No two trees are the same, yet one word covers them all. Is there a real “treeness” they share? Dignāga and Dharmakīrti said no: the word “tree” works by excluding everything that is not a tree. This lesson compares their theory with John Locke’s nominalism, Ferdinand de Saussure’s “differences,” and Maimonides’ negative theology, and asks why Buddhists wanted to see through concepts at all. 沒有兩棵樹是一樣的，但一個字就涵蓋了它們全部。它們之間，有沒有一個真實的「樹性」是共有的？陳那與法稱說：沒有；「樹」這個字的作用，是把一切「非樹」排除掉。這一課把他們的理論，跟洛克的唯名論、索緒爾的「差異」、邁蒙尼德的否定神學對照，並且問：佛教徒為什麼要看穿概念？
D15 Dharma and the West · 佛法與西方哲學Knowledge and Logic · 知識與邏輯Advanced · 高級about 30 min
遮詮・遣餘
— Pāli apoha, anyāpoha Sanskritexclusion; meaning by excluding what is other
The verdict · 判定
Closest on what exists: John Locke, for whom general and universal things are “the inventions and creatures of the understanding,” with a foundation in the likeness of things. Closest on how a word means: Saussure, for whom a sign is defined by what it is not. The likeness ends at the purpose. Locke and Saussure wanted to describe how understanding and language work. Dharmakīrti wanted to show that every concept is in one respect a mistake, a useful one, and that the path runs through concepts and then beyond them.
中文翻譯Every paragraph has a Chinese translation. Tap 中譯▾ under a paragraph to open that one, or use Show Chinese · 顯示中譯 at the top to open them all.每一段都有中譯：點段落下方的「中譯」只開那一段，或按上方的「顯示中譯」一次全部打開。
1 · What the Dharma says · 佛法怎麼說
A word that excludes 一個排除的字
Indian philosophers argued for centuries about universals. The Nyāya and Mīmāṃsā schools held that “cowness” is a real, eternal thing present in every cow, and that this is what the word “cow” names. The Buddhist logicians could not accept this. For them what is real is the particular, each thing in its own unique character at its own moment, and nothing real is shared by many things. That left them a problem. If there is no cowness, what does the word “cow” mean, and how can one word apply to many different animals? Dignāga answered in a chapter of his Compendium of the Means of Knowledge whose title the Chinese renders as “Examination of Exclusion.” The verse that opens it, in Lü Cheng’s Chinese translation from the Tibetan, reads: “Knowledge arising from words is no different from inference. Like ‘produced,’ it shows its own meaning by exclusion.”
Dharmakīrti developed the theory with the example of a tree. Every tree is different from every other. What they have in common is only that each is not a non-tree. The word “tree” draws a line around them by excluding everything else, and it is set up for those things that differ from non-trees. The mind that thinks “tree” sees many different things as one kind, and Dharmakīrti calls this cognition mistaken. A Tibetan commentary on his work, translated into Chinese by Fazun together with the verses, puts it plainly: the thought that takes separate trees as one thing is an inverted cognition. Yet it is not groundless. Dharmakīrti gives an analogy: “Medicines differ, yet taken together or one at a time they are seen to cure fever and other illnesses, and other things do not.” Different particulars can do the same work. That shared ability is the real basis of the concept, and there is no shared essence behind it.
The Chinese term for the theory, zhequan, “expression by negation,” took on a second life in China. The ninth-century master Zongmi used it for a style of teaching. “Negation removes what it is not. Affirmation shows what it is.” When the sūtras say “not arising, not ceasing, not defiled, not pure,” they speak by negation, and he gave a homely example: “To say salt is not bland is negation. To say it is salty is affirmation.” He also gave a warning. Students of his day thought negation was profound and affirmation shallow, and so they valued only phrases like “not mind, not Buddha.” Zongmi said this came from loving words of negation and being unwilling to recognize the thing itself. The Indian theory and the Chinese teaching style are not the same idea, and this lesson treats both because they share a name and a lesson. Concepts are drawn by exclusion, and exclusion is not yet seeing.
Locke asked the Buddhist question in almost the same words. “Since all things that exist are only particulars, how come we by general terms; or where find we those general natures they are supposed to stand for?” His answer was that “general and universal belong not to the real existence of things; but are the inventions and creatures of the understanding, made by it for its own use, and concern only signs.” Like Dharmakīrti he did not make the sorting arbitrary. The kinds are “the workmanship of the understanding,” he wrote, “but have their foundation in the similitude of things.” The difference lies in how the general idea is formed. Locke thought the mind builds it positively, by leaving out the particular circumstances of time and place until a common core remains. Dharmakīrti thought there is no common core to keep and that the mind gets its general idea by excluding what the thing is not. Berkeley later attacked Locke’s abstract ideas as impossible, and the Buddhist theory avoids that problem because it never posits them.
