Zhuangzi on ‘happy fish’ and the limits of human knowledge

ABSTRACT The “happy fish” passage concluding the “Autumn Floods” chapter of the Classical Chinese text known as the Zhuangzi has traditionally been seen to advance a form of relativism which precludes objectivity. My aim in this paper is to question this view with close reference to the passage itself. I further argue that the central concern of the two philosophical personae in the passage – Zhuangzi and Huizi – is not with the epistemic standards of human judgements (the established view since Hansen, “The Relatively Happy Fish”), but with the more basic problem of species-specific perspectives. On my reading, Zhuangzi’s emphatic positionality in the passage – on the dam, accompanied by his friend Huizi – plausibly suggests a circumspect reflection on the limitedness of human knowledge. It is significant that Zhuangzi’s knowledge of fish happiness is avowedly from a certain place, and not absolute. But there is still a sense in which this view is objective: namely, insofar as it adequately accounts for an inherently human perspective on the world. I call this modest form of relativism ‘Species Relativism’, which, crucially, leaves room for objectivity, even though a fully objective (i.e. absolute) view of the world is not accessible to humans.


Introduction
The "happy fish" passage concluding the "Autumn Floods" chapter of the Classical Chinese text known as the Zhuangzi (henceforth 'Zh. 17') 1 has traditionally been seen to advance a form of relativism which precludes objectivity. My aim in this paper is to question this position with close reference to the passage itself. I further argue that the central concern of the two philosophical personae in the passage -Zhuangzi and Huiziis not with the epistemic standards of human judgements, but with the more basic problem of species-specific perspectives. 2 I contend that Zhuangzi's emphatic positionality in the passage, accompanied by his (human) friend Huizi, plausibly suggests a circumspect reflection on the limitedness of human knowledge. It is crucial that Zhuangzi's knowledge of fish happiness is avowedly from a certain place, and not absolute. But there is still a sense in which this view is objective: namely, insofar as it adequately accounts for an inherently human perspective on the world. 3 I call this modest form of relativism 'Species Relativism' ('SpR'), which, crucially, leaves room for objectivity, even though a fully objective (i.e. absolute) view of the world is not accessible to humans. 4 SpR says that: It is unfounded to favour human points of view over the conflicting points of view other animals have on the same things. 5 I understand SpR in terms of the following specifications: (1) There are no second-order relativities. It is an absolute fact that x is p for Species S (e.g. it is an absolute fact that a damp environment is healthy philosophically consistent unit in spite of these difficulties, see e.g. Mair ("Chuang Tzu and Erasmus", and Van Norden ("Competing Interpretations",247). Note that the "happy fish" story presents a character called Zhuangzi (hence the absence of italics in my discussion of that passage). To date, no critical edition of the Zhuangzi has been produced. In this paper, I take as my base text a Song version -the Nan hua zhen jing 《南華真經》edition of the Xu gu yi cong shu 《續古逸叢書》 . The latter also forms the basis of the Harvard-Yenching Index. 2 I am not alone in adopting such an approach. Kuwako ("Knowing the Joy of Fish", 142) argues that the passage is concerned with "the problem of interspecies understanding of minds" as opposed to the more general problem of other minds (which typically dominates discussions of the passage). Perkins ("Of Fish and Men") also highlights "species difference" as the core issue, and incisively discusses the role that the example of fish plays (here and elsewhere in the Zhuangzi) in deflating the primacy of humans. However, neither analyses the passage in great detail. For disagreements between their views and my own, see fn. 27 below. 3 The conception of objectivity I have in mind here has much in common with that of Thomas Nagel. I assume that objectivity must comprise the facts about individual humans and other creatures, and what it is like to be them; and if that is the case, then it must account for the differences arising from species-specific sensory apparatuses. As Nagel puts it ("Subjective and Objective", 206), the more one abstracts from one's: specific spatial, temporal, and personal position in the world, then from the features that distinguish him from other humans, then gradually from the forms of perception and action characteristic of humans, and away from the narrow range of a human scale in space, time, and quantity, the more objective one's view of the world becomes. Even if, as Nagel grants, it is difficult to envisage an "end-point" to such a process, we can still achieve a more objective view of the world. 4 My thesis regarding the possibility of objectivity in the Zhuangzi is similar to Connolly's ("Perspectivism",487), who uses Nietzschean perspectivism to show that the Zhuangzi "aims at attaining a greater level of objectivity", rather than collapsing into hard relativism. 5 This formulation is inspired by Burnyeat's attribution of a similar view to the Greek thinker Heraclitus in terms of "conflicting appearances" between species ("Conflicting Appearances", 104).
