Happy fish and philosophical skepticism

by Dave Maier

Tao Most Westerners think of Taoism, if at all, as a form of Eastern mysticism, popular with hippies and new-agers. So interpreted, Taoism is a form of skepticism: our beliefs about the world are falsified by the ineffable wholeness beyond our conceptual grasp, as represented by the famous yin-yang symbol. This interpretation is not completely wrong, but anyone looking past that ubiquitous icon into the texts themselves will find that most of what lies there is hard to fit into that simple mold.

Zhuangzi in particular is a puzzle. The text which bears his name, which he may or may not have had a lot to do with, is a compendium of practical advice, obscure parables, evocative imagery, rigorous philosophical argument, and flat-out weirdness. In this post I'd like to look closely – risking, as usual, spoiling the joke with heavy-handed overanalysis – at the relatively famous story of the happy fish.

Zhuangzi (Z) and Huishi (H), a frequent interlocutor, are walking above the Hao river.

Z: Look how the fish are swimming: those are some happy fish!

H: You are not a fish. How [or whence] do you know fish’s happiness?

Z: You are not me. How do you know that I don’t know?

H: I’m not you, so I don’t know about you. You’re not a fish, so you don’t know about fish: Q.E.D.

Z: Let’s go back to where we started. When you said “whence do you know fish’s happiness?”, you already knew I know it before asking the question. I know it from up above the Hao river.

Ha! (Wait, what?) A lot of the book is like that: it sounds like there was a good zinger there, but who or what got zung?

Now that we are safely below the fold, let me back up a bit to set the philosophical stage. Historically, philosophers have divided naturally into two opposed camps. Dogmatists defend the “default” position by trying to show that our beliefs can, and sometimes do, transcend appearances and thus make contact with the real. Skeptics use thought experiments or rhetorical “tropes” to bring out the flaws in particular dogmatist views or claims, or, more radically, display the hopelessness of the project of knowledge in general.

In Western philosophy, skepticism is usually considered in, if not simply identified with, its Cartesian form: an attack on the very idea of empirical knowledge. This attack exploits the metaphysical gap between subjective appearances (sensations or mental representations) and the objective reality (the “external world”) which causes them. If Western philosophers agree on anything, however, it is that the Cartesian skeptical threat is empty bluster (disagreeing, naturally, on why that is). Of course there's a world out there!

In Eastern philosophy, however, skepticism is seen less as a threat to our common sense than as a live philosophical option. As I mentioned above, we can see at least a thread of mystical skepticism in Taoism, especially in Laozi; but the playful, ironic skepticism of Zhuangzi is trickier to understand, as well as more philosophically sophisticated. On my reading, the story of the happy fish is a vivid demonstration of the difficulty of properly applying the skeptical rejection of belief.

(Z1) Look how the fish are swimming: those are some happy fish!

This utterance looks like it contains a factual claim about the world: that the fish are happy. Different responses are possible depending on how one takes it. How should a skeptic respond? For instance, what would Zhuangzi say if Huishi were the one to say it instead? Let’s see what happens.

(H1) You are not a fish. How [or whence] do you know fish’s happiness?

-happy-fish_001 Not surprisingly, Huishi takes Zhuangzi’s assertion as a claim to knowledge and disputes it in typical, even predictable, skeptical fashion, pointing to the epistemic gap between Zhuangzi’s subjectivity and the internal states, if there even are any, of fish. What could bridge this gap and justify the assertion? Without further argument, there is no reason to take Zhuangzi as having bridged it and indeed obtained knowledge of how the fish are in reality. As his longtime philosophical sparring partner, Huishi has undoubtedly heard Zhuangzi make similarly skeptical points against him; this seemingly questionable claim about fish’s happiness gives him a chance to give Zhuangzi a taste of his own medicine.

(Z2) You are not me. How do you know that I don’t know?

