Introduction

Gongsun Longzi (Kung-sun Lung-tzu), who lived around 300 B.C.E., was the author of only two or three brief philosophical works that have survived. The most famous is this dialogue, which argues that "a white horse is not a horse." The so-called "White Horse Dialogue" is fascinating, because in it we see subtle rational argumentation used to try to establish a conclusion contrary to common sense. This combination is reminiscent of much of Western philosophy, going back to the ancient Greeks Parmenides and Plato. However, it is uncharacteristic of much of Chinese philosophy, which tends to either be rational but common-sensical (like the Mohists or Confucians) or anti-rational and mystical (like Zhuangzi). The White Horse Dialogue is also notoriously difficult to interpret, though. No doubt the subtlety of its argumentation made it hard even for Gongsun Longzi's contemporaries to understand. The situation is even worse for us today. We no longer understand much of the intellectual context in which the dialogue was written. Furthermore, the original Chinese text seems to have suffered corruption at the hands of generations of scribes, who did not understand the text that they were re-copying. (I have identified in the notes the more controversial amendations I have made to the text.) Finally, some obscurities may be inherent in Gongsun Longzi's argument itself. For instance, I am not sure that Gongsun Longzi is always clear himself about when he is talking about the words "white" and "horse," and when he is talking about the color white and the animal horse.

Nonetheless, the White Horse Dialogue is extremely important in the history of Chinese philosophy. Its argument is referred to by a number of early Chinese philosophers (see, e.g., Readings, p. 213). Gongsun Longzi was a central figure in what has been called a "language crisis" in Chinese philosophy. Because of arguments like those given in his dialogue, Chinese philosophers began to call into question the adequacy of ordinary language for describing reality. In response, the later Mohists and the Confucian Xunzi tried to protect language from what they saw as the sophistries of thinkers like Gongsun Longzi. (See Xunzi, "On Correct Naming," in Readings, pp. 278-84.) Others philosophers embraced both the rational argumentation and the paradoxical conclusions of the new style of argumentation. For example, Huizi defended paradoxes such as "Heaven is as low as earth, the mountains are level with the marshes," and "I go to Yue today yet arrived yesterday." However, Huizi apparently saw the same line or reasoning that led to these paradoxes as having important ethical consequences, for he also said, "Let concern spread to all the myriad things; heaven and earth count as one unit." (Translations from Huizi by A.C. Graham, Disputers, p. 78. For more on Huizi, see Readings, pp. 229-30 and 240.) Finally, texts like the Daodejing and the Zhuangzi embrace paradox, but evince much less confidence in the power of rational argumentation than did Huizi.

Have fun trying to figure out for yourself what the argument is for the thesis that a white horse is not a horse. Presumably, you disagree with the conclusion. But can you figure out how the argument is mistaken? And what should the "objector" have said in response to the argument? Here is a hint to one possible interpretation. The expression "X is not Y" (like the Chinese X fei Y ye) is ambiguous. It could mean "X is not a kind of Y" or it could mean that "X is not identical with Y." (We see both uses in Readings, p. 240, where we find both "You are not a fish" and "You are not me.") Which sense of "is not" does the "advocate" in the White Horse dialogue use? Which sense does the "objector" have in mind? The Selective Bibliography below will allow you to follow up on some of the other major lines of interpretation of the White Horse Dialogue.

A.

Can it be that a white horse is not a horse?¹

Advocate: It can.²

Objector: How?

Advocate: "Horse" is that by means of which one names the shape. "White" is that by means of which one names the color. What names the color is not what names the shape. Hence, I say that a white horse is not a horse.

B.

Objector: If there are white horses, one cannot say that there are no horses. If one cannot say that there are no horses, doesn't that mean that there are horses? For there to be white horses is for there to be horses. How could it be that the white ones are not horses?

Advocate: If one wants a horse, that extends to a yellow or black horse. But if one wants a white horse, that does not extend to a yellow or black horse. Suppose that a white horse were a horse. Then what one wants [in the two cases] would be the same. If what one wants were the same, then a white [horse] would not differ from a horse. If what one wants does not differ, then how is it that a yellow or black horse is sometimes acceptable and sometimes unacceptable? It is clear that acceptable and unacceptable are mutually contrary. Hence, yellow and black horses are the same [in that, if there are yellow or black horses], one can respond that there are horses, but one cannot respond that there are white horses. Thus, it is evident that a white horse is not a horse.

C.

Objector: You think that horses that are colored are not horses. In the world, it is not the case that there are horses with no color. Can it be that there are no horses in the world?