The Swiss linguist Ferdinand de Saussure founded modern linguistics on the idea that a sign has no meaning on its own. Its value comes from its place in a system, from its differences from all the other signs. In the English translation of his lectures, “A linguistic system is a series of differences of sound combined with a series of differences of ideas.” The word “sheep” means what it means partly because English also has “mutton,” and a language with one word for both carves the world differently. Modern scholars have often compared Saussure with the apoha theory, and the likeness is real. Both say that a word is defined by what it excludes. The difference is in what each is explaining. Saussure was describing a social system of signs and set aside the question of how words connect to things. Dharmakīrti was explaining how concepts connect to real particulars, and his answer depends on the work those particulars do. Saussure stays inside language. Dharmakīrti is always asking whether a concept will get you to the jewel.
Zongmi’s sense of the word has its own Western parallel in negative theology, and its sharpest statement is by the Jewish philosopher Maimonides. In the Guide for the Perplexed he wrote, “Know that the negative attributes of God are the true attributes.” His illustration is a ship. One person knows only the word. Others learn, one step at a time, that it is not an accident, not a mineral, not a plant growing in the earth, not a flat object like boards or doors, not a sphere, and so on, until the tenth “has almost arrived at the correct notion.” In the same way one comes nearer to knowing God by learning what God is not. Then comes a warning very like Zongmi’s: “You must be careful, in what you negative, to negative by proof, not by mere words.” Negation, for both, is a discipline and should not be treated as a mood or a style. The difference is the goal. Maimonides held that God’s essence stays beyond human understanding however far the negations go. Zongmi held that after the negations one must recognize the thing itself, the knowing that is the nature of mind.
Who is closest, and where the likeness ends 誰最接近，相似到哪裡為止
On what exists, the closest is Locke. Both deny real universals, both make kinds a work of the mind, and both say the work is not arbitrary because it follows real likeness or real capacity in things. On how a word means, the closest is Saussure, since both define a word by what it excludes. On the Chinese sense of zhequan, the closest is Maimonides, who like Zongmi treats negation as the right way to approach what cannot be grasped, and also warns against negation for its own sake.
The likeness ends at the purpose. Locke wanted to map the limits of human understanding, and Saussure wanted a science of language. Both describe concepts and leave them in place. Dharmakīrti calls every concept mistaken, because it presents many things as one, and still useful, because it can lead to the jewel. He says the whole framework of conventional thought is set up by the wise “in order to enter the ultimate.” Concepts are a means. The practitioner uses them to reason, to teach, and to act, and comes in the end to a perception that does not need them. Neither Locke nor Saussure thought there was any such perception to come to.
Mistaken about sameness, reliable for action對「相同」錯亂；對行動可靠
True reflections of reality實在的真實反映
Useful tools of understanding知性的有用工具
Units of a system of signs符號系統的單位
Inadequate when positive正面說時都不充分
Purpose目的
To use concepts and see through them使用概念，並看穿概念
To defend the Veda and right knowledge護持吠陀與正知
To map the limits of understanding畫出知性的界限
A science of language一門語言的科學
To come nearer to God更接近上帝
5 · The step the Dharma takes further · 佛法多走的一步
A useful mistake, and the warning against negation 有用的錯誤，以及對否定的警告
The first further step is the idea of a useful mistake. Western philosophy has usually divided concepts into true and false, or into those that match reality and those that do not. Dharmakīrti put them in a third place. Every concept misrepresents, because it shows many things as one. Some concepts still lead reliably to what we want, because they track real differences in what things can do, and others lead nowhere. This is the two truths again, applied to logic. Conventional thought is not true in the final sense, and it is not to be thrown away, because it is how anyone reasons toward the final sense at all. A teacher can say this plainly to a Western student: every word you use is a simplification, and some simplifications are maps that get you home.
The second step is Zongmi’s warning, which applies to every teacher who works in English. Western audiences drawn to Buddhism often love the language of negation: no self, emptiness, not this, not that. It can become a style, a way of sounding deep, and Zongmi saw the same thing among the Chan students of his day. They valued only “not mind, not Buddha” and would not recognize the thing itself. Maimonides gave the same warning in another tradition: negate by proof, not by mere words. The lesson for a teacher is to use negation the way Dharmakīrti and Maimonides used it, as a disciplined way to remove a specific wrong idea, one at a time. It should never be a mood. After the negations there is still something to see, and the point of all the clearing is to see it.
