Articles & Publications

AI, Finance, Federalism, and Legal Commentary

A curated archive of articles and publications covering artificial intelligence, agentic platforms, private credit, leveraged finance, high yield, constitutional law, family law, federalism, restructuring, and practical legal issues affecting businesses and institutions.

Articles & Publications

Selected writing across constitutional law, family law, artificial intelligence, agentic systems, private credit, leveraged finance, restructuring, blockchain, and related business and technology topics.

More Articles & Publications

The prior feature remains elevated, followed by external publications and hosted articles organized for quick scanning.

Back to Featured
New Article • Private Credit / Restructuring / Corporate Finance

The Captive Tranche

When Majorities Rule: Efficiency or Theft

A bond, a loan, a convertible note is sold into the market on a shared premise: the holder will act, when called upon, in the interest of the instrument. Company debt is typically divided into layers, called tranches, ranging from senior secured loans down through junior debt, unsecured bonds, and convertible notes. Each tranche has its own contract, its own voting rules, and its own economic logic. Holders within a tranche are presumed to share an economic interest, because they paid for the same exposure and face the same downside. That presumption is the load-bearing wall of corporate finance. Trust indentures, credit agreements, and the class-voting mechanics of Chapter 11 all rest on it. A majority can bind a minority within a tranche because the majority is assumed to be motivated by the same logic as the minority. The system gives up unanimity in return for this efficiency, and the price is generally worth paying. Without majority binding, a single small holder could veto an otherwise sensible restructuring and extract rents from everyone else. Most consents, amendments, and plan votes that bind minorities are routine and proper.
Read Full Article
New Article • AI / Legalweek / Legal Practice

Recent AI Conferences and the Practice of Law

Legalweek 2026, agentic AI, reliability, governance, and lawyer training.

Legalweek 2026 and last week’s AI Agent Conference both declared the experimentation phase over. The vendor market has converged. What remains is the operational scaffolding: governance, security at scale, and a training model that does not hollow out the next generation of supervising lawyers. The technology is running ahead of the institutional capacity to supervise it. These are a few thoughts on what that means for firms and in-house teams. Legalweek 2026 closed at the Javits Center in March with a keynote titled “The Reckoning.” Last week, the AI Agent Conference at the Hilton Midtown drew over a thousand executives. The two events covered different audiences and different layers of the technology, but the underlying picture was the same.
Read Full Article
Hosted Article • Constitutional Law / Family Law / Federalism

Are You My Mother? The Federal Courts and the American Family

The Domestic Relations Exception, federal jurisdiction, civil rights enforcement, and constitutional family-law disputes.

Read Full Article
I have written about the federal government's nineteenth-century relationship with the Church of Jesus Christ of Latter-day Saints, a history in which federal authority over the church's most intimate institutional practices was exercised without restraint and without principled limit. That history coincided almost exactly with the years in which the Supreme Court was constructing the Domestic Relations Exception, the doctrine that purports to place domestic relations beyond federal court jurisdiction. The juxtaposition has never been adequately examined. The same courts that announced in Barber v. Barber and In re Burrus that domestic relations belong categorically to the states were simultaneously sustaining federal prosecutions of marriages, federal dissolution of a church's corporate charter, and federal disenfranchisement of believers, all on the basis of how a specific religious minority organized its households. What the Mormon cases and the DRE share is not a constitutional principle. It is a record of the federal courts making choices: showing up as aggressive institutional parents when the political culture demanded intervention, and claiming no jurisdiction when it did not. This article is an argument that the disclaimer has never been honest, that the current moment has made the inconsistency visible in a way it has not been before, and that the exception should be closed.
LinkedIn • Finance / Blockchain

The Stock Market is Going On-Chain

Tokenized stocks explained — and the harder question of what the holder actually owns.

If you left the financial world for even a few years, you would come back to a new vocabulary: real-world asset tokenization, on-chain settlement, and digital ownership. This piece explains tokenized stocks in plain terms and then moves to the harder question: what exactly does the holder own? The article frames the core legal issue clearly. A platform may hold the actual shares while the user holds only a token tied to them. The strength of that position depends on custody, disclosures, platform structure, and contract terms — not just on the stock price chart.
Open Article
LexBlog • AI / Legal Ethics

The Misdraft: How Ungoverned AI Use Can Undermine Contract Drafting

A warning about what happens when legal drafting speed outruns governance, confidentiality, and review discipline.

This article focuses on a quiet but growing law-firm risk: associates using generative AI under time pressure without secure infrastructure or meaningful governance. The result is not just bad drafting — it is confidentiality exposure, process drift, and a false sense of reliability. The piece positions AI misuse in drafting as an operational and professional-responsibility problem, not merely a productivity issue.
Open Article
Lawgaze • AI / Future of Work

Everyone Prefers Humans: Weighing the Real Impact of AI on Jobs

A measured argument that AI changes work, but does not erase the enduring premium on judgment, trust, and human skill.

This piece examines the employment conversation around AI without collapsing into hype or panic. Its central point is simple: automation can reshape tasks and workflows, but markets still reward human accountability, persuasion, empathy, and judgment. The framing makes it especially useful for business and legal readers trying to separate labor-market theatrics from real-world role redesign.
Open Article
TechBullion • AI / Productivity

The 15 Most-Common AI Prompt Time Wasters

A practical guide to the prompting mistakes that create rework, slow decision-making, and waste professional time.

This article argues that the cost of weak prompting is not abstract. It shows up as billable-time waste, slower deal cycles, generic analysis, and compliance review built on unverified assumptions. The theme throughout is that better prompting is less about clever phrasing and more about clarity, context, objective, and subject-matter competence.
Open Article
LinkedIn • Debt / Restructuring

What Is an LME? — Managing Corporate Debt Outside Bankruptcy

An accessible introduction to liability management exercises and why they matter for borrowers and lenders under refinancing pressure.

The article explains how companies use liability management exercises to extend maturities, adjust covenant flexibility, exchange debt, and raise fresh liquidity without entering formal bankruptcy. It is especially strong on the tension between flexibility for participating parties and risk for non-participating creditors, including priming, sacred-rights disputes, and litigation exposure.
Open Article
LinkedIn • Receivables / Structured Finance

When Is Factoring Really Debt? Understanding True-Sale Risk in Receivables Financing

A focused treatment of when receivables financing is a true sale and when it is really a secured loan wearing different clothes.

This piece walks through the core recharacterization questions: who bears default risk, whether recourse exists, and how much control the seller keeps after the transfer. It is useful because it treats the accounting and legal analysis as inseparable. Labels alone do not decide the issue; economic substance does.
Open Article
LinkedIn • Inventory Finance

Off-Balance Sheet Inventory Financing—A Growing Tool for Corporate Liquidity

How inventory monetization structures can create liquidity while preserving leverage ratios — if the structure is real.

This article explains why borrowers under working-capital pressure are exploring inventory structures that avoid booking traditional balance-sheet debt. The value of the piece is its insistence on structure over labels: derecognition depends on whether ownership and the risks and rewards of ownership have genuinely shifted.
Open Article
Programming Insider • Autos / Commentary

You Don't Need a Race Car for Your Daily Commute

A consumer-facing reminder that practical, everyday utility usually matters more than aspirational overkill.

This piece uses the contrast between fantasy-performance branding and actual day-to-day driving needs to make a broader point about fit, function, and decision-making. It works well in an archive because it adds a lighter consumer-technology angle alongside the finance and AI writing.
Open Article
Amazon • External

Amazon Feature

An Amazon-linked item included in the archive as provided.

This entry is preserved as an external Amazon link in the site archive. The destination can be swapped for a fuller title, product name, or description once you want a more specific presentation.
Open Link
Featured Article • Shareholder and Member Disputes / New York Business Law

Breaking Up Is Hard to Do

Shareholder and Member Disputes: A New York Perspective

April 2026Full Article

Business partnerships, like marriages, often look stable until they don't. A minority shareholder in a closely held corporation (one with a small number of owners and no public market for its shares) may discover that the majority has been paying itself undisclosed compensation for years. A member of a limited liability company may find that the managing member has diverted company revenues into a side venture. A co-founder may wake up to the fact that her equity interest is being methodically rendered worthless while her partners take home salaries, distributions, and perquisites she never approved.

New York law provides remedies in each of these situations. But the remedies are not self-executing, and navigating them requires understanding what rights exist, how to enforce them, and what the realistic sequence of events looks like from the first demand letter to a final judgment or negotiated exit.

This article focuses on equity-level disputes: the rights and remedies of shareholders, LLC members, and partners as owners of their businesses. Employment claims, officer compensation disputes, and director liability questions that arise independently of an ownership interest involve different legal frameworks and are not covered here.

Much of my practice has developed around artificial intelligence and emerging technology companies. When I began advising in this space, I expected the calls to be about their business: what they were building, how to protect it, how to comply with evolving regulatory frameworks, how to structure the agreements that would govern their products and relationships. A significant part of the work turned out to be about the business of their business instead: who owns what, who controls what, and what happens when the people who built something together no longer agree on what it is worth or who should be running it.

The disputes that arise in AI and technology companies are not categorically different from those in any closely held business, and I can find myself on either side of them. Sometimes I represent the founder being pushed out; sometimes the majority trying to remove someone who has stopped contributing or crossed a line. The legal frameworks are the same. The facts determine which side of the table I am sitting at. What is distinctive about AI and technology companies is the consistency of when the disputes arise. In a startup, the operating agreement and the shareholder agreement are negotiated early, often quickly, and frequently with more attention to the optimistic provisions than the protective ones. The equity splits, the vesting schedules, the buyout mechanisms, the information rights: these are the sections that feel like lawyer boilerplate when the company is two people in a room with a good idea. They become something else entirely when the company has a valuation, a product, and parties who no longer agree on where it is going. The provisions that were left to the lawyers become the tools for discipline or advantage. That is when I get the call.

The Anatomy of a Business Dispute

Most disputes among owners of closely held businesses follow a recognizable pattern. One party (usually the one with operational control) begins to act unilaterally. The others notice. Requests for information are deflected or ignored. Financial statements become unavailable or unreliable. Meetings are not held. Distributions (profit payments to owners) stop. Salaries paid to the controlling party or to companies he controls increase. Assets may move.

The minority owner is left with a choice: accept the situation, negotiate an exit, or litigate. New York law is reasonably well-developed here, but it is unforgiving about procedure. The sequence in which rights are asserted matters. The documentary record matters. So does delay.

The Right to Inspect Books and Records

The single most important initial right for any owner in a business dispute is the right to see the books. Without financial records, allegations remain allegations.

For New York corporations, Business Corporation Law Section 624 gives shareholders of record the right to examine the corporation’s minutes and record of shareholders, and, on five days’ written notice, to examine the balance sheet and profit-and-loss statement for the prior fiscal year. For a more comprehensive examination, the shareholder must make written demand and establish a proper purpose. Proper purpose under BCL 624 is broadly construed to include any purpose reasonably related to the person’s interest as a shareholder, including investigation of potential mismanagement or fraud.

When the corporation refuses, the shareholder may bring a special proceeding in New York Supreme Court to compel production. Courts treat refusal to honor a proper demand seriously, and the burden shifts to the corporation to justify denial.

For LLCs, the governing provision is New York's Limited Liability Company Law, known as the LLC Law. Section 1102 gives each member, regardless of the size of their interest, the right to inspect and copy the LLC's articles of organization, operating agreement (the contract governing the company's internal affairs), tax returns, financial statements, and records of contributions and distributions. On reasonable demand, a member is also entitled to any other information concerning the company’s business and affairs reasonably required for the proper exercise of the member’s rights and duties. The legislature intended it that way.

Operating agreements sometimes attempt to limit inspection rights, and New York courts have enforced such limitations in some contexts. But limitations that would effectively prevent a member from investigating suspected fraud or misappropriation are unlikely to survive judicial scrutiny. Courts have little patience for governing documents that appear designed to insulate wrongdoing from discovery.

Where the Company Is Formed Matters

A business dispute in New York often turns out to be governed, at least in part, by the law of a different state. Many closely held businesses operating in New York are organized in Delaware, Wyoming, Nevada, or another state with favorable corporate statutes. The choice of formation state was made at the beginning, frequently without much input from the minority, and its implications may not be understood until something goes wrong.

Delaware is the most common state of organization. The Delaware General Corporation Law and the Delaware LLC Act are the most developed corporate statutes in the country, with decades of Court of Chancery decisions interpreting every significant provision. The Court of Chancery is a dedicated business court with specialist judges. When Delaware law governs, New York courts apply it, and the Court of Chancery's body of decisions provides the interpretive framework. For businesses that expect complexity or outside investment, Delaware offers a level of predictability other states do not match.

