Summary. A derivative suit is a stockholder borrowing the corporation's claim because the board will not sue itself, and nearly all of them are decided on procedure rather than on whether the conduct was wrong. The plaintiff must own stock at the right times, must characterize the claim correctly, and must either make a demand on the board or plead with particularity why demand was futile. This article walks each gate, explains the universal futility test Delaware adopted in Zuckerberg, covers special litigation committees and two-step Zapata review, and addresses settlement approval, fee awards, mootness fees, and preclusion among competing plaintiffs. It closes with what boards and defense counsel should do in the first ninety days, when most of the outcome is set.

Two things about derivative litigation surprise people the first time they encounter it.

The first is that the stockholder is not the plaintiff in any meaningful sense. The claim belongs to the corporation. The stockholder is a self-appointed representative asserting it because the people who would normally decide whether to sue are the people who would be sued. Any recovery goes to the corporation, not to the stockholder, and the stockholder's economic interest is limited to the indirect increase in the value of the shares — which is why the real economics of these cases run through the fee award.

The second is that the merits usually do not get litigated. The overwhelming majority of derivative actions are resolved at the pleading stage on demand futility, or on a motion to dismiss following a special litigation committee's investigation, or in a settlement that changes governance rather than transferring money. Understanding the procedure is not preliminary to understanding these cases. It very nearly is the case.

Step one: is the claim actually derivative?

Get this wrong and everything downstream is wrong. A direct claim belongs to the stockholder and avoids the demand requirement entirely; a derivative claim belongs to the corporation and does not.

The test. Tooley v. Donaldson, Lufkin & Jenrette, Inc., 845 A.2d 1031 (Del. 2004), asks two questions: who suffered the alleged harm — the corporation or the suing stockholder individually — and who would receive the benefit of any recovery. The old "special injury" formulation is gone.

Ordinarily derivative:

  • Waste, mismanagement, and excessive compensation.
  • Oversight failures under Caremark.
  • Usurpation of a corporate opportunity.
  • Insider trading by fiduciaries (under Delaware's Brophy line, the corporation's claim).
  • Overpayment by the corporation in an acquisition.

Ordinarily direct:

  • Denial of voting rights, or a disclosure claim in connection with a stockholder vote.
  • Dilution or improper transfer of voting control from the minority to a controller — the Gentile line, which Delaware narrowed considerably in Brookfield Asset Management v. Rosson by overruling Gentile and returning most equity-dilution claims to derivative status.
  • Denial of contractual rights attaching to a class of stock, including preferred rights, which are contract claims rather than fiduciary claims at all.
  • Breach of a stockholders' agreement.

Both, sometimes. A single transaction can support a direct disclosure claim and a derivative waste claim. Plead them separately and be precise about which recovery goes where, because a court that finds a claim mischaracterized will dismiss it rather than recharacterize it for the plaintiff.

Why plaintiffs push for "direct." No demand requirement, no continuous-ownership problem, no Zapata committee, and — critically — survival through a merger. That is not a small consideration. A stockholder cashed out in a merger loses derivative standing; a direct claim rides through.

Step two: standing

Contemporaneous ownership. 8 Del. C. § 327 and Fed. R. Civ. P. 23.1 require that the plaintiff owned stock at the time of the transaction complained of, or acquired it by operation of law from someone who did. A plaintiff who bought after the alleged wrong cannot sue over it. Where a course of conduct spans years, the date the plaintiff bought becomes a merits-adjacent fight about when the wrong "occurred."

Continuous ownership. The plaintiff must hold through the litigation. Selling is fatal. So, generally, is a merger in which the plaintiff's shares are converted into cash or into stock of an acquirer.

The merger exceptions. Delaware recognizes two: where the merger is itself the subject of a claim of fraud perpetrated merely to deprive stockholders of the derivative claim, and where the merger is in reality a mere reorganization that does not affect the plaintiff's ownership of the enterprise. Where a company becomes a subsidiary of a new holding company and the plaintiff receives stock of the parent, the claim may continue as a double derivative action — see Lambrecht v. O'Neal, 3 A.3d 277 (Del. 2010), which also clarified that a double-derivative plaintiff need not have owned parent stock at the time of the original wrong.

