Wild Fox Confucianism: Obscuration (Bi 蔽), Self-Certification, and the Problem of Discernment in Late Ming Confucian Religiosity

Abstract

Wang Yangming 王陽明 (1472–1529) held that innate moral knowing (liangzhi 良知) knows every concern that arises, yet a person can obscure it, so that “though he may be said to know, he still does not know.” Studies of this puzzle have explained how liangzhi detects self-deception. I take up a harder question, asked by Wang’s student Ouyang De 歐陽德, who feared that a person might “take the thief for one’s son.” Even if liangzhi never errs, a person’s claim to speak from it can be wrong, and his certainty cannot settle the matter. Authenticating such a claim is a religious problem of discernment, distinct from the philosophical problem of self-deception and from the assessment of conduct. Chan Buddhism met it with the master’s seal, which gave another person standing to judge such a claim, a standing I call adjudicative externality. Wang’s teaching made room for such checks, but in the “ready-made liangzhi” of Wang Ji 王畿 a claim to awakening could become a claim to be beyond correction, which late Ming critics called kuangchan 狂禪. Ledgers of merit, and the societies that reviewed them, met this danger without certifying anyone’s awakening. Whatever a person claimed to have realized, his acts remained on the record.

1. Introduction

In his Inquiry on the Great Learning (Daxuewen 大學問), Wang Yangming 王陽明 (1472–1529) holds that innate moral knowing (liangzhi 良知) misses nothing, since “of every concern that issues forth, there is none that the liangzhi of my heart-mind does not itself know” (凡意念之發，吾心之良知無有不自知者). Yet a person whose liangzhi knows a concern to be good can still fail to love it sincerely and turn away from it, and so comes to “obscure the liangzhi that knows the good” (自昧其知善之良知). In Wang’s own words, “though he may be said to know, he still does not know” (雖曰知之，猶不知也; Y. Wang 2011, Daxuewen).

The formula invites a philosophical question, how a person can possess genuine moral knowledge and still deceive himself, and how liangzhi exposes the failure. It also raises a religious one. A person may claim that his judgment comes from liangzhi, or that he has realized it, and such a claim concerns a transformed state of the person and carries authority. The religious question is how that claim is to be authenticated when the claimant is certain and others are not. A letter from Wang’s student Ouyang De 歐陽德 (1496–1554; courtesy name Chongyi 崇一) leads to it. Writing of his own practice, Ouyang feared that a person might “take the thief for one’s son” (認賊作子), mistaking a private intention for liangzhi without knowing it. Even if liangzhi never errs, a person’s claim to speak from it can be wrong, and his certainty cannot settle the matter. This article calls it the problem of discernment.

Existing scholarship has concentrated on the philosophical question, which is older than Wang. Zhu Xi 朱熹 (1130–1200), the Southern Song master whose method of learning Wang followed as a young man and later rejected, feared that the promptings of desire could pass for Pattern (li 理), so that a person might be wholly self-deceived (zi qi 自欺) while “absolutely convinced” of what he believes (Zheng 2015, pp. 351–52; Zheng 2018, pp. 1350–51). Zheng Zemian argues that Wang’s doctrine of extending liangzhi (zhi liangzhi 致良知) answers this puzzle (Zheng 2018, 2022). In what he calls the structural model, liangzhi stands at a different level from a thought and sees its good and bad directly. In the genetic model, a deviant thought arises out of liangzhi as a wave rises out of water and betrays itself by its restlessness (Zheng 2018, pp. 1357–60). Zheng also marks where the answer runs out. Wang refers misjudgment to self-examination, and self-examination can only refer it back to judgment, in a circle Zheng describes as an “infinite regress” (Zheng 2018, p. 1361). Harvey Lederman reads genuine knowledge (zhen zhi 真知) as liangzhi’s introspective knowledge that one’s own inclination is good (Lederman 2022), and he later judges that this model fares better than perceptual, practical, and normative ones (Lederman 2023). On his reading the knowledge is private, since others “cannot acquire this knowledge about me at all” (Lederman 2022, p. 182 n. 24). Philip J. Ivanhoe reads Wang’s moral knowing as a form of perception (Ivanhoe 2011), and Huang Yong reads it as a “knowing-to” that inclines its possessor to act (Y. Huang 2017). These studies explain how moral knowledge works from the standpoint of the knower, and where that knowledge reaches its limits they leave the question with him. Wang’s own circle also asked how a disputed claim to that knowledge was to be assessed by others, and that question has received far less attention.

The article’s contribution is to treat discernment as a problem in its own right. Discernment, in the sense used here, is the authentication of a claim to moral realization. It differs from moral self-knowledge, which concerns what a person knows of his own mind, and from the assessment of conduct, which concerns what he has done. For the checks that make such authentication possible, the article proposes the term adjudicative externality, meaning that a check’s standing does not depend on the claimant’s present self-assessment. The late Ming sources show that Wang learning debated the problem in its own terms and found a practical response to it that did not require deciding who was awakened. The case also bears on the old question of rujia 儒家 and rujiao 儒教. Discernment is one of the functions that religions perform, and Confucian self-cultivation can be examined for it without first deciding whether Confucianism is a religion.

Chan Buddhism offers the closest comparison, since its vocabulary ran through late Ming Confucian debate. There, a master’s seal (yinke 印可) stands between a practitioner’s certainty and the tradition’s recognition of his awakening. The tradition also has a name for the counterfeit, wild fox Chan (yehu chan 野狐禪), and the title of this article borrows it. In the koan from which the name comes, a teacher becomes a fox for exempting the person of great cultivation from cause and effect. Late Ming writers, Buddhist and Confucian alike, had a related name for the same move, kuangchan 狂禪, or “wild Chan.”

The central claim is that discernment becomes a structural difficulty wherever a claim to attainment is allowed to certify itself. Wang’s teaching made room for teachers, shared standards, and institutions, and at Tianquan Bridge 天泉橋 Wang described how he received students of different capacities by different methods. The difficulty arises in a doctrine of attained spontaneity, such as the “ready-made liangzhi” of Wang Ji 王畿 (1498–1583), where a claim to be awakened can also become a claim to be beyond correction. In what follows, the article first defines discernment and adjudicative externality, then traces the difficulty from Wang’s warning at Tianquan to the polemic against kuangchan, tests it against the Taizhou 泰州 criterion of unobstructed attunement, and concludes with the ledgers of merit and demerit and the societies that reviewed them.

2. Discernment as a Religious Problem

2.1. From Self-Deception to Discernment

Wang’s account refuses to make knowing the good and doing it two separate achievements. When knowing fails to issue in action, the two have been “cut off by selfish desire” (被私欲隔斷; Y. Wang 2011, Chuanxilu, pt. 1, “Xu Ai lu”), and Dong Ping observes that Wang offered the unity of knowing and acting both as “medicine for the disease” of his time and as a description of “the original condition of knowing and acting,” so that the teaching was “both expedient and ultimate” (Dong 2025, p. 6). The same account gives urgency to a further question, which is who genuinely knows. The hard case is a person for whom selfish desire (si yu 私欲) has come, over time and quite systematically, to pass for liangzhi. When a concern stirs, he has no verdict to consult beyond his own present assessment, and that assessment is the thing in doubt. The infallibility of liangzhi does not tell him, or anyone else, whether his judgment is an exercise of it. This entrenched self-misrecognition is the case in view throughout the article. Passing bad thoughts and ordinary lapses of will are different matters. The case is also distinct from the familiar paradox of self-deception that Alfred Mele contests, since it does not require attributing contradictory beliefs to the agent (Mele 2001).

Ouyang De raised this case with his teacher. He wrote to ask how deliberation that issues from liangzhi is to be told apart from private contrivance in its clothes:

但思索亦是良知發用，其與私意安排者何所取別？恐認賊作子，惑而不知也。

But deliberation too is the functioning of liangzhi; how is it to be distinguished from the arrangements of private intention? I fear one may take the thief for one’s son, and be deluded without knowing it.

(Y. Wang 2011, Chuanxilu, pt. 2, “Reply to Ouyang Chongyi”)

A thief installed in the household as a son will not be exposed by the household’s assurance that he belongs there, because the household’s own power of recognition is what has been compromised. The final clause, “deluded without knowing it” (惑而不知也), makes the failure second-order. It escapes the very self-assessment that the person takes to express liangzhi. Wang answered with a mark of the genuine and referred the rest back to liangzhi:

良知發用之思，自然明白簡易，良知亦自能知得。若是私意安排之思，自是紛紜勞擾，良知亦自會分別得。蓋思之是非邪正，良知無有不自知者。所以認賊作子，正為致知之學不明，不知在良知上體認之耳。

A deliberation that is the functioning of liangzhi is naturally clear and simple, and liangzhi itself can know it. A deliberation arranged by private intention is naturally confused and laborious, and liangzhi will itself tell it apart. For whether a deliberation is right or wrong, perverse or correct, there is none that liangzhi does not itself know. That one takes the thief for one’s son is precisely because the learning of extending knowledge is unclear, because one does not know to realize it in liangzhi.

