but Informants deny this. In later Japanese times and especiaUy in the war years, when natives were impressed as laborers, there was Httle opportunity for feasting and for cultlvatmg kava; thus in 1947 kava was relatively scarce; but this was a temporary condition. During the war a large bush of kava, large enough to be carried by ten men, was worth a hundred yen. In 1947 many people were buymg kava for use at ?2 The Christian Endeavor movement has made a considerable number of converts among the Protestants; its members are pledged not to dance, smoke, or drink. feasts or privately. Enough kava for one stone, which would be about half of the roots from an average bush, sold for $5; a large bush, providing enough roots for five to 10 stones, was worth $50 or more. (Twenty people can drink for a whole evening of the kava produced at one stone.) By 1963 kava was much more plentiful. Earfier, the Protestant missionaries had made a deter? mined effort to eradicate the drink and the customs as? sociated with It, One Nahnmwarki convert in the 1870s is said to have uprooted large quantities of the plant. Never? theless, Joe Kehoe, an American trader in the 1880s and 1890s, accorduig to Christian, found it profitable to ex? port kava from Ponape to Fiji, so it must stUl have been in plentiful supply. By 1910 when Hambmch was at Ponape Its use had greatly declined, and he predicted that It would soon disappear entirely. But the efforts of the missionaries were destined to fail; they achieved perma? nent results only on Kusaie, the ordy other Micronesian island where kava and the complex of traits accompany? ing it existed aboriglnaUy. Kava and Feasting At a traditional feast, with fuU development of cere? monial, the kava bushes are carried to the community house In a procession of men singing a type of song called ngis and blowing conch tmmpets. There Is much noise and talk and frequent emission of the typical falsetto ululation of Ponape. The plants, carried root foremost, are brought up to the main platform, where the Nahn? mwarki sits under two taro leaves bound to a post that provide spiritual protection. The kava plants are placed with the roots and the ends of the carrying poles and fitters on the edge of the platform, the branches down in the central area. The kava bushes are frequently decorated with orna? mental plants. Large bushes carried on a fitter may have a stalk of croton or breadfmit inserted in them. A femlike plant {tehnlik) that cHmbs on kava branches is artificIaUy encouraged on the growing bush by tying it on the new shoots annuaUy, A man is proud of a kava plant he brings to a feast when each branch (very old plants may have as many as 100 or 150 branches) bears this fern. The prac? tice, of course, is somewhat akin to that of adding un? necessary bearers to a fitter in order to exaggerate the apparent weight of the yam, kava plant, or whatever ob? ject of prestige value is being carried. People consider It evidence of a man's wealth to be able to keep a kava bush without having to use it tiU it is old and very large. When the plants are removed from the maui platform to be pounded the roots are cut off and the stems taken away for replanting. The earth is knocked off the roots but there is no washing. The roots are not dried, as in 104 SMITHSONIAN CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 parts of Polynesia, but used fresh. In very large plants the lowest part of the stem can also be used. One whole bush was formerly placed in the beams above the maui plat? form of the community house as an offering to the spirit of the buUding,^^ and no one might remove It except the MwarekehtUi ("Httie Nahnmwarki," A12 to A16 in dif? ferent tribes), who alone has the privilege of reaching for it above the Nahnmwarki's head. But this is seldom seen nowadays. The roots after being cut Into smaU pieces are placed on the pounding stones and, especially at a large feast, they are covered over with taro leaves untU the signal is given to begin pounding. The taro leaves are then placed on the ground around the stone so that any kava that falls on them wdU not get dirty. The pounduig stones {peitehl) are large, flat or sHghtly hoUowed, basalt slabs. Irregular in outline and three or four feet in diameter (Plate 12a). They are propped off the ground by coconut husks, which help to bring out the clear, bell-like tone produced during the pounding. The pounders {moahl) used on them are cobblestones large enough for one hand to grasp convenlentiy. During the first part of the feast the stone ovens are arranged In the community house in two rows, from front to back between the side platforms, and occupy most of the central area, leaving room orUy for two kava stones at the front of each row, in Hne with the ovens, just below the main platform. These four stones are active in the preliminary kava preparation, caUed audida, before the ovens are opened. (This is apart from the "early" kava, sakau en menseng, mentioned previously,) After the food is divided and the ovens are cleared away, more kava stones of indefinite