Law 19 does something the rest of the book almost never does. Before any case, under the heading "OPPONENTS, SUCKERS, AND VICTIMS: Preliminary Typology," Greene stops and enumerates.
"In your rise to power you will come across many breeds of opponent, sucker, and victim. The highest form of the art of power is the ability to distinguish the wolves from the lambs, the foxes from the hares, the hawks from the vultures."1
Then five named types, "the five most dangerous and difficult types of mark in the jungle, as identified by artists—con and otherwise—of the past."2
That framing matters. This is not Greene's observation; he presents it as practitioner lore, inherited from confidence men, which makes it a compiled field guide rather than a theory.
The Arrogant and Proud Man. "Although he may initially disguise it, this man's touchy pride makes him very dangerous. Any perceived slight will lead to a vengeance of overwhelming violence." And the instruction is unusually blunt: "There is no sanity behind his overreaction, so do not waste time trying to figure him out. If at any point in your dealings with a person you sense an oversensitive and overactive pride, flee. Whatever you are hoping for from him isn't worth it."3
The Hopelessly Insecure Man. Related but "less violent and harder to spot." His ego is fragile; if he feels deceived, the hurt simmers. "He will attack you in bites that will take forever to get big enough for you to notice. If you find you have deceived or harmed such a man, disappear for a long time. Do not stay around him or he will nibble you to death."4
Mr. Suspicion. "A future Joe Stalin. He sees what he wants to see—usually the worst—in other people." And then the surprise: "Mr. Suspicion is in fact the least dangerous of the three: Genuinely unbalanced, he is easy to deceive, just as Stalin himself was constantly deceived. Play on his suspicious nature to get him to turn against other people."5
The Serpent with a Long Memory. "If hurt or deceived, this man will show no anger on the surface; he will calculate and wait." Recognisable by calculation and cunning across the different areas of his life, and usually cold and unaffectionate. "Either crush him completely or get him out of your sight."6
The Plain, Unassuming, and Often Unintelligent Man. The one that reads as an easy target and is not. "Falling for a ruse often takes intelligence and imagination—a sense of the possible rewards. The blunt man will not take the bait because he does not recognize it. He is that unaware." The danger is not revenge but waste: "he will waste your time, energy, resources, and even your sanity in trying to deceive him."7
Read the five together and they are not five points on one scale. They resolve onto two dimensions, and the taxonomy is better than Greene's presentation of it.
How they retaliate — immediately and violently (the Proud Man), slowly and cumulatively (the Insecure Man), or coldly and eventually (the Serpent).
How deceivable they are — Mr. Suspicion is easy, the Blunt Man is impossible, and everyone else is somewhere between.
Those axes are independent, and the useful predictions come from the combination. Mr. Suspicion is the least dangerous precisely because high paranoia and high deceivability co-occur — a man who sees the worst everywhere can be pointed at anyone. The Blunt Man is dangerous for the opposite reason: undeceivable, and therefore a sink for effort rather than a threat.
Greene states the resolution once — "Mr. Suspicion is in fact the least dangerous of the three" — and never generalises it. The chapter has a two-dimensional model and presents it as a list.
Line up what you are told to do about each type, and something becomes visible that the framing conceals.
Proud Man: flee. Insecure Man: disappear for a long time. Serpent: crush him completely or get him out of your sight. Blunt Man: continue at your own risk.
Four of five prescriptions are withdrawal, and the fifth is a warning.
Only Mr. Suspicion gets an offensive instruction — "play on his suspicious nature to get him to turn against other people" — and he is the one classified as least dangerous.
So a chapter presented as know who you are dealing with is, operationally, a list of people to not deal with. The value of the typology is not that it tells you how to handle these five. It is that it tells you to identify them early enough to leave, which is a much narrower and more honest claim than the framing suggests.
That also explains the chapter's placement of the Blunt Man, which otherwise looks odd. He is not dangerous in the sense the other four are — nobody is coming for you. He is on the list because the taxonomy is really about where to spend effort, and he is the largest sink.