For equity owners in a dispute, Section 220 of the Delaware General Corporation Law is a powerful tool. It gives stockholders the right to inspect corporate books and records for a proper purpose, and the Court of Chancery enforces it efficiently. A refused Section 220 demand can produce board communications, internal documents, and officer materials that would take years to obtain in New York civil discovery. The parallel provision for Delaware LLCs is Section 18-305 of the Delaware LLC Act, which gives members inspection rights covering financial statements, tax returns, and records of contributions and distributions. Both rights are subject to modification by the operating agreement, within limits: courts will not permit the agreement to be used to block investigation of suspected fraud. The scope of what Section 220 compels has been the subject of ongoing litigation and legislative attention in Delaware, and recent amendments have tightened the standard in some respects. The tool is still underused by counsel who assume New York law governs because the company operates here.

Wyoming has become a serious alternative, particularly for LLCs and holding structures. Its charging order protection is among the strongest in the country, limiting a creditor's remedy against a member's interest to a charging order rather than a forced sale. There is no state income tax and no franchise tax. Annual fees are low and the statute is flexible. For closely held businesses and family structures, Wyoming is worth considering at formation.

Part of the appeal of out-of-state formation is New York's own disclosure requirements. Corporations formed in New York are subject to BCL Section 1315, which gives shareholders of record the right to obtain the names and addresses of other shareholders on demand in certain circumstances. That is a meaningful transparency obligation for the company's principals. New York LLCs have historically been subject to the publication requirement under LLC Law Section 206, which required notice of formation to be published in two newspapers in the county of the LLC's principal office for six consecutive weeks. In New York County, that requirement has been expensive. Reform has been discussed for years and partial changes have been made, but the general framework remains. Forming in Delaware or Wyoming avoids it.

One qualification on the publication point: a foreign LLC registering to do business in New York is subject to its own publication requirement under LLC Law Section 802, similar to the domestic formation requirement. The savings are real but not as clean as they first appear. The shareholder and member disclosure requirements of BCL Section 1315, however, do apply only to domestic corporations, and that distinction holds.

An entity formed outside New York that does business here must register as a foreign entity. Corporations register under BCL Article 13; LLCs under LLC Law Article 8. The registration requirement is more than administrative. Under BCL Section 1312 and LLC Law Section 808, a foreign entity doing business in New York without authorization cannot maintain an action in New York courts. In a dispute where the company needs to sue a former partner, a departing employee, or a business counterparty, an unregistered entity may find itself unable to proceed until it regularizes its status. Courts have discretion to stay rather than dismiss, but the gap in registration is a vulnerability that opposing counsel will find.

Governing law analysis comes first. An operating agreement that designates Delaware law means Delaware fiduciary duty standards, Delaware inspection rights, and Delaware dissolution statutes apply, regardless of where the business operates. Counsel who assumes New York law controls because the company is in New York will get the analysis wrong. The operating agreement says which state's law governs. It should be the first document reviewed.

Paper Tiger or Opening Shot: The Demand Letter

The demand letter is the first formal act in almost every business dispute, and it separates practitioners who understand the instrument from those who treat it as a formality. A vague or legally unfounded demand is worse than no demand at all. It signals weakness, gives the other side a roadmap of what not to produce, and forecloses options. A precisely drafted, legally grounded demand can compel production without litigation and in some cases resolve the dispute before it escalates.

Specificity is the first requirement. A books-and-records demand under BCL Section 624 or LLC Law Section 1102 must identify the records sought with reasonable particularity and state the purpose for which they are needed. A demand that asks for "all financial records" invites noncompliance. The stated purpose matters equally: courts ask whether it is a proper purpose, meaning one reasonably related to the person's interest as a shareholder or member. Investigation of suspected misappropriation or self-dealing qualifies. That purpose must appear in the letter itself. A purpose supplied for the first time in the enforcement proceeding is less likely to be credited by a court.

Tone and deadlines are strategic choices, not formalities. Some situations call for measured language that signals intent without provocation; others call for a direct statement of what the party believes is happening and what will follow from noncompliance. Response deadlines should be specific and observed. If the deadline passes without substantive compliance, the enforcement proceeding should follow promptly. A pattern of missed deadlines met with inaction trains the other side to ignore demands. Courts notice that pattern.

In derivative litigation, the pre-suit demand on the board may be excused when a majority of directors are interested in the challenged transaction or lack independence from the wrongdoer. Demand is futile when the people who would receive it are the people whose conduct is challenged. In the closely held business context, where the board often consists entirely of the controlling party, futility is typically not difficult to establish, but it must be pleaded with specificity.

The demand letter also starts contractual clocks. If the operating agreement or shareholder agreement imposes a cure period before expulsion can be effectuated, or requires mediation before suit can be filed, the demand letter triggers those requirements. Failure to comply, even when the underlying claim is strong, can result in dismissal or significant delay.

The demand letter creates the record that the rest of the litigation runs on. Every subsequent document will be evaluated against what was demanded, when, and on what basis. A specific, well-grounded demand that was ignored is powerful evidence the refusing party had something to hide. A belligerent or legally unfounded one is powerful evidence for the other side. It should be drafted as though a judge will read it. One will.

What the Records Actually Show

When records are produced, the picture is frequently worse than what the minority suspected.

Unauthorized compensation is the single most common form of minority oppression. The controlling owner puts himself, a family member, or a friendly entity on the payroll at a salary that was never approved and bears no relationship to services rendered. Sometimes it is a management fee paid to a related company. Sometimes it is rent paid to a building owned by the controlling party personally. The structure varies. The economic effect is the same. Money that should flow to all owners as profit flows instead to one owner as expense.

Related-party transactions present overlapping issues. A managing member who directs company business to a vendor he controls has a conflict of interest that may or may not have been disclosed or approved. Under both the BCL and the LLC Law, a transaction in which a fiduciary has a personal financial interest is subject to challenge unless it was properly disclosed and approved, or can be shown to have been fair to the company.

Misappropriation of funds is more straightforward but harder to prove without records. Cash businesses are especially vulnerable. So are businesses where one person has sole signatory authority over accounts and no one else reviews bank statements. When books and records reveal unexplained transfers to the controlling party’s personal accounts, payments to entities that cannot be identified, or a pattern of expenses with no supporting documentation, the inference of misappropriation is not difficult to draw.

Forged documents arise more often than most people expect. In partnership and LLC disputes, the most common forgeries are guaranties purportedly signed by a minority member, loan documents, and amended operating agreements. If a document exists that the minority does not recognize (a supposed personal guarantee of company debt, an amendment stripping minority rights), its authenticity can and should be challenged through handwriting analysis and forensic document examination. Courts have not been reluctant to impose sanctions, including adverse inference instructions, when forgery is proven.

Minority Oppression: The Legal Framework

New York Business Corporation Law Section 1104-a is the centerpiece of minority protection in the corporate context. A shareholder who holds at least twenty percent of all outstanding shares of a closely held corporation may petition for dissolution (a court-supervised winding up of the company) on the ground that the majority has engaged in illegal, fraudulent, or oppressive actions toward the petitioner.

Oppression under Section 1104-a has a specific meaning in New York courts. The leading case is Matter of Kemp & Beatley, Inc., decided by the Court of Appeals in 1984. The court defined oppressive conduct as behavior that substantially defeats the reasonable expectations of the minority shareholder held when she became involved in the business. In a closely held corporation, those expectations typically include employment, a voice in management, and participation in profits. When the majority systematically denies all three, oppression exists even if no single act is technically illegal.

Courts look at the totality of conduct. Removal from employment, exclusion from decisions, cessation of dividends, dilution of the minority’s interest: each is relevant, and together they state a claim for dissolution or a forced buyout.

The court has discretion under BCL 1104-a to order dissolution or, in the alternative, to order the majority to purchase the minority’s shares at fair value. The buyout option is frequently pursued by majority shareholders who want to preserve the business. It produces litigation over valuation. Valuation methodologies vary, and whether a minority discount applies has been contested. The Court of Appeals held in Matter of Friedman v. Beway Realty Corp. that discounts for lack of control and lack of marketability are not appropriate in a Section 1104-a proceeding, because the very point of the remedy is to compensate the minority for being squeezed out.

The twenty percent threshold under Section 1104-a is a real limitation. A minority shareholder holding less than twenty percent cannot petition for dissolution on oppression grounds. That does not leave smaller holders without options. Breach of fiduciary duty claims, derivative actions, and the right to an accounting are available regardless of ownership percentage. And if the conduct rises to fraud or illegality rather than oppression, BCL Section 1104 provides a separate dissolution remedy with no percentage requirement.

For LLCs, the picture is more favorable. LLC Law Section 702 has no ownership threshold. Any member, regardless of the size of her interest, may petition for judicial dissolution when it is not reasonably practicable to carry on the business in conformity with the operating agreement. New York courts have interpreted this provision strictly. A deadlock among members may satisfy the standard; so may a demonstrated pattern of fraud or misappropriation. But mere discord or financial losses do not.

The Squeeze-Out and the Freeze-Out

A squeeze-out is the set of tactics by which a majority attempts to force the minority to sell. A freeze-out more specifically refers to a transaction, typically a merger or recapitalization, designed to eliminate the minority interest entirely.

In the closely held business context, the squeeze-out is usually informal. The minority is removed from employment and thereby denied her salary. Distributions are stopped. The majority begins paying itself through salaries, bonuses, and management fees rather than through dividends or distributions that would benefit all owners proportionally. The minority is effectively locked into her investment with no return and no exit, while the majority prospers.

This conduct is actionable. It fits squarely within the Kemp & Beatley framework for corporations and within the mismanagement and breach of fiduciary duty theories available to LLC members. The difficulty is not the law; the difficulty is evidence and endurance. Squeeze-outs are often gradual, and litigation is expensive.

The freeze-out merger presents a different set of issues. In a formal freeze-out, the majority causes the company to merge into a new entity or to recapitalize on terms that eliminate the minority. New York’s appraisal remedy under BCL Section 623 allows a dissenting shareholder to receive fair value for her shares rather than the merger consideration. The appraisal proceeding is separate litigation, and it too turns heavily on valuation.

One common tactic that courts have addressed is the dilutive issuance. The majority causes the company to issue new shares, at a favorable price, to the controlling shareholder or a related entity, thereby diluting the minority’s percentage interest. If the issuance was not made on fair terms and was not approved by disinterested parties, it may be challenged as a breach of fiduciary duty. The same analysis applies to LLC membership interest dilutions effected through amendments to the operating agreement.

Derivative Actions and Direct Claims

When the wrongdoing is to the company (funds taken from the treasury, contracts diverted to the controlling party), the injured party is technically the company, not the individual member or shareholder. The owner’s loss is derivative of the company’s loss.

In New York, a derivative suit under BCL Section 626 requires that the plaintiff have been a shareholder at the time of the alleged wrong and remain a shareholder throughout the litigation. For LLCs, the derivative mechanism is less clearly codified, but courts have recognized the right, applying the BCL by analogy.

Before filing a derivative suit, a shareholder must make a demand on the board of directors to take action or plead with particularity why such demand would be futile. Demand futility exists where a majority of the board is interested in the challenged transaction or where the board lacks independence from the alleged wrongdoer. In the closely held company context, where the alleged wrongdoer is the controlling shareholder and the board is either the wrongdoer himself or his appointees, demand futility is often not difficult to establish.

Direct claims are those personal to the individual owner. A claim that the majority has wrongfully excluded the minority from management, wrongfully terminated her employment, or wrongfully refused to pay distributions owed under the governing documents is a direct claim. So is a claim for breach of a shareholder agreement or operating agreement provision that directly benefits the minority member.

The distinction between direct and derivative claims matters procedurally and in terms of where the recovery goes. Derivative recoveries go to the company. Direct recoveries go to the individual plaintiff. In a case involving both categories of wrongdoing, as is common, the pleadings must be carefully structured to preserve both.

Fiduciary Duties in the Closely Held Business

Majority shareholders in a closely held corporation owe fiduciary duties to minority shareholders under New York law. The Court of Appeals confirmed this in Ingle v. Glamore Motor Sales, Inc., and the principle is well established. The duty runs between the owners directly, not merely through the corporate form, and it imposes obligations of good faith, fairness, and loyalty.

