In MemoriamJean-Claude Rivierre, 1938–2018
We would have celebrated his 80th birthday in June 2018.
He left us without warning, to join David, one of his sons, and his wife Françoise Ozanne-Rivierre (see Tryon 2008), an eminent linguist like himself.
At his funeral, one of his nephews aptly characterized his personality in terms of four qualities: timeless, balanced, humble, and straightforward, as somebody who never praised himself for what he had done for the Kanaks, whose gratitude and admiration of him has never diminished over the years. With his in-depth analyses of several Kanak languages of the Far South and of the Isle of Pines, in the form of dictionaries (Paicî, Cèmuhî, Numèè, and Bwatoo), grammars (Cèmuhî), and analyses of sound systems, Jean-Claude Rivierre gave invaluable gifts to the whole New Caledonian community. Equally remarkable and valuable, however, is the legacy he left to the entire linguistics community, in his studies on the genesis of tone, in his comparisons of tonal and accentual systems, and in his work on the appearance of certain complex sounds, rarely attested in the languages of the world, such as aspirated nasal consonants, postnasalized consonants, and labiovelars. Many of these studies were jointly written with Françoise: their cooperation and the passion they shared for their work inspired great respect in those who knew them.
A very important part of Jean-Claude's work was devoted to the transcription, translation, and analysis of orally transmitted stories, myths, or historical tales about the organization and creation of social groups headed by a chief. Most of these recordings were collected by himself during his long stays in New Caledonia, mainly in the central part of the Mainland. Over several decades he also engaged in close collaboration with the anthropologist Alban Bensa, with whom he published several books based on Paicî and Cèmuhî stories.
Mention must also be made of his generous availability for students from different disciplines, who often came from far away just to have the opportunity to meet him and exchange ideas on phonological or ethnolinguistic topics with him.
Jean-Claude Rivierre was born on the 18th of June 1938, in Le Sap, a small village in Normandy. He met Françoise during his last year of high school; they both pursued their education at the University of Caen, where Jean-Claude took a course in Classics (mainly Greek and Latin studies). Then he went to Paris where he took Master's courses in sociology at the École Pratique des Hautes Études. This is where he met the two well-known [End Page 248] personalities who introduced him to linguistics and to Oceanic studies: André-Georges Haudricourt (see Ozanne-Rivierre and Rivierre 1997), an ethnoscientist and linguist, and the ethnolinguist Jacqueline Thomas (see Motte-Florac and Guarisma 2004). They both used to recall how they became immediately aware of Jean-Claude's extraordinary accurate ear and interest for languages, and introduced him to linguistics through André Martinet's courses at the Sorbonne.
In 1964, Jean-Claude became a member of the French National Centre of the Scientific Research (CNRS), first as a member of A.-G. Haudricourt's Cooperative Research Program, which a few years later merged with J. Thomas's own team to give rise to the "Oral languages, field work and description" research team (ER 74). And in 1976, Jean-Claude was one of the founders of LACITO (Langues et Civilisations à Tradition Orale), a department of CNRS specializing in language description and documentation that he directed from 1991 to 1995.
Jean-Claude's first fieldtrip, which he undertook with Françoise and Nicolas, their one-year-old first child, required a month-long journey by boat from France to New Caledonia, and lasted for more than two years, from January 1965 to May 1967. Under Haudricourt's supervision, Jean-Claude undertook an extensive phonological study of the languages of the Far South of New Caledonia's Grande Terre and the Isle of Pines. This study was the basis for his PhD dissertation, completed in 1970 (EHESS-Paris), and published in 1973. On the same trip, Jean-Claude also spent some time studying some languages of the New Hebrides (now Vanuatu). But crucially, this is also when he started exploring the tonal languages of the Center of the Mainland. At that time he played the trumpet, and had already developed a great passion for jazz: he liked to compare tone languages and the improvisations of jazz. With his wife and A-G. Haudricourt, Jean-Claude contributed substantially to establishing the Kanak languages as a subgroup of the Oceanic branch of the Austronesian family, by providing clear evidence for complex but regular sound correspondences with the help of the comparative method.