(Y. Wang 2011, Chuanxilu, pt. 2, “Reply to Ouyang Chongyi”)

Ease against agitation is the mark, and Zheng Zemian has shown why it cannot settle Ouyang’s question. A practiced schemer can deliberate for his own advantage with perfect composure, while a beginner sincerely trying to follow liangzhi may appear confused and laborious, so the state of a deliberation cannot confirm its source (Zheng 2022, pp. 106–7; cf. Zheng 2018, p. 1360). Zheng draws the consequence for the agent, who is left without guidance in sorting his own deliberations. The same consequence holds for a teacher. When the person he corrects takes the challenged judgment to be the voice of liangzhi itself, the teacher has no mark to point to that the student cannot claim as well.

Carried beyond the self, Ouyang’s question separates authenticating a claim to moral realization from recognizing moral knowledge in oneself and from judging conduct. Recognizing genuine moral knowledge in oneself is what the studies of Zheng Zemian and Harvey Lederman examine, and it belongs to moral epistemology. Judging whether a particular act was right belongs to ethical accountability, and Wang’s Nangan community covenant of 1518 (Nangan xiangyue 南贛鄉約; Y. Wang 2011, juan 17), with its registers of commendation and correction, did precisely this. Authentication asks whether a person who claims to act from liangzhi, or to have attained it, in fact does so. These activities are connected, since a failure of self-knowledge produces false claims, and conduct is among the evidence for them. They can also come apart, as in the late Ming fellowships that reviewed their members’ faults without ruling on anyone’s attainment. This article uses discernment for authentication alone.

Discernment in this sense is a religious problem because of what the claim is about and what it confers. The claim concerns a transformed state of the person, which Wang calls the realization of liangzhi and Buddhists call awakening, and the state is held to give its possessor authority. Such a person may teach, may be trusted where others are checked, and at the limit may be thought to need no correction. A tradition that recognizes such states has to decide who may certify them. Wang’s teaching makes the question acute, because it holds liangzhi to be infallible, treats its operation as accessible to the subject in a way that others do not share, as in Wang’s language of solitary knowing (du zhi 獨知), and allows a person to take the apparent character of a judgment as sufficient assurance of its source. Infallibility and interiority do not by themselves create the difficulty, since an infallible authority can be invoked mistakenly and an inward source can coexist with outside assessment. The difficulty arises with self-certification, when the mark offered to distinguish a genuine exercise from its counterfeit is the same self-assessment whose reliability is in question. Robert Sharf has made the general point about religious experience. An appeal to experience does not become independent evidence simply by being placed beyond another person’s inspection (Sharf 1995, 1998).

2.2. The Seal and the Wild Fox

Through the master’s seal or confirmation (yinke), Chan Buddhism separates awakening from a person’s claim to have awakened. The seal places another judgment between the claimant’s certainty and the tradition’s recognition of his claim. Yuanwu Keqin 圜悟克勤 (1063–1135) stated the principle plainly. In the primordial age before the first buddha, he wrote, one could awaken without a teacher, and awakening was “whole and ready-made” (渾圝成現), needing no tempering. Since then, even one who has taken up the matter “to the ground where no doubt remains” must still “rely on a teacher’s examination and seal” (直下承當到無疑之地，要須依師決擇印可; Yuanwu Keqin 2026, X69, no. 1357, p. 468b–c). Yuanwu grants that awakening can be whole and denies that the practitioner’s certainty can certify it. Bernard Faure has shown that appeals to unmediated realization in Chan were accompanied in practice by mediations of lineage, ritual, text, and institution (Faure 1991). Procedures differed from one community to another, and an authorized master could be wrong. Across these differences, a claim of inward realization did not have to be accepted on the claimant’s assurance, because the practice recognized someone else as entitled to contest it.

In practice the seal was bound up with office and faction. In the Song, the certificate of dharma transmission also qualified a monk for the abbacy of a public monastery, abbots were confirmed by officials, and transmission could be irregular. The Caodong line, for instance, was revived through a “transmission by proxy” from a master who had died before his heir was born (Schlütter 2008, pp. 38–40, 63–66, 79–89, 161). In the seventeenth century the seal itself became the object of open dispute. Hanyue Fazang 漢月法藏 (1573–1635) had come to awakening largely through his own practice and reading, and accepted transmission from Miyun Yuanwu 密雲圓悟 (1566–1642) only reluctantly. Hanyue held that a seal must answer to stated criteria, the “tenets” (zongzhi 宗旨) of the classical houses, while Miyun trusted the living encounter. When they fell out, Miyun charged that Hanyue had “appealed to the empty sky and claimed Linji as his true master” across the Five Dynasties, the Song, and the Yuan, and turned against him an older saying, which Yuanwu Keqin had echoed, that after the first buddha “whoever awakens without a teacher is a naturalist heretic” (望空啟告，攀臨濟為真師……威音王以後，無師自悟，盡是天然外道; Miyun Yuanwu 2026, X65, no. 1280, p. 166a). The two sides called each other wild foxes, and in 1733 the Yongzheng 雍正 emperor (r. 1722–1735) branded Hanyue’s line demonic and ordered it struck from the lineage (Lien 1996, pp. 170–89; J. Wu 2008, chaps. 4–6; Miyun Yuanwu 2026, X65, no. 1280, pp. 143a–b, 166a, 177a; Yongzheng 2026, X65, no. 1281, pp. 191a, 193b). The seal could be wrong, and it could be partisan. Even so, Hanyue’s claim to awakening was argued before others and judged by others, and his own certainty did not settle it.

The tradition also has a name for the counterfeit. Case 2 of the Gateless Barrier (Wumenguan 無門關), “Baizhang’s Wild Fox,” tells of an old man who appears at the assembly of Baizhang Huaihai 百丈懷海 (749–814) and confesses that in a former age he had been asked whether a person of great cultivation still falls subject to cause and effect. He replied that such a person “does not fall into cause and effect” (不落因果), and for that answer he was reborn five hundred times as a wild fox. He now asks Baizhang for a turning word, and Baizhang gives him one, “does not obscure cause and effect” (不昧因果). At these words the old man is released (Wumen Huikai 2026, T48, no. 2005, p. 293a–b; my paraphrase, cf. Aitken 1990).

The story turns on a teacher’s mistaken answer and another teacher’s correcting word. The old man’s error lay in exempting cultivation from the order that binds everyone else, and Baizhang’s word leaves the cultivation intact while withdrawing the exemption. The case serves here as a warning about false teaching and about the authority to correct it. The expression wild fox Chan names the lesson. Speaking from the position of a teacher does not make what one says true, and the difference is settled neither by the position nor by the confidence with which it is occupied.

The word on which Baizhang’s liberating answer turns is mei 昧, “to obscure” or “to darken.” It is the same character Wang uses in the Daxuewen for the failure of liangzhi, when he speaks of a person who comes “to obscure the liangzhi that knows the good” (自昧其知善之良知). The echo lies in a common word. Wang himself likened liangzhi to sunlight, which even “when clouds and mist block every side” still lets shapes be made out (雖雲霧四塞，太虛中色象可辨，亦是日光不滅處; Y. Wang 2011, Chuanxilu, pt. 3). Obscuration in his sense leaves the sun shining, and to reaffirm that it shines does not show whether a given person sees his own condition clearly.

2.3. Adjudicative Externality in Wang’s Teaching

The Chan seal gives another party standing to assess a claim that its subject holds to be inwardly certain. The decision need not persuade the claimant, and it keeps its authority within the practice even when he disagrees. A check of this kind has what this article calls adjudicative externality, meaning that its standing does not depend on the present self-assessment of the person whose claim it tests. The concept concerns standing and leaves correctness aside. A doctrine may leave room for such a check, and a community may give it institutional form, yet neither guarantees a correct verdict.