Adequacy. Rule 23.1 requires that the plaintiff fairly and adequately represent the interests of similarly situated stockholders. Challenges succeed rarely, but they succeed where the plaintiff has an economic conflict, is dominated by counsel to a degree that becomes evident at deposition, or has a vindictive personal agenda. Defense counsel should take the plaintiff's deposition on adequacy early in a case where the named plaintiff owns a de minimis position.

Federal courts apply state substantive law. Kamen v. Kemper Financial Services, Inc., 500 U.S. 90 (1991), holds that the demand requirement's content is supplied by the law of the state of incorporation, even in a federal action. Rule 23.1 supplies the particularity pleading standard; the substance comes from Delaware, Nevada, California, or wherever the entity is organized.

Step three: demand, or futility

This is where derivative cases live and die.

The rule. A stockholder must first demand that the board pursue the claim, unless demand would be futile. The premise is 8 Del. C. § 141(a): the board manages the corporation's affairs, and the decision whether to bring a lawsuit is a business decision like any other.

Making demand concedes board independence. Spiegel v. Buntrock, 571 A.2d 767 (Del. 1990), holds that a stockholder who makes a demand tacitly concedes that the board is disinterested and independent for purposes of that demand. Having made it, the plaintiff cannot later argue futility. That is why demand is almost never made in a case worth bringing, and why the futility pleading is the whole opening move.

If demand is made and refused, the plaintiff's only path is wrongful refusal — that the board's refusal was not a valid exercise of business judgment. Grimes v. Donald, 673 A.2d 1207 (Del. 1996), confirms the difficulty: the refusal decision itself gets business judgment protection, and the plaintiff must plead particularized facts showing the investigation was unreasonable or made in bad faith. Plaintiffs who make demand should at minimum use § 220 first to see what the board actually did with it.

The universal test

For decades Delaware ran two tests. Aronson applied where the challenged transaction was a decision of the current board; Rales applied where it was not — because the board had changed, or because the claim was about inaction rather than a decision, which is the Caremark situation.

United Food & Commercial Workers Union & Participating Food Industry Employers Tri-State Pension Fund v. Zuckerberg, 262 A.3d 1034 (Del. 2021), replaced both with a single three-part test applied director by director. For each director on the board at the time the complaint was filed, ask:

  1. Did the director receive a material personal benefit from the alleged misconduct that is the subject of the complaint?
  2. Would the director face a substantial likelihood of liability on any of the claims?
  3. Does the director lack independence from someone who received a material personal benefit or who would face a substantial likelihood of liability?

If the answer is yes for at least half of the board, demand is excused. Aronson and Rales remain useful as sources of reasoning; the test is now unified.

What each prong actually requires

Material personal benefit. Not a benefit shared with all stockholders. Compensation from the company does not disqualify an outside director as a matter of course, but compensation that is material to that individual — a retired director for whom board fees are a substantial part of income — can. Delaware courts consider director wealth, and plaintiffs increasingly plead it.

Substantial likelihood of liability. This is where exculpation does its work. Where the charter contains a § 102(b)(7) provision, a duty-of-care claim cannot create a substantial likelihood of liability, so the plaintiff must plead a non-exculpated claim — loyalty, bad faith, or improper personal benefit — against a majority of the board. That is a demanding standard, and it is the reason so many derivative complaints are pleaded as Caremark claims: the oversight theory is inherently a bad-faith theory and therefore not exculpated.

Independence. Beam ex rel. Martha Stewart Living Omnimedia, Inc. v. Stewart, 845 A.2d 1040 (Del. 2004), held that mere social or business acquaintance is not enough, but Delaware has since taken a more contextual view. Facts that have mattered: decades-long close friendship, employment of a family member, a director's primary employment controlled by the interested party, significant charitable contributions to an institution the director leads, and business relationships material to the director rather than to the counterparty. The inquiry is fact-driven and holistic, and it is the prong most often litigated now.

The books-and-records prerequisite in practice

Delaware courts have said repeatedly that a stockholder should use 8 Del. C. § 220 to obtain books and records before filing. It is not a formal prerequisite, but the practical consequences of skipping it are severe: a complaint drafted from press reports rarely satisfies particularity, and the court will not permit discovery to cure it — Rule 23.1 pleading must stand on its own.

