Greene gives the career in five sentences, and the compression is the point — this is a man who changed sides twice and ended up courted by everyone.
"It was Alcibiades who inspired and led the massive Athenian armada that invaded Sicily in 414 B.C. When envious Athenians back home tried to bring him down by accusing him of trumped-up charges, he defected to the enemy, the Spartans, instead of facing a trial back home. Then, after the Athenians were defeated at Syracuse, he left Sparta for Persia, even though the power of Sparta was now on the rise."1
And the result: "Now, however, both the Athenians and the Spartans courted Alcibiades because of his influence with the Persians; and the Persians showered him with honors because of his power over the Athenians and the Spartans. He made promises to every side but committed to none, and in the end he held all the cards."2
Look at the second defection, because it is the one that does not follow from self-interest as ordinarily understood.
He left Sparta for Persia "even though the power of Sparta was now on the rise."
That is the operative sentence. A man optimising for the winning side stays with Sparta after Syracuse. He left the strong horse.
Greene does not explain it, and the explanation is available in his own account of the outcome. Alcibiades' value to each party was his influence with the others — Athens and Sparta courted him for his standing in Persia, and Persia valued him for his leverage over Athens and Sparta.
That value requires him to be attached to none of them. The moment he settles anywhere, he stops being a bridge and becomes an asset of whoever holds him, at which point his price collapses to whatever a defector is worth.
So the departure from Sparta was not a bet on Persia. It was maintenance — the periodic move required to prevent any one party from acquiring him.
Elizabeth ran non-commitment across many suitors. Kissinger ran it across two. Alcibiades ran it across three mutually hostile powers, and the structure behaves differently.
With two parties, your value to each is your potential defection to the other — a single threat, and each side can price it.
With three, your value to each is your standing with the other two, and that quantity is invisible to any of them. Athens cannot verify how much influence he really has in Persia. Persia cannot verify his leverage over Sparta. Each party's estimate is generated from the fact that the others are courting him, which is the social-proof engine of Law 20 running at the level of states.
Three parties is where the estimate becomes unfalsifiable, and Greene's summary — "in the end he held all the cards" — is a description of that condition rather than of any resource he actually possessed.
Take an inventory of what Alcibiades controlled at the peak Greene describes.
No army of his own. No territory. No treasury. No formal office in any of the three states — he was an exile from one, a guest in another, an adviser in the third.
What he had was three parties' beliefs about his standing with the other two.
It is genuinely unusual, and both properties matter.
It costs nothing to hold, which is why it can be sustained by a man with no resources. Elizabeth needed a throne; Kissinger needed genuine access to a negotiating team. Alcibiades needed only to be plausibly welcome in three places.
And it is not convertible. He could not spend it — the moment he used the influence, its extent became measurable, and a measured influence is smaller than an imagined one. Which means the position, for all Greene's "he held all the cards," produced no outcome he could bank.
The account stops at the peak. The Athenians recalled him, he lost their confidence again after a naval defeat he was not present at, and he was murdered in Phrygia — none of which is in the chapter, and all of which follows from an asset that cannot be converted and must be continuously maintained.
"If you aspire to power and influence, try the Alcibiades tactic: Put yourself in the middle between competing powers. Lure one side with the promise of your help; the other side, always wanting to outdo its enemy, will pursue you as well. As each side vies for your attention, you will immediately seem a person of great influence and desirability."3
Note "you will immediately seem a person of great influence."
Greene's own verb is exact and he does not follow it. The tactic produces the appearance of influence, and the appearance is the entire product — there is no underlying capability being signalled.
Which makes this the clearest case in Law 20 of the mechanism operating with nothing behind it, and it is why Alcibiades is a better illustration of the law than Elizabeth. She had a real throne and a real marriage to withhold. He had the fact that three parties were watching each other watch him.
You are between two or three parties who are rivals, and each of them thinks you have standing with the others.
The temptation is to let that stand, because it is doing work for you and correcting it would be tedious and self-diminishing.
Alcibiades' case is worth holding here for one specific reason, and it is not a moral one. The position cannot be spent.
Test it directly: name the thing you would actually do with this influence. Not the deals it makes available — the concrete action you would take, that requires the standing to be real. If you can name one, the standing is real and this page does not apply to you.
If you cannot, then what you have is three parties' estimates of each other's estimates, and the moment you attempt to use it, the extent becomes measurable and every measured influence is smaller than an imagined one. The asset survives only while unexercised, which means it is not an asset in any sense that pays.
And notice the maintenance cost, which is the part that ended him. Alcibiades left Sparta while Sparta was rising, because staying would have converted him into somebody's man. A position made of not-being-attached requires periodic detachment, and each move burns a relationship in a world where the number of available parties is finite.
Elizabeth converted her ambiguity into a signed treaty in 1582. Kissinger converted his into a cabinet post. Alcibiades never converted, and that — not the switching — is what the case actually demonstrates.
