# When a DAO Ends: Five Ways Token Governance Actually Terminates
*Probity Institute — demonstration series. Data-only case records; descriptive comparisons, no ratings, no advice. Sources are linked or on-chain; verification method noted per case.*
Every governance token eventually faces the same question. What happens to holders when the organization around the token changes shape: when it winds down, gets bought, fails, or decides it no longer wants to be governed? Traditional markets answer with a body of law. Liquidation preferences, appraisal rights, dissolution waterfalls. DAOs are writing the answer in governance proposals, one case at a time. This page collects the live cases in one place, so the emerging vocabulary is visible.
Five distinct endings have now occurred live in 2026. They are not variations of one phenomenon. Each runs on a different mechanism, and each exposes a different missing token right.
---
## 1. Orderly winddown — Balancer (BIP-928)
**What happened:** Tokenholders voted on Sep 29, 2026 (12.18M BAL for) to wind the DAO down and distribute the treasury in-kind, pro-rata, to holders, through an audited claim contract. Distribution is targeted from Q2 2027.
**The right being exercised:** Exit at fair value. Every holder gets a pro-rata claim on an audited treasury. It is the closest thing on-chain to a statutory liquidation waterfall.
**What it proves:** A DAO *can* choose an ending that pays everyone symmetrically. The commitment is soft while it lives in governance, and hard once the claim contract deploys. After deployment, verification is a one-call read on the claim contract.
**Record:** `BIP-928` — vote chain-verified on Snapshot; claim contract pending.
---
## 2. Distress liquidation — Abracadabra
**What happened:** After protocol distress, holders voted on an orderly winddown. It passed, 100.05M to 0.52M. The process is discretionary: MIM claims rank senior to SPELL holders, and the June 2026 stewardship transfer (5.51B votes to 0) sits outside the redemption base.
**The right being tested:** Seniority, and the integrity of the distribution base. When claims conflict, who is ahead of whom, and what is actually in the pot at the moment of division?
**What it proves:** An orderly winddown can still be unfair. Without a hardcoded waterfall, whoever writes the proposal sets the base and the seniority. Two addresses supplied a supermajority of the voting weight on the transfer.
**Record:** paired with Balancer in the dissolution comparison (Oct 2, 2026).
---
## 3. Capture-and-drain — Neutron (proposal 9)
**What happened:** An adversarial actor acquired enough voting power for $20,199 and steered $9.4M of treasury. The interval between vote close and execution was effectively zero minutes.
**The right that was missing:** Cost-of-capture as a property of the mechanism. When quorum is low and execution is instant, the assumption that "the community will stop it" has a price. Here the price was four figures.
**What it proves:** A right that exists on paper still has a price to exercise against. That price is a measurable quantity, and our exercisability work measures it.
**Record:** capture-and-drain archetype (Oct 3, 2026), chain-anchored.
---
## 4. Post-attack governance revocation — Astroport
**What happened:** After a governance attack, the project did not repair its governance. It removed it. Voting, proposals, and tributes ended. The proposal quorum was set to 100% of voting power while emissions went to zero, so no quorum can ever again be met. We verified this with a single smart-contract query: `quorum == 1`. Contributors now operate the protocol through on-chain organizations. Holders keep only the contractual unlock.
**The right that was missing:** Any exit with value. Compare Balancer. Two "ends of governance": in one, holders receive a pro-rata treasury claim; in the other, they receive nothing but their unlock. Both are exits. One pays.
**What it proves:** "Is this DAO still governed?" is now a machine-checkable question. One query to an assembly contract returns the quorum parameter. A governance surface can be switched off, and a token can keep trading without one.
