Picture a man in the dark, a few feet behind a sentry. He has one job: end the watch before the watch ends him. Every instinct says throw the blade — it is fast, it crosses the gap, it ends the matter from a safe distance. Haha Lung's manual says the exact opposite, and the reason is brutally simple. A thrown knife that hits its mark still leaves a dying man with a mouth and a few seconds of air, and a dying man who can scream is a man who sounds the alarm.1 Worse, the thrower has just handed his only weapon to the enemy and stood there empty-handed. So the first rule of the close-quarters blade is a rule about not using it the obvious way: never throw your knife.2
This is not squeamishness. It is the logic of an operator who treats noise as the true enemy and his own equipment as something he cannot afford to surrender. The blade is for reaching, holding, and finishing in silence at arm's length — not for gambling across a room.
The never-throw doctrine is a small cluster of rules governing edged and strangling weapons in the close-quarters sentry-removal context Haha Lung lays out in Knights of Darkness.3 At its center sits the prohibition: a knife should never be thrown at an enemy, because even a hit is no guarantee of silence, and a throw disarms you in the same motion.4 Around that core orbit a set of tools and their proper uses — the throwing weapons that are purpose-built (and their sharp limits), the modern engineered blades, the strangling cord, and the counter-armor a smart target wears.
Lung is explicit that purpose-built throwing weapons exist, the best known being the shuriken, the so-called ninja star.5 But he immediately deflates the romance: traditional shuriken were thrown to dissuade pursuit, not to kill, and unless poisoned a small one has "harassment value only."6 The exception is improvised — a shuriken held like a razor blade can slash a throat at contact range.7 He then catalogs the modern variants: an "X-knife" of two four-to-eight-inch stainless blades that fold closed for carry and lock open into a four-pointed X that is, in his words, almost impossible not to stick into a target;8 and the Soviet Spetsnaz "shooting knife," a blade propelled several feet, originally spring-loaded and later driven by compressed gas.9 For strangulation he names the garrote — a length of cord, wire, or stick.10 And he closes the whole survey with the one truth under all the hardware: the body is always your first weapon and your last line of defense.11
Everything here runs on a single hidden currency: silence. The operator is spending it, and the target is trying to spend it back. Read the doctrine that way and every rule snaps into place.
Throwing the knife spends silence recklessly. The blade leaves your hand, and now two bad things can happen at once — the wound might not be instantly fatal, leaving a throat that can shout, and the weapon is gone whether the throw landed or not. You have bought a chance at a quiet kill by guaranteeing you are unarmed if it fails. That is a terrible trade, and the doctrine forbids it.
The garrote, by contrast, is silence made into a tool. A cord around the windpipe does not just kill; it kills the sound first. There is no edge to miss, no air for a scream, and the weapon never leaves your grip. The X-knife earns its place by the same logic from the opposite direction: it is engineered so the geometry does the aiming — four points mean one almost always bites — which lowers the odds of the half-hit that leaves a man gurgling instead of gone.12 The shuriken's demotion to "harassment only" is the same accountant talking: a tool that wounds without silencing is, for this job, nearly useless unless you tip it with poison or press it to the throat by hand.13
The final line — the body as first and last weapon — is the logic taken to its floor. Strip away every tool and the silence budget still has to be balanced with hands, an arm-bar across the throat, a bludgeon of rock or heavy stick to the skull.14 The hardware is optional. The requirement to be quiet is not.
This page hands the rest of the vault a clean, ugly principle: the weapon you commit is the weapon you no longer have, and the kill that makes noise is the kill that fails. That is a doctrine about resource commitment and signal discipline dressed up as knife-work, and it generalizes well beyond a sentry post.
