What This Concept Is About
You may have encountered expressions like 生生世世 (life after life) or 生生不已 (ceaseless arising) in Buddhist scriptures, and assumed that 生生 simply means the continuous flow of life, one rebirth after another. But the 生生 discussed here is another 生生 — one that is more fundamental, more abstract, and more vexing than 生生 in the sense of rebirth.
Here is how it arises.
Any 有為法 (saṃskṛta-dharma, conditioned phenomenon) — that is, anything that is produced and subject to arising and ceasing — is analyzed by Buddhism as having three basic characteristics at the moment it arises: 生 (utpāda, arising), 住 (sthiti, abiding), and 滅 (bhaṅga, ceasing). 生 means it has appeared; 住 means it persists; 滅 means it dissolves. These three are called the 有為三相 (the three characteristics of the conditioned).
Here is the problem. Are 生, 住, and 滅 themselves conditioned dharmas?
If they are, then 生 itself is also within arising and ceasing, and it too should have its own 生, 住, and 滅. But what about its 生? What is that "production of production" — namely, 生生 (jātijāti, "production of production")? Is it also a conditioned dharma? If so, it needs yet another "production of production of production" to produce it… and so on without end, never finding a first term.
This is a classic problem in Buddhist philosophy: the "infinite regress" problem. It is not merely a logical puzzle; it touches a fundamental question: when a thing is about to begin to exist, where does this "beginning" itself come from?
Within Buddhism, two major camps give diametrically opposed responses to this problem.
說一切有部 (Sarvāstivāda) — the school represented by the 《俱舍論》 (Abhidharmakośabhāṣya) — proposes this solution: to each of 生, 住, and 滅, assign a 隨相 (anulakṣaṇa, secondary characteristic). 生 is paired with 生生, 住 with 住住, 滅 with 滅滅 (in the Sarvāstivāda four-characteristic system, 異 (jarā, "decay") is paired with 異異, making four 本相 (mūla-lakṣaṇa, primary characteristics) and four secondary characteristics). 生生 means "the production of production" — it is the dharma that allows the primary characteristic 生 itself to arise.
The 《俱舍論》 states this very clearly: when one dharma arises, nine dharmas arise together with it — the dharma itself, the four primary characteristics (生, 住, 異, 滅), and the four secondary characteristics (生生, 住住, 異異, 滅滅). Among these, the primary 生 can produce the other eight dharmas, while the secondary 生生 can produce only the single dharma 生 among the primary characteristics. The treatise uses a simile to illustrate this difference in function: the primary 生 producing eight dharmas is like a hen that can hatch many chicks; 生生 producing only one dharma is like a hen that hatches but one chick. Both are real functions, differing only in breadth.
This leads to the core definitional statement of the 《俱舍論》: 「生能生所生，非離因緣合」— the function of the 生 characteristic lies in producing the dharmas that are produced, but this "ability to produce" is not self-subsistent; it is established only in dependence on the convergence of causes and conditions.
The 中觀 (Mādhyamaka) school — represented by the 《中論》 (Mūlamadhyamakakārikā) and the 《十二門論》 (Dvādaśamukhaśāstra) — fundamentally questions this solution. Nāgārjuna's refutation is extremely subtle, advancing on several levels.