The chapter supplies one actual test, and it comes attached to the insecure type: "To see if you are dealing with such a type, test them first—make, say, a mild joke at their expense. A confident person will laugh; an overly insecure one will react as if personally insulted."8
It is cheap, fast, and it reveals something the person cannot easily fake in the moment.
But note what it can and cannot separate. It distinguishes secure from insecure. It does not distinguish the Insecure Man from the Proud Man — both react badly to a joke at their expense — and it is useless against the Serpent, who "will show no anger on the surface" by definition, and against the Blunt Man, who may simply not follow the joke.
So the one test in the chapter covers two of five types and cannot tell them apart, and the two it misses are the ones the chapter says will destroy you slowly.
There is a deeper difficulty and Greene's own closing cautions walk straight into it.
"In judging and measuring your opponent, never rely on your instincts. You will make the greatest mistakes of all if you rely on such inexact indicators. Nothing can substitute for gathering concrete knowledge."9
And: "never trust appearances. Anyone with a serpent's heart can use a show of kindness to cloak it; a person who is blustery on the outside is often really a coward. Learn to see through appearances and their contradictions. Never trust the version that people give of themselves—it is utterly unreliable."10
Now put that against the typology. Four of the five types are defined by internal states — touchy pride, fragile ego, suspicion, a calculating memory. None is directly observable. The Proud Man "may initially disguise it." The Serpent shows nothing on the surface.
So the chapter says: classify people by their interiors, do not trust appearances, do not trust instinct, and gather concrete knowledge — without saying what concrete knowledge of an interior state would consist of.
The honest answer is available in the chapter and unstated: it is behaviour over time, toward other people. How this person responded to the last slight, how they treated someone who crossed them two years ago. That is checkable, it is not an appearance, and it does not require instinct. Greene's Serpent diagnostic gestures at it — "recognize this man by his calculation and cunning in the different areas of his life" — and the chapter does not generalise the method.
You are about to do something that will cost somebody else something.
Not a betrayal — a normal competitive act. Taking a role they wanted, arguing against their proposal, declining to back them. Something that will land as a slight whether or not you intend it.
Before you do, spend twenty minutes on one question that is not about them and is: what happened to the last person who crossed them?
You can usually find out, and the answer is worth more than any impression you have formed in a room. It is behaviour, it is historical, it is about someone who is not you, and it cannot be performed at you. Did they let it go? Did it come back eighteen months later in an unrelated decision? Did they simply become permanently unavailable to that person?
That is Greene's concrete knowledge, and it is the only version of it available for an interior state.
Then run the cheap test where it applies — mild joke, watch the recovery time, not the first reaction. Everyone bristles; the tell is whether they come back to it later.
And take the chapter's actual advice seriously, because it is the part people skip in favour of feeling strategic. Four of five prescriptions are to leave. If you identify one of these types, the recommendation is not a cleverer approach. It is that the thing you wanted from this person is probably not worth what collecting it will cost, and that judgement should be made before you are owed a response rather than after.
The typology's strength is that it is presented as compiled practitioner lore rather than theory — five types "identified by artists—con and otherwise—of the past" — which is a claim about accumulated field experience, and the types are individually recognisable.
Tension: the chapter forbids the only instruments available for using it. Do not trust instinct, do not trust appearances — and four of five types are defined by unobservable interiors. Greene demands "concrete knowledge" of internal states and never says what that would be.
Tension: the one diagnostic covers two types and cannot separate them, and is useless against the two the chapter calls most durable.
Tension: the operational content is avoidance, and the framing is mastery. "The highest form of the art of power is the ability to distinguish the wolves from the lambs" — and four of five prescriptions are flee, disappear, get him out of your sight, or continue at your own risk.
Tension: the Reversal forecloses examination. "What possible good can come from ignorance about other people? Learn to tell the lions from the lambs or pay the price. Obey this law to its fullest extent; it has no reversal—do not bother looking for one."11 Verified in the source: Law 19 is one of the six self-declared exceptionless laws (5, 10, 19, 35, 37, 43) standing against Law 48's doctrine of having no fixed anything.