What a § 220 demand should ask for, and what companies should expect to produce:

  • Board and committee minutes and materials on the subject, over a defined period.
  • Committee charters and any compliance reporting cadence.
  • Materials showing what management reported to the board about the risk, and when.
  • Communications of officers and directors on the subject where formal materials are inadequate — Delaware now permits this in appropriate cases, and companies with a thin formal record have effectively invited it.

Companies respond to these demands in one of two postures. The productive posture is to negotiate scope and produce a clean, complete set with an appropriate confidentiality and incorporation-by-reference agreement. The unproductive posture — resisting everything, producing three sets of minutes, and litigating for a year — usually ends with a broader order and a court predisposed to believe the record is thin because it is.

Step four: the special litigation committee

When demand is excused, the board is not out of options. It may form a special litigation committee of independent directors, delegate to it full authority over the claim, and have it investigate and decide whether pursuing the litigation serves the corporation's interests. If the committee concludes it does not, the corporation moves to dismiss.

The Delaware standard. Zapata Corp. v. Maldonado, 430 A.2d 779 (Del. 1981), created a two-step review that is unusually intrusive by Delaware standards:

  1. The corporation bears the burden of proving the committee's independence, good faith, and the reasonableness of its investigation. The plaintiff gets discovery on these questions.
  2. Even if the corporation carries that burden, the court may apply its own independent business judgment to decide whether the motion should be granted, "giving special consideration to matters of law and public policy in addition to the corporation's best interests."

Step two is discretionary and rarely reached, but its existence changes the settlement dynamics: the corporation cannot be certain that a perfect investigation ends the case.

Where committees fail. In re Oracle Corp. Derivative Litigation, 824 A.2d 917 (Del. Ch. 2003), is the cautionary tale. The committee consisted of two Stanford professors; the defendants included a Stanford professor, a large Stanford donor, and a director who had made significant donations, and the company itself had donated substantially. The court found the committee had not carried its burden on independence, emphasizing the "thicket of social and institutional relationships" and that human beings are influenced by more than economic self-interest. The committee had not disclosed the ties in its report — which the court found telling.

Building a committee that survives.

  • Appoint directors who joined after the challenged conduct wherever possible, and who have no relationship — professional, social, institutional, or philanthropic — with any defendant.
  • Delegate real authority by board resolution: to investigate, to determine, and to take any action including dismissal or prosecution of the claims. A committee that must report back for board approval is not a Zapata committee.
  • Retain independent counsel, not the company's regular outside counsel, and not counsel that represents the individual defendants.
  • Investigate the whole claim. Interview the accusers as well as the defendants. Review documents rather than relying on management summaries. Address each theory in the complaint separately.
  • Write a report that discloses relationships, describes the methodology, identifies what was reviewed and who was interviewed, and reaches conclusions supported by cited evidence. Committees lose on reports that read like advocacy.
  • Consider partial recommendations. A committee that pursues one claim and dismisses another is far more credible than one that finds nothing anywhere.

Not all states follow Zapata. Some apply pure business judgment review to the committee's decision (the New York approach in Auerbach v. Bennett, examining only independence and the adequacy of procedures, not the substance). Others follow the Model Business Corporation Act's structured approach. A few, like Massachusetts and North Carolina, have their own frameworks. Determine the standard before deciding whether the committee route is worth its cost, which for a genuine investigation is substantial.

Where the money is: fees, mootness, and settlement approval

Nobody is doing this for the stockholder's pro rata share. Plaintiffs' counsel are compensated under the corporate benefit doctrine: a plaintiff who confers a substantial benefit on the corporation may recover reasonable attorneys' fees from it, whether or not the benefit is monetary.

Monetary recoveries. Fee awards are typically a percentage of the recovery, adjusted for stage and risk. Delaware uses the Sugarland factors — results achieved, time and effort, complexity, contingency, and counsel's standing and ability — and applies percentages that decline as recovery size increases. Americas Mining Corp. v. Theault, 51 A.3d 1213 (Del. 2012), affirmed a fee award of roughly $304 million on a $2 billion judgment, confirming that Delaware will pay percentage fees on very large recoveries rather than reverting to a lodestar cross-check.

Therapeutic and governance benefits. Many derivative settlements provide no money at all: board committee formation, independent director additions, compliance program changes, clawback policies, term limits. Courts award fees for these, but the amounts are far smaller and the scrutiny is greater, because the benefit is hard to value and the incentives are obvious.

