CASE 01
The DAO Hack and the Ethereum Hard Fork
“Code Is Law” versus Community Justice in a Leaderless Organization
In June 2016, an unknown party moved roughly one third of the largest crowdfunded venture in history out of its treasury — without stealing a password, forging a signature, or breaking into a single computer. Every instruction the attacker sent was executed exactly as the organization’s own published rules permitted. What followed was not a lawsuit but a referendum: an organization with no board, no executives, and no legal personality had to decide whether to rewrite its own accounting records to undo a transaction it had itself authorized.
Course anchor
Ch. 6 (required) + Ch. 7–18 frameworks
Case period
Apr 2016 – Jul 2017 (legacy to today)
Decision point
20 July 2016 — fork, freeze, or do nothing
What this case is designed to teach
After analyzing this case, you should be able to:
- Separate three questions that non-specialists routinely collapse into one: Was it technically permitted? Was it legal? Was it ethical?
- Apply the Chapter 6 apparatus to a genuine dilemma — social obligation vs. responsiveness vs. responsibility; the classical vs. socioeconomic view; the ethics tests in Exhibit 6.4; the determinants of ethical behavior in Exhibit 6.5; Kohlberg’s stages in Exhibit 6.6; issue intensity in Exhibit 6.7.
- Map stakeholders in an organization with no employees, no customers in the ordinary sense, and no legal personality — and explain why the conventional stakeholder map partly breaks down.
- Diagnose an organizational design that is simultaneously the most formalized (every rule is executable code) and the least formalized (no manager can approve anything) structure you will study this semester (Ch. 11).
- Evaluate a control system using feedforward, concurrent, and feedback control (Ch. 18), and explain why all three had already failed before a single malicious instruction ran.
- Argue and defend a recommendation in English under adversarial questioning, in a situation where every available option imposes a serious cost on someone.
Part A — The Case
Read the whole narrative before looking at the questions. The facts you need are in this section and in the exhibits that follow.
A.1 Background: an organization made of code
Ethereum, launched in 2015, is a public blockchain: a shared ledger maintained by thousands of independent computers (“nodes”) that no single party owns. Its distinguishing feature is the smart contract — a program stored on the ledger that executes automatically when its conditions are met. Once deployed, a smart contract runs exactly as written. It cannot be telephoned, persuaded, sued, or instructed to make an exception.
This property produced a governance philosophy that circulated widely in the early Ethereum community under the slogan “Code is Law.” The claim was not merely technical; it was normative. If participants voluntarily join a system whose rules are published, auditable, and enforced without discretion, then the outcomes that system produces are legitimate — including outcomes nobody intended and some participants find outrageous. The attraction was the removal of human discretion, which in conventional finance is also the channel through which favouritism, bailouts, and political pressure flow.
A DAO — a Decentralized Autonomous Organization — pushes the idea to its limit. It is an organization whose charter, treasury, voting rules, and payout logic are entirely smart-contract code. There is no CEO, no board of directors, no registered office, and in most jurisdictions no legal personality. Members hold tokens; tokens carry votes; votes release money.
A.2 “The DAO”: design and crowdsale
In April 2016 a small German startup, Slock.it — founded by Christoph Jentzsch, Simon Jentzsch, and Stephan Tual — published code for a general-purpose venture fund built on Ethereum. It was called, with no false modesty, “The DAO.” The concept was a decentralized venture-capital fund: anyone could contribute Ether (ETH) during a creation period and receive DAO tokens in proportion; token holders would then vote on funding proposals, and approved proposals would be paid automatically. Slock.it wrote and released the code but did not own or control the resulting fund.
Three design features matter for your analysis:
- Curators. A group of well-known community figures held the keys to a whitelist. They did not judge whether a proposal was good; they verified that a proposal’s stated recipient address really belonged to the people it claimed to. This was a deliberately minimal human role — but it was a human role, inside an organization advertised as needing none.
- Splitting. Any member who disagreed with the majority could “split” — move their share into a child DAO under their own control rather than be outvoted and trapped. This was the minority-protection mechanism, the code’s substitute for shareholder appraisal rights. Crucially, funds moved into a child DAO were locked for roughly 27 days before they could be withdrawn.
- The code as the contract. The DAO’s published materials stated that the terms of the offering were set by the smart-contract code itself, and that plain-language descriptions elsewhere were explanatory only and did not override the code. Participants were, in effect, told in advance: the program is the agreement.
The creation period ran from 30 April to 28 May 2016. It raised approximately 12.7 million ETH — on the order of US$150 million at prevailing prices, and roughly 14% of all Ether then in circulation. It was, at that moment, the largest crowdfunding event ever conducted. Estimates of the number of participating investors vary across sources (commonly cited figures range from about 11,000 to more than 20,000 holding addresses); no authoritative register exists, which is itself part of the case.
