Stanley Starosta passed away on July 18th, in Hawai`i, from heart complications. He was 62 years old. To general linguists he was well-known as the founder and tireless advocate of lexicase, a theory of grammar he described in a nutshell as a "panlexicalist monostratal dependency variety of generative localistic case grammar": to specialists of Asian languages he was best-known for many contributions to the grammar of languages of the region.

He was a grammar man; he wrote his dissertation on Sora, an Austroasiatic language of India, on which he did extensive fieldwork, learning the language as he went (he knew many, including Mandarin and Minnan, learnt them easily. His best was German). At the University of Hawaii, where he was offered a a teaching position in linguistics, and where he taught until his death, he became interested in Austronesian: this led him to the mountains of central Taiwan, for fieldwork on Saaroa. In a short time he became a leading authority on Austronesian grammar. He encouraged his students to do fieldwork and to produce synchronic descriptions in the lexicase framework, mostly of languages of Asia and the Pacific. He spent an enormous amount of time with his students, many of whom have since achieved eminent positions. His ideas live through their work, too.

Stan did not limit his field of investigation to synchronics. In Austronesian linguistics his main concern was the reconstruction of early Austronesian grammar. He introduced the notion that Austronesian grammar was ergative, instead of a sui generis "focus" system without equivalent elsewhere in the world. He was one of the authors of the famed "SPQR" theory (SPQR stands for Starosta, Pawley and Reid 1982, with Q added as a joke to make it sound like "Senatus Populus Que Romanus", the Roman Senate's authoritative signature): that theory states that the main "focus", or voice markers of Austronesian grammar began their careers as nominalizing devices. Lately he was concerned with the higher Austronesian subgrouping, trying to demonstrate Harvey's 1982 conjecture that Malayo-Polynesian was actually a subgroup of one of the Formosan branches of Proto-Austronesian, on the basis of uniquely shared innovations in grammar.

He kept an open mind. When around 1988 I told him about my ideas about the relationship of Chinese and Austronesian, he did not laugh, but offered constructive criticism. He believed in testing hypotheses againt facts. His last paper presents a hypothesis that all language families of East Asia (Sino-Tibetan, Austronesian, Miao-Yao, Austroasiatic, Tai-Kadai) are in fact genetically related. In it he proposes a specific historical scenario putting together facts from archaeology and linguistics, in order to make the proposal "easier to support, correct, and/or refute". He was putting the final hand to this paper on July 6, twelve days before he died, as I know from his email, the last I received from him.

He was also an extremely witty man, who loved to tell jokes, of which he possessed an unending supply, good ones too. He could do an uncanny duck imitation. He enjoyed discussing linguistics over a beer and peanuts. In scientific discussions, he was always rational and courteous. The truth mattered to him greatly. His friends, colleagues and students sorely miss him.

Laurent Sagart