Mootness fees. When a company voluntarily takes the action the complaint sought — supplements a proxy, rescinds a grant, amends a bylaw — the case becomes moot, and plaintiff's counsel seeks a fee for having caused the change. This has been a genuine abuse vector. The demand is often small enough that paying it is cheaper than fighting, which is precisely the problem. Delaware courts have grown notably less receptive to mootness fees for immaterial disclosure supplements, and companies increasingly refuse to pay them and disclose the demand instead.

Disclosure-only settlements are largely dead. In re Trulia, Inc. Stockholder Litigation, 129 A.3d 884 (Del. Ch. 2016), announced that Delaware would approve disclosure settlements only where the supplemental disclosures are plainly material and the release is narrowly tailored. The Seventh Circuit reached the same conclusion in In re Walgreen Co. Stockholder Litigation, 832 F.3d 718 (7th Cir. 2016), with Judge Posner observing that such settlements are "no better than a racket." The practical effect was migration: merger-objection suits moved to federal court as § 14(a) claims, and then largely to mootness-fee demands that never require court approval at all.

Settlement mechanics. A derivative settlement requires court approval after notice to stockholders, because it releases the corporation's claim. Expect the court to ask:

  • What is the realistic litigation value of the claim, discounted for the demand-futility and Zapata risk?
  • Is the release limited to claims arising from the transactions pleaded, or does it sweep in unknown claims?
  • Who is paying? A settlement funded entirely by D&O insurance, with no contribution from the individuals accused of disloyalty, invites the question whether the corporation is meaningfully compensated. Courts increasingly ask for individual contributions in serious cases.
  • Are the governance changes durable — embedded in bylaws or charters with a stated term — or announcements that can be reversed the following quarter?

The fast-filer problem and preclusion

When a corporate crisis becomes public, multiple stockholders file derivative complaints in multiple courts within days. Consolidation handles the same-court problem. The cross-jurisdictional problem is harder, and it created a real trap.

The trap. A plaintiff who files immediately, without a § 220 demand, gets dismissed for failure to plead demand futility with particularity. Another plaintiff who spent a year on books and records then files a well-supported complaint — and is met with a collateral estoppel defense based on the first plaintiff's loss.

Pyott v. Louisiana Municipal Police Employees' Retirement System, 74 A.3d 612 (Del. 2013), held that Delaware would give full faith and credit to a federal court's dismissal on demand futility and rejected the Court of Chancery's view that derivative plaintiffs are not in privity until a complaint survives a motion to dismiss. The Delaware Supreme Court has since moderated the harshest applications on due-process grounds where the first plaintiff's representation was grossly inadequate, but the risk is real.

Practical responses:

  • Forum-selection bylaws designating a single court for internal-affairs claims — now the standard fix, and enforced widely.
  • Coordination among plaintiffs' firms and lead-counsel structures that consolidate the § 220 work.
  • For defendants, an early evaluation of whether an easy win against a weak first-filer is actually the best outcome. Sometimes it is; sometimes it purchases a preclusion fight and a bad-faith narrative.
  • For plaintiffs, a stay of the fast-filed action pending books-and-records completion in the well-developed one.

Special contexts

LLCs and partnerships. Derivative actions are available under 6 Del. C. § 18-1001 for LLCs and analogous provisions for LPs, with the same contemporaneous-ownership requirement and a demand-or-futility pleading rule. The threshold twist is that the operating agreement may have modified or eliminated the fiduciary duties the derivative claim would enforce, and may also alter the demand procedure. Read the agreement first. Note also that Delaware LLC members do not have the same statutory books-and-records rights as corporate stockholders; § 18-305 rights are broader in some respects and more easily restricted by agreement in others.

Closely held corporations. Several states permit a court to treat a derivative claim in a closely held corporation as a direct action where there is no risk of multiple suits, no prejudice to creditors, and no unfairness to other stockholders — the Model Act takes this approach. That matters enormously in a two- or three-owner company, where insisting on derivative form means a recovery paid to an entity the wrongdoer still controls.

Insurance and indemnification. Two structural points that shape every derivative case:

  • Judgments in derivative actions are generally not indemnifiable, though expenses are, and advancement obligations are enforced aggressively. That is why individual defendants insist on advancement and why the company's cash goes out the door long before any liability is determined.
  • D&O Side A coverage is the practical source of derivative settlements. Watch the insured-versus-insured exclusion: a properly drafted policy carves out derivative actions brought by stockholders without the assistance or participation of an insured. A policy without that carve-out can leave the corporation's own claim uninsured, which is a diligence item on every renewal.

