An Analytic Philosopher Learns from Zhuangzi

1 An Analytic Philosopher Learns from Zhuangzi My aim is twofold: to reflect on the famous butterfly-dream passage in Zhuangzi, and to display the way an analytic philosopher typically thinks about metaphysical issues. Here is the passage in question, as translated by Burton Watson: Once Zhuangzi dreamt he was a butterfly, a butterfly flitting and fluttering around, happy with himself and doing as he pleased. He didn't know he was Zhuangzi. Suddenly he woke up and there he was, solid and unmistakable Zhuangzi. But he didn't know if he was Zhuangzi who had dreamt he was a butterfly, or a butterfly dreaming he was Zhuangzi. Between Zhuangzi and a butterfly there must be some distinction! This is called the Transformation of Things.1 Zhuangzi does not know whether he is Zhuangzi who dreamed he was a butterfly or a butterfly that is dreaming it is Zhuangzi. That is, Zhuangzi cannot tell which of the two scenarios is true: the one saying that he is Zhuangzi who had a dream and in the dream he was a butterfly, or the 1 The Complete Works of Zhuangzi, Burton Watson (tr.), New York and London: Columbia University Press, 1968: 49. 2 other saying that he is a butterfly that is now dreaming and in the dream it is Zhuangzi. Zhuangzi cannot tell this because the two scenarios are indistinguishable to him. This indistinguishability leads him to a kind of skepticism. Exactly what is Zhuangzi skeptical about? The wording in the text (as translated) gives the impression that he is skeptical about whether he is Zhuangzi. The text suggests that he is unsure whether he is Zhuangzi who dreamed he was a butterfly or he is a butterfly that is dreaming it is Zhuagnzi. But this is highly misleading, for it strongly suggests that being Zhuangzi and being a butterfly are mutually exclusive; if he is Zhuangzi he is not a butterfly, and if he is a butterfly he is not Zhuangzi. This cannot be right. In fact it is self-contradictory. Take the second conditional sentence, “If he is a butterfly, he is not Zhuangzi.” Who is he? Zhuangzi, of course. So, the sentence says that if Zhuangzi is a butterfly, Zhuangzi is not Zhuangzi. The consequent clause, “Zhuangzi is not Zhuangzi,” is a logical falsity. Only a logical falsity entails a logical falsity. This means that if the conditional sentence is true (when read as expressing entailment), then the antecedent clause, “he is a butterfly,” is a logical falsity; hence Zhuangzi cannot possibly be a butterfly. But then the second scenario is ruled out, which means that Zhuangzi’s skepticism is refuted. Thus, we need to depart from the literal wording of the text (as translated) in order to make sense of the text. There are two points of departure. First, the dream contents in the two scenarios should be conceived not in the third-person but in the first-person. The dream content in the first scenario should be understood as “I was a butterfly” rather than as “Zhuangzi was a butterly,” and it is better to understand the dream content in the second scenario as “I am Zhuangzi” rather than as “it (the butterfly) is Zhuangzi.” 3 The second point of departure from the literal wording of the text concerns the elimination of the notion of being Zhuangzi. We should put ourselves in Zhuangzi’s shoes and speak of the contrast between being a human and being a butterfly instead of the contrast between being Zhuangzi and being a butterfly. This involves a further modification of the dream content in the second scenario, and the result is the following reformulations of the two scenarios: First Scenario: I am a human being who is awake and who dreamed that I was a butterfly. Second Scenario: I am a butterfly that is dreaming that I am a human being. Now that the two scenarios have been clarified, let us compare them. There are two obvious differences. The first difference is inessential and rather distracting, while the second is important. The first obvious difference is a difference in tense. Did I dream a few moments ago that I was a butterfly, or am I now dreaming that I am a human being? According to the first scenario the dreaming occurred in the past, whereas according to the second the dreaming is occurring now. The past dreaming in the first scenario may play a psychologically important role to push us into conceiving the second scenario as a vivid alternative, but it is philosophically dispensable. So let us remove it from the first scenario. This brings up the crucial contrast between the two scenarios starkly: Am I a human being now who is awake, or am I a butterfly now that is dreaming? 4 Why is the alternative in the second disjunct (that I am a butterfly now that is dreaming) a live possibility for me? Does my current perceptual experience not obviously refute the claim that I am a butterfly now? No, it does not. The reason comes from the evident fact that the second scenario says not only that I am a butterfly now but also something more. My current perceptual experience refutes the claim that I am a butterfly now that is awake, but the second scenario does not say that I am a butterfly now that is awake. It says that I am a butterfly now that is dreaming. Moreover, it says that in the dream I am a human being (with the normal human perceptual capabilities). So, my current perceptual experience does not refute that scenario. The second obvious difference between the two scenarios is that they treat my current perceptual experience differently. The first scenario, in which I am a human being who is fully awake, allows my current perceptual experience to be veridical (even though it does not force it to be so), whereas the second scenario, in which I am a butterfly dreaming myself to be a human being, forces my current perceptual experience to be illusory. So, part of my uncertainty concerns the veridicality of my current perceptual experience: (1) If my current perceptual experience is veridical, then I am a human being. (2) If my current perceptual experience is not veridical, then I might be a human being and I might be a butterfly. But, (3) I am uncertain whether my current perceptual experience is veridical. Therefore, (4) I am uncertain whether I am a human being or a butterfly. 