For LLCs, the statute provides that members and managers owe a duty of loyalty and a duty of care. The duty of loyalty requires the fiduciary to account to the company and its members for any benefit derived from the conduct of the company’s business, to refrain from competing with the company, and to deal fairly in any transaction in which the fiduciary has a personal interest. These duties can be modified by the operating agreement, but they cannot be eliminated entirely. An agreement cannot immunize outright fraud.

Most minority oppression claims are built on fiduciary duty. When the majority pays itself excessive compensation, that is a breach of the duty of loyalty. When it approves related-party transactions without disclosure or fair terms, same result.

The corporate opportunity doctrine comes up often in closely held business disputes. A fiduciary who diverts to herself a business opportunity the company would have been interested in and financially capable of pursuing is liable to account for all profits derived from it. It does not matter whether the opportunity came to the fiduciary personally or through her role in the company. A managing member who bids secretly on a contract the company was also pursuing and routes it to a related entity has taken something that was not his. New York courts look at whether the company had an existing interest or expectancy in the opportunity and whether the fiduciary's involvement created a conflict.

When the business judgment rule does not protect a challenged transaction because the fiduciary had a personal financial interest or lacked independence, courts apply the entire fairness standard. The burden shifts to the fiduciary to establish that the process was fair (fair dealing) and that the price or terms were fair (fair price). This is a demanding standard, and it is frequently dispositive in closely held business disputes where related-party transactions are approved without any independent review.

The accounting remedy is especially useful in fiduciary duty cases. A court can order the fiduciary to account for all transactions in which she was involved and all benefits she derived from the relationship. The accounting remedy reaches misconduct that ordinary books and records may not reveal. It is ancient. It remains effective.

The Litigation Sequence

Preservation is critical. When litigation appears likely, both parties should be on notice of their preservation obligations. Electronic records (QuickBooks files, bank statements, email, text messages) can be deleted or altered. A litigation hold should be implemented immediately. If there is reason to believe that records are being destroyed, a temporary restraining order may be appropriate.

Motion practice in a business dispute follows a demanding rhythm. A motion to dismiss tests the pleadings. Summary judgment tests whether the documentary record resolves the case without trial. Discovery disputes over scope, privilege assertions, and production adequacy generate satellite litigation that can consume months. Forensic accounting is standard in misappropriation cases: a forensic accountant can reconstruct transactions from bank records and tax returns, identify unexplained transfers, and present damages in a form that survives cross-examination. The forensic report is typically one of the two or three most consequential documents in the case.

Most dissolution petitions resolve through a negotiated buyout before trial. Filing promptly after records are analyzed signals that the minority is serious, and the prospect of a court-ordered wind-down concentrates the other side's attention in ways that letters and demands do not.

Valuation and the Exit

Many business disputes end in a buyout rather than a trial. The minority exits; the majority pays. The fight is over price.

New York courts apply a fair value standard in statutory appraisal proceedings, which means the pro-rata value of the minority’s interest without discount for its minority status or lack of marketability. This standard is more favorable to the minority than the fair market value standard that would apply in an arm’s length sale, which would typically include both discounts.

Valuation methodologies in business disputes include the income approach (discounted cash flow analysis), the market approach (comparable company and transaction analysis), and the asset-based approach. Courts assign weight to each methodology based on the nature of the business. A service business with recurring revenue is typically valued on an income basis; a real estate holding company on its assets; a hybrid business requires a blended analysis. Goodwill is frequently the most contested component. Where goodwill is largely personal to the controlling owner, it has limited value to the minority. Where it is enterprise goodwill independent of any individual, it can dominate the valuation. Experts on each side frequently reach very different conclusions, and spreads of two or three to one are not unusual. The court is the trier of fact on value, and the quality of each expert's analysis and her performance under cross-examination typically determines which number the court credits.

Knowing what a court would actually award shapes what either side will accept. If the minority’s expert projects a value of ten million dollars and the majority’s expert projects three million, the settlement range is somewhere in between, and the parties’ litigation risk and litigation cost shape where in that range the deal gets done.

A negotiated exit should be documented carefully. A release, a covenant not to sue, a non-disparagement clause, and provisions addressing outstanding distributions and compensation are all standard. The exit agreement should address every aspect of the parties’ relationship going forward, because once the papers are signed, the right to re-open old grievances is gone.

Practical Observations

The starting point is documentation. The narrative of the dispute matters as much as the financial record: who said what, when requests were made, when they were refused, what was promised and not delivered. A coherent timeline, supported by contemporaneous records, is persuasive. Reconstruction after the fact is not.

Speed matters on preservation. Electronic financial records are fragile. Access to accounting software can be revoked. Bank portals can be changed. If the minority is being pushed out of operational control, the window to obtain unaltered records may be short.

The governing documents should be read early. The operating agreement may contain provisions more favorable than the statutory defaults, or ones that complicate the minority’s options. Either way, that review is not optional.

Limitations periods deserve attention. Claims for breach of fiduciary duty in New York are generally subject to a three-year period under CPLR 214(4), though the discovery rule can toll the clock when misconduct was concealed. Waiting too long to assert claims that were known or knowable can forfeit them entirely.

The tax dimension is worth considering. The structure of a buyout has tax consequences for both parties. A payment structured as a redemption of equity interest is treated differently from a payment structured as compensation or a settlement of claims. These distinctions affect the net economic result for both sides and should be part of the negotiation from the outset.

Assessing the counterparty realistically matters as much as assessing the claim. A judgment is not money. A majority party with hidden assets, a history of noncompliance, or a demonstrated willingness to transfer assets to evade obligations may present collection risk that affects the calculus of whether to litigate and on what terms. A negotiated exit at a modest discount may be more valuable in present dollars than a full award that takes three years and a contempt motion to collect.

When Civil Wrong Becomes Criminal Act

Some business disputes reach a point where the conduct is not a breach of fiduciary duty or a contract violation. It is a crime.

The most common financial crimes in the closely held business context are larceny, embezzlement, and scheme to defraud under New York Penal Law. Grand larceny in the third degree covers theft of property worth more than three thousand dollars; the second degree covers amounts over fifty thousand; the first degree, over one million. Embezzlement is not a separate statutory category in New York. It is prosecuted as larceny by an employee or fiduciary. A managing member who systematically transfers company funds to his personal accounts has committed larceny by embezzlement, and the dollar thresholds are easily met in a commercial dispute.

Scheme to defraud under Penal Law Article 190 is broader. It covers any systematic course of conduct intended to defraud or obtain property from ten or more persons. In a business context, this encompasses fraud on vendors, customers, investors, or lenders. A controlling owner who submits inflated invoices to the company from a fictitious vendor, diverts the payments to himself, and keeps the books to conceal the scheme has committed scheme to defraud in the first degree, a Class E felony (the lowest felony tier under New York law, but a felony nonetheless).

Federal exposure is frequently more serious. Wire fraud under 18 U.S.C. § 1343 covers any scheme to defraud that uses interstate wire communications (emails and electronic fund transfers) and carries penalties of up to twenty years per count. Bank fraud under 18 U.S.C. § 1344 applies when the scheme targets a financial institution or uses false pretenses to obtain bank funds. The Racketeer Influenced and Corrupt Organizations Act is available when there is a pattern of predicate offenses and carries civil treble damages in addition to criminal penalties, though courts scrutinize RICO pleadings closely and the pleading standard is demanding.

The minority owner who discovers what appears to be criminal financial conduct faces a decision that civil litigators sometimes handle poorly: whether to involve law enforcement, and when. There is no legal obligation on a private party to report financial crime to prosecutors. But there is a strategic dimension. A criminal referral or complaint can significantly change the dynamics of a business dispute. Law enforcement has tools civil litigants cannot match: grand jury subpoenas, search warrants, forensic resources. A federal investigation can unlock financial records that would take years of civil discovery to obtain.

The better approach in most cases is to develop the civil record as far as possible while keeping the criminal referral option open. If the wrongdoing is serious enough and the evidence is clear enough, the threat alone may resolve the dispute. A controlling party who knows the file is being assembled and the DA’s office is a phone call away frequently becomes more cooperative about a buyout at a fair price.

A separate question arises when the minority discovers ongoing criminal conduct and has to decide whether to stay. Fiduciary duties and non-compete obligations under the LLC Law or an operating agreement do not automatically dissolve because the majority is engaged in fraud. The majority could invoke them. But doing so would require affirmative litigation, and any court asked to enforce those obligations against a minority who exited a business engaged in active misconduct would hear an unclean hands defense it could not easily ignore. The more pressing concern for the minority in that situation is her own exposure. A member who remains associated with a business that is misappropriating customer deposits, for example, risks being drawn into the liability that conduct creates. At some point the safer course is to exit, document the reasons, and in appropriate circumstances report. The majority’s theoretical statutory claims do not create an obligation to remain.

The Duty to Investigate and the Danger of Ignorance

One of the more uncomfortable questions in equity disputes is how much a non-controlling owner is expected to know about the company's affairs as an owner, and what happens when the answer turns out to be: not enough.

Under New York law, a director of a corporation has an affirmative duty of oversight. The business judgment rule protects directors who act in good faith and with reasonable care, but it does not protect willful ignorance. Courts have recognized that a director who consciously disregards red flags cannot claim the protection of the business judgment rule for the resulting harm. This is sometimes called the Caremark duty, after the Delaware Chancery Court decision that articulated the standard in the corporate context and which New York courts have applied by analogy.

For LLC members who hold management roles, the duty is similar in principle if less clearly codified. A member-manager who oversees operations and ignores obvious signs of financial misconduct may be liable for the resulting harm. The standard is not perfection; it is reasonable oversight consistent with the responsibilities that come with an equity stake in the company.

The duty of oversight does not impose a duty to interrogate. A controlling shareholder who is misappropriating funds will typically have built safeguards to keep the minority from discovering it: restricted access to accounts, controlled financial reporting, a bookkeeper who answers only to him. A minority member, even one with a management title, is not imputed to have discovered what she was deliberately kept from seeing. The standard is what a reasonable person in her position, with her level of access, would have done. Courts draw a clear line between a director who ignored obvious warning signs and one who was systematically excluded from the information that would have revealed the problem. That distinction should inform how the minority owner assesses her next steps. A minority member who had no meaningful access to the books, who made requests that were refused or deflected, and who took reasonable steps to understand what was happening is in a defensible position. She should not let fear of being deemed complicit deter her from going to law enforcement, seeking judicial relief, or otherwise acting to stop conduct she did not participate in and could not have discovered without cooperation she was denied.

When wrongdoing is suspected, the right response is investigation. If access is pressed for and refused, every request and every deflection should be documented. That record may matter as much as the financial evidence if the misconduct later surfaces. Beyond the books-and-records demand, forensic accountants can analyze whatever records are available, counsel can conduct interviews, and a private investigator may be warranted. Courts look more favorably on a party who took reasonable steps to understand what was happening than on one who accepted assurances that everything was fine.

Being Blindsided: When You Discover It All at Once

The most disorienting scenario in business disputes is not the slow-developing squeeze-out. It is the sudden discovery that the business an owner thought she had a stake in has been systematically looted, that the company's financial statements were fabricated, that money believed to be in the accounts is simply gone. It happens. And how the owner responds in the first hours determines a great deal about what comes next.

The first twenty-four to seventy-two hours are critical. Evidence can be destroyed quickly. Bank accounts can be drained. Accounting records can be altered or deleted. If the wrongdoer knows that discovery is imminent, the window closes fast. The first call is to counsel, not to the partner being accused.

An emergency application for a temporary restraining order and preliminary injunction can freeze the situation while the facts are developed. A TRO application in a business dispute may seek to enjoin the transfer of company assets, to compel the preservation of books and records, to prevent the dissipation of funds, and in appropriate cases to appoint a temporary receiver to manage the business pending resolution.

Asset tracing is often necessary when funds have been misappropriated. A forensic accountant with experience in tracing can follow the money through bank accounts, wire transfers, cryptocurrency wallets, and related entities. New York courts can issue orders requiring financial institutions to produce records on an expedited basis in aid of litigation, and federal courts have similar authority. If assets have been transferred to related parties or offshore, turnover proceedings and judgment enforcement mechanisms become relevant.

One question the discovering owner must address early is her own potential exposure. An equityholder who also holds an officer title, who signed the company's tax returns, or whose name appears on representations lenders relied on, faces a more complicated position than a purely passive investor. The remedies question and the exposure question require separate analysis, and in some cases separate counsel.

Who Controls the Books: And Why It Decides Everything

In almost every business dispute involving financial misconduct, the controlling party controls the books. Control of the records is control of the narrative.