Adjudicative externality differs from an established moral standard, from interpersonal correction, and from institutional authority. A standard supplies the content of a judgment. Zhu Xi’s Pattern and the entries of a ledger tell a person what counts as good, but a standard has no external force if only the claimant may say whether he meets it. Correction is the usual Confucian form of such a check. Admonition that a student may set aside by appealing to his own certainty leaves the decision with him, and correction acquires adjudicative externality when the practice treats it as counting after he rejects it, as the Nangan covenant did. Institutional authority is one way of securing the standing at issue, as the Chan seal shows, and not the only one. A recognized teacher can possess it without an office, and an institution can lose it if it defers entirely to the claimant’s account of himself.

The features that let a check carry this standing can be stated from within Confucian discussion itself. In Zhu Xi’s arrangement, inquiry into the Pattern furnishes a standard that an agent can bring to later self-examination and that teachers and friends can discuss with him (Zhu 1983, 1986). Zhu knew that the Pattern must in the end be grasped and applied by an agent who may himself be obscured. He warned that where “the learning is not discussed and the Pattern not made clear, there are also places where one mistakes human desire for Heavenly Pattern, and this cannot but be examined” (學未講，理未明，亦有錯認人欲作天理處，不可不察; Zhu 2002, juan 54, first “Reply to Xiang Pingfu”). In this passage, he located the remedy in discussion (講) and in clarifying a Pattern held in common, and he did not look for it in sharper introspection. The Pattern so understood is prior, determinate, and public. It is prior because it is made out in a stage of inquiry that precedes the present act of self-assessment, determinate because it can be stated and accumulated piece by piece, and public because it is lodged in the classics and in the discussion of teachers and friends, so that another person can point out, on specified grounds, a transgression that the agent’s present heart-mind will not own. These features belong to a check in relation to the judgment it assesses, and they admit of degree. Zhu Xi’s Pattern is determinate without being numerical, while the late Ming ledgers of merit assigned numbers to deeds. The criterion of adjudicative externality is thus drawn from a Confucian account of correction and does not need to be imported from elsewhere.

Wang Yangming directed his students to resources beyond the self. A person, he said, “must be polished in affairs” if he is to stand firm (人須在事上磨，方立得住), and even a thief, when someone calls him a thief, still blushes (喚他作賊，他還忸怩; Y. Wang 2011, Chuanxilu, pts. 1 and 3). The Daxuewen opens with the great person who “forms one body with heaven, earth, and the myriad things” (大人者，以天地萬物為一體者也; Y. Wang 2011, Daxuewen). The difficulty has to be located more exactly, by distinguishing the senses in which a check can be “external.”

(a)

Normative ground. The ultimate basis for judging good and bad, located either within the heart-mind or beyond it.

(b)

Epistemic activity. The classics, teachers, and engagement in affairs through which a moral judgment is reached.

(c)

Causal occasion. The force, coming from others or from events, that rouses a person to act.

(d)

Adjudicative authority. The standing of a check that does not depend on the claimant’s present self-assessment, that is, adjudicative externality.

The argument here concerns (d), and its distinction from (a) is essential. To locate the normative ground within the heart-mind is not yet to settle whether others can challenge a particular person’s judgment with an authority that he need not first acknowledge. Chan shows that the two can come apart. It located awakening in the practitioner’s own mind and still required him to “rely on a teacher’s examination and seal.” A teacher who can hold a student to a shared account of conduct, or a covenant whose judgments stand despite a member’s disagreement, can have adjudicative externality within Wang’s teaching. The question is whether that standing survives when the student claims to have attained liangzhi already.

Wang’s own Nangan community covenant gives the distinction concrete content. It remarks that people see others’ faults while remaining obscure about their own, appoints officers, and keeps three registers, including separate records of commendation and correction. At the assembly, reports of conduct are put before the members for further testimony, and persistent wrongdoing can be referred to the authorities. The covenant also invokes divine punishment for outward goodness joined to hidden evil (Y. Wang 2011, juan 17, “Nangan xiangyue”). Here are shared standards, records, other judges, and consequences that do not await a member’s private agreement. The covenant shows that an inward ground of morality did not make an external check impossible in Wang’s teaching. Its scope, however, is conduct, and it does not decide whether anyone has realized liangzhi.

The difficulty is therefore structural wherever self-certification is made final, and the Nangan covenant shows that it did not mark every practice associated with Wang’s teaching. A tradition can join an inward ground of moral authority to external procedures of assessment, as the Chan seal shows. It cannot make every challenge depend on the claimant’s present ratification and still treat the challenge as an independent test of that ratification. The problem becomes acute where a doctrine of attained spontaneity makes further correction appear unnecessary, and the late Ming supplied such a doctrine.

3. Self-Certification in Late Ming Wang Learning

A structural difficulty of this kind should leave traces in the tradition’s own discussions. At Tianquan Bridge in 1527, Wang Yangming warned his students against an emptiness that could pass for awakening. Wang Ji taught that the discipline of extending liangzhi was “set up for those who have not yet awakened.” The late Ming–early Qing historian Huang Zongxi 黃宗羲 (1610–1695) judged that Wang Ji’s teaching “cannot but approach Chan,” and his contemporary, the classical scholar Gu Yanwu 顧炎武 (1613–1682), described scholars who had fallen into Chan “without themselves knowing.” Late Ming polemic had already given the counterfeit a name, kuangchan. A teacher’s warning, a doctrine, two scholars’ verdicts, and a polemical label together show that the problem of distinguishing attainment from a claim to it was argued over inside the tradition.

Huang Zongxi’s Records of Ming Scholars (Mingru xue’an 明儒學案), still the organizing framework for the period, places Wang Ji, known as Longxi 龍溪, in the Zhezhong 浙中 branch of Wang’s school (juan 12) and not among the Taizhou masters, who descend from the salt merchant Wang Gen 王艮 (1483–1541). Huang classified them separately, although his preface to the Taizhou records names them together as the channel through which Wang’s teaching both spread and was lost (陽明先生之學，有泰州、龍溪而風行天下，亦因泰州、龍溪而漸失其傳; Z. Huang 2008, p. 703). Selections are available in English in T.-H. Huang (1987), although translations here are my own, and citations give the pages of the 2008 revised edition. Taizhou itself, Johanna Lidén argues, “did not form a ‘school’ in the strict sense of the word but became a ‘movement’” (Lidén 2018, abstract). Li Zhi 李贄 (1527–1602), the most controversial thinker of the late Ming, whose books were condemned and who died in prison, drew on Taizhou ideas, but Lidén describes him as “staying outside their movement,” and his writings cannot serve as evidence of what a lineage taught (Lidén 2018, pp. 18–19). Claims about doctrine therefore attach to named figures and cited texts. A collection of scandals would not establish a structural claim. Such a claim rests on what a teaching implies, as when Wang Ji held that extending liangzhi was “set up for those who have not yet awakened.” A critic such as Huang Zongxi, who argued that such a teaching cannot but drift, shows that the tradition itself saw the difficulty.

3.1. Tianquan and the Ready-Made Liangzhi

Wang Yangming himself is the most valuable witness. In the ninth month of 1527, on the eve of his departure for the Guangxi campaign, Qian Dehong 錢德洪 (1496–1574) and Wang Ji brought him a disagreement. Wang Ji held that if the substance of the heart-mind is without good and evil, then concern, knowing, and thing must equally be without good and evil, the doctrine later called the “four negations” (si wu 四無). Qian held to the four-sentence teaching (siju jiao 四句教), which begins “without good and without evil is the substance of the heart-mind” (無善無惡心之體), and argued that doing good and removing evil at the level of concerns is the very work of recovering the original substance. At Tianquan Bridge, Wang affirmed both positions as teaching suited to different capacities:

我這裏接人，原有此二種，利根之人，直從本原上悟入……其次不免有習心在，本體受蔽，故且教在意念上實落為善、去惡……汝中之見，是我這裏接利根人的；德洪之見，是我這裏為其次立法的。

In receiving people here I have always had these two kinds. A person of sharp roots enters directly by awakening at the source… Those next below inevitably have habituated mind-states, and the original substance is obscured; so for the time being I teach them to work concretely on concerns, doing good and removing evil… Ruzhong’s [Wang Ji’s] view is what I use to receive people of sharp roots; Dehong’s view is what I use to set a method for those next below.

(Y. Wang 2011, Chuanxilu, pt. 3, Tianquan Bridge dialogue, recorded by Qian Dehong; my translation, cf. W.-T. Chan 1963)

The passage speaks of obscuration, and it gives the teacher the task of sorting his students. The original substance is “subject to obscuration” (受蔽), and obscuration is what makes gradual cultivation necessary. The judgment of a student’s capacity belongs to the teacher, who makes it from outside the student. This is adjudicative externality exercised informally by a living master.

