What This Concept Is Saying
Imagine you walk from your apartment to a coffee shop. By the time you arrive, the walking has stopped. If someone now asks, "Where is the *going*?" and points to your completed journey — to the already-walked path, the already-sat-down you — what would you say? Does the "going" still lurk somewhere in the "already-gone"? This is exactly the question Nāgārjuna (龍樹, c. 150–250 CE) presses in the *Mūlamadhyamakakārikā* (《中論》, *Fundamental Verses of the Middle Way*), specifically in its sixth chapter, the "Examination of Going and Coming" (觀去來品). The chapter opens its three-time analysis with the position called 已去 (gata) — literally "already-gone" — and this is the focus of our inquiry.
The Three-Time Framework
Nāgārjuna's chapter does not analyze "going" directly. It first partitions going into three temporal positions and then dismantles each:
- 已去 (gata) — "already-gone": the time-segment in which the activity of going has already ceased. - 未去 (agata) — "not-yet-gone": the time-segment in which the activity of going has not yet begun. - 去時 (gatasama) — "the going-time itself": the supposed present moment when going is supposedly occurring.
The whole chapter's argument lives in this triptych. Each panel is examined in turn. If "going" cannot be located in 已去, then it must be in 未去 or in 去時 — but each is examined and found equally empty. The first panel, 已去, sets the tone.
The Core Claim: "已去無有去"
The chapter's banner verse (偈) is unmistakable:
已去無有去，未去亦無去， 離已去未去，去時亦無去。 "In the already-gone there is no going; in the not-yet-gone there is no going. Apart from already-gone and not-yet-gone, in the going-time there is also no going."
This is not a denial that you *arrived* at the coffee shop. It is a precise philosophical claim: if you go looking for the *going itself* in the time when going has already finished, you will not find it. The activity (作業, *karman* as observable activity) has ceased; what remains is only its trace, not its occurrence.
Piṅgala (青目), the commentator whose exegesis is preserved in Kumārajīva's 鳩摩羅什 translation (T1564), glosses this with characteristic terseness: 「已去無有去，已去故」 — "In the already-gone there is no going, *because* it is already-gone." The reasoning is straightforward: if "going" is a process whose essence is to be *in the act of going*, then by definition it cannot be in a phase that the process has already passed through. To say "the going is in the already-gone" is grammatically and ontologically incoherent — like saying "the race is over" while the race is still happening.
The Three Aspects of Going
Nāgārjuna's analysis does not stop at the time-positions. It also dissects "going" into three component aspects, and 已去 is examined in relation to each. This triadic vocabulary is a distinctive feature of Kumārajīva's translation:
1. 去法 — the *activity* of going itself (literally "going-dharma"); the process, the action. 2. 去者 — the *goer*, the subject who performs the going. 3. 所去處 — the *destination*, the place gone to.
These three are not independent entities. The activity presupposes a goer; the goer presupposes an activity; the destination presupposes both. None of them can stand alone, and this mutual conditioning (因待, *pratītya*) is the structural backbone of the whole chapter.
The Opponent's Counter-Argument
A skilled opponent does not concede easily. The classical interlocutor in this chapter argues as follows:
動處則有去，此中有去時， 非已去未去，是故去時去。 "At the moving place there is going; in it there is going-time; it is not already-gone or not-yet-gone; therefore going-time has going."
The reasoning: in 已去, observable activity (作業) has already ceased — we can no longer catch the going in the act. In 未去, observable activity has not yet begun — there's nothing happening. So *mustn't* the going actually be happening in 去時, the present moment of motion? This is intuitively compelling: when you watch someone walking, the "walking" is visibly present *now*, not before or after.
Nāgārjuna has two replies, and both bear on 已去 as well.
The First Refutation: Two Goings in One
If "going-time" (去時) is something separate from the "activity of going" (去法), then we face a problem. Where is the going? In the activity? In the time? If the going is in the activity, then the time is going too — because the time contains the activity. If the going is in the time, then the activity is going too — because the activity occupies the time. Either way, we end up with two goings: one that the time has, and one that the activity has. This is the "two goings" (二去) fault.
The chapter makes this explicit in a verse:
若言去時去，是人則有咎； 離去有去時，去時獨去故。 "If one says going-time has going, that person has a fault: apart from the going there is a going-time, so going-time goes alone."
A "going-time" that goes alone, without a going, is a meaningless concept — like a "container" with nothing in it, or "north of the North Pole." This is why the verse continues to say that going-time and the activity of going mutually condition each other; neither can be isolated.
The Second Refutation: No Goer Without Going
If the goer (去者) is taken to be an independent entity, we hit a deeper problem. Piṅgala's analysis:
去者則不去，不去者不去； 離去不去者，無第三去者。 "The goer does not go; the non-goer does not go. Apart from goer and non-goer, there is no third goer."