A.3 The warnings
The DAO was not attacked out of a clear sky. On 27 May 2016 — one day before the creation period closed — researchers Dino Mark, Vlad Zamfir, and Emin Gün Sirer published “A Call for a Temporary Moratorium on The DAO,” arguing that the mechanism contained governance flaws serious enough that no proposals should be funded until they were fixed. The call was widely read and widely dismissed as excessive academic caution.
Separately, in early June 2016, developers in the Ethereum ecosystem publicly discussed a class of smart-contract flaw then called the recursive call bug (today usually called reentrancy), and noted that contracts of The DAO’s type were exposed to it. Slock.it acknowledged that a recursive-call issue existed in the codebase and stated publicly that no funds were at risk. A fix was being prepared. It was not deployed in time.
Analytical note
The warnings are the hinge of the ethics analysis. If the vulnerability was unknowable, the case is about misfortune. If it was published, discussed, and deprioritized while money kept flowing in, the case is about organizational conduct — and the question “who is responsible?” becomes much harder to answer with “the anonymous attacker.”
A.4 The attack: 17 June 2016
On 17 June 2016 an unknown actor began draining The DAO. Over a period of hours, roughly 3.6 million ETH — about one third of the fund, worth on the order of US$50–60 million at the time — was moved out.
The mechanism, stripped of jargon:
The split function sent the member’s Ether out first and updated the member’s balance afterwards. The attacker’s receiving contract was written so that, on receiving Ether, it immediately called the split function again — before the first call had finished and before the balance had been set to zero. The DAO consulted its ledger, still saw the original balance, and paid again. And again. The loop ran dozens of times per transaction.
Analogy: a bank teller who hands over the cash and then writes the withdrawal in the ledger. A customer who can interrupt the teller between those two acts can withdraw the same balance repeatedly — and every entry in the book will be correct.
Note carefully what did not happen. No password was stolen. No server was breached. No signature was forged. No rule of the deployed system was violated. The attacker used public functions, in a published contract, in an order the authors had not anticipated. This is why the word “hack” became contested almost immediately, and why the incident is an ethics case and not merely a security case.
One design feature bought the community time: because the funds had been moved through the split mechanism into a child DAO, they were subject to the same ~27-day lock as any other split. The money was identified, visible on a public ledger, and immovable. A deadline existed, and everyone could see the clock.
A.5 Three options, none of them good
With the clock running, the Ethereum community — core developers, the Ethereum Foundation, miners, exchanges, DAO token holders, and a very loud public forum — converged on three options.
| Option | What it does | Case for | Case against |
|---|---|---|---|
| 1. Do nothing “Code is Law” |
Let the lock expire; the attacker keeps the Ether. The ledger is never altered. | Preserves immutability and neutrality — the properties that make the platform worth trusting at all. Losses fall on those who accepted the stated terms. | Roughly one third of a US$150m fund is transferred to an anonymous party by exploiting a flaw that had been publicly flagged in advance. |
| 2. Soft fork Freeze |
Miners agree to ignore any transaction touching the attacker’s funds. The money exists but can never move. No history is rewritten. | A middle path: nobody is paid, nothing is erased, and the precedent set is weaker than outright reversal. | It is still censorship of valid transactions. Researchers then identified that the proposed implementation opened a denial-of-service attack vector, and it was abandoned before activation. |
| 3. Hard fork Reverse |
Change the protocol rules at a fixed block so that DAO balances move to a withdrawal contract from which original contributors can reclaim their ETH. | Returns money to thousands of contributors; treats a systemic design failure as a collective problem; protects the platform’s reputation and future. | Human intervention overrides the ledger. It demonstrates that the chain is not immutable when enough influential people are unhappy — and invites the question: whose losses qualify next time? |
In parallel, a volunteer group that became known as the “Robin Hood Group” used the same vulnerability to drain the remaining exposed funds into contracts they controlled, intending to return them to token holders. They attacked The DAO in order to protect it — which raises, in a single move, most of the questions in Exhibit 6.4.
A.6 The decision and the split
There was no shareholders’ meeting, because there were no shareholders. Legitimacy had to be manufactured. An informal, non-binding signalling mechanism called a “carbonvote” let ETH holders vote with the weight of their holdings. The result was roughly 87% in favour of the hard fork — on a turnout of under 6% of circulating Ether. Developers wrote the fork; miners and node operators chose whether to run it; exchanges chose which chain to list.
At block 1,920,000, on 20 July 2016, the hard fork activated. The DAO’s balances were moved to a withdrawal contract, and contributors could reclaim Ether at a rate of 1 ETH per 100 DAO tokens. In financial terms, the rescue worked.
But a minority refused. They kept running the original rules, on the grounds that a ledger that can be edited by majority sentiment is not a ledger at all. That chain continues to exist today as Ethereum Classic (ETC), with its own price, its own miners, and an explicitly stated “Code is Law” identity. The forked chain kept the name and the ecosystem and is the Ethereum used by essentially the entire industry today.
The structural point
The community did not resolve the disagreement. It duplicated the organization so that both answers could be true on different ledgers. No conventional firm has this option: Toyota cannot fork into two Toyotas because the board disagrees about a recall. Ask what this tells you about the relationship between organizational structure, conflict resolution, and accountability.