The first ninety days, from the company's side

Most of the outcome in a derivative case is determined before any brief is filed.

On receipt of a § 220 demand:

  • Treat it as the opening of litigation, because it is. Issue a litigation hold immediately, covering directors, officers, and the compliance function, and covering personal devices and messaging applications used for company business.
  • Do not clean up the file. The single worst outcome in this area is a spoliation finding layered on top of an oversight claim, and messaging apps with disappearing-message settings have produced exactly that in recent cases.
  • Assess the record honestly. If board minutes on the mission-critical risk are one line per year, know that before choosing a litigation posture.
  • Negotiate scope and produce. Over-resistance to a demand that will ultimately be granted costs credibility and money.

On the complaint:

  • Separate the defendants. Directors with exculpation and no personal benefit have a different case from officers, and from anyone who traded or received a benefit. In re Cornerstone Therapeutics Inc. Stockholder Litigation, 115 A.3d 1173 (Del. 2015), confirms that independent directors are entitled to dismissal where a non-exculpated claim is not pleaded against them individually, even in an entire-fairness case. Insist on that separation early; joint representation obscures it.
  • Count the board. Demand futility is arithmetic. Identify the board as of the filing date and analyze each director under the three Zuckerberg prongs. If the plaintiff cannot get to half, the case ends.
  • Consider whether refreshing the board helps. Adding genuinely independent directors changes the futility calculus for future filings, though not for a complaint already on file.
  • Decide on an SLC deliberately. It is expensive, it waives nothing but it commits the company to a real investigation, and a half-hearted committee is worse than none.
  • Notice the insurer. Late notice under a claims-made policy is a recurring and entirely avoidable coverage disaster; a § 220 demand may itself constitute a claim or a circumstance requiring notice under the policy's terms.

On governance, regardless of the litigation:

  • Fix what the demand revealed. If there was no committee overseeing the risk, create one now, with a charter and a reporting cadence. This is worth doing on the merits, and it is also the currency in which these cases settle.
  • Improve minute practice immediately. Minutes going forward will be produced in the next § 220 demand, and their quality is entirely within the company's control.

What this body of law is for

It is easy to read the procedural apparatus above as a set of obstacles designed to protect boards. Some of it is. But the underlying problem is real and does not have a clean solution: the corporation's claim is controlled by the people accused of harming the corporation, and any rule that gives every stockholder a free hand to litigate on the company's behalf produces a great deal of litigation of very little value — as the merger-objection era demonstrated conclusively.

The compromise the law reached is to make the stockholder show, with particularity and before discovery, that the board genuinely cannot be trusted with the decision. That is a high bar and it screens out most claims. When a plaintiff clears it — usually by doing the books-and-records work and finding that the board really did have no process for a risk that mattered — the case is generally worth its cost. When a plaintiff does not, the dismissal is not an injustice; it is the system working as designed.

For boards, the lesson is the same one that runs through the substantive fiduciary law: the record you build in ordinary times is what you are judged on in extraordinary ones. A board with committee charters, a compliance reporting cadence, minutes that show engagement, and a documented response to every red flag is not merely more likely to win a derivative case. It is substantially less likely to face one, because the § 220 production will show a board doing its job, and the complaint that would have been drafted from it never gets written.

Variations outside Delaware

Delaware supplies the vocabulary, but a substantial number of operating companies are incorporated elsewhere, and the differences are not cosmetic.

Universal demand states. A large group of states — those following the Model Business Corporation Act — have abolished futility entirely and require demand in every case, coupled with a waiting period of typically ninety days before suit may be filed, subject to an exception where irreparable injury would result. In these states the Zuckerberg analysis is irrelevant. What matters instead is the quality of the board's response to the demand and whether the statutory committee determination process was followed, and the litigation shifts from "was demand futile" to "was the refusal the product of a reasonable inquiry conducted in good faith by qualified persons."