The first of Zhuangzi’s two responses shows him taking the skeptical thought at face value and examining its consequences more carefully than Huishi does. He neatly turns the tables, inviting us to see the first exchange not as a dogmatic claim laid low by virtuous skeptical scruple, but as an innocent non-dogmatic observation unjustly attacked out of the blue by dogmatic skepticism. Huishi’s own dogmatism is now itself subject to skeptical counterattack. Zhuangzi’s point here could be put like this: live by the epistemic gap, die by the epistemic gap. So we’re taking offhand remarks to express dogmatic claims to knowledge, are we? Well, the justification for your (implicit) claim (that I don’t know) is exactly the same as that for mine (that you don’t know that I don’t): that the epistemic gap between subject and object (Zhuangzi/fish; Huishi/Zhuangzi) cannot be overcome. That’s how skeptics should argue, he seems to say: by humbling real dogmatism – here ironically manifested in a skeptical attack – not by coming down like a ton of bricks on every innocent remark.

Of course, Huishi could respond dogmatically to this, by claiming that in his case he did indeed bridge the epistemic gap, and does indeed know Zhuangzi’s mind; but that just undermines his original skeptical remark. He could also cut his losses by conceding the point, if he even sees what it is yet; but he doesn’t.

(H2) I’m not you, so I don’t know about you. You’re not a fish, so you don’t know about fish: Q.E.D.

Instead, smelling victory, Huishi takes the bait. It seems to him that in attempting to parry the skeptical attack, Zhuangzi has been the one to concede the point. Huishi’s original concern wasn’t his own knowledge of Zhuangzi, which he will happily give up on the skeptical grounds Zhuangzi provides, but instead Zhuangzi’s knowledge of the fish, as claimed in (Z1); so when Zhuangzi admits that the epistemic gap cannot be crossed by argument (the skeptical “claim”), Huishi simply repeats the skeptical conclusion, now ironically presented (presumably the irony is the author’s … ?) as a substantive result conclusively established by argument. But of course this is still just as much a dogmatic result as a skeptical one. Consistent and effective skepticism cannot push its conclusions too hard lest it fall into dogmatism at the metalevel. Huishi seems not to notice this; but rather than pointing it out (and thereby risking doing the same thing himself), Zhuangzi lets it go. Instead, he tries another tack.

(Z3) Let’s go back to where we started. When you said “whence do you know fish’s happiness?”, you already knew I know it before asking the question. I know it from up above the Hao river.

First, note the ambiguity (at least in this English version) of “where we started” in the first sentence, which can mean both “I see my first reply (Z2) didn’t convince you, so forget that and let me try again” (in other words, don’t take (Z3) as a reply to (H2) but instead to (H1)) and “remember where we were standing when I made the original remark.” This cleverly sets up the ambiguity to follow, between “whence” in the figurative sense of “how/by what means” and in the literal sense of “from where.” One of Zhuangzi’s points here is that to ask how I know can be taken precisely not to shed doubt on my knowledge but to take it for granted and inquire simply into my means, and this is why that question can be answered with a clever pun: Zhuangzi provides no justification, but instead a location – that from which he made the original observation.

A743E-lg This will of course strike Huishi as mere rhetoric rather than the conclusive result he seeks. Zhuangzi has not even pretended to prove that the fish were happy, let alone that he has a demonstrably reliable method of determining the mental states of lower animals. So why does this response work, if it does? The answer is that for skeptics “mere” rhetoric is appropriate where a conclusive result is not (it is Huishi, not Zhuangzi, who declares victory with “Q.E.D.”). On the other hand, Zhuangzi seems not to have conceded an inch: in fact, while at first he simply mentioned the fish’s happiness, now he says that he knows that the fish are happy, that the question is simply how he knows, and that the answer is that he knows from above the Hao river.

Why then, if Zhuangzi insists that he knows, is his attitude skeptical at all? Zhuangzi’s skepticism holds (i.e., without asserting) that the proper attitude is active attention to one’s surroundings, literal (the world) and figurative (the conversational context), rather than dogmatic determination, by rigorous argument, of what must be the case for everyone, and therefore for us as well. With respect to the former sense, Zhuangzi is chiding Huishi for not simply looking and seeing, as he suggested; instead, Huishi disregarded the world and turned immediately to abstract epistemological considerations. Similarly, in the other sense of “where we started,” Huishi disregarded the conversational context (an offhand remark by a notorious skeptic) in his haste to utilize the watertight inference from “no non-fish knows fish’s happiness” to “you in particular don’t know fish’s happiness,” and thus finally extract a concession from Zhuangzi by force. In response, Zhuangzi reminds Huishi that he was there too, above the Hao river, and that he (Zhuangzi) was not making some universal claim for posterity but addressing him, then.