The case's strength is its structure: three parties, two defections, and an outcome Greene states clearly enough to test. The second defection — leaving Sparta while Sparta was rising — is the detail that distinguishes this from ordinary opportunism, and it is in the text.
Tension: Greene ends at the peak. The account stops at "he held all the cards." Alcibiades was subsequently recalled to Athens, lost confidence again, and was murdered in exile — an ending that follows directly from an asset that cannot be converted, and its omission makes the tactic look like a terminal position rather than a phase.
Tension: Greene's own verb undercuts his instruction. "You will immediately seem a person of great influence" — the tactic produces an appearance, and the chapter recommends it as though it produced the thing.
Tension: the case is the law's best illustration and its worst advertisement. It is the purest instance of the mechanism running with nothing behind it, and the purest demonstration that a position made of other people's beliefs has no exit.
🚩 SINGLE SOURCE · 🚩 SECONDARY WITHOUT PRIMARY — no source is named. The material derives ultimately from Thucydides and Plutarch, neither cited, and Thucydides is quoted in this same book's margins at Law 13 without being named there either. The characterisation of the charges as "trumped-up" is a contested historical judgement presented as fact — the Athenian accusations concerning the mutilation of the Herms and the profanation of the Mysteries are debated, not settled. [POPULAR SOURCE] · [CONTESTED]
Open questions. Was the Persian influence real or reputational? Greene's account requires only that Athens and Sparta believed in it, and the case reads identically either way — which is itself the finding, but it means we cannot distinguish a skilled operator from a well-positioned rumour. And: he left Sparta while Sparta was rising. Was that maintenance, as this page argues, or flight — Greene omits that he had reportedly made himself unwelcome there, which would make the departure a necessity dressed as a strategy.
Within Law 20 the three cases form a ladder of increasing detachment from any real underlying asset, and the chapter does not notice the progression.
Elizabeth withheld a real, unique, one-shot thing — the English throne's marriage — and converted it into a signed treaty. Kissinger withheld a real service and converted it into a cabinet post. Alcibiades withheld nothing that existed independently of the parties' beliefs, and converted nothing.
The pattern: the technique's convertibility tracks whether there is an actual asset under it, and Greene presents all three as the same recommended tactic.
Against Law 15's crushing, Alcibiades is a live demonstration of the problem that law is about, from the other side. He is the enemy who was not finished — Athens accused him and he defected rather than being destroyed, and he then spent years as a resource for their enemies. Law 15 would read the Athenians' failure as insufficient thoroughness. Law 20 reads Alcibiades' survival as skill. Same event, two chapters, and the corpus never sets them against each other.
Psychology — Social Proof
That page's mechanism: relabel a dish House Recommendation and sales rise with the food unchanged, because diners substitute the crowd's apparent judgement for their own when they lack the information to judge.
Alcibiades is that operation running between states rather than diners, and the pair identifies why three parties is the operative number.
Athens could not verify his standing in Persia. Persia could not verify his leverage over Sparta. Each party's estimate was generated almost entirely from observing that the others were courting him — which is social proof with no independent evidence available anywhere in the system.
What the pair produces: the substitution is strongest exactly where verification is impossible, and international standing in the fifth century B.C. was maximally unverifiable. The restaurant diner could in principle taste the dish. Athens could not audit Persian court influence.
Which gives the condition Greene's instruction omits. The tactic scales with the opacity of the environment, not with the operator's skill — and it degrades sharply wherever the parties can compare notes cheaply. That makes it a strategy for closed, slow-communicating, mutually hostile systems, and considerably less available in any environment where three rivals can text each other.
Business — You Are the Last Defensible Moat — the Swap Test
That page's instrument: swap yourself for someone with a comparable skill set and ask what actually changes. Whatever survives the swap is the moat.
Alcibiades fails and passes the test in a way that exposes its limit.
Swap him for another well-born Athenian exile of similar ability and almost everything changes — because what he held was not a skill but a specific position in a specific network of beliefs, and a replacement would start with none of it. By the swap test, that is a formidable moat.
And it is worthless, because it cannot be exercised without being measured, and measurement destroys it.
What the pair produces: the swap test measures irreplaceability and says nothing about convertibility, and those come apart completely. A moat that no one can cross is not the same as a position that pays. Alcibiades was maximally irreplaceable and realised nothing; Elizabeth was equally irreplaceable and banked a treaty.
Which suggests the swap test needs a second question that the page does not ask: not only what changes if I am replaced, but what can I do with this that produces an irreversible result. An asset that survives only while unexercised passes the first test and fails the thing the test is for.
Sharpest implication. Alcibiades' asset was three parties' beliefs about his standing with the other two — costless to hold, impossible to verify, and destroyed by use, because an exercised influence becomes a measured one and every measured influence is smaller than an imagined one. He held all the cards and could not play any of them.
Generative questions.