5 The nature of my uncertainty has little to do with being a butterfly. Essentially the same uncertainty may be expressed in any number of different ways, as follows: “I am uncertain whether I am a human being or a beetle” (i.e., “I am uncertain whether I am a human being who is awake or a beetle that is dreaming itself to be a human being”); “I am uncertain whether I am a human being or a dog”; “I am uncertain whether I am a human being or a dolphin”; etc. The general form is clear: (A) I am uncertain whether I am a human being or a [ ]. In Zhuangzi’s original story and in the three examples I have just given, the blank is filled with a term for an animal. An interesting question suggests itself immediately: (Q1) Is the same uncertainty preserved with any term for a non-animal in the blank? Consider a rock. Am I uncertain whether I am a human being or a rock? Am I uncertain whether I am a rock dreaming myself to be a human being? Not really. I am fairly certain that I am not a rock dreaming myself to be a human being. Why? Is it because I am fairly certain that dreaming requires brain activities and that a rock lacks a brain? Perhaps, but why do I think dreaming requires brain activities? Perhaps it is because physiologists and psychologists tell us so. That is not a bad reason, but do we need to rely on sophisticated empirical sciences? No, we do not. The main philosophical reason for the fair amount of certainty for the claim that I am not a rock dreaming myself to be a human being is that I am fairly certain that there is no such thing as what 6 it is like to be a rock, viz., what it is like to have the first-person perspective of a rock. Because of this, I am fairly certain that there is no such thing as what it is like to be a rock dreaming itself to be anything. So I am fairly certain that it is impossible for me to be a rock and dream myself to be anything. If there were such a thing as the first-person perspective of a rock, what it was like to have such a perspective would be almost completely mysterious to me. Underlying this is my conviction that a rock lacks any perceptual capability whatever. I am uncertain about what it is like to have the first-person perspective of an individual that does not perceive anything external to it. So in the case of a rock, my answer to (Q1) is “No.” The degree of uncertainly is significantly lower in the case of a rock than in the case of a butterfly. We should be careful to note that what is at issue is a matter of comparative uncertainty, not absolute certainty. The point is that it is less uncertain that I am not a rock dreaming myself to be a human being than that I am not a butterfly dreaming myself to be a human being, and not that I am maximally certain that I am not a rock dreaming myself to be a human being. The reason for the lack of the maximal certainty is that external perception is not the only possible mode of perception. Internal perception, viz., perception of one’s own purely mental states, is another possible mode of perception. If I know that an individual lacks both external and internal perception, then I am maximally certain that there is no such thing as what it is like to have the first-person perspective of that individual. A rock might lack external perception but might possibly possess internal perception. If so, there might be such a thing as what it was like to be a rock dreaming itself to be thus-and-so. I am not maximally certain that a rock lacks internal perception. To the extent to which I am uncertain that a rock lacks internal perception, I am uncertain that I am not a rock dreaming myself to be a human being. 7 How about a rose? Unlike a rock, a rose is a living thing. But this hardly helps me imagine what it is like to assume the first-person perspective of a rose any better than what it is like to assume the first-person perspective of a rock. I am fairly certain that a rose cannot dream, albeit not quite as certain as I am that a rock cannot dream. I think the cases of a rock and a rose generalize safely. Therefore, my general answer to (Q1) is “No.” Next take a bacterium. A bacterium is very close to a rose, and I am fairly certain that a bacterium cannot dream, albeit not quite as certain as I am that a rose cannot dream. This brings us to another interesting question concerning (A): (Q2) Is the same uncertainty preserved with every term for an animal in the blank? Take a sponge (poriferan). I am fairly certain that a sponge cannot dream (albeit not quite as certain as I am that a bacterium cannot dream), so I am fairly certain that I am not a sponge dreaming myself to be a human being (albeit not quite as certain as I am that I am not a bacterium dreaming myself to be a human being). As before, the main reason is that I am fairly certain that there is no such thing as the first-person perspective of a sponge. Therefore, my answer to (Q2) is also “No.” Thus, I am fairly certain that I am not a rock, a rose, a bacterium, or a sponge. Or at least, Zhuangzi’s dream hypothesis does not make me uncertain about the possibility of my not being any of these things. But is there anything about which I am maximally certain? Consider the schema: (B) I am maximally certain that I am [ ]. 