The warning followed, as Wang’s parting instruction:

已後與朋友講學，切不可失了我的宗旨……利根之人，世亦難遇，本體功夫一悟盡透，此顏子、明道所不敢承當，豈可輕易望人。人有習心，不教他在良知上實用為善、去惡功夫，只去懸空想個本體，一切事為俱不著實，不過養成一個虛寂，此個病痛不是小小，不可不早說破。

Hereafter, in discussing learning with friends, you must on no account lose my fundamental purport… A person of sharp roots is rarely met with in the world. To penetrate the original substance and effort completely in a single awakening is what Yanzi and Mingdao did not dare claim; how can it lightly be expected of others? People have habituated mind-states. If you do not teach them to apply real effort on liangzhi in doing good and removing evil, but only to conjure up the original substance in the void, then all their affairs and undertakings will be unreal, and they will merely cultivate an empty quiescence. This malady is no small one, and must be pointed out early.

(Y. Wang 2011, Chuanxilu, pt. 3, Tianquan Bridge dialogue, recorded by Qian Dehong; my translation, cf. W.-T. Chan 1963)

The warning does not repudiate the four negations, and Wang declined to do so, telling the two disciples that their views “should be used to complement each other” (正好相資為用). It identifies the failure with precision. Wang has in view a person who, untaught in the work of doing good and removing evil, only “conjures up the original substance in the void,” whose affairs become unreal, and who ends by cultivating an empty quiescence. He charges his disciples, as future teachers, to point this malady out early (說破). The charge assumes that a teacher can see what the sufferer cannot. Wang died in 1529. His followers taught on in many provinces, and Huang Zongxi later classified them by region. No single successor inherited the authority to correct a student’s mistaken awakening, and the question became what happens wherever a claimed awakening is allowed to exempt its bearer from the corrective work that teacher and friends would otherwise require.

Wang Ji’s mature teaching gave doctrinal form to such an exemption for the awakened. Huang Zongxi summarizes it to this effect:

良知原是無中生有，即是未發之中……當下現成，不假工夫修整而後得。致良知原為未悟者設，信得良知過時，獨往獨來，如珠之走盤，不待拘管而自不過其則也。

Liangzhi is originally something arising from nothing, which is just the equilibrium prior to arousal… It is ready-made in the immediate present, and is not obtained by first repairing it with effort. Extending liangzhi was set up for those who have not yet awakened. When one has come to trust liangzhi through and through, one goes and comes alone, like a pearl rolling in a dish, which without waiting to be restrained does not of itself exceed its bounds.

(Z. Huang 2008, p. 238)

The image of the pearl gives the doctrine in miniature. A pearl in a dish needs no added restraint because its motion already stays within bounds, and one who trusts liangzhi through and through is said to act without corrective management. The sentence 致良知原為未悟者設 makes the distinction plain. The discipline of extending liangzhi is a provision for the unawakened, and the awakened no longer need it in that form. Yuanwu Keqin granted an awakening that needs no tempering only to the primordial age before the first buddha. Here the same freedom belongs to whoever has come to trust liangzhi through and through. The passage does not say that the pearl has no bounds, since the next clause says that it does not exceed them, and it does not say that everyone who announces his awakening has achieved it. The difficulty lies between these claims, in deciding who establishes that a given person belongs to the class whose spontaneity can be trusted. If the claimant’s own trust in liangzhi is allowed to settle that, a challenge to his conduct can be answered by invoking the very status the challenge puts in doubt. The passage describes the attainment with great force and offers no way of adjudicating a disputed claim to possess it.

Huang records a contemporary judgment on Wang Ji’s self-trust. He reports the words of Tang Shunzhi 唐順之 (1507–1560; Jingchuan 荊川) and draws his own conclusion from them:

唐荊川謂先生篤於自信，不為形迹之防，包荒為大，無淨穢之擇，故世之議先生者不一而足。

Tang Jingchuan said that the master was firm in self-trust and took no guard in respect of outward traces, that he treated all-embracing tolerance as greatness and made no discrimination between pure and defiled; hence the criticisms of the master in his age were more than one could enumerate.

(Z. Huang 2008, pp. 238–39)

“Firm in self-trust” (篤於自信) and “taking no guard in respect of outward traces” (不為形迹之防) are Tang’s characterization, and the conclusion that criticism of Wang Ji was innumerable is Huang’s. Tang describes a disregard for appearances and states no doctrine that conduct is inadmissible as evidence. Wang Ji himself treated conduct as evidence. In a self-accusation he wrote that heart and traces are never divided, so that where the traces give cause for suspicion the heart still has places that cannot be fully trusted (心迹未嘗判，迹有可疑，畢竟其心尚有不能盡信處; Z. Huang 2008, p. 251). The admission is his own, and it leaves to him the judgment of what the suspicion shows. Outward traces are among the things an interlocutor can examine, and a cultivated indifference to them makes criticism easier to dismiss. Whether that indifference expresses freedom from conventional display or an exemption from accountability cannot be settled by the phrase alone. It is the distinction a practice of discernment would have to make.

3.2. The Critics and the Name of the Counterfeit

Huang Zongxi’s editorial verdict on Wang Ji is the most important retrospective judgment on the problem:

夫良知既為知覺之流行，不落方所，不可典要，一著工夫，則未免有礙虛無之體，是不得不近於禪。流行即是主宰，懸崖撒手，茫無把柄，以心息相依為權法，是不得不近於老。雖云真性流行，自見天則，而於儒者之矩矱，未免有出入矣。

Since liangzhi is taken as the flowing of awareness, not lodged in any place and not reducible to a constant rule, then the moment one applies effort, one cannot avoid obstructing the empty substance; hence it cannot but approach Chan Buddhism. Since the flowing is itself the ruler, one lets go of the cliff-edge with nothing to grasp, and takes the interdependence of mind and breath as an expedient; hence it cannot but approach Daoism. Though it is said that the true nature flows and of itself sees the norm of Heaven, with respect to the measuring square of the Confucians there is inevitably some departure.

(Z. Huang 2008, p. 239)

The modal force of “cannot but” (不得不) carries the judgment. Huang asks what follows from the teaching’s own claims. If liangzhi is a flowing awareness that admits no fixed rule (不可典要), deliberate effort threatens to obstruct the empty substance, and the means of correction appear to undermine what they are meant to cultivate. His verdict is a historical criticism of one teacher’s doctrine, and its value as evidence lies in where it places the difficulty, inside the teaching itself.

Huang’s closing measure, “the measuring square of the Confucians” (儒者之矩矱), answers the “cliff-edge with nothing to grasp” (懸崖撒手，茫無把柄) of the preceding clause, a Chan phrase that Huang turns against a Confucian. A measuring square exists before the work, it is determinate, and anyone who holds one can check the work against it.

Huang’s censure is not wholesale. He acknowledges that Wang Ji “received Yangming’s final instructions in person” (親承陽明末命), and he sets him beside an earlier disciple. After Lu Jiuyuan 陸九淵 (1139–1193; Xiangshan 象山) there could not but be a Yang Jian 楊簡 (1141–1226; Cihu 慈湖), and after Wang Yangming a Wang Ji, and the rise and fall of their learning turned on these men. Yet where Yang Jian breached Lu’s dykes, Wang Ji dredged the channel and “brought much of Wang Yangming’s learning to light” (象山之後不能無慈湖，文成之後不能無龍溪……慈湖決象山之瀾，而先生疏河導源，於文成之學，固多所發明也; Z. Huang 2008, p. 239). Eric Nelson has read Wang Ji, alongside Wang Yangming, as articulating an existential Confucianism of self-awareness and nothingness (Nelson 2024). That reading brings out what is philosophically serious in the four negations, and it sharpens the question of how a claim to realization prior to determinate good and evil is to be assessed when the claimant’s conduct is disputed. Lidén, for her part, finds Huang’s parallel charge that Wang Gen added Chan to Wang Yangming’s philosophy reasonable, and adds that the problem with the charge is that “it pertains also to Wang Yangming himself” (Lidén 2018, abstract). An affinity with Chan and the absence of a workable check are, however, different claims, and the Chan seal itself shows that a language of immediacy can coexist with elaborate authorization.

The seventeenth-century reaction against late Ming learning is often summarized as the charge that Wang learning caused the fall of the Ming. The texts are more specific than the summary, and more useful. Gu Yanwu’s entry in the Rizhilu 日知錄 under the heading “The Master’s words on nature and the Way of Heaven” (夫子之言性與天道) reads in part:

今之君子學未及乎樊遲、司馬牛，而欲其說之高於顏、曾二子，是以終日言性與天道，而不自知其墮於禪學也。

The gentlemen of today, whose learning has not reached that of Fan Chi or Sima Niu, wish their doctrines to stand higher than those of Yan and Zeng; and so they talk all day of nature and the Way of Heaven, without themselves knowing that they have fallen into Chan Buddhist learning.

