Dharma · AI Companion

持誦法

dhāraṇasvādhyāyadharma

「持诵法」指真言行者以口诵、心持或随息念诵真言,观字印本尊三密平等之修行法门。AI-generated

Type PracticeDifficulty AdvancedTruth-level span IV · beyond wordsInitial AI estimate · evolves with use

A Contemporary ReadingAI-generated

What this concept is really saying

持誦法 (pronounced *chísòng fǎ*) literally means "the dharma/法 of holding-and-reciting." Its Sanskrit name, *dhāraṇīsvādhyāyadharma*, already tells you exactly what it is and what it is *not*: it is the dharma of *svādhyāya* — that is, structured, self-cultivating repetition — applied to a *dhāraṇī*, a "总持" (zǒngchí), the formula that gathers and holds a field of meaning. The principal classical source for this entry is 《大日經疏》(T1796), the commentary attributed to 善無畏 (Śubhakarasimha) on the 《大毘盧遮那成佛神變加持經》— the *Mahāvairocana Sūtra*, the foundational scripture of Chinese esoteric / 真言宗 (Zhēnyán zōng, Mantra Vehicle) Buddhism. The very chapter that treats this practice is named the 世出世間持誦品 ("Chapter on Worldly and Transcendent Recitation"), and the śāstra declares it 祕密中之祕密 — "the secret within the secret" — because, in the commentator's reading, the entire 大日經 exists to teach bodhisattvas who are cultivating mantric conduct *how to enter the path through recitation*. So this is not an auxiliary technique. It is, in this tradition, the doorway.

The first thing to fix precisely is what is being recited. The śāstra uses three Chinese terms — 持誦, 念誦, 持真言 — as mutually defining synonyms. They all mean the *recitation of a 真言 (mantra)*, specifically. They do not mean the chanting of a sutra passage (讀誦), and they do not mean the casual muttering of a blessing. A 真言 here is a syllable-string (or seed-syllable, 種子字) that names or embodies a particular 佛 / 菩薩 / 本尊 — a deity-aspect. So 持誦法 is a修行法門 (xiūxíng fǎmén, a cultivation practice) in which the body, the breath, the sound, and the visualized image all converge on that one 字-印-尊 (character-mudra-deity) field.

The three forms, kept distinct

The śāstra is unusually specific about what counts as this practice and what does not. There are three forms of 念誦:

1. 一一念誦 (yīyī niànsòng), "syllable-by-syllable recitation." This is the oral form: every syllable, every vowel length, every 點 (the dots, diacritical marks that change meaning) is pronounced deliberately and completely. The practitioner "斷除攀緣" — cuts off grasping after other objects — and so each voiced sound is also an *attentional* event, not a lip movement. This is the form most easily conflated with simple *chanting*, but the classical text insists it is not vocalization for its own sake; it is the synchronization of voice with unbroken presence.

2. 作意念誦 (zuòyì niànsòng), also called 心念誦 or 意念誦, "intentional / mental recitation." The mantra is held silently in mind, no sound issued. It is *not* "thinking *about* the mantra." It is the mantra-as-object rested in by attention. The text groups this together with form (3) and calls these two the "最相應之二法" — the two most inwardly congruent methods — and warns the practitioner not to slip into stray images or thoughts ("不得另起異緣異想"). This is a strong claim: even a beautiful visualization of the deity that is not anchored in the syllable, in this framework, counts as a different object.

3. 出入息念誦 (chūrùxī niànsòng), "recitation on the in-and-out breath." The syllables travel with the breath: a long mantra is broken into units, each unit riding one inhalation or exhalation; a seed-syllable may be repeated at each breath. The classical commentary calls this 最上 — the highest of the three. The reason given is functional: 念念相續而不間斷 (sustained, unbroken, moment-to-moment continuity), and it使心不散亂 (settles the mind so it doesn't scatter), and from that settling one can 恬然易入三昧 (calmly, easily slip into *samādhi*). Here the breath is not a separate object of meditation; it is the *vehicle* that keeps the syllable from being an isolated mental event.

These three are not stages in a hierarchy of better-or-worse, except that (2) and (3) are flagged as "most congruent" — i.e., more directly aligned with what 持誦法 is ultimately for. A practitioner can move between them, but they must know which they are doing.