Start with the old puzzle, which Westerners know from Plato and Locke. Point to two very different trees and ask what makes both of them trees. Most people will say they share something. Then give the Buddhist answer: they share nothing, except that neither is a non-tree, and both do what trees do. Mention Saussure for anyone who has studied linguistics, since the comparison will help them. Then tell the story of the two lights, because it carries the main point better than any argument. Every concept is a little mistaken, and the question is whether it gets you to the jewel. End with Zongmi’s salt, which is easy to remember. “Not bland” clears away an error, and “salty” is what you taste.
Buddhist logicians held that only particular things are real. There is no treeness shared by all trees.
佛教的邏輯學家主張：只有個別的東西是真實的。沒有一個被所有樹共有的「樹性」。
The word “tree” works by excluding everything that is not a tree.
「樹」這個字的作用，是把一切不是樹的東西排除掉。
What grounds the word is that different trees do the same work, as different medicines all cure a fever.
使這個字有根據的，是不同的樹做同樣的工作；就像不同的藥都能治發燒。
Every concept is a little mistaken, because it treats many things as one. Some of them still lead you to the jewel.
每一個概念都有一點錯，因為它把許多東西當成一個。其中有些仍然能帶你拿到寶珠。
Saying salt is “not bland” removes an error. Saying it is “salty” tells you what it is.
說鹽「不淡」，是去除一個錯誤。說它「鹹」，是告訴你它是什麼。
Negation is a tool for clearing away wrong ideas. It should not become a way of sounding deep.
否定是清除錯誤觀念的工具。它不應該變成一種聽起來很深的方式。
Phrases to avoid 要避開的說法
“Buddhists think words are meaningless.” 「佛教徒認為字是沒有意義的。」
They had a detailed theory of how words mean and why they work.
他們有一套詳細的理論，說明字怎麼有意義、為什麼行得通。
“Concepts are just illusions, so ignore them.” 「概念只是幻覺，所以別理它們。」
Dharmakīrti called them mistaken and reliable. The jewel story shows both.
法稱說它們是錯亂的，而且可靠。寶珠的故事兩面都顯示了。
“This is just Saussure.” 「這不過就是索緒爾。」
The likeness is real, but Dharmakīrti ties words to what things do, and Saussure did not.
相似是真的；但法稱把字連到事物所做的事，索緒爾沒有。
“The deepest teaching is pure negation.” 「最深的教法就是純粹的否定。」
Zongmi called that a mistake: negation clears the way to recognizing something.
宗密說那是個錯誤：否定是為認得某樣東西清出道路。
Three languages · 三語對照
The vocabulary, in Chinese, Pāli, Sanskrit, and English 術語：中文、巴利文、梵文、英文
中文
Pāli
Sanskrit
English
遮詮・遣餘
—
apoha, anyāpoha
exclusion; meaning by excluding the other
表詮
—
vidhi
affirmation; positive expression
自相
salakkhaṇa
svalakṣaṇa
the particular; a thing’s own character
共相
sāmaññalakkhaṇa
sāmānyalakṣaṇa
the general character; the universal
分別
vikappa
vikalpa
conceptual construction
錯亂
—
bhrānti
mistaken cognition
不欺誑
—
avisaṃvāda
non-deceptiveness; reliability
能作義
—
arthakriyā
causal efficacy; the work a thing can do
聲量
—
śabda-pramāṇa
verbal testimony as a means of knowledge
唯名論
—
—
nominalism
否定神學
—
—
negative theology; via negativa
Reading check · 理解測驗
Did it land? 讀懂了嗎
1. Why did the Buddhist logicians need a theory like exclusion?
2. According to Dharmakīrti, what grounds the concept “tree”?
3. What is the point of the two lights, one from a jewel and one from a lamp?
4. How does Zongmi illustrate negation and affirmation?
5. Where does the lesson say the likeness with Locke and Saussure ends?
Write · 你怎麼講？
Choose a word for a kind of thing, such as “chair” or “game.” Try to define it by what it excludes. Then say what work the things it covers have in common. About 150 words.
Name a concept you use every day that is a useful simplification. When does it lead you to the jewel, and when does it lead you to the lamp?
說出一個你每天使用、而且是有用的簡化的概念。什麼時候它帶你拿到寶珠？什麼時候它帶你跑向燈光？
Zongmi and Maimonides both warn against negation for its own sake. In 200 words, describe a time when saying what something is not was more honest than saying what it is, and a time when it was an evasion.