number are set up behind the first two stones of each row in place of the ovens, and the second stage of the kava preparation, caUed audsapahl, begins. But when kava is dmnk privately the arrange? ments are much less formal than in the feasting situa? tion and a single stone may be set up, perhaps outside of a dwelfing or in a cookhouse. OccasionaUy now? adays a sheet of iron is seen in place of a stone in such circumstances. Sometimes the four stones of the first phase are not seen, and audida does not commence untU aU the ovens are cleared away. The leader of the kava ceremony (the soun dei sakau) is usuaUy a man of high title. In Net nowadays the B2 is, because of his abUitles, designated for the job, but it does not seem that there is a connection with a particular title. In Madolenihmw he is said always to be a lesser Royal ChUd, The major requirement Is that he be weU In? formed on the proper procedure. He may not sit but stands '^ Hambruch (1932 II, p. 136) has a reference to an offering of kava in this manner to a deity at a regular priestly ceremony in Kiti, contuiuously at the central post of the main platform with his back to the notables who sit on this platform, facmg the workers and caUing out the various steps in the kava ritual; there Is no prompting from any of the chiefs behind hun. If someone drinks without his permission, he may knock the cup out of the man's hands; if any faux pas occurs, he scolds the guUty person pubHcly. A person called up from the central area to drink a cupful must come promptiy. The fiUed cup is supposed to be passed up to the leader by the kava pounders and he distributes it according to a sequence that, as wUl be detaUed, differs from one tribe to another. In Uh he gives the cup du-ectiy to the Nahnmwarki, but elsewhere he is supposed to give it to one of the servitors sitting around the Nahnmwarki who passes it on to the latter^* (Plates 11a, l i b ) . With "early" kava, however, at least in Madolenihmw, he always hands the cup to the Nahnmwarki; the servitor sits in the usual position but is bypassed. The writer has also sometimes seen one of the pounders pass a cup di? rectiy to a servitor instead of to the leader. There is no passing of the kava cup from one man to another; a man drinks and hands the cup back to his servitor, to be refiUed and passed to another. A man who can drink a whole cupful is greatiy admired. (This is indeed a feat, for the roots are pounded fresh and the drink is strong; the first pressings, which contain slime from the hibiscus-bast strainer,*? are of high flavor; and whatever earth adheres to the roots adds its own quafity to the concoction. The writer found that in Samoa, where the dried root produces a weaker and more watery fluid, he could easUy drink a whole cup of kava but he could manage only a few sips of the Ponapean variety.) It is considered good form to refuse a cup of kava when It is first offered and to point to someone else as the proper recipient, in keeping with the Ponapean pattem of mod? esty. This may explain the failure of a very rigid sequence of servmg to develop. The Polynesian traits of macerat? ing the root by chewing instead of pounding, the use of a mixing bowl, the kava circle, the formalized motions of preparing and servmg the drink, and the custom of pour? ing some of the liquid on the ground or flicking a drop into the air as an offering to the gods are aU absent on Ponape (though Hbations poured on the ground are men? tioned in some of the early fiterature). ??Fischer's notes seem to imply that direct passing of kava to the Nahnmwarki in Uh occurs only at the kava first fruits and applies only to the first four cups. "5 The use of fresh hibiscus bast for the strainer, which results in the slime in the first cups, is said to have begun about 1915. Sigismundo, who was later Nahnmwarki of Kiti, was the first to experiment with it, and it caught on because it produced a stronger drink and because it was less work to prepare the bast, which hitherto had to be scraped with clam shells and then dried. Even earlier the strainer is said to have been made of the "cloth" from the base of the coconut leaf. But hibiscus bark was in use in the 1850s (Gulick, 1858c, p. 27). PRESTIGE COMPETITION 105 There are, or formerly were, a number of minor offi? cials with duties connected with kava ritual. One of them, the Sapadan Sau,?" for example, was charged with clean? ing the stones used in the community house. Another official stands outside of the buUding and awaits the bringers of the hibiscus poles used for making the bast strainers. The kava workers (the tohn wie sakau) are commoners and holders of lesser A-titles. During the process of pounding, water is added to the roots from time to time by means of a half coconut sheU, which is replenished by dipping into a vessel of water (formerly the Ponapean boat-shaped kasak, of wood; nowadays, often, a bucket). Although the stone on which the pounding is done is flat, there is Httie spilling over, for the shredded roots absorb most of the water. When the roots are weU macerated their