Why completeness matters — 支分 (the parts)

Here is a feature that often surprises contemporary practitioners. The śāstra lays unusual emphasis on 闕支分念誦 (quēzhīfèn niànsòng) — "recitation with parts missing." If a syllable has a 點 (dot) and you don't sound it; if a character is shortened when it should be long; if you skip a phrase in the sequence, the practice is "功用不具" — the function does not arise, even if you have done thousands of repetitions. So this tradition refuses the modern folk-idea that *effort and quantity* are what matter. Accuracy of the formula matters first; integrity of form is treated almost as a precondition. (In Chinese esoteric lineages this precision was carried over into romanization systems like 悉曇字 / *siddhaṃ* script instruction and the use of 念誦次第 diagrams.) The point is not magical correctness. The point is that the practice is supposed to be a *coordinated* event of voice, breath, mind and visualized image — and any leak in the syllable is a leak in the coordination.

Two registers: 世間 vs. 出世間

Within 持誦法 there is a very important inner split that the śāstra names clearly. The practitioner is not told to abandon one for the other, but to know which register they are working in.

世間持誦 (shìjiān chísòng), "worldly / relative recitation." Its signature is 有緣相續 — sustained continuity *with a support-object still operative* (緣, *pratyaya*, the condition/anchor of attention). Here the practitioner takes a 字 (the seed-syllable, the first character of a mantra, or the full text of a short mantra) and visualizes it arranged "布列於本尊心上之圓明中" — laid out within the luminous circle of the deity's heart, like rings in a chain (如連環相續). Or the practitioner may take the 印 (mudrā) as the object, or the 本尊 (the deity's form) itself. The instruction is that whichever is chosen, it must be 宛然分明 — vividly, distinctly, clearly present. This is the more imagistic, more "having-an-object" mode.

出世間念誦 (chūshìjiān niànsòng), "transcendent / supra-worldly recitation." The shift is from *having* the object to *seeing through* it. It is identified primarily with 作意念誦 (silent mental holding), and the śāstra uses a phrase that has caused confusion for centuries: 離於文字 — "apart from / free of letters." A modern reader can easily take this as advice to *drop the mantra and just sit*. The commentary *immediately* contradicts that reading: "此非排除或捨棄真言文字,而是通達字之本性為圓明、本來不生" — "this does not mean to exclude or abandon the mantra characters, but to penetrate to the characters' *own nature*, which is 圓明 (perfect luminosity) and 本來不生 (originally unborn)." So the syllable does not disappear; what disappears is your grasping at it as an object.

Then comes the second layer of progression, the 誦/照 pair. 誦 (sòng) is recitation proper — engaging with sound or breath. 照 (zhào) is *照見*, illuminating-and-seeing, contemplating the nature of the character. Practice can include both, can include only contemplation, can switch between them. The trajectory is: from a clear, image-rich visualization of the deity's heart-circle → to *seeing* that this circle was never a thing apart from mind → to resting in that seeing.

Three-mystery equality — 字、印、本尊 ↔ 語、身、心

This is the heart of how the śāstra frames 持誦法 as a complete 入道次第 (rùdào cìdì, ordered path of entry). The classical equating is precise:

- 真言字 ↔ 語密 (yǔ mì, "speech-mystery" / vāg-guhya) - 身印 (mudrā) ↔ 身密 (shēn mì, "body-mystery" / kāya-guhya) - 本尊心上的圓明 ↔ 心密 (xīn mì, "mind-mystery" / citta-guhya)

In 世間 recitation you can take any *one* of these as your anchor and let the others fall into the background; you 觀 (visualize) character, or seal, or deity. In 出世間 recitation this division collapses: 字, 印, and 尊 are seen as 平等一相 (píngděng yīxiàng, "equal, of one suchness") — three faces of one event, not three objects to choose between. And there is a further turn: 觀本尊與自身相同 — the practitioner sees the deity as not different from themselves. The technical term 一合相 (*yīhé xiāng*, "one-merged appearance") and 不壞 (*bùhuài*, "not destroyed") are explicitly clarified in the śāstra as *not* destruction or fusion of forms, but the *non-discrimination* (不作分別取著) that no longer holds the three apart.