(Gu 2006, p. 401; my translation)

The operative phrase is “without themselves knowing” (不自知). Gu charges his contemporaries with having fallen into Chan without knowing it, which presents error as invisible to the party committing it. The charge recalls, without repeating it, Ouyang De’s fear that one might take the thief for one’s son, “deluded without knowing it” (惑而不知也). Gu carries a related anxiety to the level of a learned culture. Whether his targets understood themselves as badly as he says is a further historical question, since the passage gives his diagnosis and no direct access to their state of mind. Gu then continues with the passage most often quoted:

昔之清談談老莊，今之清談談孔孟，未得其精而已遺其粗，未究其本而先辭其末。不習六藝之文，不考百王之典，不綜當代之務，舉夫子論學論政之大端一切不問，而曰「一貫」，曰「無言」，以明心見性之空言，代修己治人之實學。股肱惰而萬事荒，爪牙亡而四國亂，神州蕩覆，宗社丘墟。

The pure conversation of former times discoursed on Laozi and Zhuangzi; the pure conversation of today discourses on Confucius and Mencius. Not having grasped the refined, they have already discarded the coarse; not having examined the root, they first take leave of the branch. They do not study the texts of the six arts, do not examine the institutions of the hundred kings, do not comprehend the affairs of the present age; of the great heads of the Master’s discussions of learning and of government they ask nothing at all, but say “one thread” and say “without words.” With empty talk of illuminating the mind and seeing the nature they have replaced the substantial learning of cultivating oneself and governing others. Limbs grew idle and the myriad affairs ran to waste; claws and fangs were lost and the four quarters fell into disorder; the divine land was overturned and the ancestral altars became a wasteland.

(Gu 2006, p. 402; my translation)

The rhetorical force of the passage is unmistakable, and Gu does connect a style of learning to dynastic ruin. In this entry he names no one. His targets are “the gentlemen of today” (今之君子) and “the pure conversation of today” (今之清談), and the passage is governed by an explicit analogy to Wei–Jin pure conversation (qingtan 清談). A few lines earlier Gu observes that the disorder of the Five Barbarians had its root in the calamity flowing from qingtan, a thing everyone knows (五胡亂華，本於清談之流禍，人人知之), and he closes with Wang Yan 王衍 (256–311), the Jin minister who excelled at abstruse discourse and, facing execution, regretted not having labored to save the realm (Gu 2006, p. 402). Elsewhere Gu names the man. Among those who single-handedly changed the fashion of the world for more than a century, he lists “the pure conversation of Wang Yifu, the new doctrines of Wang Jiefu, and in our own day the liangzhi of Wang Bo’an” (王夷甫之清談，王介甫之新說，其在於今，則王伯安之良知是也; Gu 2006, p. 1068), that is, of Wang Yan, Wang Anshi 王安石 (1021–1086), and Wang Yangming. The verdict is a historical judgment on a teacher’s influence, delivered by a polemicist who had survived the conquest, and it rests on the same analogy with the Jin.

The entry on nature and the Way of Heaven establishes something narrower than the usual summary. A formidable critic writing from inside the tradition described a style of learning as unselfconscious drift, and joined that diagnosis to the charge that concrete study and public affairs had been abandoned. He objected both to what the learned had become and to their failure to recognize it. The second objection bears on discernment.

The late Ming had a name for the condition these critics describe, kuangchan, “wild Chan.” The first half of the word has a good Confucian pedigree. Confucius said that if he could not find companions who walked the middle way, he would settle for the kuang 狂, who “press forward” (Analects 13.21). Mencius ranked them far above the village worthy, though their conduct, when examined, did not live up to their words (Mencius 7B37; Zhu 1983). Wang Yangming claimed this disposition for himself. In 1523, a question in the metropolitan examination attacked his teaching without naming him and asked whether such books should not be burned and banned (Gu 2006, p. 1068). That year, when his disciples discussed why criticism of him kept growing, Wang gave his own answer. He now trusted what liangzhi judged right and wrong and acted on it without concealment. “Only now,” he said, “have I attained the spirit of the kuang; let everyone in the world say that my conduct does not live up to my words” (我今纔做得個狂者的胸次，使天下之人都說我行不揜言也罷; Y. Wang 2011, Chuanxilu, pt. 3, and Nianpu, entry for 1523; cf. W.-T. Chan 1963, pp. 238–39). This is courage against conformity, and its structure is the one examined here, in which trust in liangzhi is set against the verdict of others. In Wang the trust was joined to long cultivation and, as the Tianquan warning shows, to a sharp eye for its counterfeit. Cut loose from these, the same posture is hard to tell from a conviction that reads public disapproval as proof of its own integrity.

Kuangchan is the name critics gave to that posture cut loose, and the critics came from both sides of the line between Confucian and Buddhist. Tang Hezheng 唐鶴徵 (1538–1619), Tang Shunzhi’s son, traced it to a doctrine he blamed on the Chan school, that “a single awakening settles everything once and for all.” Once that doctrine spread, he wrote, those with the slightest insight thought themselves sufficient, so that “Confucians became wild Confucians, and Chan became wild Chan” (謂一悟便一了百當，從此使人未少有見，輒以自足，儒為狂儒，禪為狂禪; Z. Huang 2008, p. 611). Tang’s complaint is aimed at the exemption itself. On the doctrine he attacks, a single awakening ends the need for further effort, and so a modest insight could be presented as a finished attainment.

Yuan Hongdao 袁宏道 (1568–1610), a lay Buddhist and a friend of Li Zhi, made the same diagnosis from inside Buddhism. One kind of Chan, he wrote, is “wild,” for it “makes some chance entry into the original substance and then takes everything as ready-made” (於本體偶有所入，便一切討現成去), and he cites Dahui’s warning that the quick-witted, finding the matter easy, stop cultivating. Yuan grants the glimpse and faults what is claimed for it. The remedy he records came from Li Zhi’s companion the monk Wunian Shenyou 無念深有 (1544–1627), who quoted an older saying that “after awakening one must still meet a person” (悟後須遇人始得). Whoever has done so, the saying continues, will not “blind the eyes” of his students when he comes to teach (不瞎卻學人眼; Yuan 2026, J20, no. B098, p. 524a–b). The saying places the test of an awakening in someone other than the one awakened, and it gives the reason in terms of the authority that awakening confers, since an untested awakening does its harm when its bearer begins to teach.

Critics also attached the word to persons. Huang Zongxi, who gives Li Zhi no entry of his own, records that when Li Zhi “drummed up wild Chan” and scholars followed him “as grass bends before the wind,” Geng Dingxiang 耿定向 (1524–1596) tried earnestly to set them right, but, being himself half drawn to Buddhism, “could never subdue Zhuowu” (先生因李卓吾鼓倡狂禪，學者靡然從風……終無以壓服卓吾; Z. Huang 2008, p. 815). A friend’s correction, offered without a standing that Li Zhi had to acknowledge, could simply be refused. The Qing imperial catalogers later made the word a verdict on a whole age, charging the late followers of Wang Yangming with reading the classics as wild Chan (Yongrong and Ji 1983, juan 1) and counting Wang Ji among those who, “brilliant to excess,” had become “wholly given over to Chan” (高明之過，純涉禪機; juan 93; on the label’s later history, see Lidén 2018, pp. 48–49). The label could also be turned against those who used it. Guan Zhidao 管志道 (1536–1608) declared that what most needed refuting in his day was “wild Confucians spilling over into wild Chan” (狂儒之濫狂禪; Qian 1985, juan 49, “Guan gong xingzhuang”), yet the catalogers, judging by the lineage of his learning, placed him in “the wild Chan faction of the late Ming” (明季狂禪一派; juan 37). Like a disputed seal, the label records a contest and does not settle one.

Taken together, these sources form a connected set of problems and do not report a single causal sequence. Wang Yangming identified a mistaken claim to awakening and exercised the authority to challenge it. Wang Ji distinguished an attainment that no longer needs corrective management from the condition of those who still require effort, leaving open how a contested claim to the former was to be judged. Huang Zongxi traced the difficulty of reconciling that spontaneity with deliberate correction to the teaching itself, and Gu Yanwu described a learned culture that failed to recognize its own departure from substantial learning. The polemic over kuangchan shows the question fought in the open, with Confucian and Buddhist critics naming the same exemption and reaching, as the monk Wunian did, for a test of awakening that the awakened could not supply for themselves.