for fish). 6 Relative facts are relative to a framework (that of e.g. the human species), but the facts themselves are nonrelative.
(2) SpR itself is nonrelative. SpR is not a view which strictly belongs to a human perspective: SpR ipso facto denies the priority of any human viewpoint. To say that there is not SpR absolutely is to think that other animals don't experience the world in a way that is radically different from us, or indeed to deny that they experience the world at all. 7 However, while SpR is absolute, the above formulation of it is from the human perspective: x is p (e.g. 'a damp environment is healthy') for S1 and not-p (e.g. 'nothealthy') for S2, etc., presupposes the relevance of p; but p is not necessarily relevant beyond the human purview, as will become clear. 8 In what follows, I argue that the "happy fish" anecdote powerfully illustrates the consequences of SpR. 9 In doing so, I aim to elucidate how an objectivist conception of knowledge is actually compatible with the epistemological picture we get in this passage. 10 The species-specific indexical: "this is fish being happy" (shi yu le ye 是魚樂也) Zhuangzi and Huizi were strolling (you 遊) on the dam of the Hao River. Zhuangzi said, "How these minnows jump out of the water and play about (you 游) at their ease (cong rong 從容)! This is fish being happy (le 樂)! " 6 'For' is here understood not in the sense of 'from the perspective of', but rather as 'in relation to'. The point is that it is a fact that water is good for the survival of fish, rather than that, from the perspective of fish, water is conducive to their survival. In Zh. 2, the character Wang Ni points out that a fish obviously thrives in a damp environment, while a human develops pains in her joints, and even faces the prospect of death, when sleeping in a damp environment. 7 I intend this argument to be flexible around the specific details of species individuation, and leave this matter to the biologists and philosophers of biology. For my purposes, "species" may be read as shorthand for an appropriately individuated biological kind. 8 Biological facts (e.g. 'fish thrive in a damp environment') seem to hold independently of what humans think or say. But other more evaluative facts (e.g. 'these fish are happy') don't obviously hold outside a human purview. This, I argue, is the problem which the "happy fish" passage raises. 9 I grant that SpR is not the only type of relativity the Zhuangzi is interested in, since different perspectives within the same species or even the same individual are often considered. However, SpR deserves to be investigated thoroughly in its own right, since it is particularly relevant to the question whether the Zhuangzi can accommodate both objectivism and (epistemological) scepticism. 10 Van Norden ("Competing Interpretations"), Schwitzgebel ("Zhuangzi's Attitude"), Graham ("Taoist Spontaneity",12ff;Disputers,193), and Allinson ("Relativism") have all tried to limit the scope of the Zhuangzi's scepticism. The more convincing of these attempts, in my view, are Van Norden's ("Competing Interpretations", 258) and Schwitzgebel's ("Zhuangzi's Attitude", 69), both of whom attribute "therapeutic scepticism" to the Zhuangzi. However, they do not adequately account for passages which specifically call the reach of language into question. My strategy is rather to take seriously the Zhuangzi's scepticism about language, which I take to be a consequence of a broader criticism of speciesism.