**Record:** fourth archetype (Oct 10, 2026), chain-verified.
---
## 5. Negotiated self-concentration — Orca ("Resourcing Orca for Its Next Phase")
**What happened (live as of Oct 12, 2026):** A proposal from inside the Governance Council itself would cut staker fee share from 40% to 10%, move the entire community treasury (~14.2M ORCA — verified on-chain: 14,200,768.34 in the proposal-named wallet) to a team-controlled acquisition account, and dissolve the Council, leaving tokenholder governance limited to approving token dilution. The acquisition has since been publicly identified: Orca announced a merger with the fixed-rate lending protocol Loopscale into "Formation" (New York, Loopscale's CEO leading) on Oct 7 — four days into the original vote and after the proposal was filed. Financial terms remain confidential under the definitive agreement; disclosure commitments to holders are after-the-fact (14 days post-close, semi-annual reports). The proposal now live (account created 2026-10-13, closing 2026-10-18 18:38 UTC) therefore ratifies a deal that was struck and announced before voting began.
**The right being tested:** Information rights at the moment of decision, and the mechanics of consent itself. The original vote would have passed at 3,000,000 Yes votes with no requirement that Yes exceed No. Community members identified the asymmetry on the forum before the vote closed, and the proposal was re-posted with a Yes-greater-than-No rule. The live proposal closes 2026-10-18 18:38 UTC, with a cooldown through 2026-10-20 18:38 UTC.
**What it proves:** Concentration does not always arrive as an attack. The operators themselves can propose it, with a supermajority of the existing governance body endorsing it. Vote mechanics — threshold design, majority requirements, who is excluded — are themselves an auditable surface, not a technicality. One further chain-readable fact: the Strategic Account is a program-derived address governed through a multisig, and it already holds the upgrade authority of the protocol's core on-chain programs, including the main liquidity contract. That authority was transferred before the vote concluded. Holders voting on the transfer can verify, from the chain itself, that the destination account already governs the protocol's upgrade path.
**Record:** fifth archetype (Oct 11, 2026; Formation identity + account structure verified Oct 12). The vote's closing window on this venue is recorded as a mechanical check in [Three Clocks](/docs/terms-integrity.md). The DAO's stated close and the program's last accepted instruction differ by 48 hours. A successful `FinalizeVote` receipt was recorded 22,276 seconds later; the proposal account now reads `Executing`. Subsequent relinquishment and withdrawal did not change its recorded tally.
---
## The comparison is the determination
| Archetype | Case | Treasury outcome for holders | Vote mechanics | Verification hook |
|---|---|---|---|---|
| Orderly winddown | Balancer BIP-928 | Pro-rata audited in-kind distribution | Pass threshold; executed by contract | Claim-contract read (pending) |
| Distress liquidation | Abracadabra | Discretionary; MIM senior, base narrowed by author | Two addresses supplied supermajority | Redemption-contract read |
| Capture-and-drain | Neutron p9 | $9.4M steered for $20,199 of control | Close-to-execution ≈ 0 min | Cost-of-capture computation |
| Post-attack revocation | Astroport | Unlock only — no treasury claim | Quorum set to 100%, unreachable | One smart query (`quorum == 1`) |
| Self-concentration | Orca (live) | ~20% of supply to team-controlled account | Threshold asymmetry, amended mid-arc | Wallet read; close 18 Oct, cooldown to 20 Oct |
Three patterns cut across all five:
1. **Exit quality is the differentiator.** Balancer and Astroport are both "the DAO ended." The difference between holders receiving a pro-rata treasury claim and receiving an unlock is the presence or absence of an enforceable redemption path. That single variable, not the story around it, is what a rights determination should capture.
2. **Mechanics are the contested surface.** Twice this quarter, the vote rules themselves became the live dispute, not the proposal content: Astroport's unreachable quorum, Orca's threshold-without-majority. Pass-mechanics symmetry is a checkable property. Does a threshold rule let a proposal pass against a No majority? Does the mechanism exclude the largest stakeholder class?
3. **Everything above maps to a missing standard.** Redemption path, distribution-base integrity, cost-of-capture, quorum reachability, mechanic symmetry. No contract standard enforces any of them today. That gap is what our standards exist to measure.
*Case records live in the Institute's determination dataset. This page is descriptive; it is not a rating of any protocol, not voting guidance, and not investment advice.*
---