The critics’ verdicts and the argument of this article are claims of different kinds. The critics judged particular teachings and teachers, often from partisan positions. The argument here is structural. Where corrective effort belongs to the unawakened, a mistaken claim to awakening can also become a claim to exemption from correction, and once that exemption is secured by self-certification, no challenge can test it independently. The criticisms show that the tradition itself perceived this liability and argued about it. The same tradition also produced the Nangan covenant, which kept correction in force.

4. The Taizhou Criterion and Its Limits

The most explicit criterion offered within Wang learning came from Taizhou. Luo Rufang 羅汝芳 (1515–1588), known as Jinxi 近溪, stated it in a passage that Huang Zongxi recorded in the third Taizhou chapter, in reply to an objection raised by a contemporary:

一友自述其平日用工，只在念頭上纏擾……方悟心中靜之與動，明之與暗，皆是想度意見而成……於是一切醒轉，更不去此等去處計較尋覓，卻得本心渾淪，只不合分別，便自無間斷，真是坦然蕩蕩，而悠然順適也。或詰之曰：「如此渾淪，然則善不消為，惡不必去耶？」友不能答。羅子代之答曰：「只患渾淪不到底耳。蓋渾淪順適處，即名為善，而違礙處，便名不善也。故只渾淪到底，即便不善化而為善也，非為善去惡之學而何？」眾皆有省。

A friend described his daily effort, saying it consisted only in getting entangled at the level of thoughts… He then awoke to the realization that stillness and motion within the heart-mind, brightness and darkness, are all constituted by conjecture and opinion… Thereupon he turned about entirely, no longer calculating and searching in such places, and obtained the undivided wholeness of the original heart-mind; simply by not engaging in discrimination it was of itself without interruption, truly open and vast, and easefully unobstructed. Someone challenged him: “If it is thus undivided, then good need not be done and evil need not be removed?” The friend could not answer. Master Luo answered on his behalf: “The only trouble is that the wholeness has not been carried to the bottom. For where it is undivided and flows unobstructed, that is what we call good; and where there is obstruction, that is what we call not good. So if one only carries the wholeness all the way to the bottom, then the not-good is transformed and becomes good; and what is this if not the learning of doing good and removing evil?” All present had a realization.

(Z. Huang 2008, pp. 770–71)

Here the criterion is stated outright. Good is to be “undivided and flowing unobstructed” (hunlun shunshi 渾淪順適), and not-good is “obstructed, snagging” (wei’ai 違礙). Chi-keung Chan reads the passage as identifying the not-good with a departure from the effortless coping of young children (C.-K. Chan 2025, p. 553). Stated in a text and taught to a company, the rule has a publicity that a private sensation lacks. Its application is another matter. Calling a movement of the heart-mind unobstructed is already an interpretation of it, and the passage gives another person no ground for challenging that interpretation that the practitioner must accept. The questioner in the text saw where this leads, since a wholeness that settles everything seems to leave no good to be done and no evil to be removed. Luo’s answer, that the wholeness has not been carried to the bottom, can correct a student who knows his wholeness is incomplete. Against a person who believes he has carried it to the bottom it has no grip, unless something can count as evidence that his own account of his attainment is wrong.

The hard case is again the person in whom selfish desire has come to pass for liangzhi. Long habit makes the desire-serving impulse familiar and fluent, while the surviving scruple is felt as an interruption. Zheng Zemian’s practiced schemer, who deliberates for his own advantage with perfect composure, shows what such a person can look like, and Zheng adds that what habit has made familiar can come to seem “clear and simple” to the one who errs (Zheng 2022, pp. 106–7). If he identifies fluency with attunement, Luo’s criterion is inverted. The corrupt disposition receives the name of good, and the warning against it receives the name of obstruction. Luo saw a related danger. He warned that those who cling to a “bright vision” (guangjing 光景) of the heart-mind take a construct for its substance, and that the harder they strive, the further they stray (C.-K. Chan 2025, p. 546). To a student troubled that his practice kept breaking off, he counseled turning the thought “lightly and quickly” (輕輕快快轉個念頭) and returning to the flow of daily life (C.-K. Chan 2025, p. 559). The counsel assumes that the practitioner can tell the fixation from the warning. A person who has taken his scruple for the fixation will turn away from the scruple, lightly and quickly.

Chi-keung Chan’s reading of Luo through embodied cognition offers the strongest reply. Chi-keung Chan writes against critics from Mou Zongsan to Chen Lai who charge Taizhou with letting natural impulse pass for liangzhi, and he grants that intuition can mislead, as when disgust toward people of another race or gender issues in discrimination (C.-K. Chan 2025, p. 545). His answer is a practice that balances reflection with spontaneity (C.-K. Chan 2025, pp. 556–59). It is embodied, and the body, in Luo’s words as Chi-keung Chan renders them, is “rooted in my parents, connected with my siblings, and accompanied by my spouse” (C.-K. Chan 2025, p. 547). It also lives in a world that answers back, since violence draws the disdain of others and, in grave cases, exclusion from society (C.-K. Chan 2025, p. 554). Luo said as much of the household. Asked how to hold to the center and how to regulate the breath, he dismissed both techniques, and asked how the heart-mind could be made harmonious, he answered, “be harmonious with wife and children, fitting with brothers, compliant with parents, and the heart-mind will be harmonious” (和妻子，宜兄弟，順父母，心斯和矣; Z. Huang 2008, p. 247). Among his recorded answers to Geng Dingxiang, Wang Ji took up the exchange, defended holding to the center and regulating the breath, and judged that resting on harmony of heart, qi, and body, “the commonplace talk of worldly Confucians,” left “no opening for awakening” (世儒常談……無入悟之機; Z. Huang 2008, p. 247). Luo was not Wang Ji, and his cultivation answered to the household. The household can still be read two ways. A parent’s opposition may expose a son’s selfishness, and it may be the price of doing right, as when Shun married without telling his parents and Mencius defended him (Mencius 5A2). Receiving an answer from the world and accepting another person’s authority to interpret it are different achievements, and Chi-keung Chan’s account secures the first. Where a tradition gives the second a determinate procedure, it has acquired the corrective resource called here adjudicative externality, and the achievement is its own.

An inner criterion does not make a person’s application of it self-warranting. The ledger practices of the late Ming show what gave that distinction practical force, in stated grounds of correction, records that could be revisited, and persons whose adverse judgment had consequences, all in a setting where the authority of the heart-mind was retained.

5. Ledgers of Merit: Recording, Supervision, and Adjudication

The earliest surviving ledger of merit and demerit dates from the Song. The Tract of Taishang on Action and Response (Taishang ganying pian 太上感應篇, c. 1164) set out the workings of supernatural retribution, and the Ledger of Merit and Demerit of the Taiwei Immortal (Taiwei xianjun gongguoge 太微仙君功過格, 1171), a handbook of a Daoist sect, was the first text to assign points to good and bad deeds, so that a person could keep score of his own merit. By the late Ming the two texts had together become, in Cynthia Brokaw’s phrase, the “bible” of the system (Brokaw 1991, pp. 26, 35).

Members of the Wang school took up ledgers of merit and demerit (gongguoge 功過格), and some explained why a doctrine of spontaneous goodness could be practiced alongside a written schedule of conduct. Their reception within these circles brings moral record keeping into the same intellectual world as Wang Ji’s ready-made liangzhi (Brokaw 1991, pp. 110, 113–20). The practitioners themselves felt the difficulty of judging their own claims. Wang Fan-sen describes their predicament as that of “a madman curing his own madness” (狂人自醫其狂), and shows how some of Wang Yangming’s heirs turned to records and to societies that could play the accuser (F.-S. Wang 1993, pp. 679–80, 701–8). Wang Fan-sen calls these gatherings societies for the examination of faults (xingguohui 省過會), and the language in which they worked, of examining and amending faults (省過, 改過), was the tradition’s own.

Ledger keeping in these circles took the form of private accounting, communal supervision, and adjudication, and each did something different. In private accounting, as Yuan Huang 袁黃 (1533–1606) practiced it, a person recorded his own deeds under a religious sanction. In communal supervision, as in the fellowship of Zhou Rudeng 周汝登 (1547–1629), members reviewed one another’s records. In adjudication, as in the society of Yan Maoyou 顏茂猷 (d. 1637), a body with rules could investigate an account and impose consequences that did not wait for the member’s agreement. None of these arrangements authenticated a claim to moral realization. Each recorded or judged conduct. Their bearing on discernment is therefore indirect, and it lies in what they did with a claim to attainment.