8 The lesson we have learned from Zhuangzi is that (B) is false if the blank is filled with “a human being.” It is easy to see that many other fillers will do as well: “human-shaped,” “sitting in a chair,” “talking to people,” “awake,” etc. (B) is also false if the blank is filled with “not a human being,” “not human-shaped,” “not sitting in a chair,” “not talking to people,” “not awake,” etc. But (B) is true if the blank is filled with “not number 3” and “not a triangle.” I am maximally certain that there is no first-person perspective of number 3 or a triangle, or any other abstract object. In addition, I am maximally certain that number 3, a triangle, or any other abstract object could not possibly be a human being. For these reasons, I am maximally certain that I am not number 3, a triangle, or any other abstract object that is dreaming itself to be a human being. But some might complain that such a scenario in which I am an object of pure mathematics or some other abstract object is hardly intelligible to begin with. Are there other fillers of the blank in (B) than those involving terms of pure mathematics or terms for other abstract objects which make (B) true? Yes, there are. (B) is true if the blank is filled with “not you,” “real,” or “I.” Let us start with “not you.” Going back to the case of a butterfly, I said the following: I am uncertain that I am not a butterfly. But what does this really mean? Does it mean that there is some particular butterfly such that I am uncertain that I am not that butterfly? No. It means rather that I am uncertain that I am not some butterfly or other, i.e., that I do not have the property of being a butterfly. My uncertainty concerns my possession of a certain general property, and not my being identical with a particular 9 individual. Suppose that my current perceptual experience is such that I see a particular butterfly --- call it “Buddy” --- two meters in front of me. Is it coherent to ask whether I am Buddy and am dreaming myself to be looking at Buddy two meters in front of me? The answer is “No.” The reason has much to do with a special status of identity. To be identical with a particular individual is not at all like possessing a general property. Being identical with Buddy is not at all like possessing some general property, like the property of being a butterfly. This is demonstrated by the following argument: (5) If I am Buddy dreaming myself (Buddy) to be standing here and looking at Buddy, then in that dream I am Buddy and Buddy is two meters in front of me. (6) If Buddy is two meters in front of me, then I am not Buddy; for no individual can be two meters in front of itself/herself/himself. Therefore, (7) If I am Buddy dreaming myself (Buddy) to be standing here and looking at Buddy, then in that dream I am Buddy and I am not Buddy. Obviously, any dream in which I am Buddy and I am not Buddy is a dream with an incoherent content. This clearly undermines the interpretation of Zhuangzi’s butterfly-dream hypothesis in terms of identity with a particular butterfly. This reasoning applies not only with respect to an arbitrary butterfly but also with respect to any arbitrary individual other than me, including you. Suppose we replace Buddy with you and I reason in the parallel manner as follows: 10 (8) If I am you dreaming myself (you) to be standing here and looking at you, then in that dream I am you and you are two meters in front of me. (9) If you are two meters in front of me, then I am not you; for no individual can be two meters in front of itself/herself/himself. Therefore, (10) If I am you dreaming myself (you) to be standing here and looking at you, then in that dream I am you and I am not you. The content of the dream is again incoherent, just as before. In this case, however, there is one small difference from the case of Buddy. Here, unlike in the case of Buddy, there is no general property for me to fall back on. The general property corresponding to Buddy is being a butterfly. The general property corresponding to you is being a human being, but it does not take a skeptic to suggest that I might possibly be a human being. My current perceptual experience already overwhelmingly confirms that I am a human being. Another way to argue that Zhuangzi’s dream hypothesis does not establish the possibility of the identity of you and me is by arguing directly for the incoherence of that identity: (11) You and I are two, i.e, two individuals. (12) It is mathematically impossible for two to be one. Therefore, (13) It is mathematically impossible that you and I are one, i.e., identical to each other. 11 Here, we should take care not to confuse “X and Y are one individual” with “X and Y become one individual.” Two individuals might well become one individual, e.g., by fusing with each other. There is nothing mathematically impossible about that. Another famous passage from Zhuangzi, the one on the Happiness of Fish, might be expanded to shed light on the issue at hand: Zhuangzi and Huizi were strolling along the dam of the Hao Waterfall when Zhuangzi said, “See how the minnows come out and dart around where they please! That’s what fish really enjoy!” Huizi said, “You’re not a fish --- how do you know what fish enjoy?” Zhuangzi said, “You’re not I, so how do you know I don’t know what fish enjoy?” Huizi said, “I’m not you, so I certainly don’t know what you know. On the other hand, you’re certainly not a fish --- so that still proves you don’t know what fish enjoy!” Zhuangzi said, “Let’s go back to your original question, please. You asked me how I know what fish enjoy --- so you already knew I knew it when you asked the question. I know it by standing here beside the Hao.”2 This passage seems to contain a number of logical loose ends, but let us focus on the last thing Huizi said to Zhuangzi: “... you’re certainly not a fish --- so that still proves you don’t know what fish enjoy!” If Huizi had heard Zhuangzi’s musing about his being a dreaming butterfly, Huizi might have said to Zhuangzi, “You are certainly not a butterfly --- so that proves you don’t know what it is like to be a butterfly, hence you don’t know what it is like to be a butterfly dreaming 2 Ibid.:188-9. 