Interesting Lecture on an Intricate Subject An ai tive audience listened toa very interesting lecture delive: ed at the Royal Asiatic Society jon April 24 by Mr, T. Chu, 31.4. (Columbia), on "A new! cataloguing system for old Chinese characters." The Rev. F. L. Hawks Pott presided.
After explaining the term "word-cataloguing," Mr, Chu said that the great value of a good cataloguing system lay Tu the fact that it was the one point hat made a dictionary helpful. The formation of a good cataloguing system for the Chinese dictionary was of great difficulty owing to the lack of an alphabet. Many systems had been devised but few were really satisfactory, due to the great num-
Having described. the various! systems which had been invented, the speaker said that he was much interested in the four-corner method which, taken as a whole, was an improvement over the one it sought to replace. Unfortunately, it was little too complicated and the four corners were difficult to determine.
The speaker started the compilation of his catalogue of words with the study of a single stroke, the first one at the outset and then the last one.
After many difficulties, he employed two strokes simultaneously, both the first and the last, using one Arabic numeral to represent each as they appeared. During the initial stages of his work, Mr. Chu used two! Arabic numerals to represent the strokes, the first digit for the first stroke and the last digit for the last. The Arabic numerals were used for the sake of orderly enumeration, thus:—00, 1, 02, 03, 29, there were altogether 100 niches for the placement of words.
After his work had progresse/, Mr. Chu discovered that the two Arabic numerals were not. sufficient to take care of the whole of the [Chinese vocabulary. Taking the vocabulary to be 10,000 and placing every word into its proper' niche a determined'by its first and last strokes, there would be an Javerage of about 100 words to each night. This was a good deal too and he then sought to solve the difficulty, so he employed two more for the orderly enumeration] 0001 to 9999 there would be of characters; thus from 0000,
altogether 10,000 niches or an inlerease of 9,000 niches over the total used before, and there would-be an] average of one word to each niche. This addition of two more digits was .ccomaplished by splitting a word inti its two natural parts, usingsthe frst present the frst and last strokes of the first part and the second two digits to represent the first and last strokes of the second part. This method, he corisidered, was the correct path to pursue 20° per cent. of tie Chinese characters were formed by two com-to. two] parts. He maintained this idea until the complete cataloguing of the whole active vocabulary was accomplished.
Mr. Chu said his plan provided for the filflment of the two most important requirements of a word cataloguing system, easy to find and easy to learn. His main idea was to provide a simple, fool-proof method for the looking up fof any word without, at the same time, requiring mach time for its study. This was achieved. after a careful study of the construction of Chinese strokes and their manners of combination, After a series of Jexperiments, he classified the strokes into ten convenient groups, 1, 2, 3, 4, 5, 6, 7, 8 9, 0. He then described this' system' at length and showed the different strokes. By this method, the task of radical. determination was entirely discarded and, ju its stead, the easier way of into two" was sub-
The task of stroke counting also was completely discarded, inasmuch the order of precedence was set by the decimal. progression of Arabic numerals representing the strokes. It was quite evident that this new method was much superior to the old one which called for the Jcounting process, as. it gave no chance for errors and saved much time. By far the most scientific part lof it was the great increase of the number of niches which worked for a more citable, distribution of words, derstand that taking the number of words of a language to be a constant, the number of words: contained in each group was ly proportional to the number of riches the system provided, To express it in another way, said the speaker, the present system propos-8 much renter umber tore place the 214 radicals of the Hal system by ao. many of the number keys (2,543 in all), each re-Rroup of words believe" concluded the speaker, "that with the spread of this system, there will be a side by side develop: ment in the realm of Chinese index science. Cyclopadias, library cards, books, telegraphic codes, business directories, magazine digests, book-back indexes, ete,, could all be compiled upon this new basis, and a general increase of efficiency' in most kinds of index service would: be the natural result, Its use in lexicon graphy represented but one phase of ity usefulness."
Text read from the printed page and cleaned; old type leaves the occasional error.