Private accounting is the best documented. Brokaw’s study gives the practice its historical setting and describes how the records were kept. Most, though not all, ledgers assigned numerical values to specified deeds and provided a calendar for daily entries and monthly totals (Brokaw 1991, p. 4). Yuan Huang, her principal late Ming example, described recording a deed as soon as he performed it, while his wife marked her own contributions with vermilion circles. In office he kept a book at the place where he heard cases, and in the evening reported his conduct to heaven. This was at once a discipline of recollection, a household practice, and a religious undertaking, accompanied by vows, confession, and consultation with Buddhist monks (Brokaw 1991, pp. 82–88). Yuan studied with Wang Ji and Luo Rufang and converted two of Luo’s disciples, Zhou Rudeng and Yang Qiyuan 楊起元 (1547–1599), to ledger use (Brokaw 1991, p. 113). Luo himself had kept a ledger in his youth and later destroyed it, for fear that it would turn his thoughts to profit (Brokaw 1991, pp. 62–63). The ledger was thus a practice argued over within Wang learning itself.

A ledger gave conduct an account that did not depend on the keeper’s mood of the moment. Its schedule described in advance which deeds counted, even though applying the description still required judgment (Brokaw 1991, pp. 3, 106). Listed deeds and assigned values made some disagreements specific enough to pursue, although intention, context, and consequences could alter an act’s worth, and Yuan expected the user to exercise a demanding moral judgment in reading the numbers (Brokaw 1991, pp. 98–102, 106). A circulating schedule made the descriptions of conduct public, while the entries stayed with their keeper. The ledger’s most important material feature was the persistence of the record. A deed entered yesterday could be returned to next month, compared with other entries, or shown to another reader. Paper does not make judgment unnecessary. It makes certain judgments recoverable.

The religious premise gave the private account a limit. In Yuan Huang’s practice someone else was also keeping the account, and the other account was the one that finally settled the matter, since only heaven and the spirits fully know the purity of a person’s intentions (Brokaw 1991, p. 108). The Buddhist master Yunqi Zhuhong 雲棲袾宏 (1535–1615), who compiled a ledger of his own, the Zizhi lu 自知錄, whose preface is dated 1604 (Yunqi Zhuhong 2026, J32, no. B277, p. 681b), put the point in one image. Even if he does not write it down, he says, the celestial scribes “have already recorded it, finer than silk thread and the tip of an autumn hair” (政使我不書，彼之書固已密繭絲而析秋毫矣; Yunqi Zhuhong 2026, J33, no. B277, p. 92b–c). Vincent Goossaert’s study of the ritual foundations of morality books locates this appeal in a wider religious culture, in which divine codes revealed to ritual adepts, and later morality books produced through spirit-writing, carried a divine warrant that their human recipients did not claim to have created (Goossaert 2020). The ledger Yuan received from the Chan master Yungu Fahui 雲谷法會 (1500–1575) claimed no divine authorship and played down the role of the gods (Brokaw 1991, p. 92), and Brokaw calls the system a “Confucianized” version of one “originally associated principally with Buddhism and Daoism” (Brokaw 1991, p. 107). A practitioner who believes that heaven knows his motives has a reason to distrust a favorable entry of his own. The belief supplies a sanction for scrutiny. It does not make the divine verdict publicly inspectable, and it gives no human party authority over the account.

Private accounting does not cure the person whose selfish desire has come to pass for liangzhi. He still selects what to record, interprets what he has done, and may mistake the motives with which he did it. Brokaw explains that Yuan Huang’s judgments could depart from conventional moral opinion, and that his practice demanded a sensitive interpretation of conduct from within the heart-mind (Brokaw 1991, pp. 101–2). Those who accepted the ledger while holding a doctrine of spontaneous goodness had to explain the combination. Zhou Rudeng had opposed using moral rules to guide conduct, but after a night-long conversation with Yuan Huang he began keeping a ledger of merit and demerit of his own. In a preface he contributed to the 1607 edition of Yuan’s work, he argued that a mind which does not grasp goodness need not be one from whose actions goodness is absent. Grains of rice exist separately even when the person eating does not attend to each grain, and particular acts can likewise be distinguished without making goodness a collection of consciously sought merits (Brokaw 1991, pp. 117–19). A defense of ledger keeping in the Gongguoge lun 功過格論 argued that without stated measures there is no reference for judging excess and insufficiency, and described nightly inscription under the gaze of the spirits (Brokaw 1991, p. 120). The manuscript Brokaw used names Tao Wangling 陶望齡 (1562–1609) as its author, while Sakai Tadao attributes it to his brother Tao Shiling 陶奭齡 (Brokaw 1991, p. 120 n. 25). Within these circles the need for a measure that could be brought to bear on a performance was recognized, while the moral value of the performance still depended on the state of the heart-mind.

Communal supervision added readers. Zhou Rudeng’s teaching left room for it. Replying to the objections of Xu Fuyuan 許孚遠 (1535–1604), he allowed that those who hide their wrongdoing should be taught to do good and remove evil “so that they have something to hold to” (使有所持循; Z. Huang 2008, p. 864). Before he took up Yuan’s ledger, he and seven companions had kept a register of their faults and met monthly to review it. By Zhou’s own account the meetings ran for twenty years, and by 1589 the first register was full and a second begun (F.-S. Wang 1993, pp. 703–4). In such a fellowship a member’s account of his conduct was no longer his alone to read. Other members could hold the account up against him, and accepting correction became a condition of belonging. In the society he founded, Lü Weiqi 呂維祺 (1587–1641) kept separate registers for moving toward the good and for amending faults, and gave as his reason the very danger Ouyang De had named. Without friends to admonish one’s faults, he warned, one runs the risk of taking the thief for one’s son (F.-S. Wang 1993, p. 704). The evidence shows peers reviewing recorded faults. It shows no member ruling on another’s attainment.

Adjudication appears most clearly in the Yunqi She 雲起社, which Brokaw calls the Yunqi Hui, a society organized by Yan Maoyou, who declared himself a follower of Wang Yangming (Brokaw 1991, p. 162). Its rules required each member to keep a register of merits and faults and to produce it at the monthly assembly “for mutual examination,” and in the circle that aimed at sagehood, conduct at home and toward others was “checked item by item” (會日出之，以相儆察……須一一對勘). Monitors (siju 司糾) were appointed to “examine the idle and assign penalties” (糾察怠惰量罰), and a member who refused to keep his register, or who relapsed into grave wrongdoing, was expelled (Z. Wu 2010, pp. 70–72). Brokaw reads the rules as also requiring a completed ledger to be investigated before admission (Brokaw 1991, pp. 222–23), while Wu Zhen finds the conditions of entry fairly lenient (Z. Wu 2010, p. 71). Here the other members could investigate an account and act on their judgment of it, and a declaration of inward certainty did not exhaust the conditions of membership. These provisions come from the society’s own rules, and how fully they were carried out can no longer be determined (Z. Wu 2010, p. 72). Zhou Rudeng’s register is known from a member’s own account of twenty years of meetings, while the Yunqi She is known from what its rules authorized its officers to do. Wang Yangming’s Nangan covenant, with its officers, registers, and referral of persistent wrongdoing to the authorities, belongs with this form.

This is the form of externality that record keeping can support. Its force comes from the conjunction of a stated expectation, a surviving record, and another party entitled to inspect it. A member’s account of what he has done can be questioned at a specified point, and the question can have consequences. A society can misjudge and a member can fail to reform, and a fallible capacity for correction is still a real addition to his situation.

None of these arrangements decided who had attained the Way, and the omission was deliberate. The question had been put to Zhou Rudeng directly. Writing against Wang Ji’s account of the Tianquan exchange, Xu Fuyuan complained that Wang Ji talked of the four negations incessantly and every clever scholar could repeat them, “yet who has heard the Way, no one knows” (然而聞道者，竟不知為誰氏). The complaint amounts to a demand for discernment, a way to tell the attainment the doctrine described from the talk it had generated. In his reply Zhou turned the question back on each person and closed it to everyone else, holding that “whether one has heard the Way, each must answer for himself; one may not doubt others” (聞道與否，各宜責歸自己，未可疑人; Z. Huang 2008, p. 867), and that one who aspires to the Way and “trusts himself sufficiently” may well “set his claims high and take on much” (一時自信得及，誠不妨立論之高，承當之大; Z. Huang 2008, p. 867). The warrant for a high claim is thus the claimant’s own trust, the self-trust (自信) that Tang Shunzhi had seen in Wang Ji. On attainment Zhou left no one but the claimant entitled to judge, the condition under which discernment becomes a structural difficulty. On conduct he did the opposite, keeping a register of faults that his companions reviewed every month. The societies thus divided the problem. A claim to attainment remained the claimant’s own, while his acts were entered and examined like everyone else’s, and no attainment exempted him from that examination.