Huizi said: "You, sir, are not a fish, how (an 安) do you know (zhi 知) what the happiness of fish is?" Zhuangzi replied: "You, sir, are not me, how (an 安) do you know (zhi 知) that I do not know (bu zhi 不知) what the happiness of fish is?" Huizi said: "I am not you, sir, so I inherently don't know you; but you, sir, are inherently no fish, and that you don't know (bu zhi 不知) what the happiness of fish is, is [now] fully [established]." Zhuangzi replied: "Let's return to the roots [of this conversation]. By asking "how (an 安) do you know (zhi 知) the happiness of fish," you already knew (zhi 知) that I know (zhi 知) it, and yet you asked me; I know (zhi 知) it by standing overlooking the Hao River." (Zh. 17. Trans. Meyer,"Truth Claim",335,modified.) The limits of the human purview A great deal has been written about the "happy fish" story concluding Zh. 17. The picture that emerges from the bulk of the discussions in the secondary literature is that Zhuangzi: either (1) commits himself to fallacious argumentation (whether willingly or unwillingly); or (2) endorses a relativistic theory of reality, in which objective truth, and with it the idea of a metaphysical absolute, is ruled out.
The strongest grounds for (1) consist in Zhuangzi's inference at the end of the passage, since, prima facie, it is at best a non-sequiturand at worst sophistical to conclude that because Huizi asked for Zhuangzi's grounds for knowing p, therefore he already knew that Zhuangzi knew p. I submit that commentators are right to worry about the logic of the passage. However, I will argue that the inference is not only valid but also gestures at a deep insight in the Zhuangzi's epistemology on the issue of species-specific perspectives. Let us consider perhaps the most influential attempt at saving Zhuangzi from the charge of sophistry or irrationality: namely, Hansen's (followed in many respects by Teng's), who proposes an interpretation along the lines of (2).
Hansen claims that, having deemed any reflection on an absolute view of the world to be irrelevant, Zhuangzi "consistently resists … invitations to state the "view from nowhere"" ("The Relatively Happy Fish", 159). It is not always clear whether (a) Hansen thinks that the absolute view is non-existent because it is ultimately inaccessible to humans; or whether (b) he takes it to be irrelevant (though real) inasmuch as it remains unknowable to humans. Option (a) is fallacious, since it assumes that because something cannot be accessed (an epistemological claim), therefore it isn't real (an ontological claim). 11 Though (b) is 11 This might seem to leave a third optionnamely, that the absolute view is simply non-existent (and hence inaccessible to anyone). But this is not an option which Hansen considers, since he does not reject the absolute view per se, but only that it is intelligible to us: "The cosmos does not make judgments. The less problematic, it also presumes a degree of anthropocentrism which is far removed from the spirit of Zh. 17. Indeed, given what precedes the "happy fish" passage, it is not the case that the Zhuangzi is unconcerned with whatever is not directly relevant to an intrinsically human experience of the world. 12 But given how Hansen elsewhere ("A Tao of Tao") elaborates on the Zhuangzi's concept of dao and perspectival relativism, it looks like he actually does think the Zhuangzi is committed to (a). 13 On his unorthodox line of interpretation, dao 道 merely refers to "discourse", 14 and not to a metaphysical absolute, as scholars have usually understood it. One major worry I have with his reductionist view is that it ignores the sense in which it is precisely discoursean inherently human method of conceiving of the worldwhich is repeatedly shown to be limited in the Zhuangzi, including earlier on in Zh. 17. 