Nevada. Nevada codified a director-protective standard requiring a plaintiff to rebut the business judgment presumption by showing a breach involving intentional misconduct, fraud, or a knowing violation of law, and its statute makes that protection automatic rather than dependent on a charter provision. Nevada also has no analogue to Delaware's developed Caremark jurisprudence. Companies reincorporating there are, among other things, buying a harder derivative-pleading standard, and stockholders evaluating such a move should understand that is what is being traded.

California. California imposes its own verification and security-for-expenses regime, and Cal. Corp. Code § 800 permits a court to require a plaintiff with a small holding to post a bond of up to $50,000 for the defendants' expenses on a showing that there is no reasonable possibility of benefit to the corporation. California also applies its own long-arm provisions to certain "pseudo-foreign" corporations under § 2115, which can displace the internal affairs doctrine for companies with dominant California contacts — a provision other states have declined to honor and that has been narrowed but not eliminated by litigation.

New York. Auerbach v. Bennett, 47 N.Y.2d 619 (1979), applies business judgment review to a special litigation committee's determination, examining the committee's disinterested independence and the adequacy of its investigative procedures but declining to review the substantive conclusion. New York also requires the plaintiff to plead with particularity the efforts made to obtain board action, under N.Y. Bus. Corp. Law § 626.

The practical instruction is the same as everywhere else in corporate law: identify the state of incorporation before analyzing anything, and do not assume that a memo written about a Delaware corporation transfers. Two companies with identical facts, identical boards, and identical conduct can reach opposite outcomes on the motion to dismiss based solely on where the certificate was filed.

A candid word about value

Boards receiving a derivative complaint tend to react in one of two unhelpful ways. Some treat it as an outrage and instruct counsel to fight everything, which converts a manageable governance problem into two years of fee-generating procedure and a public record of resistance. Others treat it as a nuisance to be paid off, which invites the next one and leaves the underlying governance defect unaddressed.

The useful posture is neither. Evaluate the claim the way the special litigation committee would even if you never form one. Was there a system for the risk? Did red flags reach the board? Did anyone benefit personally? If the honest answers are good, the procedural defenses will work and they should be pressed. If the honest answers are bad, the company's interest is in fixing the governance failure quickly, capturing that fix as settlement consideration, and getting a release — not in spending three years proving that a majority of the board was technically independent.

That evaluation should happen in the first month, conducted by counsel who did not advise on the underlying conduct, and it should be written down. It is the same discipline the substantive law asks of directors in every other context: get informed, address the conflict, and make a decision you can explain later.

Primary authority

Derivative practice is unusually rule-bound for a body of equity, and the rules below carry most of the weight in a motion to dismiss.

  • Fed. R. Civ. P. 23.1 and its state analogues — the verified-complaint, contemporaneous-ownership, and particularized-demand requirements.
  • 8 Del. C. § 220 — the books-and-records demand, and in practice the first step in any derivative case worth filing.
  • Aronson v. Lewis, 473 A.2d 805 (Del. 1984) and Rales v. Blasband, 634 A.2d 927 (Del. 1993) — the two historical demand-futility tests, for board decisions and for boards that never decided anything.
  • United Food & Commercial Workers Union v. Zuckerberg, 262 A.3d 1034 (Del. 2021) — collapses Aronson and Rales into a single three-part, director-by- director test. Cite this one; the older framing is now shorthand at best.
  • Zapata Corp. v. Maldonado, 430 A.2d 779 (Del. 1981) — the special litigation committee, and the two-step review that preserves the court's own business judgment.
  • Tooley v. Donaldson, Lufkin & Jenrette, Inc., 845 A.2d 1031 (Del. 2004) — direct versus derivative, decided by who suffered the harm and who would recover.
  • Kamen v. Kemper Financial Services, Inc., 500 U.S. 90 (1991) and Burks v. Lasker, 441 U.S. 471 (1979) — federal courts borrow state demand law even for federally created claims.
  • Cal. Corp. Code § 800 and N.Y. Bus. Corp. Law § 626 — the security-for- expenses and demand provisions that make forum selection consequential.
  • 8 Del. C. § 115 — forum selection bylaws for internal-affairs claims, now routinely enforced.

Related articles

This article is provided for general informational purposes and does not constitute legal advice. Demand requirements, special litigation committee standards, and derivative procedure vary by state of incorporation, and Delaware law in this area continues to evolve. Consult qualified corporate litigation counsel before making a demand, responding to one, or forming a committee.