A.7 Aftermath
- Regulation arrived. On 25 July 2017 the U.S. Securities and Exchange Commission issued a Report of Investigation under Section 21(a) of the Securities Exchange Act concluding that DAO tokens were securities under U.S. law and that the offering was subject to federal securities regulation. The SEC brought no enforcement action, but the message was explicit: calling an organization “decentralized” and “autonomous” does not place it outside the law. The idea that the code could be the only applicable rulebook was, as a legal matter, rejected.
- Attribution was never established. The attacker was not identified through any legal process. In 2022 a journalistic investigation named a specific individual, who publicly denied it. Treat this as unproven; the case does not depend on it, and your report must not assert it as fact.
- Industry practice changed. Reentrancy became the textbook smart-contract vulnerability. The “checks-effects-interactions” pattern (update your ledger before you send money) became standard. Independent third-party code audits, bug bounties, staged deployments, and formal verification grew into a professional sector — a striking real-world example of feedforward control being built only after a feedback-control catastrophe.
- The precedent is still argued about. Every subsequent large exploit reopens the question the fork appeared to settle: when, if ever, may a community rewrite its own record? In practice, no comparable reversal has been executed on Ethereum since — a fact you should weigh when you assess whether the “slippery slope” argument turned out to be correct.
Part B — Exhibits
You may reproduce these exhibits in your report, but you must extend them with your own analysis. Copying them unchanged earns no analytical credit.
Exhibit 1 — Timeline of the decision
| Date | Event | Why it matters for the analysis |
|---|---|---|
| 2015 | Ethereum main network launches with smart-contract capability. | Establishes the technological and ideological context (Ch. 3: external environment). |
| 30 Apr 2016 | The DAO creation (crowdsale) period opens. | Funds are raised before any operating history exists. |
| 27 May 2016 | “A Call for a Temporary Moratorium on The DAO” published. | Documented advance warning — central to assigning responsibility. |
| 28 May 2016 | Creation period closes: ~12.7m ETH (~US$150m). | Scale creates issue intensity (Exhibit 6.7: greatness of harm). |
| Early Jun 2016 | Recursive-call (reentrancy) vulnerability class publicly discussed; Slock.it says funds are not at risk. | A public risk statement that later proved wrong — disclosure ethics. |
| 17 Jun 2016 | ~3.6m ETH drained via reentrancy into a child DAO. | The act that was permitted by code, contested in law, and condemned by most of the community. |
| Jun 2016 | “Robin Hood Group” uses the same exploit defensively. | Same act, opposite intent — a clean test of consequentialist vs. rule-based reasoning. |
| Late Jun – Jul 2016 | Soft-fork proposal developed, then abandoned after a DoS vector is identified. | Shows how technical constraints narrow the ethical option set. |
| Jul 2016 | Carbonvote: ~87% in favour, turnout under 6% of circulating ETH. | The legitimacy problem (Ch. 13: group decision making; Ch. 17: leadership authority). |
| 20 Jul 2016 | Hard fork activates at block 1,920,000; withdrawal contract created. | The decision itself — the object of your recommendation in Question 10. |
| Jul 2016 onward | Ethereum Classic (ETC) continues the unforked chain. | Organizational fission as conflict resolution (Ch. 7: managing change and resistance). |
| 25 Jul 2017 | SEC Section 21(a) Report: DAO tokens were securities. | The legal answer to “code is law” — and evidence for the classical/socioeconomic debate. |
Exhibit 2 — Stakeholder inventory (starting point only)
This list is deliberately incomplete and unranked. Question 3 requires you to add missing stakeholders, assess power and legitimacy, and rank them.