15 I am also unpersuaded that the Zhuangzi's scepticism entails the view that all "evaluative distinctions made in prescriptive discourse" are equally valid (pace Hansen, "A Tao of Tao", 33; see 35). Importantly for our purposes, the notion of happiness (le 樂) in the Zhuangzi is usually taken to amount to an point of view of the cosmos (or any absolutist point of viewincluding the perfect man) is so irrelevant as to be unintelligible to us as we seek ways in this world." ("The Relatively Happy Fish", 159 Similarly, in Zh. 2, disparate types of animals are invoked (men, monkeys, fish, deer, centipedes, snakes, crows, mice, etc.) and, given their disparate behavioural patterns and natural habitats, it is suggested that their views of the world must be radically different from one another. But at the same time, such passages emphasize that each species has a perspective, which is on an equal footing with that of human beingsa point which Perkins ("Of Fish and Men", 121) has argued is especially apparent in the Zhuangzi's repeated use of the example of fish. Parkes ("The Wandering Dance", 238) has more generally stressed the significance of non-mammal perspectives in the Zhuangzi: "mammals are the animals closest to human beings and those with whom it is easiest to fall into the pathetic fallacy; it is far less easy to project human feelings and emotions onto creatures such as insects and fish". As will become clear, this remark supports my reading of the "happy fish" passage, since I take it to make a point about the inescapably human point of view from which we judge fish to be happy. 13 Hansen ("A Tao of Tao", 24) contends that "Chuang-tzu's tao is a linguistic rather than metaphysical object … and that his doctrine is relativist rather than absolutist". His case for dismissing more orthodox 'metaphysical' readings hinges on (1) supposed inconsistencies inherent in these views and (2) the allegedly alien (either 'Western' or Buddhist) modes of reasoning they attribute to the Zhuangzi (see 28). Yet metaphysical readings of the Zhuangzi are not particularly guilty of these tendencies. Of course any modern reading of the Zhuangzi presupposes importing elements of contemporary (including 'Western') frames of referenceyet, ironically, this is a disclaimer which Hansen ultimately makes with regards to his own interpretation in the same article (51). But note that in his later monograph (Daoist Theory, 297), he reads Zhuangzi as "taking a position that does not reject metaphysical realism. He is merely skeptical about the usefulness of the realist, regulative ideal of one right answer"a position which instead favours interpretation (b). Apart from the objection I have raised above against (b), I fail to see how Hansen's reading of dao as a strictly linguistic object in "A Tao of Tao" can be squared with the metaphysically loaded reading of the Zhuangzi which he concedes in Daoist Theory. 14 Note that Hansen is not always consistent in this respect. In Daoist Theory (265-303) he sometimes appears to use the term in the broader sense of 'perspective', which he applies to (non-human) animals as well (see e.g. 290, 294). 15 See fn. 12. See also Ivanhoe, "Was Zhuangzi a Relativist?", 201.
ideal of human flourishingimplying the permissibility of at least certain kinds of prescriptive discourse. 16 In addition, surely the Zhuangzi would not want to say that all prescriptive judgments are reducible to the social, 17 since there are also more basic biological facts which determine how we humans view the world, including, most broadly, those which set us apart from other species. As Ivanhoe points out, epistemological scepticism does not entail that all use of language is meaningless or "equally valid": "To deny that language is adequate to the task of providing us with a full and faithful understanding of the world is not to say it is of no use in this task" ("Was Zhuangzi a Relativist?",199). To recognize that language is always relative to a human perspective on the world is not to say that it does not reach the world at all. 18 This raises the question: what is the relevance of an inherently human evaluative judgement (signalled by the indexical shi 是) with regard to nonhuman creatures, such as "this is fish being happy"? I will now show why it is significant that Zhuangzi knew fish happiness only from a certain place, which he occupied with his fellow human friend, Huizi, as they walked along the dam.
The basic problem of species-specific perspectives My thesis, then, is that the central concern of the two philosophical personae in the passage relates to the issue of species-specific perspectives.