In most closely held businesses, financial records live in one place: the accounting software. QuickBooks is the dominant platform for small and mid-sized businesses. It maintains a complete record of all transactions, journal entries, payroll, accounts payable, and accounts receivable. Access is controlled by the administrator. In a business where one owner is the administrator and the others are not, those others have no practical ability to see what has been recorded or detect whether entries have been altered.

The same is true of bank accounts. A single signatory on the company’s bank accounts can authorize transfers, write checks, and access online portals without any other owner’s knowledge or approval. Many closely held businesses operate this way as a matter of efficiency. They discover the problem only when the relationship breaks down.

When a dispute arises, the controlling party can move quickly. QuickBooks access can be revoked. Bank portals can be reprogrammed. Prior period entries can be altered, since the software does not prevent retroactive modification without a password most users never set. Physical records can disappear. The longer the dispute goes without legal intervention, the more shaped the record becomes.

A word on the limits of ordinary accounting. A routine audit or compilation takes the underlying data as presented by management. The outside accountant works from the QuickBooks file or equivalent that the controlling party controls, reconciles it against bank statements provided by management, and certifies numbers that reflect what she was given. She is not conducting a forensic investigation. If entries have been manipulated, if transfers have been characterized as legitimate expenses, if related-party transactions have been structured to look arm's-length, a routine audit will generally not catch it. Forensic accounting is a different discipline: it tests the source data, traces cash flows independently, reconstructs transactions from primary records, and asks whether the picture the books present is consistent with the underlying reality. That is the tool that matters in a dispute.

If the company maintains its books through an outside accountant or bookkeeper, that professional has an independent duty of competence and in some cases a duty to report irregularities. A demand on the outside accountant for records is sometimes more productive than a demand on the controlling party directly. The accountant’s files are also subject to subpoena in litigation.

A forensic image of accounting data (a bit-for-bit copy of the database at a given moment) can capture the record before alteration. This requires either access to the system (which the minority may not have) or a court order compelling production. Courts grant such orders when prior noncompliance or record alteration is documented.

The audit trail function in QuickBooks and similar platforms records every change to the database: who made it and when. If entries have been backdated or amounts altered, the audit trail reveals it. A forensic accountant can extract and analyze this data, and courts have treated audit trail evidence of intentional manipulation as highly probative. Bank records present a parallel question: a court order can compel the institution to produce records directly, without the controlling party's cooperation. In an emergency, an account freeze combined with forensic imaging, both obtainable on short notice, can secure the evidentiary foundation before the wrongdoer has time to react.

What You Can Say to Clients and Customers, and What You Cannot

One of the most practically urgent questions in a business dispute is what the departing owner can say to the company’s clients and customers. The answer is governed by defamation law, non-disparagement agreements, and unfair competition principles, and requires careful thought before any communication goes out.

The foundational principle of defamation law is that a false statement of fact, communicated to a third party, that damages someone’s reputation is actionable. Truth is an absolute defense. A statement that is accurate, even one that reflects badly on the controlling party, is not defamation. The challenge in a business dispute is that the facts are often contested, and a statement made in good faith based on available information may later be shown to be inaccurate in some particular.

Communications with clients and customers must be evaluated on two tracks. First, are they true? Second, are they necessary? A communication that says simply that the owner has left the company and clients are welcome to continue working with her directly is generally permissible and legally safe, provided it does not violate any non-solicitation covenant. A communication that attributes financial wrongdoing to the controlling party by name, in a letter sent to every company client, is a different matter entirely.

Trade libel (also called product disparagement or injurious falsehood) covers false statements about a business rather than an individual. A statement that the company's accounts are frozen or its principals are under investigation, when untrue, may give rise to a trade libel claim. New York courts require proof of actual damages, but reputational harm to a business can be substantial. Non-disparagement clauses add a contractual layer: a party bound by such a provision cannot make disparaging statements about the company or its principals even if those statements are true. Before any communication reaches a client or customer, counsel must review whatever agreements are in place and evaluate compliance.

The privilege doctrine provides important but limited protection. Statements made in judicial proceedings (complaints, testimony, motion papers) are absolutely privileged under New York law and cannot form the basis of a defamation claim. Statements in good faith pre-litigation communications to opposing counsel are conditionally privileged. These protections do not extend to statements made to clients, to the press, or to the general public outside the litigation context.

When clients ask what happened, the answer should be accurate and short. Confirming that there is a dispute, that the owner has departed or is in the process of departing, and that she is available to serve them directly is sufficient for most business purposes. The impulse to explain in detail why the other party is a bad actor is understandable but legally dangerous. Clients do not need the full story to decide whether to follow the departing owner. And the risk of a defamation counterclaim, even a meritless one, adds complexity and cost to an already expensive proceeding.

Publicity: Strategic Tool or Loaded Weapon

Business disputes become public in different ways: a filed complaint, a press call from one side, a story that writes itself. Getting it wrong can damage both sides.

New York courts have strong norms against litigation by press release. A party that takes its case to the media before it is fully developed in the courtroom runs several risks. Statements to the press can be used against the party at trial. A public accusation that turns out to be wrong, or that overstates the evidence, undermines credibility. Judges notice when parties use publicity as a tactical weapon, and some take it badly. The litigation context is the right forum for adjudicating financial misconduct; the court of public opinion is unreliable and its verdicts unpredictable.

That said, there are circumstances in which publicity is not just unavoidable but strategically valuable. A controlling party who is engaged in ongoing financial misconduct and believes the minority is isolated and powerless may recalibrate that assessment when a lawsuit is filed and covered in the trade press. Publicly held companies, regulated businesses, and businesses with significant reputational stakes are especially sensitive to adverse publicity. A complaint that documents financial misconduct in specific detail, filed in a public court and covered by a reporter, can produce a settlement conversation that was previously unavailable.

Regulatory disclosure is a related but distinct lever. If the company operates in a regulated industry and the misconduct involves regulatory violations, a referral to the relevant regulator may be both appropriate and strategically significant, producing license revocations or industry bars that civil litigation cannot. The decision to refer should be made deliberately with counsel, as it transfers a degree of control to the government and can complicate the timing of parallel civil proceedings.

The controlling party will typically attempt to use publicity preemptively, framing the minority as a disgruntled ex-partner with a meritless vendetta. The best counter to that narrative is not a competing press campaign. It is a factual record. Press coverage of business disputes tends to follow the documents. A party with a strong, publicly filed evidentiary record is better positioned than one relying on oral representations.

Non-disparagement clauses apply equally to public statements. A party who speaks to a reporter in violation of such a clause has created a separate breach of contract claim regardless of the truth of what was said.

The practical prescription is straightforward: the strongest possible complaint should be filed, the factual record made in court, and the documents left to speak. If the story is covered, let it be covered on the basis of the court file rather than inflammatory statements made to reporters outside the courthouse.

The Other Side of the Table: Removing a Partner Who Is Not Pulling Weight

The preceding sections address the minority owner's position. But the law of business divorce is not one-sided. Businesses also face the problem of the partner who has stopped contributing, the member who has gone dark, the co-founder who is drawing a salary and a distribution while adding nothing, or the managing partner whose conduct has crossed the line from disappointing to actionable. Removing that person, or compelling an exit on fair terms, is its own body of law, and it is harder than most clients expect.

A well-drafted operating agreement anticipates these situations. It defines what constitutes a default, specifies the process for removal, establishes whether removal triggers a buyout and at what price, and sets out the dispute resolution mechanism. A poorly drafted agreement, or none at all, leaves the parties with the statutory defaults, which are designed to protect the member being removed rather than the members seeking removal.

Expulsion of an LLC Member: The Contractual Framework

New York’s LLC Law does not provide a general statutory mechanism for expelling a member from an LLC against her will. Unlike states that have adopted the Revised Uniform LLC Act (which permits expulsion by unanimous vote in certain circumstances), New York’s statute is largely silent on the subject. The right to expel a member in New York is almost entirely a matter of contract, and it exists only if the operating agreement creates it.

A well-structured expulsion provision will identify the triggering events precisely. Common triggers include: breach of the operating agreement that is not cured within a specified period; conviction of a felony; conduct that constitutes fraud, gross negligence, or willful misconduct in connection with the company; material failure to perform agreed-upon services or capital contributions; bankruptcy or insolvency of the member; or conduct that causes material harm to the company’s business or reputation. The specificity of these definitions matters. A vague trigger ("conduct detrimental to the company") invites litigation over whether it applies. A precise trigger, defined with examples and supported by a clear process, is far harder to contest.

The expulsion process must also be defined. Who makes the determination that the trigger has been met? Is it a vote of the non-defaulting members? A vote of the board of managers? Is notice required? Is there a cure period? Is there a right to be heard before the vote is taken? Absent a defined process, a purported expulsion is vulnerable to challenge as a denial of the member’s contractual rights. Courts take procedural defects seriously. The consequence of expulsion is loss of an ownership interest.

The price is where the real negotiation happens. Options include: fair market value as determined by an agreed methodology; book value; a formula price based on a multiple of earnings; the lower of fair market value and original contribution (for cause situations); and zero (in the most extreme for-cause expulsions, sometimes called a “dilute to nothing” or “forfeit” provision). Courts have enforced contractual forfeitures of equity as a remedy for breach, but such provisions must be clear and unambiguous, and courts will scrutinize them carefully when the conduct at issue is disputed.

Removing an Officer or Director

The removal of an officer or director from a corporation is governed by the Business Corporation Law and by the certificate of incorporation and bylaws. Under BCL Section 706, a director may be removed for cause by action of the shareholders, and a corporation’s certificate of incorporation may permit removal without cause. Officers serve at the pleasure of the board and may generally be removed at any time, with or without cause, unless an employment contract specifies otherwise.

The distinction between removal from a corporate role and removal of an ownership interest is critical and often misunderstood by clients. Removing someone as CEO does not remove her as a shareholder. Removing someone as a director does not reduce her equity interest. In a closely held corporation, the owner who is removed from her operating role still holds her shares, still has inspection rights, still has the right to vote on major corporate actions, and still has a claim to any distributions declared. The removal from management, if not accompanied by a buyout, simply converts a participant into a passive investor who may then assert oppression under BCL 1104-a.

For LLCs, the removal of a manager is similarly governed by the operating agreement. A manager who is not also a member can be removed pursuant to whatever provisions the agreement establishes. A manager who is also a member presents the more complex situation: removal from the managerial role is possible if the agreement permits it, but the membership interest remains unless separately addressed. Many disputes that appear to be about management control are really about the equity, and the parties benefit from counsel who can identify which problem is actually being solved at each step.

The Non-Participating Partner: Capital, Sweat Equity, and the Failure to Deliver

One of the most common disputes in early-stage businesses involves the founder or partner who received equity in exchange for a promise of capital, services, or connections that was never fulfilled. The other owners find themselves doing the work, funding the business, and watching a colleague take a distribution on an interest that was never truly earned.

The legal treatment of this situation depends heavily on how the equity was structured. If the interest was granted as fully vested consideration for a promise that was not performed, the remedy is generally a breach of contract claim, not a right to claw back the equity automatically. Unless the operating agreement contains a vesting schedule, a repurchase right, or an express condition that the equity is contingent on performance of specified obligations, the equity is the equity. A partner who promised to introduce investors and never did is liable for breach of the underlying promise, but she may still own her membership interest.

Vesting schedules are the standard solution, and the failure to include them in founding documents is one of the most consequential mistakes early-stage businesses make. Without a vesting schedule, the analysis reverts to breach of contract principles and the economic outcome is far less predictable.

Capital call failures present a related but distinct issue. Most operating agreements include provisions requiring members to make additional capital contributions if called by the managing member or by majority vote. The consequences of failing to meet a capital call vary by agreement: some impose dilution (the non-contributing member’s interest is reduced proportionally); some treat the shortfall as a loan from the contributing members, bearing interest; some permit the other members to make the contribution on behalf of the defaulting member and treat it as a priority return; and some permit expulsion for failure to meet a capital call.

The sweat equity dispute (where the promised contribution was services rather than cash) is harder to resolve because the value of services is inherently subjective and the failure to perform is harder to quantify. The most defensible position for the party asserting non-performance is a contemporaneous record: emails requesting that the partner perform her obligations, documented instances of the work not being done, communications from clients or vendors about the absence of promised services. Without that record, the claim devolves into a dispute over subjective assessments of effort, which courts find difficult to resolve on motion and expensive to litigate at trial.

Removing a Partner for Malfeasance: The For-Cause Expulsion

When the partner being removed has not merely failed to contribute but has actively harmed the business through fraud, theft, competing conduct, or disclosure of confidential information, the equities shift considerably. A for-cause expulsion, properly documented and properly executed, is among the more defensible actions the remaining owners can take. Poorly executed, it is an invitation to counterclaims.

