Dharma · AI Companion

聖言

ārya-vacana

佛陀及圣者宣说的正教AI-generated

Type FrameworkDifficulty IntroductoryTruth-level span outside the four-fold two-truths frameworkInitial AI estimate · evolves with use

A Contemporary ReadingAI-generated

What This Concept Is Actually Saying

The first thing to clear up is the most common misreading. When many people hear the Chinese term 聖言, the word 聖 (*ārya*, "noble" or "sage") carries such strong religious weight that the phrase automatically gets translated in the mind as "the Buddha's words" or "the teachings of enlightened beings." The *Yogācārabhūmi-śāstra* (《瑜伽師地論》, T1579) — the systematic treatise where this technical term gets its sharpest definition — does not use 聖言 in that way at all. In that text, 聖言 is not a label for sacred doctrine. It is a technical category within the analysis of sound and verbal action (*karmic speech*), and its criterion has nothing to do with who is speaking and everything to do with whether what is spoken matches the speaker's own actual experience.

The definition is disarmingly simple: 依自身實際見、聞、覺、知而作的如實言說 — speech that corresponds to one's own actual seeing, hearing, sensing, and knowing. That is the entire backbone of the term. "Sage speech" is not sage because a sage uttered it; it is sage because it accurately reports what is present in the speaker's own awareness.

The text then builds a four-fold classification on this single criterion, and it is important to preserve each of the four legs precisely. The complete term in the treatise is 四聖言聲 ("the four kinds of sage speech / sage utterance"), and each individual instance is referred to simply as 聖言:

- 見則言見,不見則言不見 — regarding *seeing*: if you have seen, say you have seen; if you have not seen, say you have not seen. - 聞則言聞,不聞則言不聞 — regarding *hearing*: if you have heard, say you have heard; if you have not heard, say you have not heard. - 覺則言覺,不覺則言不覺 — regarding *sensing / feeling*: if you have sensed or felt something, say so; if you have not, say so. - 知則言知,不知則言不知 — regarding *knowing*: if you know, say you know; if you do not know, say you do not know.

These four categories correspond to four distinct modes by which a person has access to experience. *Seeing* (見) covers direct perceptual contact — what is visually, evidently before you. *Hearing* (聞) covers what has reached you through testimony, report, or auditory contact with others. *Sensing / feeling* (覺) covers the felt dimension — bodily sensation, affect, the raw "this is how it is for me" register that lives below concepts. *Knowing* (知) covers conceptual cognition — the conclusions, understandings, and judgments you actually hold. The point is not that these four are philosophically airtight epistemologies; the point is that they collectively cover the full range of ways a human being can have any experience at all, and the demand of 聖言 is that whatever you say about your experience be accurate to one of these four registers.

Then comes the strict mirror image, the 四非聖言聲 ("four kinds of non-sage speech"), which is every bit as important as the positive list. Misalignment with one's experience goes in both directions, and the text is careful to enumerate both:

- 不見而說見, 見而說不見 (claim to have seen what you haven't; deny what you have) - 不聞而說聞, 聞而說不聞 (claim to have heard what you haven't; deny what you have) - 不覺而說覺, 覺而說不覺 (claim to feel what you don't; deny what you do) - 不知而說知, 知而說不知 (claim to know what you don't; deny what you do)

This bidirectional structure is the heart of the concept. 聖言 is not just "don't lie"; it is equally "don't hide, don't perform, don't curate." It forbids both the inflation of one's experience (claiming more than is there) and the deflation of it (denying what genuinely is there). Saying "I've read that book" when you haven't is non-sage speech; so is saying "I don't really mind" when in fact you do. The criterion is alignment, period. Direction does not matter; only the gap between word and actual experience matters.