On the first level, he attacks the direction of "mutual production." The Sarvāstivāda setup is: 生生 can produce the primary 生, and the primary 生 can also produce 生生 — the two produce each other. Nāgārjuna asks: if 生生 is produced from the primary 生, then as "that which is produced," how can it in turn produce the very 生 that produced it? That would be like a child who has not yet been born somehow being his own father. Conversely: since the primary 生 comes from 生生, on what grounds can it produce the 生生 that produced it? The verse says:
「若謂是生生，能生於本生；生生從本生，何能生本生？」 若谓是生生，能生于本生；生生从本生，何能生本生？ *"If it is said that 生生 can produce the primary 生 — but 生生 arises from the primary 生; how can it produce the primary 生?"*
On the second level, he attacks via "not yet produced." Even setting aside the question of which produces which first, just ask: has 生生 itself arisen? If it has not yet arisen, it has no 自體 (svabhāva, self-nature); without self-nature, what does it use to produce the primary 生? The 《中論》 verse presses hard:
「若生生生時，能生於本生；生生尚未有，何能生本生？」 若生生生时，能生于本生；生生尚未有，何能生本生？ *"If, while 生生 is being produced, it can produce the primary 生 — but 生生 does not yet exist; how can it produce the primary 生?"*
"While 生生 is being produced" — this seemingly subtle moment — is precisely the most incoherent: it is in the process of being produced, meaning it has not finished being produced; not finished, it has no complete self-nature; no self-nature, no capacity to produce. How can something not yet formed bring about something else?
The 《十二門論》 also records a rejoinder: the opponent offers the simile of a lamp. When a lamp is lit, it illuminates both itself and other things. Likewise, 生生, while being produced, can produce both itself and the primary 生. Nāgārjuna's reply is four characters: 是事不然 ("this is not so"). Why not? Because a lamp can illuminate itself and others precisely because the lamp and its light already exist simultaneously; but "生生 while being produced" means it has not yet completed itself — this is not the same as "the lamp has already been lit." The lamp illuminates because it has already ignited, not because it is "in the process of igniting"; if it has not yet ignited, it cannot even illuminate itself, let alone others.
On the third level, he summarizes into a dilemma: if every production requires yet another production to produce it, then we have infinite regress, never reaching a first term; if production does not require another production and can produce itself, then all dharmas should be able to produce themselves — but clearly this is not the case. A dilemma.
The Mādhyamaka conclusion is therefore: the mutual-production architecture of 生生 and the primary 生 does not hold. The characteristic 生 is not, as the Sarvāstivādins imagined, a real entity that can itself be produced by another 生.
The two schools differ in their commitments: the Sarvāstivādins want to uphold the reality of the three characteristics 生, 住, and 滅, and so posit 生生 to stop the regress; the Mādhyamikas want to refute the reality of the three characteristics, and so use the regress to break 生生. But on one point both sides agree: without 生生 to underwrite it, 生 itself cannot stand.
This seemingly hyper-technical debate actually concerns a very deep intuition: how does "beginning" itself begin?
A Walkthrough in Everyday Life
Let us pull the lens back from the treatises to everyday life.
Scene one: you set a goal for yourself.
"From tomorrow on, I will get up at six and go running every day." — This is a 生. A new habit, a new self-appointment, has arisen in you. This 生 itself also needs a 生生 to drive it — that is, the power that makes you "begin."
The question is: where does this 生生 come from?
You say: I have willpower. But doesn't willpower itself also need to be activated? You cannot first have willpower and then use it to activate willpower — in that case, the very act of activating your willpower would require yet another willpower to activate it… It is like trying to install a "start button" for yourself, but the button itself also needs to be pressed. You either have an infinite chain of buttons, or a button that does not need to be pressed — but then it would not be a button.
So in reality, when a goal collapses, it is not because you "lack willpower" — it is because the 生生 that would have activated willpower never arose. It is something more fundamental than willpower; it has no cause, no prior step; it is simply that one moment. If that moment does not come, everything after it is just spinning in neutral.
Scene two: scrolling on your phone.
You pick up your phone and open a short-video app. The first video appears. It hooks you; a second, a third… an hour passes. Which one is the 生 of "you started scrolling"?
Is it the moment of opening the app? Yes. But that moment of "opening the app" itself also needs a 生 — the impulse that made you "decide to open the app." That impulse also has its 生 — the one that made you "decide to decide"… At this point you realize that asking "what was the very first beginning" is much harder than asking "how long have you been scrolling."
Worse, you often feel "just one more and I'll stop" — this "stopping" is also a 生, and it too needs its 生生. And that 生生 never comes. It does not come because you do not want it to come; it does not come because "wanting it to come" is itself just another thought, and it too must be produced. The infinite-regress dilemma plays out for real in every "just one more."