Advocate: Horses certainly have color. Hence, there are white horses. If it were the case that horses had no color, there would simply be horses, and then how could one select a white horse?³ A white horse is a horse and white. A horse and a white horse [are different]. Hence, I say that a white horse is not a horse.

D.

Objector: "Horse" not yet combined with "white" is horse. "White" not yet combined with "horse" is white. If one combines "horse" and "white," one uses the compound phrase "white horse." This is to take what is not combined and combine them as a phrase.⁴ Hence, I say that it cannot be that a white horse is not a horse.⁵

Advocate: You think that there being white horses is there being horses. Is it acceptable to say that there being white horses is there being yellow horses?

Objector: It is not acceptable.

Advocate: If you think that there being horses is different from there being yellow horses, this is for yellow horses to be different from horses. If you differentiate yellow horses from horses, this is to think that yellow horses are not horses. To think that yellow horses are not horses, yet to think that white horses are horses -- this is to turn things upside down and inside out!⁶ This is the most incoherent doctrine and confused discourse in the world!

E.

Objector: If there are white horses, one cannot say that there are no horses, because of what is called "the separability of white."⁷ Only according to those people who do not separate can having a white horse not be said to be having a horse.⁸ Hence, the reason we think there are horses is only that we think that "horse" is "there are horses." It is not that we think "there are white horses" is "there are horses." Hence, because of the reason that there are horses, one cannot say that a [white] horse [is not] a horse.⁹

Advocate: "White" does not fix that which is white. It ignores that. The expression "white horse" fixes that which is white. That which fixes what is white is not white. "Horse" is indifferent to color. Hence, [if you were only looking for a horse,] a yellow or black horse would each be appropriate. "White horse" does select for color. So [if you were looking for a white horse,] a yellow or black horse would be rejected on account of its color. Hence, only a white horse alone would be appropriate. That which does not reject is not what does reject. Hence, I say that a white horse is not a horse.

1. As the translation suggests, it is possible that the issue is not whether it is always true that "a white horse is not a horse," but whether it is possible for it to be true.

2. The two speakers in the dialogue are not named. I have labeled them "advocate," who defends the thesis that a white horse is not a horse, and "objector," who argues against this thesis. The section headings ("A," "B," etc.) are also not in the original.

3. Following this sentence in the original Chinese, there is a sentence that reads, "Hence, white is not horse." This does not seem to make any sense in context, so I have left it out.

4. I believe that this sentence is a defense of saying that "a white horse is a horse." (See the next note for my interpretation.) However, following this sentence in the original Chinese, there is a sentence that reads, "That is not acceptable." I do not think this makes any sense in context, so I have left that sentence out. However, many translators leave that sentence in, which would mean that the sentence immediately prior to this note is an objection to saying that "a white horse is not a horse."

5. The argument may be that, since "horse" refers to horse when it is used as a simple expression, it must continue to refer to horse when it becomes part of a compound expression. Since we can obviously say that "a horse is a horse," we can also say that "a white horse is a horse."

6. Literally, "…this is for flying things to enter the water, and for the inner and outer coffins to be in different places!"

7. "Separability" seems to have been a technical term in ancient Chinese philosophy of language. It apparently referred to the possibility of discussing separately two terms that were used in a compound expression. For example, one Chinese commentator observes that "There must be a shape corresponding to a name, and the best way to examine the shape is to distinguish the colour from it…" (translation from A.C. Graham, Later Mohist Logic [London: University of London, 1978] , p. 175.)

8. Translation of this line follows the Dao zang version of the text.

9. The last part of this sentence literally reads, "…one cannot say that horse horse." I have amended the text to make sense out of it.

Graham, Angus. Disputers of the Tao. Chicago: Open Court Press, 1989. Excellent popular discussion of Gongsun Longzi and his general intellectual context on pp. 75-95. Note that Graham radically rearranges the received text of the White Horse Dialogue, whereas the translation above follows the received text as closely as possible.

-----. Studies in Chinese Philosophy and Philosophical Literature. Albany: State University of New York Press, 1990. Includes three important, scholarly articles on the writings attributed to Gongsun Longzi, including the White Horse Dialogue.

Hansen, Chad. Language and Logic in Ancient China. Ann Arbor: University of Michigan Press, 1983. Controversial but important discussion of many aspects of ancient Chinese philosophy of language, including the White Horse Dialogue.

Harbsmeier, Christoph. Language and Logic. Volume 7, Part I of Joseph Needham, Science and Civilisation in China. New York: Cambridge University Press, 1998. A magisterial overview.

© 2001 Bryan W. Van Norden. All rights reserved.