fluid content is wrung out by means of the hibiscus-bast strainer. Previously poles of hibiscus wIU have been brought and the bast stripped off and distributed among the workers. The bast in lengths of three or four feet Is knotted in the middle to form a bundle (nimakale) ^^ and keep the fibers from part? ing. This bundle is the strainer. GeneraUy two such bimdles of bast are apportioned to each stone. The bundle Is combed out with the fingers and laid out flat on the stone, doubled, with the knot at one end. Then the mace? rated kava root is heaped along the surface of the bast and two of the four workers each take one of these strainers, twist the ends to enclose the kava in a cylinder of bast, then raise it above the stone and wring it out tightly over a cup made of a half coconut sheU {ngarangar^^). The other two workers each hold one such cup In which they catch the streams of Hquid that run out. These two men are caUed respectively longpeik and soupeik. When not In use the cups should be kept in a doulong, which is a stick of hibiscus wood driven In the ground close by the stone, spfit in four from the upper end to about its middle, with two wedges to hold the four parts separate so that the vessel can rest in the spread upper end; but it Is seldom seen nowadays. Although the formal motions of Polynesia in straining and wringing are lacking, certain other be? haviors are required, such as keeping the two hands at the same level and keepmg the back of the fist tumed up, and ??Hahl (1901, p. 3) refers to him as a priest, as does Hambruch (1932 II, pp. 131-132), Hambruch gives a whole series of priestiy tides, each assigned some duty in connection with kava preparation and feasting. " Given as kots by Hambruch (1932 II, p. 246) and by some^ of my informants, but I also recorded the word kot as an altemative name of the bast itself. "8 A half coconut shell is used as a container for all sorts of liquids, especially medicines, but except when it is used for kava it is called a pohndal. (Some of Hambruch's texts, however, use this latter term for kava cup.) These cups are highly polished with a stone. Occasionally a china bowl is used nowadays, but in? formants say the flavor is then inferior. the workers are supposed to maintain proper decorum and not wear European clothing (Plates 12b, 12c, 12d). Kava Ritual Pounding of kava in the commuruty house Is done in unison and according to various rhythms. At one stone four workers normaUy do the pounding. Two of these men wUl pound wokpekid, which is a rocking motion of the hand, hittuig altemately with each end of the pounder projecting from the sides of the closed fist; the others pound tempil, a straight up-and-down motion. The term for pounding, sukusuk, includes both of these beats, but Kiti informants apply it also to the method caUed wokpekid elsewhere, Wokpekid is a sort of minor beat, heard throughout the strong, steady tempo of tempil; it is said to be not native In origin but an Imitation of the sound produced by the coopers aboard the whaling ves? sels of the last century; It Is used when there are a large number of kava stones being pounded at one tlme,?^ The pounding, as observed in Net, begins with reidi, a 4-note beat consisting of three short and one long beats. This is done only once,^ ?? on a bare part of stone. Then there is a short pause, after which pounding of the root begins in earnest, with some of the men pounding tempil, some wokpekid, on each of the stones. The rhythmic pounding on the bare stone is not done for the "early" kava. When the kava roots are weU pounded, another tempo begins called sokemwahu; this consists of three pairs of short beats foUowed by slow single beats that quicken rapidly and continue for a variable length of time, depending on the degree of maceration of the kava. This is repeated three tunes more, making four such rhythms In aU; but the fourth time I have occasionaUy heard four instead of three pairs of short beats. Soke? mwahu is the signal whereby people outside the commu? nity house are supposed to know that the kava Is almost ready and that the Nahnmwarki Is within. When the kava is aU pounded, there is a final rhythm called kohdi; this is again on the bare stone and consists of seven rapid beats, foUowed by a short pause, two rapid beats, another pause, two long beats, a pause, two rapid, and one long. After this the wringuig-out of the kava begins. In Kiti the reidi rhythm Is used again to end the pound? ing."^ In Awak, Uh, the final rhythm is caUed pedidi, but it was not ascertained whether this is the same rhythm as that caUed kohdi In Net, If the Nahnmwarki and B? Hambruch (1932 II, p. 246) refers to wokpekid as "Begleit- musik beim Stampfen" and to tempil as "Dampfen der Musik." '<"> Hambruch (1932 II, p. 242) gives this same rhythm, but more than once, as occurring only if the Nahnmwarki is present, and a different rhythm if only a lower chief is present. "1 Hambruch (1932 II, p. 246) likewise calls this "Schluss- musik." 