Table 1 brings the distinctions together. It separates claims about moral capacity from arrangements for examining conduct, since a doctrine and a procedure cannot fairly be measured as though they were the same kind of thing, and it gives private accounting, communal supervision, and adjudication separate rows.

Table 1. Adjudicative externality: sources of correction and their limits.

Standard or Practice	Prior Standing	Determinate Grounds	Public Access	Record of Conduct	Independent Check
Zhu Xi’s Pattern (li 理)	Classics and inquiry	Articulated Pattern; judgment required	Teaching and discussion	Not entailed by theory	Shared grounds; procedures depend on institutions
Wang Yangming’s liangzhi 良知	Teachers and shared norms	Judgment in practical affairs	Teaching and visible conduct	Not entailed by theory	Teachers can challenge claims of attainment
Nangan covenant	Established offices and rules	Specified conduct and procedures	Reports tested in assembly	Registers of conduct, merits and faults	Officers, correction and referral to authorities
Wang Ji’s ready-made liangzhi	Original goodness affirmed	Self-regulation asserted	Conduct visible; self-trust invoked	Not specified in the passage	Certification of disputed attainment unresolved
Luo Rufang’s unobstructed attunement	Stated character of goodness	Embodied and relational judgment	Family, teachers and other responses	Not specified by the criterion	Disputed attunement still requires adjudication
Private ledger (Yuan Huang)	Schedules and prescriptions	Listed deeds and values; interpretation remains	Schedules circulate; entries kept by the user	Dated entries	Heaven’s account; no human reader with authority over it
Fellowship register (Zhou Rudeng)	Agreement among eight members	Recorded faults	Reviewed at monthly meetings	Register of faults	Peer review of conduct; attainment left to each
Society review (Yunqi She)	Society rules and offices	Ledgers and group records checked	Leaders report to the society	Members’ registers of merits and faults	Monthly review, monitors, penalties, expulsion

The columns describe available resources, not guarantees of correct judgment. “Prior” concerns a check’s standing relative to a disputed verdict; “determinate” concerns grounds for a specified challenge; “public” concerns access within the relevant community. For the covenant, see Y. Wang (2011, juan 17, “Nangan xiangyue”); for the private ledger, Brokaw (1991, pp. 82–108); for Zhou Rudeng’s register, F.-S. Wang (1993, pp. 703–4), and for his views, Z. Huang (2008, pp. 864, 867); for the Yunqi She, Z. Wu (2010, pp. 70–72) and Brokaw (1991, pp. 222–23).

Wang Yangming’s heirs took up a practice the Song had produced, debated its compatibility with inner moral authority, and gave record keeping a place in organized moral life. How widely ledgers were used is harder to say, and Brokaw warns that wide publication is not proof of widespread use among the unlettered population (Brokaw 1991, pp. 225–26). The documented cases are enough to show the demand for a check taking practical form within the same world that had pressed the authority of the heart-mind so far.

The arrangement had a cost, and the contemporary criticism was more searching than an objection to mercenary virtue. Liu Zongzhou 劉宗周 (1578–1645) worried that merits used to cancel demerits would conceal the faults that needed correction, that a printed list might omit a practitioner’s particular errors, and that writing down a score could be mistaken for having changed one’s conduct (Brokaw 1991, p. 124). An instrument introduced to expose self-misrecognition can become a new way of sustaining it, especially if the practitioner treats a satisfactory account as a certificate of an accomplished moral self. Private accounting, in other words, can turn into self-certification on paper.

Liu’s response altered the practice. In the Renpu 人譜 he retained the recording of transgressions, removed the positive merits and their numerical exchange, and joined the record to an examination of the beginnings of error in the heart-mind, so that the account could not cover the faults it was meant to disclose (Brokaw 1991, pp. 128–36). His preface, written as a rejoinder to Yuan Huang, gives the reason in a phrase. It “speaks of faults, not of merits, to keep far from profit” (言過不言功，以遠利也; Liu Zongzhou, “Renpu zixu” 人譜自序, quoted in Brokaw 1991, p. 129; my translation). A tally of merits lends itself to being read as a certificate, and a record of faults much less so. This was an argument over the design of moral scrutiny, conducted by thinkers who continued to take inward cultivation seriously.

The ledgers thus show a practical recognition that confidence in moral capacity leaves work to be done in assessing its exercise. That work was distributed across a written record, a religious sanction, and the judgment of others, and each arrangement could itself require correction, as Liu Zongzhou’s revision of the ledger shows. Once the distinction between genuine authority and a claim made in its name became socially consequential, the tradition had to decide who could question the claim and on what grounds. For conduct, the late Ming societies supplied an answer. For attainment they supplied none, and the problem of discernment remained open.

6. Conclusions

Wang Yangming held that liangzhi knows every concern and that a person may nonetheless obscure his own knowing of the good. Studies of this puzzle have explained how genuine moral knowing works. This article has pursued a further question, which grows out of the one Ouyang De put to his teacher and concerns how a tradition assesses the person who takes himself to possess that knowing. The question is a problem of discernment, the authentication of a claim to moral realization, and it is distinct from both moral self-knowledge and the assessment of conduct. Comparison with the Chan seal shows what is at stake. The seal was no infallible procedure, but it refused to let attainment exempt its bearer from examination.

Adjudicative externality names the standing of a check that does not depend on the claimant’s present self-assessment. A shared standard, a recognized teacher, or a procedure can supply grounds for challenging a person that his present conviction does not settle, and an inward ground of moral authority need not exclude any of them. Wang’s own teaching cannot be reduced to an absence of external correction, as the Tianquan exchange and the Nangan covenant show. The difficulty becomes sharp in claims of attained spontaneity, where the person who says he no longer needs correction can appeal to the very attainment whose presence is disputed. Wang Ji’s ready-made liangzhi supplied a doctrine that such a claim could invoke. Huang Zongxi, Gu Yanwu, and the polemic against kuangchan show that the tradition perceived the liability and argued over it in the open. Luo Rufang’s criterion gives the argument its strongest test, since a stated distinction between good and bad is not yet an answer to the person who misidentifies his own state through that distinction.

The ledgers of Yuan Huang, Zhou Rudeng, and their associates, and the societies that reviewed their members’ records, supplied a historical response of a particular kind. They did not adjudicate attainment, and Zhou Rudeng left the question of attainment to each person. They kept every claimant’s acts on the record, before heaven in the private ledger and before other members in the fellowship and the society. Recording conduct did not authenticate realization, yet it denied the claim to realization an exemption, and in that limited sense it answered the fox’s move. Liu Zongzhou’s criticism shows that the answer carried its own risk, since a satisfactory account could itself become a certificate.

These findings bear on the question of rujia and rujiao without deciding whether Confucianism was a religion. Examined for the functions that religious formations perform, Confucian self-cultivation shows discernment to be one of them, and an unavoidable one wherever a tradition accepts that claims to inward authority can be mistaken. Such a tradition has to say how the genuine state is told from its counterfeit, including when the person who claims the state is certain that he possesses it. Wang learning is a revealing case. Its identification of moral authority with liangzhi gave individual knowing extraordinary standing, and that standing became a liability where a person’s claim to exercise it was allowed to certify itself. In the koan that gives this article its title, the fox’s error was to exempt great cultivation from the order that binds everyone else.

The general claim is conditional, and it reaches beyond Confucianism. When every challenge to a person’s claim of attainment must finally be ratified by the same present self-assessment, the challenge cannot independently test that assessment. Correction remains possible, but its authority is left unexplained at the point where the person rejects it. A master’s seal, a covenant’s officers, and a record of conduct each answer part of that difficulty in different ways. The ledgers show how ordinary practices of recording and revisiting conduct can place a judgment at a distance from the certainty of the moment, and the belief that heaven keeps its own account gives that distance a religious warrant without making the account publicly verifiable.

Further inquiry could extend the institutional question to the Confucian academies (shuyuan 書院) and to the community covenants (xiangyue 鄉約) that spread in the late Ming (F.-S. Wang 1993, p. 703 n. 124), asking when they sustained a check on claims of attainment and how they related that check to liangzhi. The same question could be brought to modern New Confucian accounts of inner transcendence, which have to say who is entitled to tell a person that he has taken the thief for his son.