12 itself to be a human being. Therefore, you can’t even understand the thought that you are a butterfly dreaming yourself to be a human being.” To this, Zhuangzi might have replied, “You are not certain that I am not a butterfly. It might be that you and I are both butterflies and that you are dreaming we are human beings having this conversation.” Huizi might have retorted, “That would make no sense. I am not you, so I don’t know what it is for me to dream you are a butterfly. I can’t tell the difference between my dreaming myself to see a butterfly that is you and my dreaming myself to see a butterfly that is not you. I can’t tell the difference between my dreaming you are a butterfly dreaming yourself to be a human being and my dreaming someone else is a butterfly dreaming himself to be a human being who looks like you. So I don’t know what it is for me to dream you are a butterfly dreaming yourself to be a human being.” The point Huizi would have been making is that dreaming oneself to be a butterfly is intelligible owing to the first-person perspective one in fact has but that dreaming of some particular person other than oneself that he is a butterfly is not, because the specific first-person perspective of someone who is not in fact oneself is impossible to adopt. Any putative such perspective would be indistinguishable from another specific first-person perspective of someone else indistinguishable from the person in question. The other two fillers of the blank in (B), “real” and “I,” yield the clearest cases. Unlike all the preceding cases, Zhuangzi’s dream hypothesis does not even appear to get off the ground at all: (C) I am unreal and am dreaming I am real. (D) I am not I and am dreaming I am I. 13 (C) and (D) are utterly implausible, for they are clearly self-refuting. If I am unreal, then my dreaming should also be unreal, which destroys the dream hypothesis itself. If I am not I, then the law of self identity is violated, which destroys the logical coherence of the dream hypothesis itself. We have learned that Zhuangzi’s story about the butterfly dream vividly points to a wide variety of possibilities for the self which the first-person perspective allows. I might be a human being, a butterfly, a dolphin, a dog, or a beetle. On the other hand, we have seen that for any individual other than myself, it is impossible --- indeed incoherent --- for me to be that individual. We have also noted that the first-person perspective does not permit me to be unreal or non-identical with me. We may take the skeptical lesson learned from Zhuangzi one step further and question the skeptical lesson itself. It will be meta-skepticism, or higher-order skepticism about the original skepticism. Is it truly possible for me to be a butterfly? Does Zhuangzi’s butterfly-dream hypothesis really show such a possibility? Can I really imagine or dream myself to be a butterfly? Do I really know what it is like to fly from one flower to another while having the brain of a butterfly and the perception of a butterfly, as opposed to what it is like to be a human being acting like a butterfly? Is the first-person perspective of a butterfly truly comprehensible to me? Does a butterfly really have the first-person perspective at all? Let us distinguish four theses: (E) If I can dream that P, then it is possible that P. (F) If I cannot dream that P, then it is impossible that P. (G) Even if I can dream that P, it may be impossible that P. 14 (H) Even if I cannot dream that P, it may be possible that P. According to (E) - (F), my ability to dream and the possibility of the content of my dream dovetail nicely. According to (G) - (H), there is a rift between my ability to dream and the possibility of the content of my dream. If (G) is true and I can dream something impossible, then the feasibility of a dream hypothesis is not a sufficient condition for the possibility of the dreamed content. If (H) is true and some possibility cannot be dreamed of, then the feasibility of a dream hypothesis is not a necessary condition for the possibility of the dreamed content. Either way, reasoning based on a dream hypothesis generally will be of little use to us when we consider matters of possibility and impossibility. In order to take Zhuangzi’s butterfly-dream passage with metaphysical seriousness, we need to assume not only that either (E) or (F) is true but more specifically that (E) is true. How are we to justify the assumption of the truth of (E)? Unfortunately, a proper treatment of this exciting topic must wait until another occasion. Let me conclude with a brief thought which might be somewhat discouraging. The last sentence of Zhuangzi’s butterfly-dream passage is as follows: This is called the Transformation of Things. I have said nothing about this sentence, for unfortunately I have nothing at all to say. I see no way to incorporate this sentence into our discussion of the rest of the passage. For all I can tell, the correct interpretation of this sentence --- if there is one --- might be contrary to the main thrust of our discussion, and if it is, we should be prepared to overhaul our thinking about the entire 15 passage. But for now, our discussion ends with the simple admission that the last sentence of Zhuangzi’s butterfly-dream passage remains mysterious.3