To the stealth-tactics cluster it offers the close-range terminal node — where the broader stealth-weapons matrix (this batch's parent page) sorts tools by distance and noise, this page lives at the zero-distance, zero-noise corner and explains why that corner is where the work actually gets done. To the influence and persuasion pages it offers a structural mirror: the throw that disarms you is the same shape as the manipulator who fires his best leverage too early and is left with nothing. To the history pages on shadow warfare it offers a corrective — the shuriken was a deterrent, not an assassination weapon, which trims a lot of pop-culture fat off the ninja myth.15
Run the forbidden play in full and watch why Lung bans it. An operator crouches behind a sentry, blade in hand, the gap maybe ten feet. He throws. Best case, the point buries itself between the shoulder blades. The sentry staggers — and in that stagger, before the body understands it is dying, the lungs are still full and the throat still works. He shouts. The alarm goes up. The operator, now weaponless, has converted a stealth approach into an open fight he is badly equipped to win.16
That is the entire case against the throw, and it is worth sitting with because the failure is not in the aim. The throw can be perfect and still fail the mission, because the mission was never "hit the sentry" — it was "remove the sentry without a sound." A perfect throw that buys three seconds of consciousness has failed at the only thing that mattered.
Now contrast the sanctioned approach. The same operator closes the last ten feet, gets a cord or a forearm across the windpipe, and the sentry's first attempt to shout dies in a crushed throat. No blade has left his hand. If the cord slips, the knife is still there. If the knife is lost, the hands remain. Each layer falls back on the next, and at no point does the operator surrender his last option to a gamble.
Lung's own modern-tool catalog is really a series of attempts to fix the throw's flaw without abandoning reach. The X-knife's four points are an admission that a single-point throw misses too often; multiply the points and you multiply the odds of a sticking hit.17 The Spetsnaz shooting knife is the engineer's dream of having the reach and keeping a blade — but note it is a state intelligence service that built it, not a lone operator, because only an institution can afford that kind of single-purpose hardware.18 The case study's lesson is that every clever device in the chapter is a workaround for the one rule a careful man simply obeys: keep the edge in your hand.
You are behind him now. The wire mesh — you noticed it earlier, wrapped at his throat and again at the small of his back — tells you he has thought about this night too, that someone trained these men against exactly the cord and the rear blade you carry.19 So you do not reach for the knife at his spine; the mesh would turn it. You read the gap between collar and mesh, the bare inch of neck where the cord still bites.
Your weight is on the balls of your feet. The cord is already doubled in your fists, the stick threaded through for the twist. You do not rush the last steps — rushing scrapes gravel, and gravel is sound, and sound is the only thing that can beat you here. You breathe out slow. The hands stay loose until the instant they are not.
The cord goes over and crosses behind the nape in one motion, and you drop your weight rather than pull with your arms — let the body do it, let gravity spend the silence for you. His first sound is already gone; there is no air behind it. The knife stays sheathed at your hip the whole time, unspent, because the moment the cord works you never needed it, and the moment it fails it is still there. When it is done you are still armed, still quiet, and still the only thing in the dark that knows what just happened — unless, as Lung warns, you were never the only predator out here.20
You can spot a man who has lost the thread of this doctrine before he ever gets caught. The first sign is the itch to throw — the operator who treats the blade as a projectile is a man who has confused reaching with finishing, and he will end the night disarmed and exposed.21 He has forgotten that the throw spends two things at once and recovers neither.
The second sign is romance about the shuriken. A practitioner who believes the star is a killing weapon has swallowed the myth Lung explicitly punctures; he will rely on harassment tools for terminal work and be shocked when a wounded man stays loud.22 The third is tool-worship — the operator so in love with his X-knife or his exotic shooting blade that he forgets the body is the first weapon and the last. When the clever device jams, fails to deploy, or is simply left behind, the tool-worshipper has no floor to fall back to.23 The healthy operator carries the hardware and trusts the hands; the failing one carries the hardware and needs it.
The last sign is assuming you are alone. The man who never imagines a second predator wears no mesh, watches no back, and dies to the very rear-knife and strangle he was busy preparing for someone else.24
Lung's account is internally consistent but rests almost entirely on assertion — this is a practitioner manual, not a documented study, and claims like the X-knife being "almost impossible not to stick" are sales-pitch confident rather than tested.25 He cross-references his own other titles (Assassin! for edged weapons, The Ancient Art of Strangulation for the garrote) as the evidentiary backstop, which keeps the corroboration inside the same author's corpus rather than reaching outside it.26 [POPULAR SOURCE]
The one genuine tension is between the never-throw rule and the catalog of throwing tools that follows it. Lung forbids throwing the knife, then spends paragraphs on shuriken, X-knives, and a Soviet blade designed precisely to be thrown.27 He resolves this only partly — by demoting the shuriken to harassment and framing the engineered tools as exceptions that fix the throw's silence problem — but the page-level message is mixed: never throw, except with the specific hardware built to make throwing work. That tension is worth preserving rather than smoothing.