Finally, the treatise draws a careful terminological perimeter around 聖言 that we must not violate. The same passage also mentions 四善語業道 ("fourfold path of wholesome verbal action") and 四不善語業道, but it does not state that the four sage speeches correspond item by item to the four wholesome verbal actions. The text places these groups side by side without equating them, so any tidy mapping ("sage speech = truth-telling = wholesome speech") would be an over-reading of the source. Likewise, the term 賢聖諦語 ("the truthful speech of the noble ones") appears in the text with only a partial context — the phrase "與此相違,當知即是賢聖諦語" tells us it denotes speech opposite to some preceding non-truthful pattern, but does not give us a full independent definition. We can register the term; we cannot redefine 聖言 through it. And, critically, the well-known verse 「諸惡者莫作,諸善者奉行,自調伏其心,是諸佛聖教」 ("Do no evil, do all good, tame your own mind — this is the sage-teaching of all buddhas") uses the term 諸佛聖教 (*sages' sacred instruction*), not 聖言. 諸佛聖教 is the doctrinal content taught by buddhas; 聖言 is a discipline of accurate speech open to any practitioner. These two share the character 聖 but address entirely different things.

Walking Through the Concept in Contemporary Life

The most striking thing about this concept for a modern reader is how it cuts across so many scenes that have nothing to do with Buddhism. Let me walk through each of the four legs — 見, 聞, 覺, 知 — and then through the bidirectional trap, using situations you have probably been in this week.

見 (seeing) — what is directly before you. The contemporary version of "見則言見, 不見則言不見" plays out every time someone asks "Did you actually read that article / watch that show / look at the data?" The social pressure to have consumed the same content as everyone else is enormous. In a group chat, the conversation is about a trending series; everyone is dissecting it. You haven't watched it. The cheap move is to drop a half-line that implies you have. The 聖言 move is to say "I haven't seen it, but I'd like to hear what you thought." This sounds almost embarrassingly simple, but notice how rare it is. The same pattern shows up in workplace contexts: "Have you reviewed the contract?" "Have you seen the report?" — and the temptation to say yes because the alternative feels like admitting incompetence. 聖言 does not call this humility; it calls it accuracy. You are protecting the integrity of what *seeing* means for you and for the people relying on your report.

聞 (hearing) — what has reached you through others. "I heard that the company is laying people off next quarter." Did you actually hear it, or did you hear a cousin of a friend mention it at a dinner? Hearsay has a natural tendency to compact itself: "people are saying" becomes "I heard" becomes "it's basically confirmed." 聖言 on the 聞 side asks you to mark the difference between what reached your ears as actual testimony and what arrived as rumor, gossip, or vague social-media noise. In close relationships this matters even more: "I heard you were upset with me." Did you actually hear that, in those words, from them? Or did you infer it from a tone, a sigh, a three-day gap in texting? Inferences are not hearing. The text is not asking you to suppress inferences — it is asking you to label them accurately when you speak.

覺 (sensing / feeling) — what is happening in your body and affect. This is where the concept does some of its most radical contemporary work. "覺則言覺, 不覺則言不覺" sounds straightforward until you realize how much of contemporary emotional culture trains people to perform or suppress felt experience. Someone asks "How are you?" and the fluent response is "I'm fine." Often, you are not fine — you are tired, anxious, grieving, lonely, restless — but the script offers "fine" as the easy door out. That is the deflation direction of non-sage speech: you have sensed something, and you deny it. The inflation direction is equally common: "I'm so excited about this!" "I'm devastated!" — when the actual felt sense is closer to mild interest or quiet disappointment. Social media, in particular, rewards affective inflation; algorithms register strong feelings. 聖言 asks you to check: is that the actual *覺*, the raw felt-tone of this moment, or is it a performed version of it? The body is often a better lie-detector here than the mind. If your stomach is knotted but you are saying "I'm thrilled," the 覺 register is being lied about.