Scene three: you tell someone a joke.
The punchline lands on the last line. But for the punchline to land, the setup must already be fully in place. You cannot deliver the punchline first and then set it up — that would not land. And you cannot wait until all the setup is complete and then "begin" to deliver the punchline — because the act of delivering requires a "now" in which to happen. Where does this "now" come from?
The question sounds absurd, but it is structurally identical to the 生生 dilemma: a process needs a starting point, and the starting point is not itself within the process. If you treat "delivering the punchline" as a process that needs preparation, then it too needs its own "delivering the punchline" — and you would be preparing forever, and it would never land. In fact, the moment the punchline lands is precisely the moment when you do not prepare the "delivering" but just deliver it — it does not need another "delivering" to push it.
This is where 生生 sits in life: it is not something that can be prepared, manufactured, or derived. It is that one moment of "coming directly." The more you try to find a cause for it, the less it comes; the less you look, the more it comes.
The most common place to misproject here is:
People often think 生生 is "the power that makes things begin" — such as resolve, motivation, willpower. But strictly by the definition in the treatises, 生生 is the production of 生 itself — it is second-order, not the first-order motive power itself. It is like resolving to go running — "making the resolution" is a thought, and it too must be produced. You cannot first "resolve to make a resolution" and then make the resolution. That moment of "directly having resolved" is 生生 — it has no cause; it is simply that one moment. You think it is your willpower at work, but willpower itself is only something that gets produced; the moment that actually produces it is precisely where willpower has no jurisdiction.
Why Contemporary People Need It
One: to recognize that "beginning" cannot be derived.
Contemporary people live in an era where everything must have a "cause," a "method," a "path." You ask "how do I start running," "how do I start writing," "how do I start changing," and the answers you get are always a pile of methods: start with five minutes, set a small goal first, first… Do these methods work? They work. But you will find that the more methods you use, the better you get at "preparing to begin" and the worse you get at "beginning." You hoard vast amounts of knowledge about "how to begin" and never begin.
The concept of 生生 tells you: beginning, structurally, cannot be driven by another "beginning." This is a logical impossibility, not a deficiency of method. What you need is not more "how to begin," but the recognition that "beginning" itself is uncaused, unconditional, unprepared — it can only come by itself. This recognition itself is a kind of liberation.
Two: to understand the predicament of self-reference.
The problem of 生生 is, in essence, the problem of self-reference: can a system explain its own origin? Can a thought think its own beginning?
Contemporary AI, algorithms, and systems theory keep hitting this wall. Can an AI model "improve itself"? Theoretically it can recurse, but each recursion requires an externally set objective function — that objective function is where 生生 sits: it is not generated by the system itself. Can an algorithm "self-initialize"? No — initialization must be performed externally. You cannot write code that says "I start myself," because the act of starting is not inside the code.
These contemporary technological predicaments are structurally identical to the logic of Nāgārjuna's refutation of 生生 two thousand years ago: a system cannot complete an account of its own origin within itself. What is distinctive about the Buddhist approach is that it does not try to solve this predicament; instead, by revealing its necessity, it points to a deeper insight: the starting point is not inside the system, and not outside the system either; rather… 生 itself is a phantom without self-nature.
Three: to loosen the fixation that everything needs a reason.
The ultimate conclusion of the 生生 debate — at least from the Mādhyamaka angle — is not "we need to find a 生生," but "we do not need 生生, because 生 itself is not substantially real." This conclusion is extremely radical, but also very healing.
We live in an age of over-rationalization. Everything must be given a reason. Love must have a reason, hatred must have a reason, choice must have a reason, existence must have a reason. But have you considered that the act of "giving a reason" itself also needs a reason? If it does not, then is it not itself a 生生? — a reason that needs no reason.
The wisdom of Buddhism is not to make you a person without reasons; it is to let you recognize: the most fundamental acts of life — arising, beginning, existing — need no reason. They are their own reason. This recognition is more direct than any project of "searching for the meaning of life."