Before any expulsion vote is taken, the evidentiary record should already be assembled: the financial records showing the unauthorized transfers, the communications evidencing the disclosure of trade secrets, the evidence of the competing business, the records of the self-dealing transaction. The expulsion should not be the opening shot in the dispute; it should be the culmination of an investigation that has already established the facts.

Notice and an opportunity to be heard, even if not technically required by the operating agreement, are advisable regardless. A member who can demonstrate that she was expelled without warning and without any opportunity to respond to the allegations will have a more sympathetic position before a court reviewing the process. Notice of the specific allegations, a defined period to respond, and a recorded vote with minutes that document the basis for the decision are all protective steps. They add time. They add protection.

The intersection of expulsion and concurrent civil claims requires careful sequencing. The expulsion and any civil claims for damages will proceed together. The expelled member will assert that the expulsion was wrongful and seek reinstatement or damages. The company will assert that the expulsion was justified and press its affirmative claims for the harm caused. Each side’s claims are the mirror image of the other’s. The resolution is usually a global settlement that addresses both the ownership question and the damages question together.

Non-compete and non-solicitation obligations often come into focus at expulsion. A member who is expelled, especially one who has been competing with the company, may be subject to post-expulsion non-compete obligations under the operating agreement. New York courts are skeptical of broad non-compete clauses, applying a reasonableness standard that evaluates geographic scope, duration, and whether the restriction is necessary to protect a legitimate business interest. A non-compete clause that is drafted overly broadly may be unenforceable in its entirety under New York’s “blue pencil” doctrine, or the court may narrow it to what is reasonable.

Won't Leave, Won't Deal: The Immovable Partner

The hardest removal problem has nothing to do with malfeasance. It is the partner who has done nothing technically wrong, isn't in breach of anything, and has simply decided that her interest is worth more than anyone is offering. She won't sell at a reasonable price. She won't negotiate in good faith. She shows up at meetings, copies counsel on everything, and makes clear that an exit will come at a premium. This is not a legal problem in the first instance. It is a negotiating problem. But the law shapes the negotiating range, and understanding that range is the job.

If there is no expulsion clause in the governing documents, there is no mechanism to remove her involuntarily without judicial intervention. She knows this. The practical tools available to the majority are limited: dissolution proceedings, a receiver, or a buyout at whatever price she will accept. Each carries costs, and she is counting on the majority to find them prohibitive.

The dissolution threat is the most powerful lever. Filing a petition for dissolution under BCL Section 1104-a or LLC Law Section 702 puts the business itself at risk, and a majority owner who has spent years building something has more to lose from a court-ordered wind-down than a passive minority member who is already collecting nothing. That asymmetry is what makes dissolution filings so frequently effective even when the petitioner does not actually want dissolution. The petition signals a credible willingness to force a wind-down rather than continue subsidizing a holdout. Most holdouts respond to that signal.

The agitator variant is more corrosive than the passive holdout. This is the partner who isn't stealing but is making operations impossible: calling extraordinary meetings on short notice, demanding special audits of routine transactions, copying regulators on complaints, threatening personal liability for every business decision, and generally poisoning relationships with employees, lenders, and customers. The conduct may not rise to the legal definition of oppression or malfeasance, but it imposes real costs. And it may, in fact, cross a line.

Courts have recognized that the duty of loyalty and the duty of good faith run to the company and to co-owners, not just from the majority to the minority. A member who systematically disrupts operations without legitimate basis, who uses her position to extract a settlement the facts do not support, or who threatens and initiates litigation as a negotiating tool rather than as a genuine remedy is herself in breach. The same fiduciary analysis that protects the minority from majority overreach can be turned against a minority member who weaponizes her position. A running record of the conduct should be kept: the meeting demands, the regulatory threats, the bad-faith positions. As in a dissolution proceeding, that record becomes the evidentiary foundation for any counterclaim.

The appointment of a temporary receiver is an underused tool in this context. A receiver displaces both parties from operational control and places the business under court supervision. For a partner who has been using operational interference as leverage, a receiver eliminates that leverage immediately. For a majority owner who has been reluctant to file because of reputational concerns, a receiver provides a neutral buffer. Courts do not appoint receivers casually, but a demonstrated pattern of disruption combined with a genuine impasse about value can provide sufficient grounds.

In most of these situations, the majority eventually pays a holdout premium. The question is how much. A partner who understands the legal ceiling on what a court would award in a dissolution proceeding (fair value, no marketability discount, no minority discount under BCL 1104-a) can negotiate more effectively than one who doesn't. If the partner's realistic litigation outcome is seven hundred fifty thousand dollars and she is demanding a million, the cost of that gap has to be weighed against the cost of two to three years of litigation, the disruption to the business, and the uncertainty of any contested proceeding. Sometimes the majority pays. The goal is to pay once, document the release carefully, and close the relationship completely.

Deadlock: When No One Can Be Removed Because No One Has Control

A deadlock occurs when ownership is so evenly divided that no decision can be made. The classic version is a fifty-fifty split with no tiebreaker.

New York’s Business Corporation Law addresses deadlock among shareholders and among directors separately. Under BCL Section 1104, a shareholder who holds fifty percent of all outstanding shares may petition for dissolution when the shareholders are so divided in voting power that they have been unable to elect directors for a period that includes at least two consecutive annual meeting dates. For deadlock at the director level, where the board cannot act because it is evenly split, BCL Section 1104 similarly permits dissolution.

For LLCs, deadlock is addressed through the operating agreement if it has been drafted with foresight, and through judicial dissolution under LLC Law Section 702 if it has not. Courts have found deadlock to be a basis for dissolution when the members are unable to agree on any matter of substance and the impasse is genuine and not tactical. A party manufacturing deadlock as strategy will find the court unreceptive.

The best protection against deadlock is contractual. A well-drafted operating agreement will include a tiebreaker mechanism: a designated member with a casting vote on specified matters; a dispute resolution process that escalates from negotiation to mediation to binding arbitration; a mandatory buy-sell provision triggered by deadlock; or an agreement to appoint a third-party manager with authority to break ties. Any of these is preferable to judicial dissolution, which is slow and expensive.

The buy-sell provision, sometimes called a shotgun clause or Texas shootout, is the most powerful deadlock resolution mechanism in the closely held business context. One party names a price; the other chooses which side to take. The forced symmetry creates honest pricing incentives. Buy-sell mechanisms are not risk-free (a party with greater liquidity can name a price the other cannot meet), but they beat litigation.

The Wrongful Removal Claim: When the Tables Turn

A partner who believes her expulsion was wrongful has several potential claims: fabricated cause, procedural defect, or bad-faith timing designed to cut her out before a windfall.

The primary claim is breach of contract: the expulsion violated the operating agreement or shareholder agreement, either because the specified trigger did not occur, the required process was not followed, or the price paid was not what the agreement required. Courts will enforce operating agreement provisions strictly, and an expulsion that does not comply with the agreement’s requirements is void or voidable.

Breach of fiduciary duty is a second theory. If the expulsion was orchestrated in bad faith (timed to eliminate the member before a transaction closes, or fabricated to avoid paying fair value), the conduct may constitute a breach of the duty of loyalty owed by the majority to the minority. The remedy for a wrongful expulsion in breach of fiduciary duty may include reinstatement, damages for the economic loss of the interest, and punitive damages in egregious cases.

The timing of the expulsion is often probative of the majority’s motive. An expulsion that occurs immediately before a significant liquidity event (a sale, a financing, the resolution of a major contract) raises an inference that the real objective was to eliminate the minority before the windfall, not to address the purported misconduct. Courts are alert to this pattern. Documentary evidence of the expulsion decision, including communications among the remaining owners about the timing and its financial implications, is among the most important evidence in a wrongful expulsion case.

A closely related problem arises in equity and startup disputes: the forced exit or buyout timed to occur just before unvested shares mature. Vesting schedules are designed to align incentives over time, but they also create a window of vulnerability. A controlling party who can manufacture a pretextual termination or a for-cause expulsion before a significant tranche vests captures the forfeiture as economic gain. The minority loses not only her position but equity she had substantially earned. Courts examine the timing of these transactions with the same skepticism they apply to pre-liquidity event expulsions. Where the record shows that a performance concern surfaced suddenly and conveniently as a vesting date approached, or that the buyout price bore no relationship to what the shares would have been worth had the minority remained, the inference of bad faith is available. Acceleration provisions in the governing documents, if any, are also relevant: many well-drafted agreements provide for accelerated vesting on certain termination events, and a controlling party who engineers a termination to avoid triggering acceleration has compounded the breach.

A wrongfully expelled member is not limited to monetary damages. A court can order reinstatement, restore her pre-expulsion status, and require an accounting of all distributions earned during the period of wrongful exclusion. Where the business increased in value after the expulsion, that accounting can be substantial.

Drafting for the Divorce You Hope Never Happens

The most important moment in any business relationship is the beginning. The operating agreement or shareholder agreement negotiated when everyone is aligned is the document that will govern the relationship when everything goes wrong. The founders who skip this document, or who adopt a generic template without thinking carefully about their specific situation, are making a decision they may spend years litigating.

Every governing document for a closely held business should address, at a minimum: the circumstances under which a member or shareholder can be expelled and the process for doing so; the price to be paid for an interest on voluntary exit, involuntary exit, death, disability, or divorce; who has authority to make financial decisions and under what limitations; what information rights all owners have and how often financial statements must be produced; what happens on deadlock; what non-compete and non-solicitation obligations apply during and after the relationship; and what dispute resolution mechanism governs disagreements.

The arbitration clause deserves specific attention. Many operating agreements include mandatory arbitration provisions, and parties sometimes assume that a broadly worded clause will sweep in all disputes. It will not. New York courts have held that statutory dissolution proceedings under BCL 1104-a are not arbitrable, because the right to petition for dissolution is a statutory remedy that parties cannot contract away. An arbitration clause covering contract disputes and damages claims will not eliminate the court's jurisdiction over a dissolution petition. The two proceedings can run concurrently, and often do.

The price mechanism is the provision that most often generates litigation when it is absent. If the agreement is silent on price, every buyout becomes a valuation dispute. Specifying the methodology in advance, even if the resulting number is uncertain, removes a major source of controversy. Some agreements specify an annual appraisal process to keep the price current; others specify a formula; others provide that the parties will attempt to agree and submit to binding arbitration if they cannot. All of these are preferable to silence.

Conclusion

The law of closely held business disputes runs in both directions. Minority owners who have been pushed aside, exploited, or shut out have real remedies: books-and-records demands, dissolution proceedings, derivative suits, fiduciary duty claims, statutory appraisal. The process is not fast and it is not cheap. But the law does not leave them without recourse. And majority owners facing a non-contributing partner, an agitator, or a holdout have tools too, provided the governing documents were drafted with foresight and the removal is executed with care.

What minority owners often underestimate is their own leverage. The majority needs the business running. Public litigation over financial misconduct is costly beyond legal fees. The prospect of dissolution concentrates the mind. Most of these disputes settle. The goal on both sides is to settle on terms that reflect what a court would actually do, not on terms driven by who runs out of money or patience first.

Breaking up is hard to do. But knowing your rights makes it less hard.

April 2026

Michael Simon Baker is the principal of Michael S. Baker, P.C., practicing as NYBusiness.Law. He represents equity owners, closely held businesses, and institutional clients in business ownership disputes, dissolution and separation matters, commercial litigation, and business transactions, with a focus on the rights and remedies available to shareholders and LLC members in closely held companies. Before founding his practice, he was a partner and co-head of leveraged finance at Paul Hastings LLP and a partner at Shearman & Sterling LLP, where he advised investment banks, private equity sponsors, and corporate borrowers on complex financing transactions across multiple industries. He also brings experience in corporate strategy and business development. He can be reached at (646) 212-0194 or through NYBusiness.Law.

New Article • Private Credit / Restructuring / Corporate Finance

The Captive Tranche

When Majorities Rule: Efficiency or Theft

May 2026Full Article

A bond, a loan, a convertible note is sold into the market on a shared premise: the holder will act, when called upon, in the interest of the instrument. Company debt is typically divided into layers, called tranches, ranging from senior secured loans down through junior debt, unsecured bonds, and convertible notes. Each tranche has its own contract, its own voting rules, and its own economic logic. Holders within a tranche are presumed to share an economic interest, because they paid for the same exposure and face the same downside.