A "goer" who is not currently going is a non-goer. A non-goer doesn't go. So what's actually doing the going? You cannot find a third entity — a "goer-in-itself" — that is neither currently going nor currently not-going. The goer and the going are mutually constituting; neither precedes the other. This is what the closing verses affirm:
決定有去者，不能用三去； 不決定去者，亦不用三去。 去法定不定，去者不用三， 是故去去者，所去處皆無。 "A determined goer cannot use the three goings; an undetermined goer also cannot use the three goings. Whether the going is determined or undetermined, the goer cannot use the three. Therefore going, goer, and destination are all without [inherent existence]."
The "three goings" (三去) here are precisely 已去, 未去, and 去時. Both possibilities — a goer with fixed nature, or a goer without fixed nature — fail to make sense of these three times. Hence all three (going, goer, destination) lack self-nature (無自性, *niḥsvabhāva*).
The Arising (發) Probe
Nāgārjuna then adds a further probing question, asking whether "arising" (發) — the initiation, the first kick of the process — can be found in any of the three times:
已去中無發，未去中無發， 去時中無發，何處當有發？ "In already-gone there is no arising; in not-yet-gone there is no arising; in going-time there is no arising; where would there be arising?"
If you cannot find the *arising* of going anywhere — not before, not during, not after — then going has no beginning. Without a beginning, it cannot coherently exist. And without going, no goer.
Extension to All Motion and Stillness
A particularly important move comes near the chapter's end. The opponent, seeing that "going" has been thoroughly deconstructed, tries to retreat to "staying" (住) — perhaps a still point can be located?
去未去無住，去時亦無住， 所有行止法，皆同於去義。 "In already-gone and not-yet-gone there is no staying; in going-time there is no staying. All dharmas of moving and stopping share the same meaning as going."
The same prasaṅga (reductio ad absurdum) that dissolved "going" now dissolves "staying." And the conclusion is universalized: every dharma of motion and stillness (行止法) is subject to the same analysis. This is not just about walking to a coffee shop. It is about every process, every transition, every act of doing and being.
The Method: Praṣaṅga (Reductio)
Throughout, Nāgārjuna uses the prasaṅga method (歸謬法): he does not state his own positive thesis but draws out contradictions from the opponent's position. Each time the opponent asserts that something (going, goer, destination, arising, staying) has its own nature (自性, *svabhāva*), Nāgārjuna shows this leads to absurdities — two goings, three goers, infinite regress, mutual exclusion. The point is not to demolish going but to show that going, like all phenomena, exists only as a designated, mutually conditioned designation (假名因待) — not as an independent thing.
Life Walkthrough
Let us walk through 已去 in the textures of contemporary life. The temptation in each scene is to find a substantial "going" in the already-finished event; the teaching is to see how that search cannot, on close inspection, succeed.
Scenario 1: The Regret Loop at 2 a.m.
You lie awake replaying an email you sent six months ago. The phrasing. The cc list. The implicit insult. Each time you re-enact the sending, you feel the same cringe — as if the going were *still happening*. Your mind searches the past event for the going itself, expecting to find the activity in the memory.
Mapping the classical categories: this is 已去中作業已滅 — the observable activity has ceased. You are not in fact re-sending the email; the email is sitting in someone's archive. The "going" you keep finding in the replay is constructed *now* by your searching mind; it does not inhere in the past event. Piṅgala's gloss applies: the already-gone cannot carry a present-going within it. What you are actually doing in 2 a.m. is generating *new* mental activity (a new kind of going), but it is going-on *now*, not then. The past event does not contain the going; your present re-searching does.
The classical trap, and the easiest misstep: assuming that because the past event *feels* still vivid, it must still contain the going. Vividness is in the present grasping-mind, not in the past event itself.
Scenario 2: The Project That "Should Have Worked"
Two years ago, you left a job. You still occasionally run the simulation in your head: if I had stayed, if I had pitched the other way, if I had read the room better. The "going" (the leaving, the choice) is fixed in time, yet you keep trying to add a new going to it.
This is the search for going in 已去 made systematic. The leaving happened. The trace remains (a severance check, a forwarding address, a friend who is still there). But each "if only" tries to inject a fresh activity into an already-ceased event. Each attempt produces, again, *new present mental activity* — but mislabels it as a recovery of the past going. The chapter's diagnosis: there is no going to be recovered. The leaving and the leaving cannot be reunited; they are separated by the very nature of 已去.
A common map-error: believing that more research into the past (more "if"s, more rumination) will eventually find the going. But the structure of the analysis says no — not because the past is hidden, but because the search itself cannot be in what it searches.
Scenario 3: The Ex Who "Already Ended"
A relationship ended eighteen months ago. You still occasionally drive past the old apartment. Sometimes you catch yourself searching the building, the street, the remembered corner of the café, for the *relationship itself* — expecting to find "us" there.
