As we have already seen, there is no reasonable ground to suppose that Ts’ao Kung tampered with the text. But the text itself is often so obscure, and the number of editions which appeared from that time onward so great, especially during the T’ang and Sung dynasties, that it would be surprising if numerous corruptions had not managed to creep in. Towards the middle of the Sung period, by which time all the chief commentaries on Sun Tzŭ were in existence, a certain Chi T’ien-pao published a work in 15 _chuan_ entitled "Sun Tzŭ with the collected commentaries of ten writers." There was another text, with variant readings put forward by Chu Fu of Ta-hsing, which also had supporters among the scholars of that period; but in the Ming editions, Sun Hsing-yen tells us, these readings were for some reason or other no longer put into circulation. Thus, until the end of the 18th century, the text in sole possession of the field was one derived from Chi T’ien-pao’s edition, although no actual copy of that important work was known to have survived. That, therefore, is the text of Sun Tzŭ which appears in the War section of the great Imperial encyclopedia printed in 1726, the _Ku Chin T’u Shu Chi Ch’eng_. Another copy at my disposal of what is practically the same text, with slight variations, is that contained in the "Eleven philosophers of the Chou and Ch’in dynasties" [1758]. And the Chinese printed in Capt. Calthrop’s first edition is evidently a similar version which has filtered through Japanese channels. So things remained until Sun Hsing-yen [1752-1818], a distinguished antiquarian and classical scholar, who claimed to be an actual descendant of Sun Wu, [36] accidentally discovered a copy of Chi T’ien-pao’s long-lost work, when on a visit to the library of the Hua-yin temple. [37] Appended to it was the _I Shuo_ of Cheng Yu-Hsien, mentioned in the _T’ung Chih_, and also believed to have perished. This is what Sun Hsing-yen designates as the "original edition (or text)"—a rather misleading name, for it cannot by any means claim to set before us the text of Sun Tzŭ in its pristine purity. Chi T’ien-pao was a careless compiler, and appears to have been content to reproduce the somewhat debased version current in his day, without troubling to collate it with the earliest editions then available. Fortunately, two versions of Sun Tzŭ, even older than the newly discovered work, were still extant, one buried in the _T’ung Tien_, Tu Yu’s great treatise on the Constitution, the other similarly enshrined in the _T’ai P’ing Yu Lan_ encyclopedia. In both the complete text is to be found, though split up into fragments, intermixed with other matter, and scattered piecemeal over a number of different sections. Considering that the _Yu Lan_ takes us back to the year 983, and the _T’ung Tien_ about 200 years further still, to the middle of the T’ang dynasty, the value of these early transcripts of Sun Tzŭ can hardly be overestimated. Yet the