Let us note that Huizi himself starts out being concerned with the speciesspecific issue: Huizi said: "You, sir, are not a fish, how (an 安) do you know (zhi 知) what the happiness of fish is?" Hansen thinks that, from the start, Huizi is committed to a particular kind of subjective relativism, which says that the perceiver can only know whatever is true of one's inner self. On his view, Zhuangzi's distinct version of relativismwhich he calls "perspectival relativism" ("The Relatively Happy Fish", 148)does not assume an exclusively first-person epistemic standard. This, he thinks, enables Zhuangzi (unlike Huizi) to avert the charge of self-refutation. Teng's ("Happy Fish Revisited") response to Hansen rightly disputes that Huizi assumes an exclusively first-person standard from the outset, since the worry Huizi raises is the more basic problem In Zh. 2, we are told that "human speech is not just a blowing of air. Speech has something of which it speaks, something it refers to" (夫言非吹也 。 言者有言; Trans. Ziporyn, Zhuangzi, 11.) See Ivanhoe ("Was Zhuangzi a Relativist?",[199][200] for further evidence in the Zhuangzi of the idea that there is something objective about language and the "human perspective" on the world. of species-specific perspectivesin this case, whether Zhuangzi, qua human, can know fish happiness. It is worth noting that it is Huizi who explicitly broaches the generic idea of fish happiness, as opposed to speaking about particular fish being happy, as Zhuangzi does. Now it is true that the phrase attributed to Zhuangzishi yu le ye 是魚樂也is usually translated as "such is the happiness of fish", since most translators take shi 是 ("such") to focus the ye 也 (copula "is") and take there to be an implied zhi 之 which nominalizes le 樂 (the Pness of x). 19 Yet an alternative translation, which does not presume a suppressed zhi 之, and instead reads shi 是 as an indexical ("this"), and le 樂 as an adjective, is simply "this is fish being happy" (x is p). I opt for the latter translation, since I take it to be significant that throughout the rest of the passage, we find a different construction which explicitly nominalizes le 樂 and uses zhi 之, and thereby emphasizes the abstract idea of "fish happiness".
A second linguistic point to take note of is the pun on the verb you (遊 / 游), which has two different referents in the passage; it describes both (1) the friends wandering along the dam and (2) the fish meandering about in the water, which are described as doing so at leisure (cong rong 從容). Presumably the two friends, too, are at their leisure, since they are freely taking a walk together. Thus, the text composition signals that the two men are engaging in an activity which they feel to be parallel to the experience of fish; both activities are felt to be a kind of you-ing ("wandering"), and, because both appear to be leisurely, they are deemed a source of happiness in both the case of the men and that of the fish. 20 Hence what is initially at stake is not the issue of relating with other human subjectivities, but the broader issue of accessing the experience of nonpersons. Huizi is intuitively talking about the same fundamental problem as that broached by Zhuangzi; and he struggles with its implications in a philosophically sincere way. The issue is that he is not fully aware of the irreducibly human perspective from which he speaks about "fish happiness".
Two observations are worth making at this stage. First, Huizi's initial objection already betrays his assumption that fish happiness has meaning beyond the human purview, since he thinks that being a fish would in principle give one access to fish happiness.

19
The implications of the absence of the particle zhi 之 in editions based on the Xu gu yi cong shu 《續古 逸叢書》(see e.g. the Nanhua zhenjing pijiao 《南華真經批校》 , 508) have not to my knowledge been discussed in any great detail, even though many scholars follow this version at this particular juncture. Note that later Qing dynasty editions (as reproduced in e.g. Zhuangzi ji shi 《莊子集釋》 , 606) include the particle zhi 之, and thus instead read: shi yu zhi le ye 是魚之樂也. It is possible that the inclusion of this particle is a later addition, though it is not my intention to defend such a claim here. A more important motivation for adopting the Song variant, for my purposes, lies in its philosophical interest, as I discuss below. 20 See e.g. Jullien (Nourrir sa vie), Fraser ("Wandering the Way") and Virág (Emotions, on how you 遊 ties into the Zhuangzi's eudaimonistic ideal (le 樂) more generally.