That presumption is the load-bearing wall of corporate finance. Trust indentures, credit agreements, and the class-voting mechanics of Chapter 11 all rest on it. A majority can bind a minority within a tranche because the majority is assumed to be motivated by the same logic as the minority. The system gives up unanimity in return for this efficiency, and the price is generally worth paying. Without majority binding, a single small holder could veto an otherwise sensible restructuring and extract rents from everyone else. Most consents, amendments, and plan votes that bind minorities are routine and proper.

The system has also long recognized that majority power must have limits. Standard credit agreements identify certain terms as sacred rights, requiring unanimous or affected-lender consent rather than a simple majority: the principal amount of a loan, the rate of interest, the scheduled date of any payment, the currency of payment, the pro rata sharing of payments, and the release of all or substantially all of the collateral or guarantees. The Trust Indenture Act, at Section 316(b), provides comparable protection for publicly issued bonds: the holder's right to receive payment of principal and interest, and to institute suit for enforcement, cannot be impaired without the consent of that holder. Sacred rights are the architectural acknowledgment that majority binding has limits.

The presumption strains when a holder of the majority of a junior tranche also holds a larger position in the senior tranche of the same capital structure. Such a holder may be called a cross-holder. Whatever destroys value at the junior level, if it preserves or increases value at the senior level, is rational for the cross-holder so long as the aggregate position improves. Minority junior holders who bought on the premise that the majority would be aligned with them may be dispossessed by the holders supposed to represent them. The phenomenon does not appear to have a name. This article calls it the captive tranche.

The conduct admits two characterizations. From the cross-holder's chair, it is efficient capital formation: concentrated positions accept losses on one instrument because the gains on another justify the trade in the aggregate. From the minority's chair, the same conduct is theft: voting rights the minority paid for are exercised against the minority by a holder whose economic interest lies elsewhere. The thesis of this article is that both characterizations are reasonable from inside their respective chairs, that the legal system has historically been right to protect majority binding rules as essential to efficient restructuring, and that the existing doctrinal toolkit, applied with attention to alignment rather than size, can distinguish the legitimate use of majority binding from the captive tranche without disturbing the binding rule itself.

How the captive tranche is built

A captive tranche can be built along one of several paths. None is improper on its face.

The first begins on the senior side. A private credit fund or hedge fund already holds the senior secured paper, and as distress accelerates, the holder or an affiliate accumulates the junior instrument at a discount. From the senior side, this is a sensible hedge: optionality at low cost. From the junior side, the buyer is acquiring instruments whose voting rights will determine outcomes the buyer's larger position has a stake in.

The second begins from the other direction. Most syndicated credit agreements (loans made by a group of lenders) contain disqualified institution lists barring designated funds from buying into the senior tranche. The DQ architecture is a market response to aggressive distressed strategies, intended to keep particular investors out during the life of the facility. The restrictions are leaky. They typically fall away upon default or bankruptcy, and they are worked around without formal expansion. Most credit agreements permit lenders to sell participations (shares of their exposure passed through to a third party) without borrower consent, and credit default swaps and total return swaps produce comparable economic exposure entirely outside the assignment regime. The opacity is structural: administrative agents maintain a Register of Lenders that records only holders of record, the entities formally appearing on the credit agreement and any assignment agreements. The Register does not record participants, even though participation grants typically pass through voting direction rights that let the participant instruct the lender of record how to vote. The borrower, the agent, and the rest of the syndicate see the Register names. The actual beneficial holders, including any cross-holder operating through a participation with voting pass-through, do not appear. A holder formally excluded from the senior tranche can therefore hold synthetic senior exposure equal to or exceeding its junior position, exercise effective voting control, and remain entirely off the Register.

The third path involves no prior position. The fund enters during the distress window and builds both sides at once: the junior in the open market, the senior through bridge or DIP participation (the debtor-in-possession financing extended in Chapter 11), through synthetic exposure, or through secondary purchases once DQ restrictions fall away. From the senior side, this is opportunistic positioning. From the minority junior holder's chair, the same trade puts the cross-holder on both sides of decisions the minority believed would be made by adversaries.

What the holder is buying, regardless of path, is votes. Consent rights, plan acceptance rights, and the ability to bind the other junior holders to outcomes the cross-holder will negotiate from the senior chair.

The economic logic is straightforward. Each dollar of senior exposure earns a recovery near par. Each dollar of junior exposure, on its own merits, might recover thirty cents. Each dollar of junior face acquired is also a vote, and the votes can be deployed to shift recovery from junior to senior. If the senior recovery improvement exceeds the junior recovery sacrifice, the trade is profitable. The cross-holder calls this portfolio construction. The minority calls it paying the cross-holder to vote against the minority's interest.

The position is typically concealed. There is no Schedule 13D for debt. An investor amassing a controlling stake in a company's debt has no general obligation to disclose it, even where a comparable equity stake would require filing. This is the empty voting problem Henry Hu and Bernard Black identified in equity markets nearly two decades ago, transposed into corporate credit. The vote has been decoupled from the economic interest.

The agent and the intercreditor

Two pieces of standard syndicated-finance architecture become, in the cross-holding context, the operational backbone of the captive tranche strategy.

The administrative agent is the lender group's appointed representative, responsible for notices, payments, collateral, remedies, and serving as the operational counterparty to the borrower for amendments and consents. The agent's exculpation clauses are sweeping, and its economic incentive runs toward staying on side with the lenders who direct it. The required lender threshold under most credit agreements is a simple majority of commitments. A holder controlling 50.1 percent of senior commitments controls the agent, and through the agent controls the timing of default notices, the framing of amendment packages presented to the syndicate, and the conditions under which collateral is released. Minority lenders see what the agent shows them.

An intercreditor agreement governs the relationship between classes of creditors. It allocates payment priority, lien priority, voting rights, standstill obligations (junior creditor agreements not to exercise certain remedies), and credit bid rights (the right to use a claim as currency in a sale of collateral). The senior class is given control. The junior class is given protection against the worst forms of senior overreach. The document assumes two sides bargaining at arms length.

That assumption strains when the cross-holder sits on both sides. The intercreditor functions, in operation, as an agreement among the cross-holder's various accounts. Standstills are observed only if the party with standing to enforce them sees enforcement as serving its aggregate position. Credit bid rights that should have functioned as a competitive constraint can be exercised in coordinated fashion. Objection rights that should have produced contested confirmation hearings can produce stipulations. The cross-holder calls this efficient resolution. The minority calls the alignment fictional, because it is internal to a single firm rather than across the two sides the document was drafted to mediate.

How value moves

The mechanisms by which the captive tranche moves value across tranches are the standard machinery of distressed credit.

In the pre-bankruptcy window, the cross-holder consents to amendments and exchange offers that strip the junior of its covenants or collateral in return for value transferred to the senior. Liability management exercises come in two main varieties. An uptier transaction moves favored lenders into a newly created super-senior tranche, demoting the others. A dropdown transaction moves valuable collateral into a new subsidiary outside the reach of existing lenders. Both require majority consent at the affected tranche, which the cross-holder supplies. These transactions are notable for what they accomplish through the side door. Sacred rights provisions explicitly prohibit a majority from impairing principal, interest, or scheduled payments, or releasing all or substantially all of the collateral, without the affected lender's individual consent. An uptier does not formally reduce the minority's principal, but it demotes the minority's claim to a position structurally subordinate to the new super-senior tranche, reducing likely recovery to a fraction of what it would have been. A dropdown does not formally release the collateral, but it moves the collateral to a new entity outside the minority's lien grant, leaving the claim unsecured in substance even where it remains secured in form. The LME architecture reaches by indirection the outcomes the sacred rights provisions were drafted to prevent.

In the bridge to filing, the cross-holder supports forbearance and bridge financings that move cash to the senior class. Disproportionate consent fees paid only to lenders agreeing to a particular package are themselves a form of within-tranche discrimination requiring majority cover, which the cross-holder provides.

In Chapter 11, the cross-holder lends or backstops the debtor-in-possession financing on terms that include rollups (conversion of pre-petition senior debt into post-petition DIP debt with higher priority), milestones (deadlines that compress the case to favor the lender), and priming liens (senior liens that take precedence over existing junior liens).

At confirmation, the cross-holder votes the captive junior block in favor of a plan that under-pays the class in the view of those who voted against. Under the Bankruptcy Code, a class accepts a plan if two-thirds in dollar amount and more than half in number of voting holders approve it. The captive block exceeds both thresholds. The class is deemed to have accepted, and the court never reaches the cramdown analysis under Section 1129(b), which permits a court to confirm a plan over a class's objection but only after testing the plan against statutory fairness requirements. The non-captive holders never receive the cramdown protections.

After confirmation, the cross-holder collects the senior recovery on terms it negotiated with itself, plus the captive junior recovery as a discounted secondary stream. The aggregate trade returns more than the senior alone would have returned in a contested case.

Helpful binding and exploitative binding

Both readings are reasonable from inside their chairs, and the law's task is to distinguish the legitimate case from the captive case without disturbing the binding rule itself. The legitimate case rests on a good reason. Without majority binding rules and cramdown, a single small holder could hold up an otherwise rational restructuring by refusing to consent unless paid more than its share. A liquid market in junior debt requires that the instrument be capable of being restructured without unanimous consent.

The captive tranche binds minorities through the same mechanism, but the underlying situation differs. The legitimate cramdown rule binds a minority that would otherwise extract holdout rents, in a transaction whose majority support reflects genuine alignment within the class. The captive tranche binds a minority where the majority's economic interest lies elsewhere, in a transaction whose majority support reflects the cross-holder's interest in another class.

The distinguishing feature is alignment, not size. A small dissenting holder facing a genuinely aligned majority is correctly bound. The same dissenting holder facing a misaligned majority is bound by a rule designed for a different situation. The doctrinal vehicle for that distinction already exists in vote designation under Section 1126(e), which permits the court to disregard votes not cast in good faith. The tool is calibrated for the captive tranche. The application has lagged.

Two readings of the same conduct

The captive tranche admits two readings, and both warrant consideration before reaching for a label.

The senior-side reading: the cross-holder did nothing the loan documents did not permit. Each amendment was supported by the required majority. Each consent solicitation was conducted in conformity with the indenture. Each plan vote was tabulated correctly. The Bankruptcy Code says what acceptance means, and the class accepted. The minority's complaint amounts to objecting to outcomes the rules they bought into were designed to produce. Concentration of positions, including across tranches, is how restructuring gets done. Demanding that holders disgorge votes because of aggregate position would chill the capital formation distressed companies need. The fact that the trade was profitable is evidence that it was correctly sized.

The minority-side reading: the cross-holder bought their voting rights without telling them, used those rights to engineer a result the cross-holder's other holdings benefited from, and pocketed the difference. The minority paid the same price for the same instrument as the cross-holder did, on the shared assumption that majority decisions would be made in the interest of the instrument. The cross-holder violated that assumption deliberately and profited from the violation. The fact that each step was contractually permitted does not change the character of the conduct. The minority paid for one thing and received another. From the minority's chair, this is theft, and the theft was effected through the cross-holder's intentional use of votes the minority believed were being cast for the class.

The two readings cannot both be right, but each is reasonable from the chair where it is offered. The harder question is which the law should privilege when it sets the rules going forward. American commercial law has long recognized that contracts can be used in ways the legal system finds inequitable, and that courts have an equitable role in responding. The implied covenant of good faith and fair dealing, equitable subordination, the prohibition on votes cast in bad faith, the fraudulent transfer regime, and the law of constructive trusts each exists because the formal validity of a transaction has never been treated as the only relevant consideration. These doctrines do not require the law to adopt the minority's label. They require it to take the minority's experience seriously enough to ask whether the conduct sits comfortably inside cramdown's legitimate domain or whether it has crossed into territory the existing tools can reach.

Why the law has tolerated it

The doctrinal tools have been applied narrowly, in part because courts have historically, and rightly, been protective of binding-majority rules. The challenge for reform is to articulate a distinction that preserves what cramdown was designed to do while addressing what cross-holding has made possible.

Vote designation under Section 1126(e) is the most directly applicable tool. The leading case, In re DBSD North America, a 2011 Second Circuit decision, addressed a strategic competitor that acquired senior debt to block the debtor's plan and acquire its wireless spectrum, and the court affirmed designation. Courts have remained reluctant to designate votes where the cross-holder can point to any plausible economic motive. Equitable subordination under Section 510(c) has been applied principally to insider misconduct in the classical sense and rarely extended to arms-length investors. The implied covenant of good faith and fair dealing has been read narrowly in the financial-contract context, though the recent wave of liability management litigation, including Serta in the Fifth Circuit and Mitel in the New York Court of Appeals, has begun to push against that hesitation.