| Stakeholder | What was at stake | Position on the fork | Effect of the fork on them |
|---|---|---|---|
| DAO token holders (contributors) | Their invested Ether; no legal recourse, no insurance, no identified counterparty. | Predominantly pro-fork. | Recovered most of their contribution through the withdrawal contract. |
| The attacker | ~3.6m ETH obtained through permitted operations. | Reportedly asserted entitlement under the contract terms. | Lost the ETH on the forked chain; retained value on the unforked chain. |
| Slock.it and its founders | Reputation, liability exposure, the future of their product. | Supported rescue efforts. | Reputational damage; The DAO project ended. |
| The Curators | Personal reputation; an ill-defined quasi-fiduciary role they had not agreed to. | Divided. | Exposed as a point of human control in a “trustless” system. |
| Ethereum Foundation & core developers | Credibility of the platform; the norm of protocol neutrality. | Led the fork implementation. | Saved the ecosystem; permanently accused of centralized control. |
| Non-DAO ETH holders | The value and integrity of an asset they held for unrelated reasons. | Divided; most did not vote at all. | Bore the consequences of a decision taken to rescue someone else’s investment. |
| Miners / node operators | Revenue; the practical power to accept or reject the new rules. | Majority followed the fork. | Decisive actors with no formal governance role. |
| Exchanges | Customer assets, listing decisions, operational risk. | Followed demand; several listed ETC. | Effectively decided which chain had market legitimacy. |
| Future users and developers | Whether “immutable” means anything going forward. | Not represented; did not yet exist. | Inherited the precedent without ever voting on it. |
| Regulators and the public | Investor protection; the integrity of capital markets. | Absent during the decision. | Intervened afterwards (SEC 2017). |
Exhibit 3 — The three positions in the community
| Position | Core claim | Underlying ethical logic | Strongest objection to it |
|---|---|---|---|
| Technical fundamentalism “Code is Law” |
The rules were published and accepted. The outcome is binding even though it is unwelcome. | Rule-based (deontological); consent and contract; the rule of law applied to code. | Consent to a rule you cannot read is thin consent. And the same community immediately abandoned the rule when the cost was high enough. |
| Community justice Restitution |
A systemic flaw enabled mass expropriation; a community that can fix it and does not is complicit. | Outcome-based (utilitarian) plus fairness/desert; investor protection as a moral duty. | “We reverse transactions when the victims are numerous and well-connected” is not a principle — it is a description of political power. |
| Platform pragmatism Survival |
A loss of this scale in year one would have killed Ethereum’s credibility; the fork protected the platform. | Consequentialist, with the organization’s own survival as the relevant outcome. | This is self-interest described as ethics. It also gives the ecosystem’s insiders a permanent reason to intervene. |
Exhibit 4 — Key figures
~12.7m ETH
Raised in the creation period (~US$150m; ~14% of circulating ETH)
~3.6m ETH
Drained on 17 June 2016 (~US$50–60m at the time)
~27 days
Child-DAO lock period — the window in which the response had to be decided
87% / <6%
Carbonvote support vs. turnout as a share of circulating ETH
Data caution. Ether’s price moved sharply during 2016, so dollar figures quoted in different sources for the same event differ legitimately. State the ETH amount first and the dollar figure second, and always name the source and the date of conversion.
Part C — The Assignment
C.1 How this case satisfies the syllabus requirements
| Req. | Syllabus wording | How you meet it in this case |
|---|---|---|
| (1) | Select a real organization or industry case and analyze it using the frameworks covered in this course. | The DAO / the Ethereum community is a real, fully documented case. Apply Ch. 7–18 frameworks in Questions 6–9 — at minimum organizational structure (Ch. 11) and control (Ch. 18). |
| (2) | Include an explicit business ethics and social responsibility analysis: identify the ethical issues, map the stakeholders, assess conduct against ethical criteria and CSR expectations (Ch. 6). | Questions 2, 3, 4 and 5 are mandatory and carry the largest share of the marks. Section 3 of your report must be titled “Ethics and Social Responsibility Analysis.” |
| (3) | Close with a recommended course of action and the reasoning that supports it. | Question 10 — two parts: the July 2016 decision, and a forward-looking governance design. |
| (4) | Written report and oral presentation in English. | Part G provides the terminology and presentation language support. |
C.2 First decision: which organization are you analyzing?
Most weak reports on this case fail here, not in the ethics section. “The organization” is genuinely ambiguous, and an analysis that drifts between units produces stakeholder maps that contradict themselves. State your unit of analysis in the first paragraph of your report and hold it for the whole document. Choose one:
(a) The DAO itself
A fund with a treasury, members, and a governance mechanism — but no managers. Best for Ch. 11 structure and Ch. 6 duties to members.
(b) Slock.it
A conventional startup that built and promoted the code. Best for disclosure ethics, product liability, and Ch. 10 entrepreneurial ventures.
(c) The Ethereum Foundation / core developers
The actor that actually made the fork decision. Best for Ch. 17 leadership, Ch. 7 change, and legitimacy.
(d) The Ethereum community as a whole
The most ambitious option: an organization with no boundary. Only choose this if you can define membership clearly.
Whichever you choose, the other three appear in your analysis as stakeholders, not as the subject.
C.3 Three traps to avoid
- Explaining the technology instead of analyzing the management problem. You may spend at most one page on how blockchain and reentrancy work. Marks are awarded for the analysis, not for the tutorial.
- Treating “permitted by code”, “legal”, and “ethical” as one question. They have three different answers in this case, and the report must keep them apart. The SEC’s 2017 report is directly relevant evidence on the second.
- Deciding the conclusion first and then selecting the framework that flatters it. A report that applies Exhibit 6.4 and finds that all five tests conveniently agree is not analyzing — it is advertising. Where the tests disagree, say so; that disagreement is the finding.
Part D — Analysis Questions
Questions marked REQUIRED must be answered by every team — they implement syllabus requirements (2) and (3). From the questions marked CHOOSE, answer at least three, and include at least one from the Ch. 11 / Ch. 18 pair.
Context and unit of analysis REQUIRED
Define the organization you are analyzing (see C.2) and describe its external environment at the time of the decision. Which environmental components exerted the most pressure — and what does it mean that the legal/political component was effectively empty in June 2016 but not in July 2017? Characterize the organizational culture using the dimensions from Ch. 3, and explain how that culture shaped what the community believed it was permitted to do.