Counts and the full-moon accomplishment

The text mentions 三落叉 (*luòchā*, Sanskrit *lakṣa*, "one hundred thousand") — three hundred thousand recitations — as a measure of the practice, alongside discussions of 時節 (time-period), 現相 (visible signs appearing), and 增益 (gradual increase). The term 落叉 gets a second, contemplative meaning: it is also "seeing." So a "落叉" of recitation can mean *seeing* the three-mystery field as one appearance — character, mudra, and deity unified — three hundred thousand times. After preliminary offerings and signs of accomplishment appear, on the full moon (白月十五日), the practitioner selects a site appropriate to the work — mountain peak, cow-pen, forest, riverbank, crossroads, or a 天室 (celestial chamber / upper-story shrine room) — establishes a 漫荼羅 (*màntúluó*, maṇḍala) there, and performs the 持誦成就法 — the accomplishment-form of recitation. From the 佛部 (Buddha-family) down through the 八部 (the eight classes of non-human guardians), every deity's recitation follows the same rules: 字句觀 (character-and-phrase visualization) plus 出入息念誦.

So the picture, kept clean: 持誦法 is a total practice in which (a) a syllable-string tied to a specific deity is held or voiced with full accuracy, (b) on breath or silently, (c) within a visualized circle on the deity's heart, (d) while the practitioner's body takes the deity's mudrā and (e) the practitioner's mind sees itself and the deity as one appearance — and this convergence is what the tradition means by entering the path.


A walk-through in ordinary life

Most contemporary readers who encounter 持誦法 hear about it in some ān'pài / 念珠 app, or on a meditation platform with a "Buddhist mantra" track, or in a wellness-framed piece on a particular syllable (ōṃ / 唵, *nā mó* forms, six-syllable 嘛呢, etc.). What almost always gets lost in that translation is the classical structure. Let me walk the idea through scenes a contemporary reader will recognize, mapping each scene strictly onto the classical terms above, and pausing to flag where popular framings get it wrong.

1. The morning commute, alone in the car. Modern apps treat this as a moment to "play a mantra." If the practitioner is genuinely doing 一一念誦, every syllable they voice is on purpose: the breath under each one is prepared, the long vowel is held for its full length, the dot-marks are *audible* (lengthened tone, or sustained sibilant, or whatever the syllable requires). Their attention is not on the song but on the *next syllable*. They are not practicing a calming sound for cortisol reduction — they are letting a single formula occupy the center of attention until everything else (the other cars, the podcast queue in the back of the mind, the work worry) is "斷除攀緣" — cut off. If they are doing 出入息念誦 instead, each syllable rides one breath, so that when their attention lapses they notice it as a skipped syllable, not as a wandering thought. The classical 甲 difference: this is not a sound bath. It is a *coordinate* of voice and breath and name and imagined image; the silence between syllables is just as much 念 as the syllables themselves.

The place the popular framing goes wrong here: treating 念誦 as essentially a stress-relief audio track. The śāstra's entire point is that the *integrity* of the syllable matters; substitution, abbreviation, or simultaneous multitasking is treated not as a soft preference but as a practice that "功用不具" — simply does not function. A contemporary practitioner who treats the formula casually is, by this tradition's measure, doing a different activity.

2. A meditation app's "silent mantra" feature. This is where most modern users first meet 作意念誦. Many apps translate "silent mantra" as "think the mantra." The classical text is sharper: it's not "thinking *about* the mantra," but the mantra *as the resting object of attention*. When the mind wanders to the email you'd better not have forgotten, you've left 緣 — left the support-condition — and the śāstra says this is no longer the practice. The instruction 不得另起異緣異想 (do not start up another condition or another thought) is one of the few commands in the entire 持誦 chapter; it isn't decor, it's the rule. So a contemporary practitioner who notices a beautiful image of the deity appear *unbidden* during mental recitation is now possibly no longer practicing 作意念誦; the classical taxonomist would say they've shifted into 世間持誦 with the 本尊 as their object. Both are valid; they are not the same.

The place popular framings get wrong: "Visualize the deity while reciting" is taken as a single activity. In the classical structure it is a *deliberate* shift between two registers — keep them distinct until you can hold them as one appearance, which is the 出世間 register.