The disclosure regime contributes to the tolerance. The trustee does not know which holders are cross-held. The other holders do not know either. The integrated alternative asset managers active in the distressed market are repeat players in front of the same bankruptcy courts, and the community has developed norms permitting cross-holding to function as a routine feature of restructuring practice, less from improper motive than from familiarity. The harmed parties are diffuse and quiet: retail bondholders, smaller funds, insurance portfolios, and pension allocators with limited appetite for a fight against an opposing party with substantial litigation resources. They take the haircut, book the loss, and move on.

What the law could do

Vote designation under Section 1126(e) could be the default where a creditor's net economic interest in the capital structure is materially larger in a senior class than in the class being voted, with the burden on the cross-holder to demonstrate good faith. Equitable subordination under Section 510(c) could be available against cross-holders whose conduct at the junior level was driven by senior-side considerations and produced demonstrable harm to other junior holders.

The indenture market could develop standard provisions disenfranchising cross-holders for purposes of consent solicitations and plan acceptances. The mechanism exists in the corporate context, where interested-party votes are routinely excluded from approval thresholds in conflict transactions. On the senior side, disqualified institution lists could survive default rather than fall away upon it, and the prohibition could extend to participations, total return swaps, credit default swaps, and other synthetic interests that produce equivalent economic exposure. A Schedule 13D analog for debt holders, triggered at meaningful aggregate thresholds and capturing synthetic positions, would force cross-holdings into the open at the moment of accumulation rather than the moment of harm.

The role of the administrative agent and the operation of intercreditor agreements warrant reexamination in the cross-holding context. Agents could owe enhanced disclosure and consultation duties to minority lenders when the required lender bloc is materially cross-held, and minority lenders could have standing to enforce intercreditor protections directly where the junior class representatives are themselves cross-holders.

None of these reforms requires statutory innovation in the strict sense. Each can be accomplished through judicial development of existing doctrine, market adoption of standard indenture terms, and SEC rulemaking under existing authority. None of them disturbs the legitimate operation of binding-majority rules in cases where the majority is genuinely aligned with the class.

The cost of inaction

The cost of permitting the captive tranche to operate is borne by the credit markets as a whole. Every investor participating in a junior tranche must now price in the possibility that the majority of the tranche will be used to vote against the tranche's interest. The price is paid by issuers in wider spreads and tighter covenants, and by the broader economy in the higher cost of capital for businesses that depend on junior credit to fund growth. Senior lenders bear some of this cost too, because wider spreads on junior debt make the overall capital stack more expensive to assemble.

Contracts are enforced because the parties have made commitments they can be expected to keep, and because the rest of the market is entitled to rely on those commitments. When a holder of a junior instrument has made one commitment to the market and a different commitment to itself, the minority has reason to read the enforcement of the first commitment as a fiction. Whether the law finds the minority's reading persuasive, or finds the cross-holder's reading more faithful to the bargain, is a question worth asking before the answer is presumed.

The views expressed are those of the author and do not constitute legal advice.

New Article • AI / Legalweek / Legal Practice

Recent AI Conferences and the Practice of Law

Legalweek 2026, agentic AI, reliability, governance, and lawyer training.

May 2026Full Article

Legalweek 2026 and last week’s AI Agent Conference both declared the experimentation phase over. The vendor market has converged. What remains is the operational scaffolding: governance, security at scale, and a training model that does not hollow out the next generation of supervising lawyers. The technology is running ahead of the institutional capacity to supervise it. These are a few thoughts on what that means for firms and in-house teams.

Legalweek 2026 closed at the Javits Center in March with a keynote titled “The Reckoning.” Last week, the AI Agent Conference at the Hilton Midtown drew over a thousand executives. The two events covered different audiences and different layers of the technology, but the underlying picture was the same.

From the Vendors

The market has stopped competing on basic capability. Large language model plus document upload plus chat is a commodity, demonstrated credibly by every vendor of any size. What now distinguishes products is where in the workflow the intelligence is positioned.

Harvey is positioning as the reasoning layer, optimized for analytical depth in drafting and synthesis. Legora, fresh off a $550 million raise, industrializes execution across large document sets. Thomson Reuters released a rearchitected CoCounsel Legal that embeds Anthropic’s Claude Agent SDK at the model layer, collapsing research and drafting into one workflow grounded in Westlaw and Practical Law. LexisNexis is doing the same with Lexis+ and Protégé. Clio’s Vincent AI embeds drafting in matter data. NetDocuments launched Smart Answers and expanded its Model Context Protocol connectivity, positioning the document repository as a secure intelligence layer.

The common element is grounding. Vendors have accepted that ungrounded generation is unsafe for legal work. Agentic workflows, where the model plans and executes a sequence of steps, were shown in production rather than promised on roadmaps. The front of the funnel is solved, the middle is being solved, and the back end, where the lawyer is professionally accountable for what was produced, is where the work still has to happen.

The Wider Picture

The AI Agent Conference put the legal stack in its broader enterprise context. Curated by Firsthand VC with NYSE Wired, Bright Data, and theCUBE, it ran across three tracks and unveiled “The Agentic List 2026,” a ranking of 120 companies drawn from roughly 5,000 nominations. Hebbia and Harvey made the legal cut. The broader market is now tracking legal AI as a category alongside finance and healthcare.

Two figures from that stage bear on legal practice. Seventy-nine percent of organizations report some level of agent adoption. Eleven percent are running agents in production. Forty percent of agentic projects are at risk of cancellation. The framing on the panels, in the organizers’ words, had shifted from “should we deploy agents?” to “how do we govern, secure, and scale them without getting fired?”

The gap between adoption and production is the same gap stalling legal AI rollouts. The verification and governance scaffolding has not been built.

Reliability

Reliability in this domain is not a single number. It is a function of the task, the grounding, the prompt, and the experience of the reviewer.

Stanford’s RegLab work found legal hallucination rates of 58 to 88 percent across major foundation models on complex legal queries. Vendor-specific tools grounded in licensed content perform substantially better. Paxton AI reports above 93 percent accuracy on the Stanford Legal Hallucination Benchmark. May 2026 benchmark data shows legal queries running hallucination rates around 18.7 percent on average for general-purpose models, with higher rates on questions that move past common-law doctrine into novel jurisdictions or unsettled questions.

The courts have noticed. The first quarter of 2026 produced roughly $145,000 in sanctions for AI-generated false citations, the highest quarterly total on record. The single largest sanction, $109,700 against an Oregon attorney, was issued earlier this year. The Fourth Circuit publicly admonished a lawyer in April for filing briefs containing fabricated authority. In 2023, most caught cases involved self-represented litigants. By 2025, more than half were practicing lawyers. As adoption climbed, the failure modes moved up the seniority ladder.

A grounded research query against Westlaw with a competent reviewer is reliable enough to use. An ungrounded drafting request to a general-purpose model, executed by a lawyer who does not know the substantive area, is not, and may never be in the form the question is currently asked.

Preparers and Reviewers

A theme repeated across Legalweek panels was that preparers will become reviewers. AI handles the first pass. The lawyer applies judgment. The result is faster, cheaper, and arguably better work. The framing was uncontroversial. Every vendor used some version of it.

It is partly correct. It also rests on an unstated assumption.

You cannot review what you do not understand

The traditional model of legal training was built on apprenticeship. Junior lawyers reviewed documents, drafted standard contracts, ran research, and assembled due diligence. The work was repetitive. It was also the mechanism by which lawyers developed the substantive intuition that makes senior judgment possible. A lawyer learns what a contract should look like by drafting a hundred of them. A lawyer learns to spot a problematic indemnification clause because he wrote one badly and was corrected.

If AI handles the foundational work, and junior lawyers are placed into review roles without having done the underlying production, the profession produces reviewers without the substantive basis to review. The gap does not present immediately. It presents five or seven years out, when the lawyer who has only ever reviewed AI output is asked to make a judgment the AI does not know how to ground. At that point, the lawyer has the title of reviewer and the training of a user.

Most legal AI products assume a user who is competent but unsure and wants structured guidance. That maps to a junior lawyer. It does not map to a senior associate or partner, and it does not produce one. The vendor pitch assumes the senior layer keeps generating itself, while the entry-level work that historically generated it is automated away.

This is a training problem that could just as easily be viewed as a productivity gain. The short-term economics of using AI to do junior work are visible on the engagement letter. The long-term economics of not having mid-career lawyers who can supervise that work are not.

What Follows

Governance has to mean more than policy memos. It is operational, not documentary: verified-output requirements, citation checking, model-selection discipline, clear ownership of work product. Firms defensible in three years will be the ones building verification infrastructure now, not the ones buying licenses and trusting the vendor.

Training has to change deliberately, not by neglect. If junior lawyers spend less time on foundational production, firms have to design substitute experiences that build the same substantive intuition: blind drafting before AI assistance, structured rotation through unautomated work, deliberate exposure to primary materials.

Firms also need a working description of what these tools are. They are useful leverage for a lawyer who knows what good looks like. They are dangerous for a lawyer who does not. The middle case, the lawyer who knows enough to use the tool but not enough to catch its errors, is where most of the sanctions are issued. As adoption widens, that middle case becomes more common, not less.

The lawyers who will be in trouble are not the ones who refused to adopt. They are the ones who adopted without understanding what they had agreed to review.

Michael Simon Baker is the principal of Michael S. Baker, P.C. (d/b/a NYBusiness.Law and ArtificialIntelligence.Lawyer), a New York firm practicing corporate law, commercial litigation, and AI governance.

Hosted Article • Constitutional Law / Family Law / Federalism

Are You My Mother? The Federal Courts and the American Family

The Domestic Relations Exception, federal jurisdiction, civil rights enforcement, and constitutional family-law disputes.

April 2026Full Article

A Note on the Author's Interest in This Question

I have written about the federal government's nineteenth-century relationship with the Church of Jesus Christ of Latter-day Saints, a history in which federal authority over the church's most intimate institutional practices was exercised without restraint and without principled limit. That history coincided almost exactly with the years in which the Supreme Court was constructing the Domestic Relations Exception, the doctrine that purports to place domestic relations beyond federal court jurisdiction.

The juxtaposition has never been adequately examined. The same courts that announced in Barber v. Barber and In re Burrus that domestic relations belong categorically to the states were simultaneously sustaining federal prosecutions of marriages, federal dissolution of a church's corporate charter, and federal disenfranchisement of believers, all on the basis of how a specific religious minority organized its households. What the Mormon cases and the DRE share is not a constitutional principle. It is a record of the federal courts making choices: showing up as aggressive institutional parents when the political culture demanded intervention, and claiming no jurisdiction when it did not. This article is an argument that the disclaimer has never been honest, that the current moment has made the inconsistency visible in a way it has not been before, and that the exception should be closed.

Michael Simon Baker is the author of Prophets and Prejudice: Race, Power, and the Mormon Priesthood Ban (The Bakery Publishing, Warwick, NY). He is a New York-based corporate, finance, and business disputes attorney and principal of Michael S. Baker, P.C. (d/b/a NYBusiness.Law), and maintains a user-facing AI implementation and governance practice at ArtificialIntelligence.Lawyer.

Abstract: The Domestic Relations Exception to federal jurisdiction holds that federal courts have no authority over divorce, alimony, child custody, and family structure, which belong categorically to the states. This article argues that the exception has never been a principled jurisdictional rule. It is a pattern of selective federal engagement: the same courts that constructed the DRE in the 1880s simultaneously exercised the most invasive federal authority over domestic life in American history, prosecuting marriages, dissolving a church, and disenfranchising believers on the basis of how a specific religious minority organized its households. The territorial character of Utah at the time does not explain away the contradiction: the federal government embedded its position on family structure permanently in the Utah state constitution as an irrevocable condition of statehood, and the principle established in those cases has been cited as controlling doctrine in state free exercise cases ever since. The modern federal courts have continued the pattern, adjudicating marriage, custody, and parental rights under ERISA, bankruptcy, ICWA, the Hague Convention, and the Constitution itself, while invoking the DRE to bar Section 1983 claims alleging constitutional violations arising from the same proceedings. The article traces the DRE from its origins in Barber v. Barber (1859) through Ankenbrandt v. Richards (1992), which relocated the exception to the diversity statute rather than defending it constitutionally, a move the article characterizes as a disclaimer rather than a validation. It addresses the abstention doctrines, Rooker-Feldman, the efficiency and institutional competence objections, and the practical harm the DRE inflicts on litigants whose only path to federal review is through an appellate structure that produces no record, no merits ruling, and no realistic prospect of certiorari correction. The article concludes with a narrow, administrable rule: the DRE does not apply to federal question jurisdiction, and courts should hold that Section 1983 claims alleging constitutional violations in domestic relations proceedings are not barred by a doctrine whose only authority is nineteenth-century dictum and a statutory construction the text does not clearly support.