106 SMITHSONIAN CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 Nahnken have stUl not entered the community house when the pounding is almost finished, the kava makers pound reidi, the beginning rhythm, untU these chiefs appear, then they pound pedidi.^^^ If the kava is almost ready, and the men who have been dispatched to get the hibiscus bast, used for wringing out the drink, have not retumed, the workers pound a beat called pepain, to hasten the gatherers, (But Net informants deny any special rhythm to make the hibiscus gatherers hurry.) AU of the stages In the pounding are directed by the leader of the kava ritual (the soun dei sakau) who stands on the main platform of the community house and caUs out the various tempos; for example, he wiU caU out "kohdi" and, at the same time, signal by dropping his arm from its raised position. There Is no leading of the actual rhythm, the pounders taking up the tempo from each other (or, at least as observed in Madolenihmw, from the pounders around a particular stone), but one of the pounders sometimes signals for the final long beat. There are said to have been special rhythms when kava was made for drinking by a shaman before he became possessed by a spirit, and other rhythms on various reH? gious occasions. There were apparently also various rhythms for playing on the Ponapean dmm {aip) for different ceremonies, Including the kava ritual, and songs to accompany the rhythms on both the stones and the dmm. These, along with the dmm, are long extinct. WhUe the kava is being pounded, a party of men wUl have been dispatched to gather the hibiscus bcist {dipenkeleu) in order to make the strainers. If they are late in retuming, in addition to the playing of the spe? cial rhythm the leader calls to the man who has been stationed at the central pole of the front entrance to the community house and this man (in Madolenihmw) calk to the laggards, "kohteUong, kohteunwahu, kumwa tangaklhdo dipenkeleu." (The first two words are antique phraseology; the translation is approximately "come-of- heaven, come-of-good, you bring-running hibiscus-bast.") The men answer, "nindzdawa, ninpalawa, ninpalelu, let se samwehr," the last portion of this meaning "Here we come." ?^^ An Uh Informant states that the man who does the caUIng out must be a Royal ChUd, since only Royal ChUdren can raise their voices at a pubfic assemblage, and that he stands in the soun, the space beyond the front >"* Hambruch (idem) refers to this as "Eingangsmusik." But he also, in one text (1932 II, p. 239) by a chief of Net, translates what appears to be the same word as a pause in the pounding rhythm, and in amother text (1932 II, p. 241) by the same man, as the conclusion of the pounding. The "entrance music" he (1917, p. I l l ) gives quite differentiy from any rhythm I recorded; it contains four different kinds of beats. It" The second word, ninpalawa, is said to refer to inipal, the natural cloth from the base of the coconut leaf, which is used in wringing out infusions of medicines today and supposedly was also used for kava anciently instead of the hibiscus bast used today. opening of the community house. The caU, in Uh, is given as "ninpalawa, nikoneleu, nipoikot"; this is largely untranslatable. In the Luelen manuscript, which has spe? cial reference to Kiti, the caU and response are given as one and are in Luelen's spelfing as foUows: "Kotel ina, kotei Uang; nintaleue, nipaleue, nipafio; kotel ina, kotei Uang," My translators give the first and third phrases as "wringer of the earth, wringer of heaven," making "kot" to mean "hibiscus bast wringer," instead of "come" as my Madolenihmw informants gave it. Each wringing from the pounded kava roots of the first pounding has its distinctive name, as given in Table 5. The term audida, referred to previously as denoting the first part of the kava ceremonial, seems to be used coUec? tively in Kiti only for the first four wringings, but else? where for all the wringings made from the first pounding. In Kiti the fifth to eighth wringings are caUed sapwe (meaning "end"). Fischer, who does not give the term audida, uses sapwe for the fourth wringing alone In Uh, which I also got as an altemative term in Madolenihmw (see Table 5) , These various stages of the ceremony, and the remarks on first and second poundings, and various wringings from the various poundings, should not be confused with the references previously made to "early" kava. At least four expressions are used that have to do with the time of the day during which the ceremony is held. The early kava (known variously, as already mentioned, as moming kava, quick kava, and the kava of the rising sun) has its counter? parts later in the day: noon kava (fiteraUy kava under the sun), evening kava (HteraUy kava of the setting sun), and night kava (HteraUy kava under flame, i,e,, Uluml- nated by fires). I am not certain that there Is any esen- tial difference between the last three occasions except for the time of day when they occur, but early kava, which is regarded as more important, has some unique features. For example, in rolling up the bast strainer to enclose the macerated roots, one hand holding the end of the strainer must be kept palm up, whUe the other hand does the roU? ing, In twistmg the strainer to express the liquid the hand continues to be held palm up, and the strainer must not loop over the back of that hand, which Is aUowed oiJy to be raised and lowered verticaUy during the straining. The strainer cannot be allowed to form more than one loop. After the cup Is fiUed the server takes it in both hands and walks to the Nahnmwarki in a kind of zigzag step, altemate steps taken sideways, at the same time muttering a prayer (though this last is seldom heard today). He then kneels before the Nahnmwarki, who re? ceive the cup with both hands directly from the server instead of through his attendant, holding it up high and, formerly, uttering some magic words, A smaU part of the macerated root from the first pounding is put aside before the juice is expressed for PRESTIGE COMPETITION TABLE 5.?Stages of kava ceremonial 107 Net Kiti Madolenihmw Uh 1. pwel (beginning) en sakau 2. are (second) en sakau 3. esil (third) en sakau 4. epeng (fourth) en sakau 5. alim (fifth) en sakau 6. aun (sixth) en sakau dipenkeleu 7. wong lopwon dipenkeleu 8. wong kep dipenkeleu 9. wong luh dipenkeleu 1. pwel en sakau 2. tu-e en sakau 3. esil en sakau 4. epeng en sakau *> 5. pwel en sakau 6. are en sakau 7. esil en sakau 8. lopwon, wong lopwon 9. luh, wong luh 1. pwel en sakau 2. are en sakau 3. esil en sakau 4. epeng en sakau 5. lopwon 6. kep 7. luh 1. pwilin sakau 2. arien sakau 3. esilin sakau 4. sapwen (end of) sakau (5. dipen keleu) 6. wung lupwun 7. pelien (mate of) wvmg lupwun (8. dipen keleu) 9. wung kep 10. pelien wung kep (11. dipen keleu) 12. luh * Information on Uh from Fischer's notes. t" In Wene, the eastem part of Kiti, this fourth stage is called kapahrek. This difference is attributed to the deposing of the Nahnmwzirki of Kiti, in the war with Pehleng, described previously and the subsequent defeat of Pehleng by the Soukise of Wene; but exactly why the kava terminology was affected is not clear. " Also recorded as sapwe, as Fischer gives it for Uh. the first cup. This baU of root fibers Is known as lop- won.^^* It is not used untU the cup known as wong lop? won {wong, to wring) is made. Madolenihmw Inform? ants say It is used only for this cup; according to Kiti informants a Httle of it Is added to each wringing after that cup also. After lopwon come the cups known as kep ?^^ and luh,^?^ but Kiti informants referred to several cups foUowing lopwon, all known as luh or wong luh. The term dipenkeleu properly means hibiscus bast, the material of the kava strauiing device. In Table 5, however, it is used metaphoricaUy to mean general drink? ing by commoners; a cup of kava is requested by a com? moner by his asking for dipenkeleu. In Fischer's list from Uh the term occurs as the name of certain cups in the series, but hi my list from Net it is uidicated only as stages between cups; informants from Kiti and Madolenihmw probably did not consider it worthy of mention. There is no pounding again untU the stages of audida have been completed and the shredded kava roots yield only a weak drink. At this pouit the master of the kava ceremony caUs out, "audsapahl," the name of the sec? ond major phase of the ceremonial, and fresh roots are brought up for the second pounding,"^ This call was re- ?osition in competition with his feUows, and an outstanding person of low birth may sometimes surmount barriers erected by mles of descent. The equUibrlum that results Is constantiy being eroded as men strive for and achieve advancement but is as rapidly reestablished. If a person becomes dis? affected he may resign his titie or seek self-exUe; but dis? harmonies are most commonly resolved through the for? mal feast of propitiation, which, because of the power ascribed to kava to achieve reconcUiation, is a powerful stablfizing mechanism. Public opinion is a deterrent to self-aggrandizement, A vacancy In the title series is filled by the Nahnmwarki and Nahnken, and the opportunity for these two men to serve selfish purposes would seem great. But unUateral decisions are seldom made; lower titieholders are consulted, public opinion is sounded, and deference is paid to the principle of a fair distribution of titles among the sub-clans of the mling clans, or, for tities held by commoners, among aU the common clans. Also the threat of war once kept the mlers from ignoring custom, just as they seldom rode roughshod over their tenants, despite the native theory that they owned aU the land. T H E PONAPEAN POLITY 111 Conflicts did occur, as noted previously throughout this study. Their causes were several but the most significant is the inherent contradiction between the nUes of matri- fineal seniority and the various other principles that are incorporated into the pofitical system. We have seen how the expressed native theory that older brother Is foUowed by younger brother, then by oldest sister's oldest son, faUs to be borne out In practice for many reasons, among them the just noted deference to fair distribution among fam? Uies and other