Frameworks: Ch. 3 external environment components; environmental uncertainty matrix; dimensions of organizational culture; strong vs. weak culture.
Identify and rank the ethical issues REQUIRED
This case contains several distinct ethical issues, not one. Identify at least four and state each as a decision faced by a named actor — not as a vague topic. Candidates include, but are not limited to: continuing to accept contributions after a public moratorium call; publicly stating that funds were not at risk; exploiting a known flaw for personal gain; exploiting the same flaw defensively without authorization; rewriting a ledger that was marketed as unalterable; and deciding a community-wide question on a 6% turnout.
Then rank them using the six determinants of issue intensity (Exhibit 6.7): greatness of harm, consensus of wrong, probability of harm, immediacy of consequences, proximity to victims, and concentration of effect. Present the ranking as a table with a one-line justification per factor. Your ranking must explain why participants reacted far more strongly to some of these issues than to others — particularly why the failure to act on the advance warnings generated so much less outrage than the exploit itself, despite arguably causing the same harm.
Frameworks: Exhibit 6.7 issue intensity; definition of an ethical dilemma; harm and consent.
Map the stakeholders REQUIRED
Extend Exhibit 2. For each stakeholder, record: the stake, the power to affect the outcome, the legitimacy of their claim, whether their interest was actually represented in the decision, and what the fork did to them. Then answer three harder questions:
- Who had power without accountability? Miners and exchanges held decisive practical power but had no governance role and owed no duty to anyone. What conventional corporate mechanism is missing here, and what would it have done?
- Who had a legitimate claim but no voice? Consider non-DAO ETH holders who bore the consequences of a rescue they did not need, and future users who inherited the precedent.
- Is the attacker a stakeholder? Argue it both ways in three sentences each, then commit to an answer. Your answer determines whether the fork is “restitution” or “expropriation,” so do not skip it.
Frameworks: Stakeholder identification and prioritization; power–legitimacy–urgency; Ch. 6 CSR expectations beyond shareholders.
Assess conduct against ethical criteria and CSR expectations REQUIRED
This question has three parts, all required.
(a) Apply the ethics tests. Take the five tests in Exhibit 6.4 (Golden Rule, What-If-Everybody-Did-This, Family, Conscience, Front Page/Social Media) and apply them to three actors: the attacker, the Robin Hood Group, and the core developers who implemented the fork. Present this as a 3×5 matrix with a verdict in each cell. Note the What-If-Everybody-Did-This test is the most demanding one for the fork advocates, and the Front Page test is the most demanding one for the Robin Hood Group — explain why.
(b) Classify the CSR posture. For your chosen organization, classify its conduct as social obligation, social responsiveness, or social responsibility, and defend the classification with evidence from the timeline. Then use Exhibit 6.2: was the fork decision driven by ethical or pragmatic considerations, focused on ends or means, framed over the long or the short term? Be honest about the possibility that a decision widely described as ethical was in fact responsive.
(c) Classical vs. socioeconomic view. A “Code is Law” purist and a classical-view manager both argue for minimal intervention, but for different reasons. Identify the difference. Then state which view the Ethereum community actually acted on in July 2016, and whether its subsequent behaviour (no comparable reversal since) is consistent with that view or contradicts it.
Frameworks: Exhibit 6.4 ethics tests; Exhibit 6.2 responsibility vs. responsiveness; classical and socioeconomic views; Exhibit 6.1 arguments for and against social involvement.
Explain the behavior: why did ethical people act this way? REQUIRED
Use Exhibit 6.5 to explain conduct rather than merely judge it. Work through the determinants: stage of moral development, individual characteristics, structural variables, organizational culture, and issue intensity.
- Place the “Code is Law” position on Kohlberg’s stages (Exhibit 6.6) and defend the placement. A common student answer is that it is principled; a strong answer considers seriously whether strict adherence to a published rule set is conventional (law-and-order) reasoning instead. Then place the fork advocates. Neither placement is obvious — argue it.
- Structural variables: The DAO had no performance appraisal system, no reward allocation controlled by anyone, and no goals set by management — yet Ch. 6 says these are among the strongest influences on ethical behavior. What replaced them, and did the substitute work?
- Culture and leadership: Ch. 6 states that the behavior of managers is the single most important influence on whether individuals act ethically. In an organization that has no managers, where does that influence come from? Identify who actually performed this function and by what authority.
Frameworks: Exhibit 6.5 determinants; Exhibit 6.6 stages of moral development; values, ego strength, locus of control; organizational culture.
Organizational structure and governance CHOOSE
Describe The DAO using the standard design elements: work specialization, departmentalization, chain of command, span of control, centralization/decentralization, and formalization. You will find that the organization is extremely formalized in one sense (every rule is executable and admits no exception) and completely unformalized in another (no procedure exists for any human to approve, escalate, or override anything). Explain this paradox and what it cost.
Then: is The DAO a mechanistic or an organic organization? Argue that the usual dichotomy does not capture it, and propose what a third category would need to include. Finally, evaluate the Curators — an undefined human role inside a system designed to eliminate human roles. What happens to accountability when authority exists but responsibility was never assigned?
Frameworks: Ch. 11 organizational design elements; mechanistic vs. organic; authority, responsibility, and accountability; boundaryless organizations.
Control: three failures before the attack CHOOSE
Analyze the incident with the three types of control (Ch. 18). Feedforward: what should have happened before deployment — independent audit, bug bounty, staged funding caps, a formal go/no-go gate — and why did the moratorium call fail to trigger any of it? Concurrent: the fund was live for weeks with a public ledger; what monitoring and circuit breakers were absent, and was an emergency pause even possible given the design philosophy? Feedback: the hard fork is corrective action taken after the deviation — assess it against the standard of a well-designed control system, including timeliness, criteria, and who was authorized to take it.
Close on the central tension: an effective control system requires the ability to intervene, while the organization’s core value proposition was that nobody could intervene. Can a system be trustworthy and controllable at the same time? Relate your answer to Ch. 6’s treatment of independent social audits and to whistle-blower protection — the researchers who published the moratorium call were, in substance, whistle-blowers with no organization to report to.
Frameworks: Ch. 18 control process and feedforward/concurrent/feedback control; corrective action; Ch. 6 independent social audits and whistle-blowing.
Change, disruptive innovation, and strategy CHOOSE
The hard fork is organizational change executed in under five weeks by an entity with no authority to impose it. Analyze it with Ch. 7: what was changed (structure, technology, people)? Who were the change agents? Identify the sources of resistance — and note that the resistance here did not fail and did not comply; it seceded and founded Ethereum Classic. Which resistance-management techniques were used, which were not available, and would any of them have prevented the split?
Then take the strategy angle (Ch. 9): immutability was Ethereum’s claimed source of competitive advantage. The fork spent some of it to buy survival. Evaluate that trade with a SWOT, and take a position on whether it was sustainable — using the evidence that no comparable reversal has occurred since.
Frameworks: Ch. 7 change process, change agents, resistance to change, disruptive innovation; Ch. 9 SWOT, competitive advantage, sustainability of advantage.
Decision making, leadership, and legitimacy CHOOSE
Classify the fork decision using Ch. 2: programmed or non-programmed? Structured or unstructured problem? Made under conditions of certainty, risk, or uncertainty? Identify at least two decision-making biases visible in the public debate (anchoring, escalation of commitment, confirmation, availability, overconfidence) and cite the behavior that evidences each.
Then assess the decision process against Ch. 13’s treatment of group decision making: what were the advantages and disadvantages here, and is a carbonvote with sub-6% turnout better or worse than no vote at all? Finally, Ch. 17: identify the source of influence held by the leading figures in a community with no formal authority — use the bases of power (legitimate, reward, coercive, expert, referent) and say which ones were actually operating. If informal expert and referent power can move US$150 million, what does that imply about the claim that the organization was leaderless?
Frameworks: Ch. 2 decision types, conditions, and biases; Ch. 13 group decision making and groupthink; Ch. 14 communication barriers and informal networks; Ch. 17 bases of power and leadership without authority.
Your recommendation REQUIRED
(a) The July 2016 decision. Place your team on 1 July 2016, knowing only what was known then. Recommend one of the three options in Section A.5, or a fourth of your own design. Your recommendation must include: the decision criteria you applied and their weights; who holds the authority to make the decision and how that authority is justified; the stakeholders who lose under your choice and what — if anything — you offer them; the strongest argument against your position, stated fairly and then answered; and the conditions under which your answer would change. A recommendation that costs nobody anything has not understood the case.
(b) The forward-looking design. Design the governance and ethics program you would put in place before launching a comparable organization today. Address at least: a code of ethics that can bind an organization whose members are anonymous; feedforward control (audit requirements, bug bounties, phased funding caps); an emergency procedure with pre-declared criteria, a pre-identified decision body, and time limits — decided before any crisis so that it is not decided by whoever shouts loudest during one; a channel for whistle-blowers and security researchers, with protection and reward; and a disclosure standard for known vulnerabilities while fundraising is still open. For each element, state which Ch. 6 mechanism it implements and which failure in this case it would have prevented.
(c) Transferability. In one paragraph: what should managers in conventional firms — who will never deploy a smart contract — take from this case? Consider automated decision systems, algorithmic pricing, and AI-driven approvals, where the rules are also written in advance by people who cannot foresee every interaction, and where “the system did it” is already a common defence.
Frameworks: Ch. 2 decision-making process; Ch. 6 codes of ethics, ethics training, independent social audits, whistle-blower protection; Ch. 18 control design; Ch. 8 contingency planning.
Optional extensions (for teams wanting a harder problem)
- Green management (LO 6.2). In 2016 Ethereum used proof-of-work, an energy-intensive consensus mechanism; it moved to proof-of-stake in 2022, cutting energy use dramatically. Classify the community’s environmental posture using the four shades of green, and assess whether the change was a legal, market, stakeholder, or activist response.
