Encounter of Language and the Brain in My Life 　 Dr. Michio Sugeno Head, Laboratory for Language-Based Intelligent Systems I have recently been trying to illuminate the paradigm of everyday language computing. My Lab, which belongs to the Brain-Style Intelligence Research Group, is doing research on this subject. The Lab is currently made up of 15 full-time and part-time researchers, eight of whom specialize in linguistics. I first took a strong interest in language in the latter part of the 1970's when nearly a decade had passed since I started my study of fuzzy theory. The word "fuzzy" refers to uncertainty or unclearness as represented by the ambiguity that our verbal expressions entail. In the U.S., the concept of an "intelligent system" appeared around the mid 1980's. To me, however, it meant not just a logical intelligence but rather a fuzzy "one" as seen in our ordinary and ambiguous use of words. I could refer to it as "clinical intelligence" (Yujiro Nakamura(1)). I got involved in "fuzzy" theory, as I was more attracted to Pascal's "depth of subjectivity" than to the objectivism of Descartes, Marx, etc. I then got attached to Wittgenstein(2) through the "Intelligence of language." As a result of a linguistic turn in philosophy (R. Rorty(3)), the 20th century has been hailed as the "century of language." This is specifically indebted to Wittgenstein. It is now assumed the 21st century will become the "century of the brain." Following the "century of language," however, it might rather turn out to be the "century of linguistic intelligence." In those days, I had no idea about linguistics, but I gradually developed a firm belief that there must be a theory of linguistics much closer to Wittgenstein's point of view than to Chomsky's(4). I then came across Halliday's linguistic theory in the late 1980's. Halliday originated systemic functional linguistics. He belonged to the London School in the wake of Malinowski who advocated the context of situations. I attach importance to language as something almost synonymous with intelligence because I am curious about how a human brain processes intelligent information. Human beings use language to understand and fuse the information they obtain from the outside world, through their senses of sight, hearing, taste , etc. More importantly, however, we could say that the articulation of the world through language is exactly what distinguishes the intellectual recognition of humanity. In 1997, the Ministry of International Trade and Industry conducted a survey of new computer architectures, and I was appointed as general manager of a working group evaluating new technologies in the field of intelligent information processing. The working group consisted of members from the fields of AI, fuzzy theory, neuro and functional linguistics, including Professors Minoru Tsukada and Kazuyuki Aihara from the area of neuro. In March last year, the working group organized a proposal entitled, "The Development of Everyday Language Computing Technology." The purpose of this proposal was to assign language comprehension to computers and thereby "verbalize" all the information processing across computer networks. A language is to a brain what software is to hardware. A language is born in the brain and mostly develops outside it. It is said, on the other hand, that a language thus developed will backwash the development of the brain. Thus, we could infer that there is an inherent or homomorphic correlation between the structure of a language system as previously explained in linguistics (but not Chomsky's grammar) and the structure of a speech area in the human brain. I'm determined to implement a "recombination" from logical intelligence to pursue the "intelligence of language." By so doing do, I wish to discern what a brain-style intelligent architecture should be. Myself, second person from bottom left. (1) A typical philosopher in modern Japan who introduced French structuralism to Japan. (2) One of the greatest philosophers in the 20th century. Regarded as comparable to Heidegger. He discerned that philosophical issues are none other than linguistic issues. (3) A typical philosopher in modern American philosophy. He generalized 20th century philosophy by describing the Wittgenstein school of linguistic philosophy as a "linguistic turn." (4) A Cartesian linguist who claimed that an infant could learn a language readily due to the "universal grammar" inherent in human brains.