Common Misreadings and Clarifications
Misreading one: "生生" means rebirth, life after life.
This is the most widespread and the most off-target misreading. In the treatises, 生生 is a strictly technical term, referring specifically to the secondary characteristic of the 生相 (utpāda-lakṣaṇa, characteristic of arising) — that is, the production of 生 itself. It is not about sentient beings being reborn life after life; it is about the structural question of how "arising" itself arises in the very instant any dharma arises. If you read the verses of the 《中論》 with "life after life" in mind, you will find them completely unintelligible.
Misreading two: "生生" means "endless cycling"; the scriptures are presenting a circular argument.
Some say that 生生 and the primary 生 producing each other is a circular argument, and the Buddhist texts themselves cannot get around it. This is inaccurate. Mutual production is precisely the opponent's position that the treatise sets up in order to refute it — it is not the treatise's own claim. The Sarvāstivādins set up mutual production to avoid infinity; the Mādhyamikas point out that mutual production itself is equivalent to infinity — because each "production" within the mutual relation still needs to explain how it itself is produced. A circular argument does not solve the problem; it merely hides it.
Misreading three: the Sarvāstivādins established 生生 simply in order to declare 生, 住, and 滅 to be substantially real.
This is only half right. The Sarvāstivādins do indeed hold that 生, 住, and 滅 are real dharmas, and therefore need secondary characteristics to mark the conditionedness of the primary characteristics themselves. But the 《俱舍論》's formulation is more nuanced: 「生能生所生，非離因緣合」— although the function of the 生 characteristic is real, it does not exist independently of causes and conditions. So the Sarvāstivāda position is not simply "生 is real"; it is "the function of 生 is a conditioned reality" — the function is real, but not separable from dependent origination. This differs from the stance of the 經部師 (Sautrāntikas), who hold that characteristics such as 生 are fundamentally empty, mere conventional designations.
Misreading four: the Mādhyamaka refutation of 生生 is meant to say "nothing exists."
This understands the Mādhyamaka "refutation" as nihilism. When Nāgārjuna refutes 生生, he refutes "生 has 自性 (svabhāva, self-nature)" — that is, that 生 can establish itself independently without depending on other conditions. He is not saying "there is no 生"; he is saying "生 is dependently originated, not self-natured." The result of the refutation is not "nothing" but 空 (śūnyatā, emptiness) — emptiness is not absence, but the absence of independent self-nature. When the lamp simile is refuted, it is not that "there is no lamp in the world"; it is that "the lamp's self-illumination and other-illumination cannot be established by the lamp as an isolated thing on its own."
Misreading five: understanding "生生" as "生生不息" (ceaseless production).
生生不息 is a term from the context of Chinese philosophy, referring to the cosmos ceaselessly creating and continuing. The 《周易》's 「生生之謂易」 speaks of the unceasing continuity of change. This is entirely different from the "secondary characteristic that allows the 生 characteristic to arise" in the Buddhist scriptures — that concept which is logically impossible to establish and is repeatedly refuted by Nāgārjuna. The former affirms the continuation of life; the latter reveals the self-grounding predicament of 生 itself. To conflate them is to treat "proof" and "disproof" as the same thing.
The concept of 生生, repeatedly analyzed and refuted layer by layer in the two treatises, points ultimately not to an answer but to an insight: 生 itself cannot ground itself. Every beginning, before it begins, has no "self" with which to begin; after it begins, it no longer needs to begin. That moment of "currently beginning" is precisely the place that is neither on this shore nor on the far shore.
Look at your life again with this lens: what you are doing right now, your existence at this moment, everything about you right now — their 生 does not need another 生 to drive them. They have simply arisen. You do not need to find a deeper reason for your own existence, because the very act of "looking for a reason" is itself an arising that needs no reason.
This is what 生生 ultimately tells you: that which you think needs to be explained is precisely what cannot be explained; and that which cannot be explained is precisely itself.