知 (knowing) — what you actually understand. This is the category of opinions, analyses, and expertise-claims. A podcast host says "What we know is that X causes Y." Do we know that? Or do we suspect it, partially understand it, have a hypothesis about it? 聖言 on the 知 side requires marking the difference between genuine understanding and the social pressure to have an opinion. Twitter, comment sections, and even dinner-table debates reward the confident take. The opposite move — saying "I don't actually know enough about this to have a view" — feels like a kind of social suicide in many settings. And yet the text's claim is that this honest "I don't know" *is* the sage-speech version of 知. Genuinely knowing something and saying you know it is 聖言; not knowing and saying you know is non-sage speech; knowing and pretending not to (perhaps to seem humble, or to evade responsibility) is also non-sage speech.

Now, the bidirectional trap — the part many readers miss. The four 非聖言 patterns are not just "people who lie boldly." They cover two much quieter, much more common failures. The first is over-claiming: the résumé padded, the expertise inflated, the experience narrated into something grander than it was. The second is under-claiming: the quiet self-effacement, the trained "I'm not really that upset," the habit of minimizing your own perception because you have learned that your seeing, hearing, feeling, or knowing does not really "count." Both are violations of the same alignment criterion. Both produce a gap between word and actuality. The concept does not care whether you are inflating or deflating; it cares whether you are aligned.

A useful contemporary image: 聖言 is a kind of calibration between two instruments — your inner register of experience, and the words you send out into the world. The concept is asking you to keep these two instruments tuned. Most of what we call "authenticity" in modern self-help language gestures at this, but 聖言 is more austere than that: it is not about being true to your "authentic self" (which often becomes another performance); it is about the narrower, more verifiable discipline of saying what is actually the case in your experience.

Why Contemporary People Need This

We live inside an information ecosystem that systematically rewards misalignment between word and actual experience. The architecture of social media — metrics, follower counts, engagement — provides ongoing feedback for *performing* experience rather than reporting it. The economy of attention monetizes the gap: the more your words exceed your actual experience, the more engagement they tend to draw. In such an environment, the simple discipline of *見則言見, 不覺則言不覺* becomes counter-cultural almost by default.

There is a deeper layer. Contemporary culture often frames authenticity as an emotional project — "be true to your feelings" — but is comparatively thin on the *epistemic* dimension: be true to what you actually know, have actually seen, have actually heard. The four-fold structure of 聖言 insists on all four simultaneously. You cannot be authentic about your feelings (覺) while being fake about your knowledge (知). You cannot be honest about what you have heard (聞) while inflating what you have seen (見). The concept asks for a kind of total alignment across all the registers of experience, which is something the modern self almost never audits.

There is also a quietly therapeutic dimension. Much of what Buddhist psychology describes as the mechanism of suffering — the proliferation of narrative on top of raw experience — depends on a constant, low-level dishonesty about what is actually being seen, heard, felt, and known. The exaggerated inner story ("she definitely thinks I'm incompetent") is non-sage speech directed at oneself: a 知 claim that outruns the actual 知. The disciplined alternative is not positivity or reinterpretation; it is the much more basic move of saying, to oneself as much as to others, "this is what I actually see, this is what I actually hear, this is what I actually feel, this is what I actually know — and nothing more." From this kind of accurate reporting, wise response becomes possible. Without it, every further move is built on sand.

Finally, there is the communal dimension. 聖言 is a discipline that protects not only the speaker but the listener. When you say "見則言見," you are giving the people around you a trustworthy signal about the state of your experience. When you say "不知則言不知," you are inviting them into accurate collaboration rather than false consensus. In a culture where institutional trust has eroded largely because words and reality have visibly diverged — in politics, media, corporate communication — the simple, almost boring discipline of saying what is actually the case turns out to be a quiet form of resistance and care.

