David Bowie proved in 1997 that a person could be capitalized. The 2020 personal-token wave proved that issuing the claim is the easy part. On treasuries, the exit right, and what self-incorporation actually requires.
In 1997 David Bowie did something no musician had done before: he sold his future. The banker David Pullman packaged the royalties from twenty-five albums Bowie had recorded before 1990 – 287 songs – into a bond, and Prudential bought the whole issue for fifty-five million dollars. The coupon was 7.9 percent, the average life about ten years, and the structure was clean enough that the tax authorities treated the proceeds as a loan rather than income. For a decade, a stream of income that used to belong to one man belonged, in effect, to his backers. Then it reverted to him.
Bowie Bonds are usually filed under financial trivia. They are better read as a prototype. A person, treated as an institution, issued a claim on themselves, sold it to people who believed in the future value of what they made, and used the proceeds to buy back control of their own catalogue. The instrument was crude – a single buyer, a fixed term, no secondary market to speak of, no way for Prudential to change its mind – but the idea underneath it was durable: an individual can be capitalized the way a company is.
The idea kept trying to be born. Income-share agreements let a student sell a slice of future earnings to fund an education. The 2020 "personal token" wave let anyone mint a coin against themselves in an afternoon. On paper the 2020 version was the most complete: a creator issued a token, a community bought it, and the price was supposed to track the creator's rising fortunes. In practice it was the least complete thing anyone had built. The platform that hosted most of these tokens, Roll, held the keys; on 14 March 2021 an attacker emptied its hot wallet of almost 3,000 ETH, about $5.7 million, and the personal tokens minted on top of it fell by more than half in a morning. The tokens did not fail because the creators stopped being interesting. They failed because they were claims on nothing a holder actually controlled. There was no treasury the holders owned. There was no way to leave with your share, because there was no share. When trust in the creator or the custodian wavered, the only available action was to sell into a falling market. The instrument gave you upside and a sell button, and nothing in between.
The closest miss came on Zap, a protocol built entirely around bonding curves. There, in 2020, a personal token issued in the name of Ben Gravis became one of the first attempts to put a person on a curve rather than in a custodian's ledger. On its face this was a better instrument than the Roll tokens. A bonding curve is a standing counterparty – a holder who wanted out could sell back to the curve instead of hunting for a bid, so the liquidity the others lacked was built in. What the curve could not supply was a treasury the backers owned or an exit that bound the issuer to anything. The pool was the issuer's own arrangement, maintained at his discretion, and the whole record of what followed fits on one screen – which is the useful thing about it. The contract carries twelve transfer events in its entire history, the first on 3 May 2020 and the last on 8 April 2021, every one of them a mint from the zero address or a burn back to it, and still twelve when it was read again on 23 August 2026. Backers bonded and un-bonded against the curve exactly as it was built to let them; not once did a holder hand the token to another holder. The last thing that ever happened on it is not an issuer walking away – it is a backer bonding in, and then five years of silence. Zap had supplied the very thing the other 2020 tokens lacked, real liquidity into a counterparty that could not refuse, and it was still not enough. A counterparty is not a claim. There was nothing pooled behind the price for a holder to own, so there was nothing for the liquidity to be liquidity in.
What was missing was the thing that makes a company more than a fan club: a balance sheet the owners have a real claim on, and a rule for getting your claim back.
A caper supplies both. Strip away the setting and the mechanism is simple. A person opens a caper the way a company opens a cap table. Backers buy the caper's token along a bonding curve, and the money they put in does not vanish into a price – it fills a treasury. That treasury is not the founder's private account. It is the collateral behind the token, and every holder has a claim on it that the contract enforces directly. The founder earns an allocation as buys come in, so the person being backed is compensated for the thing being backed; but the pool the backers funded stays pooled, and stays theirs in proportion.
The rule for getting your claim back is the part the 2020 tokens never had. On a caper, a holder who has voted, and whose ballot has been released, can exit. Exit is not a sell order routed into whoever happens to be bidding; it is a right against the treasury itself. When you leave, you burn your vote and redeem a proportional share of what the treasury holds – the same proportion, exactly, that would have determined your weight in a vote. Your say in the thing and your severable stake in the thing are computed from one number. That number is also the condition, and the condition has a clock. Your weight comes from the tokens you have staked into the caper's ballot box, and the vote you later burn on the way out is minted when that proposal's ballots are released, not when you cast. Between the cast and the release the stake is held: the contract refuses to return tokens that still back an unreleased ballot. So a backer who never voted has nothing to hand back and the contract turns the exit away, and a backer who voted this morning waits for the count to close before the door opens (leaving a caper). A backer of a personal caper is never in the 2020 position of holding a claim on nothing. Once their ballots are released they can name their portion of the collateral and walk out with it.
This is what turns a personal token from a bet on someone's mood into something closer to a small institution. Albert Hirschman's old distinction is the right lens: a member of any organization has two levers, voice and exit, and the credibility of voice depends on the reality of exit. A shareholder's opinion matters partly because they can sell to someone who sets a floor under the price; a citizen's complaint carries weight partly because they can, in principle, leave. The 2020 personal tokens gave their holders voice with no floor and an exit that led only to the open market. A caper gives exit a floor made of the treasury the holders themselves funded, and ties the size of that exit to the same measure as the vote. Elinor Ostrom spent a career documenting the rules that let ordinary groups govern shared resources without a boss or a sale to outsiders; the common thread was always that members had a defined, defendable stake and a clear path out. A personal caper is that pattern applied to the smallest possible commons – one person and the people who back them.
Consider the worked case. A researcher who wants to work independently opens a caper. Backers who want her output to exist buy in; the treasury fills. She draws her founder allocation to live on, and she runs the treasury through the same proposal machinery any caper uses – a grant here, a commissioned piece there, each one a motion her backers can vote on with the weight their holdings earn them. If she loses their confidence, they are not trapped the way Roll's holders were. They exit, each with their share of what is left, and the price discovery happens against real collateral rather than a thin order book. She has, in the most literal sense available, incorporated herself: a token, a treasury, a set of backers with governance rights, and a door they can always use.
Bowie could do this once, with an investment bank, for a catalogue worth fifty-five million dollars. The 2020 wave tried to do it for everyone and left out the two things that made Bowie's version real – a defined asset behind the claim, and a clean way for the claim to be settled. The personal caper is the first version where both are present and neither requires a bank. The interesting question is no longer whether an individual can be capitalized. Bowie settled that in 1997. The question is what happens to a world in which the answer is available to anyone, the collateral is real, and the way out is a rule rather than a favour.
Two companion essays take the halves further: The Exit Right on why the share a member leaves with and the weight their vote carries should be one quantity, and The Standing Counterparty on the curve that stands behind the price. The industry history of the 2020 wave is on personal tokens.