TONAL LANGUAGES OF NEW CALEDONIA
Among the 28 Kanak languages, five have tones, all localized in the Center (Paicî, Cèmuhî) and in the Far South (Numèè, Drubéa, and Kwênyïï) of the New Caledonian Mainland. Some of these tonal languages had already been identified as such by George W. Grace in 1955, with additional evidence provided by A.-G. Haudricourt on the basis of field work carried out between 1959 and 1963. Shortly afterwards, Haudricourt sent Jean-Claude to this region in order to describe these tonal systems and to test his own hypotheses concerning their origin, namely that reduplication of CV syllables is a process widely used in Oceanic languages, yielding geminate consonants, which then further evolve leading either to aspirate consonants or to high tones.
Jean-Claude (Rivierre 2001) partly validated Haudricourt's hypothesis, showing the existence of regular correspondences between high tones and aspirated consonants, but proposed to reconstruct aspirates instead of geminates in Proto-Neo-Caledonian Mainland. However, reduplication, still productive in Iaai, a language spoken in Ouvéa (Loyalty Islands) and studied by Françoise, can be assumed to have evolved into aspirated consonants through gemination. Another focal point in the analyses carried out by Jean-Claude [End Page 249] was the role of stress in the genesis of tone in the Kanak Mainland languages. The tone reversal that occurred between the languages of the Center of the Mainland and the languages of the Far South, can be partly explained through the stress system of the languages spoken between these two groups (Rivierre 1978).
Here is a more detailed version of the story of this discovery: coming back from my first fieldwork in New Caledonia, I invited Jean-Claude and Françoise to my home, together with Damas Toura and his wife. Damas was the big chief of Thio, a community located on the east coast where both Xârâcùù and Xârâgurè are spoken. Listening to the couple, Jean-Claude's exceptionally good ear was immediately aware of what he then called "chutes accentuelles" (falling accents). Three months later, he spent a few weeks in Thio recording data to illustrate these falling accents, which explain the fact that tones of the Far South languages show a reverse contour compared to the tones of Paicî, one of the Center languages. In the Xârâcùù noun phrase, a "phénomène de couplage accentual" (a process of stress-based word-bonding) occurs, while in the clause, the same terms will have a disjoint stress, each term being stressed on the initial syllable, and at the same level:
(1) noun phrase clause
/ \ \ / / \
döö béchâ döö béchâ
'flat ground' 'the ground is flat'
In serial verb constructions (SVCs), Jean-Claude observed a similar phenomenon. From the point of view of accentuation, SVCs form a unit of special schema: a sequence of falling accents analogous to a waterfall can be observed: the main accent falls on the first syllable of the first verb, whereas the accent on the first syllable of the second verb of the serial construction has less prominence than the one on the first verb.
PANCHRONIC MODELS OF EVOLUTION
Internal developments and contactinduced phonological complexification (Rivierre 1994) are the main forces explaining the very different and rich phonological systems of New Caledonian languages. Diachronic aspects of nasalization processes leading to prenasalized and postnasalized consonants (Ozanne-Rivierre and Rivierre 1996), as well as inherited features vs. recent innovations as far as labiovelars are concerned, partly explain the complex phonological systems of Kanak languages. Contact between small neighboring communities, practicing "bilingualism on an equal footing" (the plurilinguisme égalitaire described by Haudricourt) accentuated the diachronic enriching tendencies. Taking into account that the Kanak social organization1 has indeed contributed to the understanding of the complexity of the phonological systems of these languages: "sectors of the phonic system which had become incomplete through the process of evolution are reactivated or completed by means of expressive lexical creations and loans" (Rivierre 1994:519). [End Page 250]
THE "TRANSITIONAL" TENSE-ASPECT MARKER
Jean-Claude had a deep respect for the languages he studied, and was always very cautious not to betray their linguistic originality in his descriptions. His Cèmuhî grammar was written using some innovative terms in order to illustrate some distinctive characteristics of the language. For example, he invented the term "transitional" (aspect transitionnel in French) to designate the curious behavior of a tense-aspect marker he had first identified in Cèmuhî, but which exists in most Kanak languages, and in some other Oceanic languages as well. The term "transitional" given by Rivierre, in his Cèmuhî grammar (1980), is due to the fact that the meaning of this tense-aspect marker relates to the notions of transition and change. And indeed, tense-aspect systems in Kanak languages have a form (Xârâcùù wâ, Drehu hë, Cèmuhî ucè, Paicî caa, Nemi nga, etc.), that associates different values such as perfective, perfect, present perfect, depending on the context and the meaning of the verbs.