Common Misreadings and Clarifications

"聖言 means 'sage-like eloquence' or 'deep spiritual words.'" No. In the Yogācārabhūmi-śāstra's usage, 聖言 is not about the profundity or sacredness of the content. A person can speak the most ordinary, mundane sentence — "I didn't see the email" — and that sentence can be 聖言 if it accurately reports their experience. Conversely, a person can chant a sutra aloud and produce *non*-sage speech if the recitation misrepresents what they actually see, hear, sense, or know. The "sage" in 聖言 refers to the *quality* of the speech (its alignment with reality), not the status of the speaker or the loftiness of the content.

"聖言 means 'speak only positive, kind things.'" No. The criterion is accuracy, not agreeability. 聖言 can include saying "no," saying "this is wrong," saying "I am hurt by this," saying "you are mistaken." None of those are kind on the surface, but all of them can be 聖言 if they accurately report the speaker's actual experience. The text is not in the business of mandating pleasantness; it is in the business of mandating alignment.

"聖言 is the same as 'never lying.'" Partially overlapping, but not identical. The concept is broader than the common understanding of lying, because it includes the entire bidirectional structure: it forbids not only claiming experiences one doesn't have but also denying experiences one does have. It also covers omissions and self-directed speech, not just other-directed deception. And it is narrower than "never lying" in one sense: it does not require you to *speak* about every experience you have. It requires that when you do speak, what you say be accurate. Silence is not the issue; misalignment is.

"聖言 is a special attainment reserved for advanced practitioners or monastics." No. The criterion is entirely accessible — and entirely demanding — for any speaker. There is no prerequisite of meditation mastery, monastic ordination, or philosophical study. A child can speak 聖言 by accurately reporting "I saw the cat; I didn't hear the bell." An adult can speak non-sage speech by claiming expertise they don't have. The concept is democratizing in a particular way: it places the demand for accurate speech on everyone equally.

"聖言 is just another name for the four wholesome verbal actions (四善語業道)." Not according to the text. The Yogācārabhūmi-śāstra lists the four 聖言 alongside the four 善語業道 in the same passage, but does not explicitly equate them item by item, and the source material flags this as something we should not paper over. 聖言 is about the *truthfulness* of speech relative to the speaker's own experience; 善語業道 concerns the wholesome karmic quality of speech in its effects on others. These overlap in some places but are not the same concept. If we collapse them, we lose the specific contribution of each.

"聖言 is just a Buddhist way of saying 諸佛聖教 (the Buddha's teaching)." No — and this is the most important terminological boundary. 諸佛聖教 refers to the doctrinal content taught by buddhas, summarized in the well-known verse: 「諸惡者莫作,諸善者奉行,自調伏其心,是諸佛聖教」 — do no evil, do all good, tame your own mind. This is *instruction*, addressed to disciples about how to live. 聖言 is a *discipline of speech*, addressed to anyone who speaks, about how their words should align with their experience. The two share a character but not a referent. Treating them as the same flattens an important distinction: one is a teaching about ethics and practice; the other is a teaching about epistemic honesty in speech.

"If I'm not sure whether something is true, I should just keep quiet." This is a reasonable inference but goes slightly beyond what 聖言 strictly requires. The text's criterion is not "say nothing unless certain"; it is "say what matches your actual experience." Uncertainty is itself an experience — 知 knows that it does not fully know — and saying "I'm not sure" is an accurate 知 claim. 聖言 is satisfied as long as your report of your epistemic state is accurate. The discipline is not silence; it is accurate reporting of whatever state — including uncertainty — is actually present.

"聖言 will make me sound rigid, unimaginative, or uncharitable." This is the worry, not the reality. In practice, 聖言 tends to *free* speech, because it removes the exhausting project of managing other people's impressions through inflated or deflated reports. When you are no longer performing experience, you have more attention available for actually responding to the situation. People who speak 聖言 often come across as unusually trustworthy and unusually present — not rigid, but clear. The concept is less a constraint on speech and more a release from the constant low-level dishonesty that most speakers carry without noticing.


A final word on placement within the path. 聖言 sits very low in the structure of Buddhist training — almost at the foundation — and that is part of its importance. It is not a high-stage insight; it is not a tantric accomplishment; it is a basic ethical-epistemic discipline that any practitioner can take up right now, today, in the next conversation. And yet almost everything the Buddhist path later asks of a practitioner — right view, right intention, the dismantling of the proliferating self — assumes some capacity for accurate self-report. You cannot see the movements of your own mind if you are constantly narrating them wrongly to yourself. You cannot loosen the grip of craving if you keep describing your experience in the inflated or deflated terms that craving prefers. 聖言 is, in this sense, the quiet floor under the entire path: not glamorous, easily overlooked, but without it the rest has nothing reliable to stand on.

Canonical EntryAI-generated

1. 定义与核心要义

《瑜伽師地論》所说的「聖言」,主要见于声与语业的分类,并非泛指佛陀或圣者宣说的一切教言。其判定标准在于言说是否符合说话者自身实际的见、闻、觉、知:已经看见便说看见,没有看见便说没有看见;听闻、觉受、了知亦复如此。凡所说与实际经验相符,即列为「聖言」;故其核心要义是如实陈述,不虚构、隐瞒或颠倒自己的见闻觉知。

论中使用的完整分类名称是「四聖言聲」,逐项则称「聖言」。这里的「四」依见、闻、觉、知四类认识活动而建立,每一类同时涵盖肯定与否定两种如实陈述。因此,「不見則言不見」与「見則言見」同属圣言,并非只有作肯定陈述才算圣言。

2. 主要阐述

四聖言可合并归纳如下:第一,关于「见」,见则如实说见,不见则如实说不见;第二,关于「闻」,闻则说闻,不闻则说不闻;第三,关于「觉」,觉则说觉,不觉则说不觉;第四,关于「知」,知则说知,不知则说不知。四者所涉及的对象与认识方式虽有差别,分判原则一致,都是使言语与自身实际状况相应。

论中又以「四非聖言聲」构成严格的反面分类:不见却说见,或已经见却说不见;不闻却说闻,或已经闻却说不闻;不觉却说觉,或已经觉却说不觉;不知却说知,或已经知却说不知。由此可知,不如实不仅包括虚称自己具有某种经验,也包括否认自己确有的经验;无论妄加还是隐没,只要所说与实际见闻觉知相违,便属于「非聖言」。

这组正反分类呈现的是言语真实性的形式:圣言如实,非圣言不如实。同一段文字随后还列出「四善語業道」与「四不善語業道」,但所据原文并未把四圣言逐项等同于四善语业道,也未把四非圣言逐项等同于四不善语业道。因此,二组概念虽同处语业论述之中,仍应依原文分别记录,不补立未经明说的对应关系。

此外,本论另见「賢聖諦語」一词,原文说:「與此相違,當知即是賢聖諦語。」这表明「賢聖諦語」用来标举与其前述反面情形相违的真实言说。然而现有引文没有保留「此」所指的完整上文,故只能确认该术语及其相反关系,不能据此进一步确定其完整分类、范围或与四圣言的逐项对应。

3. 不同部类与译本的表述差异

现有材料仅涉及瑜伽部《瑜伽師地論》(《瑜伽師地論》),没有提供其他部类经典或同经异译本,因而不能建立跨部类或异译本的词语对照。但在本论内部,可以辨明三种不可径直混同的表述:

其一,「四聖言聲」是声与言说分类的总名,逐项简称「聖言」,侧重见、闻、觉、知方面的如实陈述;与之相对的总名是「四非聖言聲」,逐项称「非聖言」。

其二,「賢聖諦語」同样与真实言说有关,但现有原文只明示它与前述某种情形相违,未提供足够上下文说明其范围,故不宜将它直接改写为「四聖言聲」的异名。

其三,本论称佛陀教法或教诫为「諸佛聖教」。偈中以不作诸恶、奉行诸善、调伏自心三项总摄其内容,并称之为诸佛的圣教。这里使用的是「聖教」,不是「聖言」;因此,本论中的「聖言/聖言聲」偏指如实言说,「諸佛聖教」则明确指诸佛教诫,二者应依原文用法分别说明。

4. 与相关概念的关系

「非聖言」是圣言最直接的反面概念。两者都依见、闻、觉、知四类建立,其区别不在言说对象本身,而在陈述与实际经验是否相符:相符者为圣言,相违者为非圣言。

「善語業道」与「不善語業道」也是本论同段所列的语业分类,但现有文字仅能证明它们与圣言、非圣言并列出现,不能证明两组概念完全同义或逐项重合。

「賢聖諦語」与真实言说具有正面关联,但因指代所需的前文不全,只宜保留原论所说的相违关系。「諸佛聖教」则属于诸佛教诫的表述,其内容是止恶、行善与自调伏心,不应仅因同有「聖」字便与「聖言」合并为一个术语。

5. 代表性原文引用

关于四圣言,其要旨可由「見則言見,不見則言不見;聞則言聞,不聞則言不聞;覺則言覺,不覺則言不覺;知則言知,不知則言不知」概括。与此相反,则是不见言见、见言不见,乃至不知言知、知言不知的四非圣言。

关于「賢聖諦語」,本论原文为:「與此相違,當知即是賢聖諦語。」由于所据文字未载完整前文,此句宜按原有指代关系保存。

关于诸佛教诫,本论偈云:「諸惡者莫作,諸善者奉行,自調伏其心,是諸佛聖教。」此处明确使用「諸佛聖教」,显示它与四圣言所讨论的如实陈述各有不同的语境与所指。

By SutraAI-generated

瑜伽師地論T1579 · 瑜伽部

《瑜伽師地論》將「聖言」置於聲之差別門中加以安立,以「四聖言聲」與「四非聖言聲」對舉,合為八種聲之一類。其核心界定極為簡明:聖言即所言與所證相符之語,非聖言即所言與所證相違之語。具體而言,四非聖言為:一、不見言見、見言不見;二、不聞言聞、聞言不聞;三、不覺言覺、覺言不覺;四、不知言知、知言不知。四聖言則恰恰相反:一、見言見、不見言不見;二、聞言聞、不聞言不聞;三、覺言覺、不覺言不覺;四、知言知、不知言不知。

就術語形式而言,本論所用為「聖言聲」與「非聖言聲」,將聖言歸攝於「聲」之分類體系下,而非獨立為一門法數。此處「聲」指語業所攝之聲,論中緊接聖言八種之後即云「又有八種,謂四善語業道,四不善語業道」,可見聖言、非聖言與善、不善語業道為同一層級之安立,皆屬語業差別。

本論論述聖言之獨特側面有二。其一,以「見、聞、覺、知」四事為聖言所涉之全部對象,涵蓋認識活動之四個面向,而不以具體戒條或業道名目為綱。其二,判定聖與非聖之標準純在「如實」與否——所言是否忠實反映自內所證,凡增益(無而說有)或損減(有而說無)皆屬非聖言。論中未另立動機、對象、果報等條件,僅以言與境之符違為唯一判準。

在安立脈絡上,聖言八種僅為本論層層展開聲之差別之一環。論中自一種聲起,經二種(有執受、無執受)、三種(可意等)、四種、五種,逐步增上至十種,聖言八種居其中。此後更以同一「或立一種……或立十種」之格式遍說香、味諸境,顯示本論以「一法多門分別」為方法論特色,聖言並非被單獨提出討論,而是作為聲境八種差別中之一組而存在。

總之,《瑜伽師地論》於此處以極精簡之對舉句式界定聖言,不設繁複名相,其要旨唯在「如實而言」四字:見則言見,不見則言不見,聞、覺、知亦復如是。此與論中其他以業道善不善為綱之語業分類互為表裏,共同構成瑜伽行派對語業之完整分別。

Related