Open: the Korean War wire-mesh detail is offered as fact without citation, and whether South Korean night fighters actually standardized neck-and-back mesh against strangulation and rear blades is unverified here.28
Because this draws on a single Haha Lung source, the live tension is Lung against himself rather than Lung against a scholar. He opens with an absolute — never throw your knife — and the rest of the passage is a quiet argument with that absolute. He knows throwing is what people want to do, so he forbids it, then concedes that real throwing weapons exist, then concedes they mostly do not kill, then concedes that institutions have engineered around the problem entirely. The doctrine is not "throwing never works"; it is "throwing is so likely to fail at silence and self-disarmament that the lone operator should treat it as forbidden, and only purpose-built tools or state hardware earn an exception." Reading the prohibition as absolute misses that Lung is really teaching risk discipline under the cover of a flat rule — the flat rule is the pedagogy, the nuance is the truth.
Here is the plain version: this is a page about not throwing away your last good option, told through knives. That shape shows up all over the vault, which is what makes the handshakes earn their keep.
The first connection is to Flags: Five Weaknesses, Lung's framework for the five exploitable human levers. The structural parallel runs through the target, not the operator. The never-throw doctrine assumes a sentry who can still react — shout, alarm, fight — for as long as he is conscious, and the whole craft is built around denying him that window. The five-weaknesses model is the inverse craft: it maps the openings through which a target can be moved before any physical contact. Put them together and you get a fuller predator's calendar — Flags works the mind at distance to make the target compliant or careless, and the never-throw blade-work finishes at zero distance once the mind has been handled. The insight that neither page reaches alone: the sentry post and the con are the same operation at different ranges, and the wire mesh around the Korean fighter's throat is the physical-layer version of a man who has refused to have his five weaknesses played. He has hardened the body because he could not be sure of hardening the mind.
The second connection is to Assassins (Nizari Ismaili), the historical order whose name became the word for the deed. The parallel is doctrinal economy. The Nizari fida'i famously favored the dagger at close range over poison or ranged weapons — a choice historians read as much for its public, witnessed, unmistakable message as for its reliability. Lung's never-throw rule lands in the same place from a purely operational angle: the close blade is the controllable blade. But the reasons diverge sharply, and that is where the handshake produces something neither page holds alone. The Nizari kept it close partly to be seen — the killing was a sermon. Lung keeps it close to not be heard — the killing is a silence. Same weapon, same range, opposite relationship to noise: one assassination wants witnesses, the other wants none. That contrast turns "the close blade" from a mere tactic into a question about what a killing is for, which is a question the operational manual never thinks to ask and the historical order answers with its whole theology.
A third, lighter handshake runs to Shinobi Infiltration of Fortified Structures. The sentry-removal scene is the missing micro-step inside the infiltration macro-task: the shinobi page describes getting into the structure, and the never-throw doctrine describes the specific human obstacle — the watchman — that infiltration must dissolve without a sound. Read together they correct the shuriken myth from both ends. The infiltration literature shows the star used for distraction and pursuit-deterrence at the perimeter, and Lung shows it demoted to harassment at the kill — two independent sources agreeing that the famous throwing weapon was rarely the thing that actually did the killing. The cross-reference does the de-romanticizing that neither page would be quite as convincing doing alone.
The Sharpest Implication: The never-throw rule is the clearest statement in the Lung corpus of a principle that has nothing to do with knives — do not convert a held advantage into a gamble, and never spend your last fallback on a single uncertain outcome. The thrown blade is every premature commitment: the negotiator who plays his leverage too soon, the strategist who commits the reserve, the manipulator who reveals his best card to win a small pot. Lung's sentry teaches that the discipline is not in the strike but in the refusal to strike the wrong way.
Generative Questions: