In lieu of an abstract, here is a brief excerpt of the content:

  • Gongsun Long on What Is Not:Steps Toward the Deciphering of the Zhiwulun
  • Jean-Paul Reding

Introduction

The Zhiwulun, chapter 3 of the Gongsunlongzi, is one of the strangest texts in the whole ancient Chinese logical tradition.1 It has attracted many scholars, who have been fascinated by the treacherous simplicity of its syntax and vocabulary. The treatise is indeed only 269 characters in length, many of which are very common: tianxia "world," wu "things," zhi "point," you "there is," wu "there is not," fei "not," wei "call." Traditionally, this text has been interpreted as providing the fundamental principles of language and logic that underlie the no less intriguing Baimalun (White horse treatise) of the famous sophist. No scholarly agreement, however, could be reached. "There are more rival interpretations of the Zhiwulun than of any other document in early Chinese philosophical literature" (Graham 1986, p. 210).

It is the term zhi "point" that has mostly retained the attention of the interpreters. "One is convinced that if only one could identify what the writer meant by the word zhi, everything he says would fall into place," wrote A. C. Graham (Graham 1955, p. 282). Zhi has been rendered, by rather sophisticated Western logical and metaphysical concepts, into such translations as "definition" (Forke 1901-1902); "attributes" (Hu 1922, pp. 126-127; Mei 1953); "universals" (Fung 1952, pp. 209-212; Hughes 1942, pp. 125-126; Needham 1956, pp. 185-186); "meanings" (Graham 1955); "marks" (Chan 1963, p. 237); "object of reference" (Cheng and Swain 1970); "signifiant-signifié" (Kou 1953); "signs" (Thompson 1995). None of these approaches has lead to a coherent reading of the treatise (Graham 1978, pp. 463-464). The paradoxical—and even, as we shall see, dilemmatic—structure of the text is mostly ignored (see below), and as a result the treatise is construed as a theoretical exposition of semantic problems. The new reading I want to propose proceeds from the hypothesis that the Zhiwulun, like the White Horse Treatise, is another logical puzzle, much in the same style as the other transmitted sayings of Gongsun Long, like "coals are not hot," "ice is not cold," "white horse is not horse," et cetera. My own approach builds upon and completes three earlier attempts made by A. C. Graham to decipher the treatise (Graham 1955, 1978, and 1986). Rather than stating the numerous points where I diverge from him, I shall try to outline the framework that I believe is common to us:

  1. a. The structure of the treatise is dilemmatic: "Kung-sun Lung [Gongsun Long] presents us with two contradictory propositions one of which entails the other …" (Graham 1955, p. 294; cf. Graham 1978, pp. 457-458; Graham 1986, p. 214);2 [End Page 190]

  2. b. The recurring notion of tianxia deserves special attention (Graham 1978, p. 461);

  3. c. Fei zhi has a technical meaning;

  4. d. The text of the treatise is sound and intact.

I shall start by discussing the problems posed by the expressions tianxia and fei zhi.

The Expression Tianxia

Graham had already noticed that the expression tianxia "world" occurs fourteen times in this short treatise: "This suggests that "world" may itself be an overlooked concept in Kung-sun Lung's argument, the cosmos as a whole in contrast with the things which compose it" (Graham 1978, p. 461). The recurring clause tianxia wu zhi can then be interpreted, according to Graham, as "nothing within the world is the meaning (scil. of 'world')." Graham's idea is the following: names point out things that are to be found in the world; the name "world" itself, however, although it is a name like "ox" and "horse," does not point out one more thing besides all the other things that make up the world. On the other hand, names like "ox" or "horse" do also point to items that make up the world, and in one sense we can say that these things are the world, or that we can name the world also through them. Pointing to an ox, we also point to the world. Ingenious as it is, Graham's solution does not seem to be wholly satisfactory. If we take a closer look at the ways in...

pdf