kinship groups, personal traits and abUIties, achievement in war, etc. In the descent group seniority of birth has no relation to relative age, but In the practical? ities of native poHtics a child Is obviously not capable of being a ruler and a junior must often be placed ahead of his senior. Again, in a strictiy matrilineal group the posi? tion of the father has no bearing on seniority of blood, but with two lines of intermarrying chiefly groups the rank of both sides must count in determination of one's hier? archical position. Thus, when an Informant relates the theory of pofitical advancement he is essentially giving the rides of descent-group seniority; but the case histories re? lated in this work reveal how these rules must be accom? modated and compromised in applying them to a com? plex pofitical system. And it Is the application of them and their delicate balancing against all the other principles previously mentioned that result in a flexible but stable and workable state. The Ponapean poHty may profitably be compared with that of other parts of the Pacific, Micronesia, in which Ponape is one of the largest and most populous islands, is usuaUy neglected when such comparisons are made? for the very good reason that It has been, untU recent years, much less weU described in the anthropological fiterature than either Polynesia or Melanesia. These two areas are often contrasted with each other as being at opposite ex? tremes in the scale of poHtical development, Polynesia Is characterized by ranked and stratified classes or social levels, with power concentrated In the hands of men of title, and with chiefs or kings exercising authority over sometimes large areas, Melanesia, on the other hand, al? though much less homogeneous culturaUy than Polynesia (or even Micronesia) is for the most part described as classless, lacking in hereditary rank and position, demo? cratic, egafitarian, and segmented. In aboriginal Polynesia the senior members of the var? ious descent groups (which are generaUy not unifinear but do have a patrUineal bias) constitute the rufing class. Social position and succession are determined through descent. With rank goes a great number of prerogatives and privUeges, honorific forms, and deferential etiquette. Control over economic resources, power of life and death over commoners, tribute giving, and confiscatory rights are frequently occurring Polynesian phenomena. On the other hand, m most of Melanesia pofitical in? stitutions are vastiy sunpler and imcentralized, A leisure class and specialized functions are lacking. Hereditary chiefs are found in only a few places; instead, opportunity to achieve prestige is generaUy open to aU, by means of dUigence, skiUs of various kinds, mifitary abUIty, or other personal traits. PrivUeges, deferential pattems, and ex? ploitation of other men's economic activities give way to egafitarian principles. The kin-group dominates social sit? uations. Societies are smaU, usuaUy confined to the level of the viUage. In this kind of comparative survey, Micronesia is usu? ally placed by students of the subject with Polynesia. But actuaUy Micronesia is extremely varied in poHtical com? plexity. In the center of the chain of islands mnning east and west that comprise the Caroline Islands Is the com? plex atoU of Tmk. This group, which Hes nearly 400 miles west of Ponape, can be characterized in very much the same sweeping terms just used to describe Melanesia, with local viUage headmen, absence of central authority, and lack of hierarchical stratification. West of Tmk, ex? tending 600 mUes nearly to Yap, is a string of atoUs and coral islets that Includes among others Puluwat, Woleai, and Ulithi. Here again at each of these places there exists a simple, Melaneslan-type pofitical structure. In contrast, at the westem end of the chain and close to the Phifippines, are Palau with two major political groupings and with high and lesser chiefs, and Yap with two caste levels sub? divided Into nine endogamous strata; both correspond more nearly to the Polynesian characterization. Likewise, Kusaie, a high Island 300 mUes east of Ponape, and the MarshaU Islands, stUl farther east, possess intricately developed governmental organizations. Ponape would seem, at least at first glance, also to have the complex coloring of Polynesian pofitical institutions. Here too we encounter ruling classes, status ascribed through descent, centralized power, royal prerogatives, elaboration of honorific usages and language (and here the honorific forms are not limited to vocabulary, as In Polynesia, but are woven into the stmcture of the lan? guage Itself), Involuted etiquette, economic control by chiefs, tributary rights, power of confiscation and banish? ment, etc. We also find at the same time, however, a cer? tain degree of social mobUity, with status achieved through individual skUls and industry, prowess In war, and pres? tige competition, as in most of Melanesia, We get an im? pression of two major principles in conflict with one another, the Polynesian and Melanesian ones, which however become