The classical structure is exact. The going (the relationship's living activity) has ceased. The destination (the apartment, the corner) is still there, but it is not the same as the going. The goer (you-as-in-the-relationship) is not the same as you-as-now. You are doing a new going (driving, remembering) and projecting it back onto an already-gone event, expecting to find the old going resident there. The chapter says: 離去法去時不可得 — apart from the activity of going, no going-time can be found. Apart from the activity of *being-in-that-relationship*, no "us-in-2019" can be found anywhere now.
The map-error: treating the destination (the place, the face) as if it contained the going (the relationship). But the destination is conventionally still there; the going is not in it. Only your present searching is.
Scenario 4: The Algorithm Analogy — Caveats
A digital analogy, used carefully. When you send a message, the "sending" event is finished. The server logs a timestamp. The packet is delivered. The "sending" itself is not present anywhere in the system — only a record, an effect, a downstream propagation. If your mind searches for "the sending" in the logged record, it will not find it; it will find only markers that the sending occurred.
This is structurally similar to 已去中作業已滅 — the activity itself has ceased; only traces remain. But note the limit of the analogy: a digital record is a stable token. A human past is not a server log. Memory is reconstructive; the past is partly constituted by present remembering. The Buddhist analysis is concerned with this very activity of present constitution — something the analogy cannot capture. Algorithms are a useful mirror for the structure; they are not a proof that 已去 is true.
Scenario 5: The "Already There" Memory on Returning Home
You return to your childhood home. Walking through the rooms, you keep searching for the *childhood itself* — expecting it to be sitting somewhere, perhaps in the corner where you used to read. But the corner has new paint. The window is the same size, but the light is different. You do not find the childhood.
This is precisely the search for going in 已去, now spatial. The going (growing up, becoming an adult, leaving) is finished. The destination (the house) is still there. But the going is not in the destination. The chapter's universalization — 所有行止法，皆同於去義 — applies: all processes of motion and stillness, including the slow processes of growing and leaving, follow the same logic. The house remains; the growing is not in it.
The map-error: thinking the destination preserves the going. Conventionally, places carry memory. Philosophically, going does not inhere in destination.
Why Contemporary People Need This
The Three Forms of Time-Suffering
Contemporary life is structured around three recognizable forms of suffering that map directly onto the three-time analysis:
- 已去 (past-regret): rumination, replay, trauma loops, "I should have said..." Searching for going in already-gone events. - 未去 (future-anxiety): anticipatory worry, planning paralysis, "what if..." Searching for going in not-yet-happened events. - 去時 (present-fetishization): productivity metrics, mindfulness apps, "being in the moment" as performance. Searching for going in the supposed present.
已去 is the most under-theorized of the three in popular discourse. We have a whole genre of future-anxiety literature and an industry around present-moment mindfulness, but the *past-regret* category — the way we treat finished events as if they still contained the activity that finished them — receives less precise analysis. Nāgārjuna's 已去 gives it a precise diagnostic: the going is not there; what is there is the present grasping-mind re-staging the past. Recognizing this is a different, and often more useful, intervention than affirmations to "let go of the past."
Identity Construction from the Past
A great deal of contemporary identity work — résumé narratives, life-storytelling in therapy, "personal brand" curation — involves treating the already-gone as a substrate with its own substance. "I am the sum of my past actions." But Nāgārjuna's analysis says that the "past action" cannot be located as an independent going in any past moment. What you actually are is the present *naming* of past events, the present *interpretation* of past traces. The "I" that is constructed is constructed *now*, not then. This is not a denial of biography; it is a clarification of how biography works.
Algorithmic Time and the "Eternal Past"
Digital systems create strange temporal artifacts. Once data is "sent" or "posted," it remains indefinitely accessible. The going (the posting) is finished, but the destination (the platform) preserves the trace with unusual fidelity. This can reinforce the illusion that 已去 contains the going — that the past event is somehow *still* present.
Nāgārjuna's analysis helps here too: the digital trace is the destination's preservation, not the going's preservation. The "sending" is not in the server log; the server log is just a place where certain information is now stably stored. Confusing the two — confusing the trace with the activity — is a contemporary form of the same error the chapter diagnoses. (The limits of this analogy: human memory is reconstructive in ways server logs are not; the analogy illuminates structure, not phenomenology.)
Trauma, Memory, and the Mind That Won't Stop Searching
Trauma can be described, with proper care, as a mind that searches for going in 已去 and cannot stop. The event is over; the activity is finished; but the searching continues, never finding the going, never accepting the absence. Cognitive-behavioral framings talk about "rumination" and "intrusive recall"; Nāgārjuna's analysis gives a more precise diagnosis: you are looking for the activity in a place where the activity, by definition, cannot be. The intervention is not to "find" the going more thoroughly; it is to recognize that the search is misdirected.