🚩 SINGLE SOURCE · 🚩 CONSENSUS MISREPRESENTATION — "as identified by artists—con and otherwise—of the past" attributes the typology to an unnamed tradition with no source, practitioner, or document. The Stalin comparison is doing analytic work ("a future Joe Stalin… genuinely unbalanced") on the basis of nothing cited. [POPULAR SOURCE] · [LOW CONFIDENCE] on the typology's provenance.
Open questions. Are these five types, or are they two axes (retaliation style × deceivability) that Greene has sampled at five points — and if the second, are there occupied cells the list omits, such as the undeceivable and vengeful person? And: the types are stated as fixed dispositions. How much of what the chapter reads as character is positional — would the same person be a Serpent in one organisation and unremarkable in another?
Within Law 19 this typology is the frame and the cases do not all fit it. Norfleet is presented as the Insecure Man, and Greene's own reading is that the con "activated their self-doubt" — but Norfleet's response was neither slow nor cumulative; it was five years of relentless, effective, coordinated pursuit that destroyed the largest con confederation in America. That is not nibbling somebody to death. The type predicts a mechanism the flagship case does not exhibit.
Ford fits the Blunt Man exactly and is the chapter's cleanest confirmation.
Against Law 14's Chosroes, the pairing exposes what this chapter is missing. Chosroes had the same problem — determine an interior state — and solved it by constructing a situation that forced the interior to become behaviour. Law 19 tells you to classify without telling you how; Law 14 shows a method and never connects it. The two laws are five apart and between them contain the problem and a solution.
The convergence with The Laws of Human Nature is direct: Greene's later psychology is essentially this typology expanded into a book, with the internal mechanisms supplied. What Law 19 asserts as fixed breeds, the later work explains developmentally — which is the same author supplying, twenty years on, the why that would make the classification usable.
Psychology — Stereotype and A Priori Judgment
That page's territory is the categorisation formed before the evidence, which then organises how subsequent evidence is read.
A five-type taxonomy of people is a licensed version of exactly that operation, and the pair makes the risk precise rather than moralistic.
The typology's value is speed — it lets you act before you have full information, which is genuinely useful. Its cost is the one that page names: once someone is filed as a Serpent, ambiguous behaviour is read as calculation, and the filing becomes unfalsifiable because "he will show no anger on the surface" means the absence of anger is confirming evidence.
What the pair produces: the Serpent category is structurally unfalsifiable, and it is the one carrying the harshest prescription — crush him completely or get him out of your sight. A category defined by the absence of visible signs cannot be disconfirmed by an absence of visible signs. So the type most likely to be misapplied is the one whose misapplication is most expensive, and nothing in the chapter's method can catch it.
Which points at the only workable discipline: classify on recorded behaviour toward third parties, never on inference from demeanour — because behaviour toward others is a fact that can come back negative, and demeanour is not.
Business — Customer-Says-It Truth Asymmetry
That page's principle is that when the operator asserts something the other party discounts it, and when the other party says it themselves it registers as true — so you engineer the conversation until the conclusion comes out of their mouth.
Greene's one diagnostic works the same way and the pair generalises it into something better than either.
The mild joke does not ask whether someone is insecure. It creates a condition and reads what the person produces, which is exactly the sales page's structure aimed at assessment instead of persuasion.
What the pair produces: the same engineering that makes a conclusion stick can be used to make an interior state observable, and this is the missing method Law 19 demands and never supplies. Greene says gather concrete knowledge of unobservable dispositions. The only way concrete knowledge of a disposition exists is if the disposition is converted into behaviour, and converting it requires constructing a situation — a small cost, a minor ambiguity, a low-stakes disappointment — and watching the response.
That reframes the whole chapter's problem as solvable and identifies its ethical edge in one move: every reliable assessment of character requires imposing a small cost on someone to see what they do with it. Chosroes imposed a fatal one. The mild joke imposes almost none. The technique is continuous between them, and the chapter offers no principle for where on that continuum to stop.
Sharpest implication. The chapter demands classification by interior state, forbids instinct and appearance, and never names what concrete knowledge of an interior would be. The only answer that survives its own constraints is behaviour over time toward other people — which is available, checkable, and the one thing the typology never tells you to go and find.
Generative questions.