3. Walking the dog / peeling vegetables / doing the laundry. This is where 出入息念誦 maps onto ordinary life most directly. The practitioner pairs one syllable (or one short phrase of the mantra) with each exhalation, or one whole long mantra with one breath cycle. The syllables become the *exhalation*. The body is doing something else; the breath is the seam between body and syllable. This is the reason the śāstra calls this 最上 (highest): it can be maintained throughout ordinary action without becoming its own event, so 念念相續 (moment-to-moment continuity) becomes possible *outside* a cushion. *Samādhi* doesn't require sitting cross-legged; it requires non-scattering.

The place this gets misframed in contemporary use: as a "breathing exercise." That is *not* 出入息念誦. Box breathing, 4-7-8 breathing, pranayama as such — these are breath-regulating techniques with their own aims and effects. In 出入息念誦 the breath is the *carrier*, and the syllable is the *content*. If you take the syllable out, what is left is a different practice. (The contemporary analogy is closest to a "paced expiral" with a word attached, but the difference is that the word is not decoration — it is the practice.)

4. The moment of grief / a hard phone call / receiving bad news. Many contemporary dharma teachers turn to 持誦 at moments of distress, sometimes with the framing of a comforting ritual. The classical text would partly agree and partly sharpen the framing. It would agree because 出入息念誦 is, functionally, a fast way of giving the scattered mind *one* anchor. It would *disagree* with the comforting-ritual framing because 持誦法 is not primarily about the practitioner's emotional state — it is about the alignment of 字, 印, 本尊, 身, 語, 心 in one coordinate. The instruction is that even if the practitioner's heart is in pieces, what they do is *not* improvise a substitute text, *not* abbreviate the syllable, *not* switch into an inner monologue. They voice the syllables as written, or hold them silently, and they let the formula's integrity do the work of holding attention — which is sometimes the only way to *not* do further harm to a mind already over-aroused.

5. Doing a yoga / qì-gōng / 太极 sequence, or a focused gym workout. Here many moderns notice that a mantra-style cadence can be added to movement and seems to deepen it. The 身密 half of the classical framework is about exactly this — the body *in* the practice. The śāstra, however, ties 身密 specifically to the 印 (mudrā), a hand-position sealed by the practice line; it is a *specific* body-shape, not "any body posture." This is one of the deepest places where the living practice today has drifted from the śāstra: contemporary "mantra + body movement" teachings sometimes treat any gesture as 身密. The classical text would say: the body half of 三密 is a *precise* form given by the lineage for that deity, and doing a different body posture is not 三密平等 with the syllable and the visualization — it is *one* mystery (speech) being held with a body-shape that is *not* on the same coordinate.

6. Lying in bed waiting for sleep. A common contemporary use of 持誦 is as a transition-into-sleep aid. 出入息念誦 maps very well here — the breath slows, the syllable slows with it, the attention stays anchored. The classical text wouldn't object to this; it's consistent with 使心不散亂, 恬然易入三昧. The śāstra would *only* note that what makes this 出入息念誦 rather than "breathing to fall asleep" is that the syllable is engaged as the *object*, not as a pleasant sound used to smooth the way down. If the practitioner notices the syllable becoming vague at the doorway of sleep, the classical instruction is: bring it back, voice it lightly, get the 點 and the long vowel right, *then* let the breath carry it again. (Compare this to the way a musician continues to play scales with attention even at the moment of falling off to sleep: the form is kept.)

A final map for this section. Every contemporary scene above runs through the same classical structure: (a) a specific syllable-string tied to a specific deity, (b) voiced or held with full accuracy, (c) on breath or in mind, (d) anchoring an image of the deity's heart-circle, (e) with the practitioner's body in the mudrā of that deity, (f) the practitioner and the deity seen as one appearance — and this is one event, not six. The contemporary scenes differ only in how loud or silent, how still or moving, the life around the practice is.


Why people today need this

It would be misleading to claim contemporary psychology "proves" 持誦法 works. I will say what can be said and what cannot.

What can be said with some care: contemporary attention research, particularly on *sustained attention* and *default mode network* activity, consistently finds that brief, repeating, structured sensory tasks (rhythmic speech, paced breathing with a verbal anchor, repetition of a known short string) reduce mind-wandering relative to open-monitoring meditation, especially in beginners. This is broadly consistent with the śāstra's claim that a syllable-and-breath anchor 念念相續 (sustains continuity) and 使心不散亂 (prevents scattering). It does *not* validate the metaphysical claims of 三密平等 — open-monitoring and focused-attention research is not in the business of vindicating or refuting 十住 (bodhisattva stages).

Contemporary mood-research on *self-distancing*, *cognitive defusion*, and *affect labeling* (naming what one feels) shows that pairing a felt state with a structured verbal anchor can reduce emotional reactivity — and many readers have noticed this overlaps structurally with 出入息念誦 in difficult moments. But affect-labeling research tends to use *descriptive* words ("I notice anxiety"); 持誦法 uses a *non-descriptive* syllable-string whose content is not a label. They are shaped differently and aim differently. The overlap is real; so is the divergence.

The strongest contemporary reason to study 持誦法 — though this is a *traditional* reason that the śāstra would put first — is that modern life is structurally scattering. It is full of 異緣, the "other conditions" the śāstra names as what should not be allowed to take over the practice: push notifications, half-finished browser tabs, the partial attention people now call 連續部分注意力 ("continuous partial attention"). What 持誦法 offers, even taken purely as a *practice* and bracketing all metaphysics, is an unusually demanding *training in un-scattered coordination*. It trains the three registers — voice, body-shape, visualized image — *simultaneously*. Few contemporary practices demand this. Most contemporary "mindfulness" practices train the attention field; 持誦法 trains *coordination across three internal substrates at once*, with a formula that *cannot be paraphrased* without losing the practice. That is a difficult and rare demand, and modern attentional life makes it unusually valuable.

There is also a non-obvious contemporary value in 持誦法's refusal of the *comfort* framing. Most things on offer as spiritual today are oriented toward the practitioner's comfort and improvement. 持誦法 in this śāstra is oriented toward an *integrity* of syllable, image, and gesture that the practitioner either meets or doesn't. In a culture where most offerings are customizable, negotiable, and self-curated, a practice that says "long-vowel here, do not shorten" is doing some quiet cultural work. It doesn't have to be claimed as metaphysical for this to matter.

A fourth contemporary reason: 持誦法 as the śāstra frames it is a *complete* path. One syllable-string, if held with all three mysteries aligned, is supposed to be enough to enter the path; the practice does not require (in this tradition) a curriculum, an extensive curriculum of scriptures, monastic vows or years of preliminary study. This is attractive to contemporary adults whose lives cannot accommodate *another* multi-year commitment. It also imposes the complementary burden that the practitioner must be honest about whether the practice is actually meeting the structure — because the very same compactness that makes it accessible also makes it easy to *think* you have been practicing when you have not.


Common misreadings and clarifications

The following are the misreadings I encounter most often, in roughly the order of how much damage they do to the practice itself.

1. "持誦 = reciting a sutra." No. The śāstra is explicit that 持誦, 念誦, 持真言 refer to *mantra recitation*, not scripture reading (讀誦). A practitioner who reads the *Heart Sutra* or *Diamond Sutra* every morning is doing a different (and also valuable) practice called 讀誦 or 誦經. The 持誦 *form* is rooted in a syllable-string tied to a deity. Confusing the two flattens a specificity the tradition cares about, and the resulting "Buddhist chanting" practice loses the 字-印-尊 structure that makes 持誦法 what it is.

2. "念誦 is basically affirmations or self-help." The self-help framing takes the syllable-string as a *content* (a statement about the world or the practitioner) and evaluates it by whether it helps. The śāstra treats the syllable-string as a *form* whose *integrity* matters independently of the practitioner's emotional state. Different goal, different measure.

3. "The mantra works because the universe is responding to a sequence of sounds." This is a popular folk reading. The śāstra is more interesting than that. It is not asserting a magical-mechanical model in those pages; it is asserting a *coordination* model — that 身, 語, 心 converge on one suchness, and the *function* of practice depends on that convergence, which depends on the syllable-string being voiced with full integrity. Whether one accepts the metaphysical claims or not, the practice's classical self-understanding is not "magic spell works if pronunciation is correct."