Happy_fish In that context, Zhuangzi hints, Huishi might easily have taken the remark not to have made a claim about the fish at all, meaning instead something like: “On days like today, when it’s so nice out, I really feel at one with nature. It is as if I see my own happiness reflected back at me wherever I look. Those fish there, for instance: of course we can’t really know what they’re thinking – if anything – but when I see them darting about like that I find it impossible not to see them as happy.” Indeed, Zhuangzi’s subsequent pique seems to say: “I thought you were there with me then – was I wrong?”

More to our original point, attention to context has an epistemological aspect as well. Zhuangzi of course demands of his own skepticism that it not be dogmatic. Like (some) ancient Greek skeptics, he allows ordinary belief and reserves his barbs for dogmatism. This is why he can continue to maintain, in the face of skeptical pressure, that he knows fish’s happiness. However, not all doubt of ordinary claims is dogmatically skeptical. If ordinary belief can be legitimate by skeptical lights, then so is ordinary doubt. Rather than confirming, with (H2), the (dogmatic) skeptical interpretation of (H1), Huishi might instead have said:

(H2*) What? I’m not trying to prove that you can’t know; I’m just not convinced that you do know. They actually look frightened to me – that’s why they scattered. That’s why I ask: where did you get the idea that they’re happy?

This is still a fairly combative response, and it certainly misses Zhuangzi’s meaning as hypothesized above, but it seems to commit no philosophical misstep by skeptical lights. In this case Huishi’s doubt that the fish are happy (or that we can know that they are) is not in spite of his strong impression that the fish are happy, given an overly scrupulous reluctance to claim his attitudes as his own. It is due to his not having that impression at all. In fact, he thinks the opposite, and says so. In American pragmatist C. S. Peirce’s anti-Cartesian terms, his doubt is not “paper” but made of sterner stuff. Such doubt is legitimated not by theory but by actual engagement with an interlocutor in a context: I don’t appeal to universal considerations but put my own beliefs about our shared world on the line next to yours. In this sense as well, with (Z3) Zhuangzi is saying: “If you doubted my original remark on your own behalf, rather than dogmatically, you should have said so then: weren’t you there too (and thus able to judge for yourself, just as I was)?”

At this point we can draw an important moral. What started out looking like a straightforwardly skeptical position – that we can never establish any belief or claim as true beyond doubt – has turned out to be perfectly symmetrical. Writing about ancient Greek skepticism, philosopher Michael Frede defends the paradoxical idea of skeptical belief with the idea that it is not belief per se that is important but the attitude, ordinary or dogmatic, with which it is held. Zhuangzi complements that skeptical point by showing that it can be made just as well by differentiating dogmatic from ordinary doubt. It is in this sense that skepticism is entirely in line with what seemed to be the anti-skeptical common sense of pragmatists like Peirce: although the latter characteristically targets illegitimate doubt rather than illegitimate belief, both locate the illegitimacy at the theoretical level, making the point symmetrical.

In the post-Cartesian context of considering radical skeptical attacks on knowledge, it is a given that belief can be illegitimate, and that if a stable position is to be found, then ordinary belief must be distinguished from dogmatic overreaching. Like Zhuangzi’s (mutatis mutandis), the pragmatist’s insight was that the same holds of doubt: that what presents itself as skeptical resistance to dogmatism can actually be itself a form of dogmatism, and that (in one formulation) doubt can only be given content against a background of shared belief and practice which renders hyperbolic doubt incoherent. Thus ordinary doubt stands with ordinary belief, and dogmatic belief falls with dogmatic doubt.

This may not seem like an exciting result. But the point can be generalized, and in a way which provides a very useful philosophical tool. In my last column I discussed the metaphysical issue of ontological realism and idealism; and we might very well see the post-Hegelian approach of McDowell et al as making another version of this same point. Realism and idealism are both right, in a way – but in picking one side and saying why and how, we find that in solidifying the very real insight we are trying to defend, we see it lose its characteristically realist (or idealist, as it may be) character dissolve before our very eyes. Surprisingly, philosophical idealism can in fact be shown to be true, after all, but only when its actual content can just as well be construed as realistic – with the same true, mutatis mutandis, of realism. But that's another story for another day.

Like what you're reading? Don't keep it to yourself!
Share on Facebook
Tweet about this on Twitter
Share on Reddit
Share on LinkedIn
Email this to someone