Second, Huizi does not speak about those particular fish being happy, as Zhuangzi seems to do at the start of the anecdote, but about the problem of not being a fish in general. The question is no longer centred on the issue of whether "these fish are happy" (shi yu le ye 是魚樂也), but on the criteria on the basis of which one thinks there is fish happiness more broadly (yu zhi le 魚之樂).
Zhuangzi's reply to Huizi's objection, however, is prima facie equally problematic: Zhuangzi replied: "You, sir, are not me, how (or "from where", an 安) do you know (zhi 知) that I do not know (bu zhi 不知) what the happiness of fish is?" At first glance, Zhuangzi seems to be casting doubt on the possibility that Huizi could know what he himself knows (or doesn't know), on the basis that he (Huizi) is not him (Zhuangzi). His assumption thus seems to be that if you are not x, you clearly don't know anything about the mental states of x. Huizi himself explicitly accepts the soundness of this line of reasoning: Huizi said: "I am not you, sir, so I inherently don't know you … " 21 Huizi's point seems to be that "I don't know you" in the sense of "I don't know what is going on in your mind". It should be noted that this view is different from other-minds scepticism, since in a perspectival way Huizi may still be said to know other people, trees and tables. But his view does imply the following: "I am not you so I can only know how you are relative to me". 22 But then his next move violates this premise, since, despite not being Zhuangzi, he claims to know at least one thing about what Zhuangzi does and doesn't know: but you, sir, are inherently no fish, and that you don't know (bu zhi 不知) what the happiness of fish is, is [now] fully [established].' Thus Huizi claims that Zhuangzi does not know what the happiness of fish is. But he cannot claim to know this, since the standard of knowing employed herei.e. knowing by inferenceis no longer open to him now that he has endorsed the first-person standard. 23 Thus Zhuangzi, with his question, has whimsically led Huizi to refute himself, and has thereby discredited the objection that he, Zhuangzi, cannot know the happiness of fish. 24 Whatever comical dimension there may be to Zhuangzi's trap, one might still protest, on Huizi's behalf, that Zhuangzi, too, in asking his question, seemed to commit himself to the first-person standard, and thereby to refute his own 21 我非子，固不知子矣. Note that this construction marks a strong statement, whereby gu 固 ("strongly", "inherently") is followed by a particle denoting the end of a process (yi 矣). For the view that Zhuangzi is the more logical of the two thinkers based on external evidence, see Alt ("Logic and Language"). claim that he knew fish happiness. However, I contend that there is a deep philosophical point at issue in Zhuangzi's question, which is not self-undermining; rather, it draws attention to the differences in points of view which arise between people when they disagree about something. Zhuangzi is giving Huizi a fair chance to defend his view on the basis of where he is saying it from.
The ambiguity on the meaning of an 安which in Classical Chinese may convey both the spatial idea 'from where' and the explanatory 'how' (much like the English 'whence') is of central importance here. 25 Given Zhuangzi's playful references to location throughout the passage, such an ambiguity seems to suggest that one's actual place in the world has implications for how one can claim to know something about that world. Indeed, having presented Huizi with the truism that "you are not me", Zhuangzi asks an (whence) he knew that Zhuangzi did not know fish happiness.
Huizi's assumptions I have been suggesting that certain judgements may count as more objective than others, insofar as they are less contingent on subjective factors or points of involvement. Thus, when Zhuangzi asks Huizi for a 'from where' clarification in terms of a 'you' / 'not me' distinction, he is granting that Huizi might see the world in a slightly different way, qua the unique individual that he is, and so have grounds, unbeknownst to Zhuangzi, for thinking that Zhuangzi doesn't know fish happiness. But as we have seen, Huizi ends up providing an explanation which is internally incoherent. Hence Zhuangzi is free to drop the person-specific issue of perspective, and return to the original issue, which, as I have been arguing, was the broader species-specific one: "Let's return to the roots [of this conversation] … ".