4. "離於文字 means drop the mantra and just sit in silence." Easily the most damaging misreading of 出世間念誦. The śāstra itself warns against it: 離於文字 does not mean to drop the letters; it means to penetrate to their 本性 (own nature) as 圓明, 本來不生 — perfectly luminous, originally unborn. The mantra is not abandoned; the grasping at it as an object is loosened. Anyone who teaches "just sit in silence, the mantra drops away on its own" is, by this śāstra, no longer doing 持誦法 in any of its classical registers.

5. "一合相 means the practitioner merges with the deity or the deity dissolves into the practitioner." No. The śāstra explicitly clarifies 一合相, 不壞 as 不作分別取著 — non-discrimination, not destruction. The forms of deity, practitioner, syllable and seal are not dissolved; what is given up is the *holding of them apart as separate objects*. This is a fine but real distinction: one can dissolve a form by mystical experience; one can stop holding it apart by 平等一相. The classical text is teaching the second, not the first.

6. "Any visualization is fine, especially a beautiful one." The śāstra is sober here. 觀 (visualization) is a *technique within a structure*: the 字 must be visualized 宛然分明 (vividly distinct) on the 心上圓明 (heart-circle), 布列如連環相續 (arranged like linked rings). The most popular contemporary mistake is to "visualize" freely — light, color, landscapes, deities in everyday scenes — and call this part of the practice. The classical framework is much more specific, and free-form visualization is closer to a different practice or to none.

7. "Silently chanting the mantra in the back of the mind while doing other things equals 作意念誦." It does not. The classical rule for 作意念誦 is unbroken attention to the syllable as the support-object, 不得另起異緣異想 (no other conditions, no other thoughts). What most people do as they move through the day is closer to a background rehearsal than to 作意念誦. The śāstra would call that something other than 念誦 — perhaps formless 串習, or simple habit.

8. "Mantras are just prayers in an unknown language." Half-right, half-wrong. The syllables are not in an "unknown language" — they are Sanskrit syllables whose *literal* meaning, sometimes sharp, sometimes pictorial, sometimes proper names, the tradition has spent centuries glossing. But the practice is *not* primarily about understanding the literal meaning, and it is *not* a prayer in the petitionary sense. It is a *form of concentration* on a syllable-string. Reading the mantra-translation is no substitute for practice; refusing to know the translation is also not required.

9. "You need a master, a lineage, an empowerment before any of this can work." In the *living* esoteric tradition, this is taken very seriously — 傳法 (lineage transmission) and 灌頂 (empowerment) are part of what makes the practice possible, and the śāstra itself takes that context as given (the chapter is addressed to 修真言行之菩薩, bodhisattvas cultivating mantric conduct). For a contemporary reader who only wants the *concept* to think with, this is not in the way; for a practitioner who wants to do this as a real cultivation line, it is not optional. I want to mark this honestly rather than paper over it.

10. "The bigger the number of repetitions, the more it works." The śāstra introduces 三落叉 (three hundred thousand) but immediately complicates it: 落叉 also means *seeing* the three-mystery field as one. A worn-down count without 三密 alignment is, by this text's measure, less useful than a smaller, fully aligned count. Effort and quantity are not zero, but they are not the primary metric either.


To close where the śāstra opens. The 大日經疏 declares the 世出世間持誦品 祕密中之祕密 — the secret within the secret — because this chapter is where an entire 大日 (Mahāvairocana) teaching of the path comes down, in compressed form, to a single instruction: hold the syllable, voice the syllable, breathe the syllable, image the syllable, seal the syllable with the body, and — without dissolving any of these — let them be one suchness. Whatever a contemporary reader takes from this on first encounter, the structure is worth knowing. It is more specific, more demanding, and more interesting than the popular imago of "mystical chant" — and more useful, because where most contemporary practices offer some piece of what 持誦法 offers whole, this one offers the unusual combination of *full integrity of form* with *single-pointed simplicity of object*. In a life otherwise strewn with half-finished tasks and partial attentions, that combination is uncommon.

Canonical EntryAI-generated

定义与核心要义

「持诵法」(dhāraṇīsvādhyāyadharma),依《大日经疏》之说,乃真言行者依真言字句而修之念诵行法。其核心在于「世出世间持诵品」,此品被视为「秘密中之秘密」,总摄全经为修真言行之菩萨开示「作持诵入道之法」。本疏中,「持诵」与「念诵」「持真言」互训同义,主要指真言之念诵,而非泛论经典之读诵。

主要阐述

念诵形式有三:一一念诵,即专心以口诵真言,每一声每一字皆谛了而不间断,并断除攀缘;作意念诵(又称心念诵、意念诵),即心中持念而不出声;出入息念诵,使真言字随呼吸出入。三者之中,作意念诵与出入息念诵被视为最相应之二法,行者应依其法则而修,不得另起异缘异想。

本疏特别重视念诵「支分」之完整。若真言字有点而未读、文字有所缺漏,或应作长声却读成短声,皆为「阙支分念诵」;如此即使费力,亦功用不具。故持诵不仅计算遍数,亦须字音、字形及念诵程序皆不阙失。

世间持诵以「有缘相续」为其特征,仍以字、句、本尊或印为所缘。行者可取种子字、真言首字,或较短真言之全句,观想其布列于本尊心上之圆明中,如连环相续;亦可观字、观印、观尊,使所观宛然分明。此类修法之中,以出入息念诵为最上:字随息出入,念念相续而不间断,可令心不散乱,恬然易入三昧。

出世间念诵以意念诵为主,称其「离于文字」,但此并非排除或舍弃真言文字,而是通达字之本性为圆明、本来不生,离开分别缘念之心及对声音之执取。持诵时可有诵、有观,或二者兼行,亦可但修观照;「诵」是缘声字或出入息,「照」则观字之体性,由有相之明晰观想进而观其本性。

本疏又以「字、印、本尊」及「身、语、心」之平等一相说明深层持诵。世间念诵可分别取字、印或本尊为所缘;出世间念诵则不将其分裂为二而执取,而观本尊心上圆明为心、身印等为身、真言字等为语,三事平等一相,并进一步观本尊与自身相同。此处「一合相」「不坏」等语,皆用以表示不作分别取着,而非毁坏所观形相。

持诵亦有遍数与成就法。原文以「三落叉」表示念诵数目,并说遍数、时节、现相、增益等皆属多种念诵规则;「落叉」在出世间义中又被解作见字、印、本尊三相,随取其一而成一合相。另于满月(白月十五日),行者先完成供养承事并得成就之相,再依上、中、下事选择山峰、牛栏、林间、河滩、四衢或天室等处,建立漫荼罗而作持诵成就法。

此外,各尊各有别行持诵法;当诸尊摄入漫荼罗会时,仍依相应真言、印记、本尊及种子字修持。由佛部直至八部,凡有念诵,字句观与出入息念诵皆依同类法则施行。故本疏之特色,在于以「持诵」为总称,以「念诵」细分口诵、心持、随息及观照,并把真言文字之准确、所缘形相、呼吸相续与三密平等联结为完整之入道次第。

不同部类与译本的表述差异

本条目之资料主要依据《大毘卢遮那成佛经疏》(《大毘盧遮那成佛經疏(大日經疏)》)一经疏,故关于部类间之异说未能备列。然就本疏自身而言,「持诵」「念诵」「持真言」三词互用同义:原文于品题总称「持诵」,于具体行法则多称「念诵」,于总摄修行时又称「持真言」。又「作意念诵」一名,亦见作「心念诵」「意念诵」,所指皆同。论场所则用「漫荼罗」(maṇḍala)一词,未见其他异译。

与相关概念的关系

持诵法与 dhāraṇī(陀罗尼) 密切相关——其梵语 dhāraṇīsvādhyāyadharma 即「陀罗尼念诵法」之义,指总持真言而念诵之修行。与 三密(身密、语密、心密) 之关系,表现于「字、印、本尊」与「身、语、心」之对应:真言字为语密,本尊形相为身密之表,而观想则通于心密。与 三摩地(samādhi) 之关系,本疏明言出入息念诵可「恬然易入三昧」,又言「诵」「照」之进阶即由有相观想趋于本性之观照。持诵法亦与 真言(mantra) 本体不可分:本疏虽言出世间念诵「离于文字」,然同时强调此非舍离文字,而是通达字之本来不生;故持诵始终以真言文字为所缘。