compatible in a pfiable yet weU-func- tioning organization through the operation of a set of checks and balances. A thoroughgoing Polynesian system, with succession of office, titie, and power through the male luie, would seem 112 S M I T H S O N I A N CONTRIBUTIONS TO ANTHROPOLOGY VOLUME 10 to requu-e a patrilineal emphasis, since its efficiency would tical function between them and serving to check each be enhanced by father-to-son uiheritance. In Ponape, others' authority; and it is further resolved by the previ- however, a "Polynesian" poHty (no knpHcation of his- ously described Melanesian egalitarian traits that func- torical connection Is intended) wears the aspect of a tion to create a state of balance between the ruling lines superimposition on a "Melanesian" social stmcture, one and the mass of the people. The effect is a flexible and based on matrlfineal clans. The conflict possibly Inherent harmonious political organization, which avoids both the in such a situation is ameliorated by the device of two excesses of power of Polynesia and the fragmentation of intermarrying clans ruling together, with division of poli- Melanesia. LITERATURE CITED ABCFM. (American Board of Commissioners for Foreign Missions.) Letters and journals from the year 1852 on, written by missionaries and others from Ponape: Gulick, Sturges, Doane, Roberts, Pierson, and Snow. (At Houghton Library, Harvard University, Cambridge.) BAKER, ROLLIN H . 1951. The avifauna of Micronesia, its origin, evolution, and distribution. Univ. Kansas Publ., Museum of Natural History, vol. 3, no. 1. BASCOM, WILLIAM R . 1948. Ponapean prestige economy. Southwestern Journal of Anthropology, vol, 4, pp. 2 1 1 - 221. 1949. Subsistence farming on Ponape. New Zealand Geography, vol. 5, pp. 115-129. 1965. Ponape: A Pacific economy in transition. Anthropological Records, vol, 22, Univ. California, Berkeley and Los Angeles. BENNIGSEN, RUDOLF VON 1900. Bericht iiber seine Raise zum Zwecke der Ubemahme des Inselgebietes der Karolinen, Palau und Marianen in deutschen Besitz. Deutsches Kolonialblatt, vol. 11, pp. 100- 112. BLAKE, P . L . 1924. Untitled reports comprising pp. 12-28 and 654-673 in Historical Records of AustraHa, Series I, Governors' Despatches to and from England, vol. 20, February 1839- September 1840. The Library Committee of the Commonwealth ParHament. GABEZA PEREIRO, A. 1893. La Isla de Ponape, Conferencia dada en Reunion Ordinaria de la Sociedad Geografia de Madrid el 24 de Noviembre de 1891. Boletm de la Sociedad Geografica de Madrid, XXXIV. 1895. La Isla de Ponape. Manila. CAMPBELL, DR. 1836. Ascension, The Colonist, New Soutii Wales, vol, 2, no. 78, June 22, 1836, pp. 193-194. C H E Y N E , ANDREW 1852. A description of the Islands in the western Pacific Ocean, north and south of the Equator. London. CHRISTIAN, F . W . 1899. The Caroline Islands. New York. DOANE, E . T . See under ABCFM. EAGLESTON, CAPT. J, H, 1832-33, Journal of the bark Peru. (Manuscript at Essex Institute, Salem, Mass.) FINSCH, O T T O 1880. Ueber die Bewohner von Ponape. Zeitschrift fiir Ethnologie, vol. 12, pp. 301-332. 1893. Ethnologische Erfahrungen und Belegstiicke aus der Siidsee. Vienna. FISCHER^ J. L. n.d. Unpublished Ponapean field notes, 1951-1953. 1958. Contemporary Ponape Island land tenure. In Land Tenure Patterns, Trust Territory of the Pacific Islands, vol, 1, office of the Staff Anthropologist, Guam, part 2, pp. 77-160. LITERATURE CITED 113 FISCHER, J O H N L . , with the assistance of Ann M. Fischer 1957, The Eastem Caroluies. Behavior Science Monographs, Pacific Science Board, in association with Human Relations Area FUes, New Haven. ERASER, CAPTAIN 1834. Discovery of William the Fourth Group of Islands, Pacific Ocean. Nautical Magazine, vol. 3, 1834, p, 74, FRITZ, GEORG 1912. Ad majorem Dei gloriam, Leipzig, GARVIN, PAUL, and RIESENBERG, SAUL H . 1952, Respect behavior on Ponape: An ethnolinguistic study. American Anthropologist, vol. 54, no, 2, pp. 201-220. GIRSCHNER, MAX 1906, Grammatik der Ponape-Sprache. Mitteilungen des Seminars fiir Orientalische Sprachen zu Berlm, vol, 9, pp. 73-126. GLASSMAN, SmNEv F. 1952. The flora of Ponape. Bernice P. Bishop Museum, Bulletin 209. GOMEZ, JUAN GUALBERTO 1885, Las Islas CaroHnas y las Marianas. Madrid. GRESSITT, J. LINSLEY 1954. Insects of Micronesia, Introduction. Bernice P. Bishop Museum, Insects of Micronesia, vol. 1. GULICK, ADDISON 1932. John Thomas GuHck, Evolutionist and Missionary. Chicago. GULICK, L U T H E R HALSEY 1852. Article in The Friend, December 17, 1852. 1853, Article in The Friend, March 1853, p. 19. 1857. Article in the Missionary Herald, vol, 53, pp. 41-48. 1858a. The fauna of Ponape, or Ascension Island, of the Pacific Ocean. The Friend, March 1858, p. 18, 1858b. The Climate and Production of Ponape or Ascension Island, one of the CaroHnes, in the Pacific Ocean. American Journal of Science and Arts, ser. 2, vol. 26, pp. 34-49. 1858c. The flora of Ponape, or Ascension Island. The Friend, April 1858, pp. 26-27. 1859. The Ruins of Ponape, or Ascension Island. Journal of the American Geographical and Statistical Society, vol. 1, no. 5, 1859, pp. 129-137. 