What is important is that the very idea of fish being happy was not questioned in Huizi's objection. In fact, it is Huizi who introduces fish happiness as an abstract concept. When Huizi challenges Zhuangzi's epistemic grounds for claiming to know fish happiness, he is already speaking from within their shared human purview. Throughout the passage, the term consistently used to refer to "knowledge" or "knowing" is zhi 知the term which the Zhuangzi often associates with imperfect, inescapably human knowledge, which has no absolute claim to truth. 26 At best, it conveys the limited knowledge which humans can claim about the world in virtue of being one of many animal species.
Huizi's truism that Zhuangzi is not a fish shows some appreciation of the problem of SpR, but the ensuing conclusion that Zhuangzi couldn't know 25 123) has established that the use of the interrogative an 安, understood most literally as "whence", implies that knowledge claims are relative to the standpoint one occupies in the world.
fish happiness assumes that there is such a thing as fish happiness outside the human-specific framework. In working with their limited, human tools of engagement with the world, Huizi (though he was reluctant to admit it) and Zhuangzi intuitively recognized the happiness of fish; but only Zhuangzi had anticipated in what sense this could be true, namely, insofar as it belonged to their shared human purview. 27 Given the emphasis on the interlocutors' particular positionality, on the dam, overlooking the Hao River, and Zhuangzi's qualification that he knew that the fish were happy from that vantage point, we seem to have a self-conscious reflection on the specifically perspectival reach of human knowledge. The dam is an apt marker of a human worldview: it is a man-made barrier imposed on the natural environment, which emphatically separates off human subjects from the world of fish. 28 Zhuangzi is embracing the 'whatis-it-like-ness' of looking at the world as a human, while also acknowledging that this vantage point is not absolute; it is unapologetically from a certain place, and not from everywhere: "I know (zhi 知) it by standing overlooking the Hao River". When it comes to the non-human realm of experience, we should not assume that we can ever reach a view from somewhere else, let alone from everywhere.

The inference
We are still left with the difficulty that Zhuangzi's inference, according to many scholars, is invalid: By asking "how (an 安) do you know (zhi 知) the happiness of fish (or: that the fish are happy),' you already knew (zhi 知) that I know (zhi 知) it, 29 and yet you asked me; In fact, from the point of view of contemporary philosophy, Zhuangzi's inference would actually count as valid according to the principle of conversation 27 Fraser ("Wandering the Way", 542-3) has characterized the Zhuangzi's conception of happiness (le 樂) as a "normative conception of a flourishing, happy, or well-lived human life". Wandering (you 遊), he argues, is the "general mode of activity" one is in when leading a happy life (541). What is important for my purposes is that when Zhuangzi speaks of fish being happy and describes them as you-ing, he is plausibly reflecting on the conditions for a flourishing "fish existence", in accordance with his own normative ideals (le 樂, you 遊). But then the standards by which he is assessing the existence of fish are inherently human. Hence the limits of the judgement that "this is fish being happy". For the alternative view that you 遊, in the specific context of the "happy fish" passage, establishes a continuity between humans and fish which allows humans to share in the fish's happiness, see Moeller ("Rambling without Destination", 249-50). Perkins ("Of Fish and Men", 128) also suggests that you 遊 is here associated with the possibility of cross-species communication. While our starting points are similar, my approach also differs from Kuwako's ("Knowing the Joy of Fish"), insofar as I do not take Zhuangzi to be making any claims about fish happiness in a way that extends beyond his specifically human purview; the point is precisely that claims about the happiness which actually arises in fish are unassertible (pace Kuwako, "Knowing the Joy of Fish", 145). 28 I thank Dirk Meyer for this observation. 29 Note that the Chinese leaves unspecified what exactly Zhuangzi knew. Indeed, zhi 之 may be taken to refer to Zhuangzi's initial contextual claim that such is fish being happy, rather than such is the happiness of fish (in the abstract), which is how Huizi frames the problem in his criticism of Zhuangzi. implicature within pragmatics. 30 For instance, in asking "how did you come home?", one is already presupposing that you came home; whether you came home is not really at issue. In the same way, by asking how he knew fish happiness, Huizi might be taken to presuppose that Zhuangzi did know fish happiness. 31 Moreover, if we take seriously that the ambiguity on the meaning of an 安 is at play in the passage, as most commentators do, the fact that Huizi is asking from where Zhuangzi knows further suggests that the fact that he knows it is not being questioned.