代表性原文引用

「世出世間持誦品」被稱為「祕密中之祕密」,說全經旨趣在為修真言行之菩薩開示「作持誦入道之法」。
三種念誦之中,「作意念誦」與「出入息念誦」被稱為最相應之二法,行者應依其法則而修,不得另起異緣異想。
若真言字有點而未讀、文字有所缺漏,或應作長聲卻讀成短聲,皆稱「闕支分念誦」;如此即使費力,亦被判為功用不具。
出入息念誦為最上:字隨息出入,念念相續而不間斷,可令心不散亂,恬然易入三昧。
出世間念誦「離於文字」,並非排除或捨棄真言文字,而是通達字的本性為圓明、本來不生。
觀本尊心上圓明為心、身印等為身、真言字等為語,三事平等一相,並進一步觀本尊與自身相同。
「落叉」在出世間義中又被解作見字、印、本尊三相,隨取其一而成一合相。
諸尊攝入漫荼羅會時,仍依相應真言、印記、本尊及種子字修持。由佛部直至八部,凡有念誦,字句觀與出入息念誦皆依同類法則施行。

By SutraAI-generated

大毘盧遮那成佛經疏(大日經疏)T1796 · 經疏部

《大毘盧遮那成佛經疏(大日經疏)》以「持誦」「念誦」「持真言」等語論述受持真言的修行,其中尤以「世出世間持誦品」為一部宗要,說全經旨趣在為修真言行的菩薩開示「作持誦入道之法」;此品則更完整地分別其宗要之行,故被稱為「祕密中之祕密」。本疏在這些原文中主要論真言念誦,而非泛論經典讀誦。

其念誦形式有三:一是「一一念誦」,即專心以口誦真言,每一聲、每一字皆須諦了而不間斷,並斷除攀緣;二是「作意念誦」,又稱「心念誦」或「意念誦」,即心中持念而不出聲;三是「出入息念誦」,使真言字隨呼吸出入。三者之中,作意念誦與出入息念誦被稱為最相應的二法,行者應依其法則而修,不得另起異緣異想。

本疏特別重視念誦「支分」的完整。若真言字有點而未讀、文字有所缺漏,或應作長聲卻讀成短聲,皆稱「闕支分念誦」;如此即使費力,亦被判為功用不具。因此,持誦不只是計算遍數,也要求字音、字形及念誦程序皆不闕失。

在「世間持誦」方面,本疏以「有緣相續」說明其特徵,即仍以字、句、本尊或印為所緣。行者可以取種子字、真言首字,或較短真言的全句,觀想其布列於本尊心上的圓明之中,如連環相續;亦可觀字、觀印、觀尊,使所觀宛然分明。此類修法之中,以出入息念誦為最上:字隨息出入,念念相續而不間斷,可令心不散亂,恬然易入三昧。

在「出世間」念誦方面,本疏以意念誦為主,稱其「離於文字」,但明言這並非排除或捨棄真言文字,而是通達字的本性為圓明、本來不生,離開分別緣念之心及對聲音的執取。持誦時可有誦、有觀,或二者兼行,亦可但修觀照;「誦」是緣聲字或出入息,「照」則是觀字的體性,由有相的明晰觀想進而觀其本性。

本疏又以「字、印、本尊」及「身、語、心」的平等一相說明深層持誦。世間念誦可以分別取字、印或本尊為所緣;出世間念誦則不將其分裂為二而執取,而觀本尊心上圓明為心、身印等為身、真言字等為語,三事平等一相,並進一步觀本尊與自身相同。此處「一合相」「不壞」等語,皆用以表示不作分別取著,而非毀壞所觀形相。

持誦亦有遍數與成就法。原文以「三落叉」表示念誦數目,並說遍數、時節、現相、增益等皆屬多種念誦規則;「落叉」在出世間義中又被解作見字、印、本尊三相,隨取其一而成一合相。另於滿月,即白月十五日,行者先完成供養承事並得成就之相,再依上、中、下事選擇山峯、牛欄、林間、河灘、四衢或天室等處,建立漫荼羅而作持誦成就法。

此外,各尊各有別行持誦法;當諸尊攝入漫荼羅會時,仍依相應真言、印記、本尊及種子字修持。由佛部直至八部,凡有念誦,字句觀與出入息念誦皆依同類法則施行。故本疏的特色,在於以「持誦」為總稱,以「念誦」細分口誦、心持、隨息及觀照,並把真言文字的準確、所緣形相、呼吸相續與三密平等聯結為完整的入道次第。

Related