Linguist James Marshall Unger in his book, The Fifth Generation Fallacy, argued that the primary motivation for FGCS reasearch was to develop NLP for the Japane

Linguist James Marshall Unger in his book, The Fifth Generation Fallacy, argued that the primary motivation for FGCS reasearch was to develop NLP for the Japanese writing system. He claimed that the project was doomed to failure because the high level of ambiguity in the Japanese writing system required contextual understanding for even simple processing tasks. In that respect, things went as he predicted. Do Japanese IMEs like ATOK have any heritage in FGCS work?