1862. Micronesia. Nautical Magazine and Naval Chronicle, vol. 31, pp. 169-182, 237-245, 298-308, 358-363, 408^17 . 1872. A vocabulary of the Ponape dialect, Ponape-EngHsh and English-Ponape, with a grammatical sketch. Journal of the American Oriental Society, vol. 10, pp. 1-109, See also under ABCFM. H A H L , . 1901. MittheUungen iiber Sitten und rechtHche Verhaltnisse auf Ponape. Ethnologisches Notizblatt, vol, 2, part 2, pp, 1-13, 1902. Feste und Tanze der Eingeborenen von Ponape. Ethnologisches Notizblatt, vol. 3, part 2, pp. 95-104. 1904. Ein Beitrag zur Kenntnis der Umgangsprache von Ponape. MitteUungen des Seminars fiir Orientalische Sprachen, vol. 7, pp. 1-30. FEALE, HORATIO 1846. Ethnology and Philology. In United States Exploring Expedition during the Years 1838-1842, under the Command of Charles WUkes, U.S.N., vol. 6, PL\MBRUCH, PAUL 1917, Die kava auf Ponape. Studien und Forschungen zur Menschen- imd Volkerkunde, vol. 14, pp. 107-113. 1932-1936, Ponape. Ergebnisse der Sudsee-Expedition 1908-1910, ed, G. ThUenius. II.B.vii. 3 vols. Berlin, vol. 1 by Hambruch alone, vol, 2 by Hambruch with A, Eilers, vol. 3 by Eilers from Hambruch's notes. (Cited as Hambruch 1932, I, I I , and I I I ) . HERNSHEIM, FRANZ 1884, Siidsee-Ermnerungen (1875-1880), Berim. JAMES, HORTON 1835, Notes on the Island of Ascension, Pacific, Nautical Magazine, vol. 4, 1835, p. 708. 114 SMITHSONIAN CONTRLBUTIONS TO ANTHROPOLOGY VOLUME 10 JORE, L E O N C E 1953. Captain Jules Dudoit, the first French consul in the Hawauan Islands, 1837-1867, and his brig-schooner, the Clementine. 64tii Annual Report of the Hawauan Historical Society, Honolulu. KITTLITZ, F , H , VON 1858, Derdcwiirdigkeiten einer Reise nach dem russischen Amerika, nach Mikronesien und durch Kamtschatka, Gotha. 2 vols. KNIGHT, J O H N B . 1834. Letter to S. C. PhUlips, dated January 18, 1834, at Peabody Museum, Salem, Mass. 1925. A journal of a voyage in tiie brig "Spy," of Salem (1832-1834). Pp. 168-207 of The Sea, the Ship, and the SaUor, Publ. no. 7 of the Marine Research Society, Salem. KOLONIALAMT, GERMANY 1911, 1912, 1913, and 1914. Die deutschen Schutzgebiete in Afrika und der Sudsee. (Four successive years). Berlin, KUBARY, J, S. 1874. Weitere Nachrichten von der Insel Ponape. Journal des Museum Godeffroy, vol. 3, part 8, pp, 261-267. LHOTSKY, J O H N 1835. Ruins of an ancient town in one of the South Sea Islands. New South Wales Literary, PoHtical and Commercial Advertiser, February 1835, LUTKE, FREDERIC 1835-36, Voyage autour du monde execute par ordre de sa Majeste I'Empereur Nicolaus I?' sur la corvette "Le Seniavine," dans les annees 1826, 1827, 1828, et 1829, Paris. 3 vols. MAHLMANN, J O H N JAMES 1918. Reminiscences of an ancient mariner. Yokohama. MAIGRET, DESIRE 1837-38, Diary. (Manuscript at Chancery Office, Roman Catholic Diocese of Honolulu.) 1839. A letter, pp. 95-101, of Lettres des premiers missionaires de la congregation des sacres- coeurs, dite "De Picpus," MARKHAM, CLEMENTS ROBERT 1904, The voyages of Pedro Fernandez de Quiros, 1595 to 1606. Translated and edited by Sir Clements Markham, London. MICHELENA Y R O J A S , FRANCISCO 1843. Viajes cientificos en todo el mundo, desde 1822 hasta 1842. Madrid. MIGUEL, GREGORIO 1887. Estudio sur las Islas CaroHnas. Madrid, Moss , FREDERICK J. 1889. Through atoUs and islands in the great Soudi Sea. London. NAVARRETE, MARTIN FERNANDEZ DE 1837. Goleccion de losViagesy Descubrimientos. Vol, 5, Madrid, O ' C O N N E L L , JAMES F . 1836. A residence of eleven years in New HoUand and the Caroline Islands. Boston. OSBORNE, ALICK 1833, Notes on the present state and prospects of society in New South Wales, with an his? torical, statistical, and topographical account of MaiuUa and Singapore. London. PIERSON, GEORGE. See under ABCFM. REICHSTAG, GERMANY 1902 through 1911, annually. Denkschrift iiber die Entwickelung der deutschen Schutz? gebiete in Afrika und in tier Siidsee. Stenographlsche Berichte iiber die Verhand- lungen des Reichstages. Berlin. RIESENBERG, SAUL H , 1948. Magic and medicine in Ponape. Southwestern Journal of Anthropology, vol, 4, no. 2, pp, 406-429. 1959, A Pacific voyager's hoax, Ethnohistory, vol. 6, no. 3, pp. 238-264. ROBERTS, EPHRAIM P , See under ABCFM, STURGES, ALBERT A. See imder ABCFM. W I N N , J O H N D . 1833-35. Log of the ship Eliza, (Manuscript at Essex Institute, Salem, Mass,) LITERATURE CITED YANAIHARA, TADAO 1940, Pacific Islands under Japanese mandate. London and New York. YZENDOORN, REGINALD 1927, History of the CathoHc Mission in the Hawauan Islands. Honolulu. T H E FOLLOWING UNSIGNED ARTICLES AND NOTES, IN : The Friend, Honolulu, September 1,1850; December 17, 1852; May 1853, The Hobart Colonial Times, May 25, 1827, The Puritan Recorder, Boston, vol. 40, no. 38, September 20, 1855, p. 149, The Sydney Gazette, June 15, 1827; May 15, 1830; November 13, 1830; April 8, 1834; May 8, 1834. The Sydney Herald, May 8, 1834; September 4, 7, and 28, 1837. Article entitled "Wreck of the Harmony," Nautical Magazine, 1838, p. 138. 115