- Entrepreneurship (Ch. 10). Evaluate Slock.it as an entrepreneurial venture: what obligations does a founding team owe to contributors when it builds the vehicle, promotes it, and then disclaims control of it?
- Motivation and behavior (Ch. 15–16). Why did thousands of people commit money to an organization with no management, no recourse, and no track record? Apply a motivation framework and consider what it implies about disclosure duties toward such contributors.
- International context (Ch. 4). The developers were in Germany, the contributors were global, the regulator that eventually spoke was American, and the ledger was nowhere. Apply Ch. 6’s treatment of ethics in an international context and the UN Global Compact: whose standards should govern, and who enforces them?
Part E — Deliverables
Written report
- Length: 3,000–4,000 words of body text (roughly 10–14 pages), excluding cover page, exhibits, and references.
- Language: English throughout.
- Format: A4, 12 pt, 1.5 line spacing, numbered pages and numbered section headings.
- Citations: Every factual claim and every number carries a source. Consistent style (APA 7th recommended). A reference list is mandatory.
- Contribution statement: a short table naming which member led which section, signed by all members.
Required structure
- Executive summary (max. 250 words, written last)
- Case background and unit of analysis (Q1)
- Ethics and social responsibility analysis (Q2–Q5) — the largest section
- Management analysis (your three chosen questions from Q6–Q9)
- Recommendation and reasoning (Q10)
- Limitations and what would change our answer
- References · Appendices · Contribution statement
Oral presentation
- Time: 15 minutes presentation + 5 minutes Q&A (confirm in class).
- Participation: every team member speaks for a meaningful segment.
- Slides: maximum 15. Any slide that is only a paragraph of text will cost you marks.
- Language: English throughout, including Q&A. Reading a script aloud is not a presentation — use notes.
Suggested timing
- 2 min — The dilemma in one sentence, and your recommendation stated up front
- 2 min — Case facts (assume the audience knows nothing about blockchain; do not exceed this)
- 6 min — Ethics and stakeholder analysis (your core content)
- 3 min — Management analysis: structure and control
- 2 min — Recommendation, costs, and the strongest counter-argument
Prepare for these three questions — they will be asked: Who pays for your recommendation? What would change your mind? Why is your answer a principle rather than a preference?
Sourcing and academic integrity
- Prefer primary sources (the SEC report, the moratorium paper, contemporaneous project posts, on-chain records) over summaries. Secondary journalism is acceptable when cited as such.
- Cross-check every number in at least two sources. Dollar figures in this case disagree legitimately because of price volatility; state the ETH amount and the conversion date.
- Where a fact is contested — the attacker’s identity, the exact number of investors, the precise dollar value — say that it is contested. Correctly flagging uncertainty earns marks; asserting a disputed claim as fact loses them.
- If you use generative AI tools, disclose what you used them for in an appendix. You remain fully responsible for every claim, citation, and number in the report — including any the tool invented.
Part F — Grading Rubric
Total 100 points, mapped to the four syllabus requirements. The final group report is worth 30% of the course grade.
| Pts | Criterion | Excellent | Weak |
|---|---|---|---|
| 15 | Case understanding & framing Req. (1) · Q1 |
Unit of analysis declared and held throughout; facts accurate and cited; technology explained briefly and only where it bears on the argument. | Drifts between organizations; long technical tutorial; numbers uncited or inconsistent. |
| 30 | Ethics & CSR analysis Req. (2) · Q2, Q4, Q5 |
Four or more issues identified and ranked by issue intensity; Exhibit 6.4 tests applied to multiple actors with honest, sometimes conflicting verdicts; CSR posture classified with evidence; behavior explained, not just judged. | Frameworks named but not applied; one-sided moral verdict; “stealing is wrong” asserted without analysis; every test conveniently agrees. |
| 15 | Stakeholder mapping Req. (2) · Q3 |
Goes beyond the given exhibit; power vs. legitimacy distinguished; silent and unrepresented stakeholders identified; the attacker question confronted directly. | Exhibit 2 reproduced with minor edits; stakeholders listed but not prioritized; conflicts among them unexamined. |
| 15 | Management frameworks Req. (1) · Q6–Q9 |
At least three frameworks from Ch. 7–18 applied to generate insight the facts alone do not give; framework limits acknowledged where they strain. | Frameworks decorate the report rather than drive it; definitions copied from slides; conclusions would be identical without them. |
| 15 | Recommendation & reasoning Req. (3) · Q10 |
Specific and actionable; decision criteria stated; costs named and allocated; strongest counter-argument stated fairly and answered; conditions for revision given. | Vague (“improve governance”); costless; ignores the opposing case; forward-looking design not linked to any failure in the case. |
| 10 | English communication Req. (4) · report + presentation |
Clear, well-organized English; management terminology used correctly; presentation delivered (not read) with all members contributing; Q&A handled with composure. | Meaning obscured by language errors; terminology misused; slides read aloud; one member carries the presentation; Q&A avoided. |
Note on fluency: this is an EMI management course, not a language examination. Grammatical errors that do not obstruct meaning are not penalized. Misusing a management concept is penalized, because that is a content error.