(This should not be mistaken for a clinical prescription. The classical analysis is philosophical; clinical work with trauma requires professional care. The parallel is structural, not therapeutic.)
Seeing the Mutual Conditioning of Self and Action
Finally, the chapter's mutual-conditioning logic — going and goer mutually constituting each other — is directly relevant to contemporary questions about agency. We often debate whether "the person" or "the situation" causes an action, as if one preceded the other. Nāgārjuna's analysis says neither precedes: the goer and the going arise together, mutually, and have no independent self-nature. This is not a denial of agency; it is a denial of *agent-as-independent-entity*. We act, and in acting, we are the actors. Neither pole of the equation stands alone.
Common Misreadings and Clarifications
Misreading 1: "Nāgārjuna denies that you actually went to the coffee shop."
This is the most common Western mistake. The whole point of the prasaṅga method is that *you did go*. Going exists as a conventional designation (世俗諦). What is denied is not the fact of going, but the claim that "going" has its own self-nature (自性) — that it exists as an independent, intrinsically characterized process apart from its conditions, its goer, its destination, and its designation. Nāgārjuna is not a skeptic about the world; he is an analyst of how the world is *constituted*.
Misreading 2: "已去 means the past doesn't exist."
No. The past has conventional reality. What 已去 reveals is that the past does not exist *as a self-standing thing*. Past events are designated, mutually conditioned, and depend on present interpretation. The denial of self-nature (自性) is not the denial of existence (有); it is a much more specific claim about the *mode* of existence.
Misreading 3: "This is just a verbal trick about tenses."
The argument is not about the word "go" but about the *phenomenon* of going. Nāgārjuna is asking: when going is examined, what is found? The analysis does not depend on English or Chinese grammar; it depends on the conceptual analysis of what it means for something to be "happening" in a particular time-segment. The argument would survive translation into any language and any tense system.
Misreading 4: "The chapter is just about literal walking."
The closing verse explicitly universalizes: 所有行止法，皆同於去義 — *all* dharmas of moving and stopping follow the same logic. The example of going is a paradigm case for analyzing *all* processes, *all* motion, *all* activity, *all* identity-formation-in-time. The chapter's reach extends far beyond locomotion.
Misreading 5: "This proves nothing ever happens."
A frequent nihilistic misreading. The chapter proves that nothing happens *with inherent self-nature*. Things happen — extensively, thoroughly, empirically — but they happen as designated, conditioned, mutually-arising phenomena, not as independent events. This is a different claim from "nothing happens," and the chapter takes great care not to slide into nihilism. Nāgārjuna is at pains in other parts of the *Mūlamadhyamakakārikā* to distinguish emptiness (空) from nothingness (無); this chapter's analysis of 已去 is part of that larger project.
Misreading 6: "已去 is a meditation on the past."
The point is not to "meditate on" the past or to remember more vividly. The point is to see through the habit of treating past events as if they still contained the activity that produced them. This is *not* an exercise in recollection; it is an exercise in recognition — seeing the structure of the search itself, and finding that what is searched for cannot, on close inspection, be there.
Misreading 7: "The three-time framework divides time into three pieces."
No. The three "times" are not three regions of a timeline. They are *positions* defined by the activity's relation to itself: the activity has ceased (已去), has not begun (未去), or is supposedly in progress (去時). They are conceptual poles for analysis, not temporal containers. Time itself is not being divided; the analysis is of how a process is conceptualized across its conceptual phases.
Misreading 8: "The goer doesn't exist."
The goer exists conventionally — there is someone walking, sending the email, leaving the relationship. What doesn't exist is the *goer-as-independent-entity*, the goer that can be located apart from the going. The mutual conditioning of goer and going does not deny either; it denies the independent self-nature of either. This is the heart of the prasaṅga method's precision.
A Final Note on the Larger Position
已去 is not a standalone teaching. It is one panel in a three-panel triptych, and the triptych itself is one chapter in a much larger work. Nāgārjuna's purpose in the *Mūlamadhyamakakārikā* is to expose, across twenty-seven chapters, the self-nature (自性) of every category of experience — cause and effect, motion and rest, self and other, time and space, the four noble truths themselves. 已去's role is to dismantle our most basic intuitions about activity in time, because those intuitions underlie every other conceptual fabrication. If we cannot find going in the going, we cannot find self-nature anywhere; and if we cannot find self-nature anywhere, we are pointed toward the realization that liberates.
The chapter does not ask us to *believe* that 已去 lacks self-nature. It asks us to *see* this — through patient, attentive analysis of how we actually constitute "going," "goer," "destination," "already-gone," "not-yet-gone," and "going-time" in our own experience. The analysis is offered as a mirror. What we see in it is up to us.