PART OF SPEECH CATEGORIZATION
In his Cèmuhî grammar, and as one of the precursors of the debate on the noun-verb opposition in Oceanic languages, Jean-Claude Rivierre wrote "[L'opposition nom/verbe] est infiniment moins tranchée que dans une langue comme le français, et il doit être entendu que ces dénominations traditionnelles s'appliquent à des réalités différentes. Loin de s'opposer terme à terme, noms et verbes ont en commun certaines propriétés combinatoires ou fonctionnelles" (1980:78–79).2 He described nominal lexemes as bases whose meaning orients them more to a nominal or a verbal use or to indeterminacy between these two uses as is the case in action nominals.
What struck Jean-Claude was the unique combinatorial capacity of nouns, and the equally great capacity of stative verbs, compared to that of active verbs.
After Françoise's death in 2007, Jean-Claude would not go back to New Caledonia, and was not very keen to go on working on purely linguistic topics. Instead, he invested his time in his ongoing collaboration with the anthropologist Alban Bensa, and in archiving his unpublished data and recordings. Jean-Claude had been responsible for the production of Kanak oral CDs delivered in May 1998 for the inauguration of the Tjibaou Cultural Center in Noumea. With the help of two technicians, he indexed many other oral texts that are accessible online on Lacito's Pangloss Collection: 57 Cèmuhî texts, 8 Numèè texts, one long text in Kwênyïï. In addition, more than 10 hours of Paicî recordings are waiting to be transcribed and translated. There is also hope that most of the missing data will be found in Jean-Claude's cupboards.
Additional major contributions concern general information on Kanak languages, their fragile political status, their neglected role in school curricula, and their final recognition as languages of France,3 after years of active political fights. Jean-Claude was the recognized expert linguist for the Paicî-Cèmuhî linguistic area of the Kanak Languages Academy. [End Page 251]
To conclude on a personal note: I certainly owe some of the richest experiences in New Caledonia to the Kanak themselves. My approach to fieldwork, however, would not have been the same without the support received from my friends and colleagues, Françoise and Jean-Claude Rivierre: they passed on to me their passion for this country, liberally sharing with me their knowledge of the field and their social network. As John Lynch recently wrote to me, "Jean-Claude was such a lovely man: a gentleman and a scholar, as is said in English." Or to reverse the roles: a scholar, and a gentleman.
JEAN-CLAUDE RIVIERRE'S MAJOR PUBLICATIONS
Full linguistic descriptions
Articles on phonology, tonogenesis, verb compounding, etc.
Publications on Kanak oral literature and social organization
Other publications concerning general information about Kanak languages, their role in education, and their political status
1. The absence of a centralized power and linguistic leveling associated with language contact as a result of exogamous marriages, was to a large extent at the origin of the diversification found in Kanak languages.
2. Translation: "[The noun/verb opposition] is infinitely less contrastive than in a language like French, and it must be understood that traditional names are applied to different realities. Far from opposing one term with the other, nouns and verbs have in common certain combinatory or functional properties.
3. "The languages of France" refers not only to French, but to regional languages as well as minority languages that have been spoken by French citizens within the borders of France long enough to count as part of the country's cultural heritage, without being the official language of any state. In 2016, 75 languages are recognized as being "languages of France."