Part G — EMI Toolkit
Terminology and language support for writing and presenting this case in English.
G.1 Key terms
| Term | Meaning in this case |
|---|---|
| Smart contract | A program stored on a blockchain that executes automatically when its conditions are met. It cannot be renegotiated after deployment. |
| DAO | Decentralized Autonomous Organization: an organization whose rules and treasury are governed by code rather than by managers or a legal charter. |
| Immutability | The property that recorded transactions cannot be altered or deleted. The fork is significant precisely because it proved this property was social, not absolute. |
| Fork | A change to the network’s rules. A soft fork tightens rules and remains backward-compatible; a hard fork creates rules that older software rejects, so the network can permanently split into two chains. |
| Reentrancy / recursive call | A flaw in which a contract calls an external party before updating its own records, allowing the external party to re-enter and repeat the operation using stale data. |
| Exploit | Using a system’s own legitimate functions in an unintended way to produce an unintended result. Note that “exploit” describes a technique, not a legal conclusion. |
| Child DAO / split | A mechanism letting a dissenting member withdraw their share into a separate contract, subject to a waiting period. Designed as minority protection; used as the escape route for the drained funds. |
| Carbonvote | A non-binding poll weighting each vote by the amount of Ether held. Used to signal community sentiment on the fork. |
| Restitution | Returning something to its rightful owner, or restoring a party to their previous position. The word the fork advocates preferred. |
| Expropriation | Taking property from its holder, usually by an authority. The word the fork’s opponents preferred. Choosing between these two words is already an ethical judgment — use each deliberately. |
| Fiduciary duty | A legal obligation to act in another party’s interest. Relevant here because no one clearly owed one — which is the point. |
| Moral hazard | Reduced incentive to guard against risk when someone else will bear the loss. The core of the “bailout” objection to the fork. |
| Precedent | A decision that shapes how similar future cases will be handled. The fork’s most contested legacy. |
| Due diligence | The investigation a reasonable party performs before committing money. Ask who performed it here, and on whose behalf. |
| Security (financial) | A tradable financial instrument such as a share or bond. The SEC concluded DAO tokens were securities, bringing the offering within securities law. |
| Whistle-blower | Someone who raises ethical or safety concerns. In this case the moratorium authors raised them publicly because there was no internal channel to raise them through. |
G.2 Useful language for the presentation
Stating a position
- “Our recommendation is X, for three reasons.”
- “On balance, we conclude that…”
- “Applying the issue-intensity framework, the most serious issue is…”
- “The evidence points in two directions, but the stronger case is…”
Conceding, then answering
- “It is true that… However, this does not outweigh…”
- “The strongest objection to our position is… Our answer is…”
- “We accept that this imposes a cost on X. We think that cost is justified because…”
Handling questions
- “That’s a fair challenge. Let me separate two things…”
- “Could you clarify whether you mean the 2016 decision or the design we propose?”
- “We don’t have evidence on that point, so we would not want to speculate.”
Avoid these
- “Everyone knows that…” (assertion without evidence)
- “The hacker is obviously a criminal.” (assumes the legal question the case asks you to analyze)
- “They should improve their governance.” (not a recommendation — say who does what, when)
Part H — Sources and Further Reading
Primary sources (start here)
- U.S. Securities and Exchange Commission, Report of Investigation Pursuant to Section 21(a) of the Securities Exchange Act of 1934: The DAO, Release No. 81207, 25 July 2017. Available at sec.gov.
- D. Mark, V. Zamfir & E. Gün Sirer, A Call for a Temporary Moratorium on The DAO, 27 May 2016.
- Contemporaneous Ethereum Foundation and Slock.it blog posts from June–July 2016 on the attack, the proposed soft fork, and the hard-fork implementation.
- The Ethereum Classic community’s founding statement setting out the “Code is Law” position.
Course materials
- Robbins, S. P. & Coulter, M., Management, 15th ed., Global Edition, Pearson, 2021 — Ch. 6 (Exhibits 6.1–6.8), Ch. 2, 3, 7, 9, 11, 13, 17, 18.
Secondary sources
Extensive journalism and academic commentary exists on this case, including book-length treatments of early Ethereum and peer-reviewed work in law and information-systems journals on “Code is Law” and DAO governance. Search your library databases for “The DAO” AND (governance OR “code is law” OR “hard fork”). Use secondary sources for interpretation and analysis; return to primary sources for every fact and figure you assert.
A note on the facts in this document. Figures and dates here reflect the widely reported record of the events. Sources disagree on several points — particularly dollar valuations, the number of investors, and exact turnout percentages — because of price volatility and the absence of any authoritative register. Verify every figure you use against a citable source and report the disagreement where it exists. Do not cite this case document itself as the source of a fact in your report.