On the other hand, one might still worry that Zhuangzi's reply eludes Huizi's objection, to the extent that he hasn't proved how he knows fish happiness. However, once we appreciate the point Zhuangzi is making about the shared human point of view from which one identifies and speaks of fish happiness, things become rather clearer.
I argued that Huizi accepted the very idea of fish happiness because he agreed with Zhuangzi's identification of minnows jumping out of water and "playing about" as an instance of fish being happy. The fundamental issue which is lost on Huizi, but which Zhuangzi playfully gestures at, is that we have no idea whether the happiness of fish would amount to anything outside our human purview. This is why Zhuangzi stresses that he knew the fish were happy from where he stood on the dam. The problem we face as humans is that our knowledge is always from somewhere, and so falls short of the absolute. Invoking the experience of fish helps bring out this important insighta fitting end to a chapter replete with sceptical remarks about the reach of human knowledge. 32 Conclusion I have argued that the problem of conceiving of the world in a way that does not presuppose anthropocentric bias is an intractable onean insight that the Zhuangzi powerfully illustrates in the "happy fish" passage. I first examined 30 See e.g. Davis ("Implicature") on Grice's theory. 31 Interestingly, Hansen himself anticipates this solution: Ordinarily, in asking "how do you X" we presuppose that you have done or can do X. "How do you make the yo-yo sleep?" Then the question seeks information about what method you used in (admittedly) actually doing X. The form of the "how" question ordinarily implies one acknowledges the ability itself.
("The Relatively Happy Fish", 155) However, he thinks it would in this case amount to "sophistry", to a merely "verbal trick", since he holds that, "in a challenge to knowledge, the phrase "how do you know" normally has the opposite presupposition" ("The Relatively Happy Fish", 155). Hansen is certainly right that, taken on its own, the "how do you know" phrase is ambiguous (in Chinese as in English), and that in this sense the implicature that one does know does not necessarily obtain. But in the context of the passage, the crucial point is that Huizi did know that Zhuangzi knows from the outset, as betrayed by his abstract recognition of fish happiness upon seeing the minnows swimming at their ease. Hence Zhuangzi's response exploits the implicature that Huizi knows.

32
See Meyer ("Truth Claim") for a detailed discussion of Zh. 17 as a whole.
Hansen's highly influential reading, which takes the passage to advance a form of relativism that precludes objectivityan implication which I questioned based on a close reading of the text. I honed the narrative setting of the anecdote, which I took to bring the issue of perspectives to the fore. On a literary level, the shared location which Zhuangzi and Huizi occupy together reflects that humans are in many respects on the same plane, and share a common separation (figuratively, the dam) from non-humans (in this case, fish). Philosophically, this setting signals that Huizi and Zhuangzi share a common experience in looking at the fish swimming at their ease. This shared lived experience allows for the possibility of making more objective claims about the fish's happiness, even if such claims remain relative to a human framework. It has often been remarked that Huizi is depicted as a debater elsewhere in the Zhuangzi. 33 In this passage, he nevertheless seems motivated by a genuine philosophical intuition regarding the gap between species, to the effect that humans, not being fish, cannot claim to know fish happiness. But Huizi does not ultimately question the basic assumption that "happiness" is relevant to fish existence outside a specifically human framework. In this sense, the Zhuangzi presents him in a similar light to the debaters invoked earlier in Zh. 17, who, in taking for granted their anthropocentric lens, fail to appreciate that their own framework is constricted. 34