Translating the Untranslatable
A solution to the Problem of Generating English Determiners

This page contains some information about my book: the blurb, the bibliographic details, a review and some errata.


Book Cover: Translating the Untranslatable One of the hardest problems for automatic translation is generating words and inflections that are obligatory in the target language, but not in the source language. This work is the first to provide a fully implemented solution to the problem of generating determiners and determining number. Using a semantic representation and a series of three heuristic algorithms, this solution provides the most probable context-sensitive translation. It gives insight into natural language processing in general, and machine translation, semantic analysis, and generation from underspecified inputs in particular. This book finishes with an extensive evaluation of the algorithms implemented in a Japanese-to-English machine translation system.

ISBN (Paperback): 1-57586-460-6

ISBN (Cloth): 1-57586-459-2


There is a
review by Shigeko Nariyama on the linguist list.

Bibliographic details:

Errata (December, 2005):

Most of the errors are typos in the Japanese.


Chapter 2: Background

Chapter 4: Semantic Representation of Reference, Countability, Number and (In)Definiteness

Chapter 5: Automatic Interpretation of Reference, Number and Determiners

Chapter 6: Evaluation and Discussion

Much thanks to Ryo Nagata, Shigeko Nariyama, and Kiyonori Ohtake for pointing out some of the errors.

Francis Bond <>
NICT Computational Linguistics Group
National Institute of Information and Communications Technology
3-5 Hikaridai, Seika-cho, Soraku-gun, Kyoto, 619-0289 Japan
Tel: 0774-98-6847 (+81); Fax: 0774-98-6961 (+81)