The Question

In P.D. Eastman's 1960 children's book, a baby bird hatches while its mother is away and sets off to find her. It approaches a kitten, a hen, a dog, a cow, a boat, a plane, a snort. To each it asks: Are you my mother? Each says no, or says nothing, and the bird moves on. The mother eventually returns, recognizes the bird, and the question is answered.

The American family has been asking the federal courts a version of the same question for 165 years. The answers have been inconsistent in ways that reveal how the federal judiciary actually works, as opposed to how it describes itself.

The Domestic Relations Exception to federal jurisdiction holds that the federal courts have no authority over domestic relations, that divorce, alimony, child custody, and family structure belong categorically to the states, and that federal courts should stay out. That is the doctrine's self-description. The federal courts' actual behavior across the same 165 years tells a different story. They prosecuted a religious minority's marriages. They dissolved a church on the basis of its members' household arrangements. They defined who could marry whom. They set the evidentiary standard for taking children away from parents. They overrode state divorce law when federal benefit plans were at stake. They mandated federal preferences in state custody proceedings involving Native American children. They told states that race cannot determine custody. They declared that same-sex couples have a constitutional right to marry.

Are you my mother? On those questions, the answer was yes. Emphatically, repeatedly, and without jurisdictional hesitation.

But when the question comes from a parent who wants federal review of a family court proceeding that violated a constitutional right, the answer changes. The court that was just here, defining the constitutional dimensions of American family life across a century and a half of jurisprudence, suddenly has no jurisdiction over domestic relations.

The Domestic Relations Exception is not a principled jurisdictional rule. It is a pattern of selective involvement, a federal judiciary that has always been the constitutional parent of American family law but has deployed a jurisdictional disclaimer to avoid the obligations that parentage entails when the family in question does not present a sufficiently salient constitutional stake.

Understanding the exception requires understanding the pattern. Closing it requires saying so plainly.

Origins Without a Source

The exception traces to Barber v. Barber, 62 U.S. (21 How.) 582 (1859), where the Supreme Court, while actually exercising diversity jurisdiction over an alimony dispute, announced in dictum that federal courts have no jurisdiction over divorce and alimony. The Court offered no statutory citation because none existed, and no constitutional text because none supports it. It was a policy judgment dressed in jurisdictional language, and the federal courts have been working out its implications ever since.

In re Burrus, 136 U.S. 586 (1890), extended the principle to child custody, again on grounds more atmospheric than analytical. "The whole subject of the domestic relations of husband and wife, parent and child, belongs to the laws of the States and not to the laws of the United States," the Court wrote. That is a statement about tradition, not about jurisdiction. Article III does not contain a carve-out for domestic relations, and Congress, in enacting the general diversity statute, did not exclude family law disputes from its scope.

Neither case explains what constitutional authority the federal courts were invoking to disclaim jurisdiction Congress had granted. Neither attempted to.

The Aggressive Parent

What the doctrinal account of Barber and Burrus obscures is the company those decisions kept.

The three decades between 1859 and 1890 were not years of principled federal restraint over domestic relations. They were years of the most invasive federal authority over marriage, family structure, and religious domestic life in American history, directed at a specific target: the Church of Jesus Christ of Latter-day Saints and the practice of plural marriage in Utah Territory.

In Reynolds v. United States, 98 U.S. 145 (1879), the Supreme Court unanimously upheld the federal prosecution of George Reynolds, secretary to Brigham Young, for bigamy under the Morrill Anti-Bigamy Act of 1862. Reynolds had contracted a second marriage in conformity with LDS doctrine and made no factual defense. His argument was constitutional: that the First Amendment's free exercise clause protected religiously motivated conduct, including the structure of his household and marriage. The Court rejected this unanimously, establishing the belief/conduct distinction that has governed free exercise law ever since, and asserting federal authority to define, regulate, and criminalize the domestic arrangements of a religious community on the grounds that polygamy was "odious among the northern and western nations of Europe" and incompatible with republican government. The Court did not pause to consider whether the domestic character of the conduct placed it beyond federal reach. It reached in and decided.

Davis v. Beason, 133 U.S. 333 (1890), the same year as Burrus, upheld an Idaho territorial law disenfranchising anyone who practiced or advocated polygamy. Murphy v. Ramsey, 114 U.S. 15 (1885), upheld the federal Edmunds Act's disqualification of polygamists from voting and holding public office in Utah Territory. Late Corporation of the Church of Jesus Christ of Latter-Day Saints v. United States, 136 U.S. 1 (1890), also the same year as Burrus, upheld the federal dissolution of the LDS Church's corporate charter and the forfeiture of its property. These are not cases at the margins of domestic life. They reach into the structure of a church, the validity of its marriages, the political rights of its members, and the organization of its households. The same Court that was telling itself in Burrus that domestic relations belong entirely to the states was simultaneously dissolving a religious institution and voiding its property holdings on the basis of how its members organized their families.

Against this backdrop, the territorial limitation argument deserves direct examination, because it is the most plausible response a defender of the DRE can offer. Utah was a federal territory, not a state. Congress has plenary authority over territories under Article IV, Section 3. The argument runs that the federal courts in the Mormon cases were applying federal statutes to federal territory, not exercising Article III jurisdiction over state domestic relations, and that the constitutional foundation is therefore different from the DRE context.

The argument is technically accurate as far as it goes. It does not go far enough. The mechanism of statehood itself refutes it. Congress did not release its grip on domestic structure when Utah became a state in 1896. The Utah Enabling Act of 1894 required, as a condition irrevocable without the consent of the United States, that the Utah constitution permanently prohibit polygamy. Article III of the Utah State Constitution reads: "Polygamy or plural marriage are forever prohibited." That provision cannot be amended without federal approval. It is there today. The federal government embedded its position on family structure permanently in the state constitution as a term of admission. Nobody at the time argued that state sovereignty over domestic relations required the removal of that prohibition upon statehood.

Nobody suggested Reynolds would be reversed once Utah joined the Union. Nobody imagined the DRE would protect Utah's domestic arrangements from federal scrutiny the moment its territorial status ended. The principle the federal courts established in those cases was not understood, by anyone involved, as being limited to territorial status. It traveled into statehood and has been cited as controlling doctrine in state free exercise cases ever since.

Employment Division v. Smith, 494 U.S. 872 (1990), decided in Oregon, a state, applied Reynolds's belief/conduct distinction to sustain Oregon's denial of unemployment benefits to employees who used peyote as part of Native American religious practice. The Court invoked Reynolds without suggesting its authority was confined to territorial contexts.

And the territorial argument does not explain the cases that arose in states. Pace v. Alabama, 106 U.S. 583 (1883), decided within years of the main Mormon cases, reached directly into Alabama's anti-miscegenation statute on constitutional grounds and upheld it, wrong outcome corrected eighty-four years later, but the jurisdiction to review a state's marriage law was never questioned.

Loving v. Virginia, 388 U.S. 1 (1967), reached into Virginia's marriage law and struck it down on Fourteenth Amendment grounds. Virginia's own courts had explicitly invoked the principle that marriage regulation "should be left to exclusive state control," citing Maynard v. Hill, 125 U.S.

190 (1888), the 1888 articulation of the same state-sovereignty principle the DRE enforces. The Supreme Court in Loving acknowledged the state's police power over marriage and reached past it without hesitation. Zablocki v. Redhail, 434 U.S. 374 (1978), struck down a Wisconsin statute conditioning remarriage on compliance with child support obligations. The DRE was in full force in 1967 and 1978. It did not stop Loving or Zablocki, because the constitutional stakes were framed in a way that made the federal interest visible enough to overcome the preference for avoidance.

What emerges is not a principled constitutional rule. It is a record of the federal courts picking and choosing. They reached into domestic and family life when the political culture demanded it, when the target was a religious minority whose practice offended the national consensus, when a state was enforcing racial hierarchy in its marriage laws, when the constitutional stake was visible enough to compel engagement. They withdrew behind the language of jurisdictional restraint when the claim was less politically compelling, when the litigant was a parent rather than a cause, when the constitutional violation was harder to see through the noise of a disputed custody arrangement. The DRE did not create this pattern. It rationalized it, gave it a name, and allowed it to harden into doctrine. The distinction between the cases in which the federal courts intervened and the cases in which the DRE forecloses intervention correlates with political salience and social consensus in a way that doctrine alone does not explain.

What Reynolds Left Behind

Federal authority over plural family life did not end with the Utah Enabling Act. Its continuation is the clearest illustration of what the DRE's selective pattern costs.

The federal government actively prosecuted plural marriage from the Reynolds era through approximately mid-century. It no longer does. Enforcement is now almost entirely state and local, and even that has narrowed. In Brown v. Buhman, 822 F.3d 1151 (10th Cir. 2016), the Sister Wives case in which Kody Brown and his plural family challenged Utah's bigamy statute under 42 U.S.C. § 1983, the Tenth Circuit found the case moot because Utah County had adopted a policy of not prosecuting polygamists in the absence of collateral offenses such as fraud, abuse of minors, or welfare violations. The district court below had struck down Utah's cohabitation provision on due process and free exercise grounds, relying significantly on Lawrence v. Texas, 539 U.S. 558 (2003), which protected consensual adult intimate relationships as a constitutional liberty interest.

The Tenth Circuit never reached those merits. The constitutional question of whether the post-Lawrence landscape permits states to criminalize consensual plural cohabitation among adults remains unresolved at the federal appellate level.

Lawrence itself shifted the terrain considerably. Justice Kennedy's majority opinion explicitly identified "personal decisions relating to marriage, procreation, contraception, family relationships, child rearing, and education" as protected liberty interests under the Due Process Clause. The Court held that moral disapproval of a lifestyle choice is not a sufficient state interest to justify criminal prohibition of private consensual adult conduct. Justice Scalia's dissent predicted that Lawrence would destabilize prohibitions on polygamy and other non-traditional family arrangements. What neither the majority nor the dissent addressed is what Lawrence means for the constitutional claims of plural or polyamorous families who are not facing criminal prosecution but are instead involved in state family court proceedings: custody disputes, visitation restrictions, determinations of parental fitness based on household structure or religious practice.

This is where the DRE completes a particularly pointed circuit. The federal courts in the 1880s reached into plural family life and criminalized it on the basis of the federal government's moral and political objections. The federal government has largely withdrawn from active enforcement.

State family courts, however, continue to encounter plural and polyamorous families in domestic proceedings, and those courts make constitutional judgments in every ruling that denies custody, restricts visitation, or assesses parental fitness based on the structure of the household or the religious beliefs that animate it. A family court that conditions custody on the dissolution of a plural household is making a First Amendment and due process determination. A court that awards custody to a conventional spouse over a parent in a polyamorous household, citing the family structure as evidence of unfitness, is exercising exactly the authority the federal courts claimed in Reynolds, specifically the authority to define which domestic arrangements are legally tolerable, but doing so now in civil proceedings rather than criminal ones.

The parent on the losing end of those determinations who seeks federal constitutional review faces the DRE. The claims are cognizable: free exercise of religion where the plural structure is religiously motivated, due process liberty under Lawrence where it is not, equal protection where similarly situated parents in conventional arrangements are treated differently, and First Amendment protections where the court's ruling penalizes the expression of heterodox beliefs about family structure. Federal courts have adjudicated all of these constitutional theories in other contexts. In the domestic relations context, the DRE turns them away before they can be developed.

The federal courts asserted jurisdiction over plural marriage when the political culture demanded suppression. They constructed the DRE to rationalize non-intervention when the political culture demanded indifference. Lawrence shifted the constitutional baseline, potentially converting what was once a valid basis for criminal prohibition into a protected liberty interest. And the DRE now prevents federal courts from working out what that shift means for the families whose household arrangements were once the target of the very federal authority the DRE claims never to have had.

The doctrine is most consequential precisely where the constitutional law is most unsettled and where federal adjudication is most needed. That is not an accident of doctrine. It is the pattern.

The Disclaimer

The leading modern authority is Ankenbrandt v. Richards, 504 U.S. 689 (1992), where the Court had a genuine opportunity to rationalize the doctrine and declined. Justice White's opinion for the Court located the DRE not in Article III itself but in the diversity statute, 28 U.S.C. § 1332, which the Court construed as codifying the historical practice of federal courts declining to issue divorce, alimony, and child custody decrees. The statutory move was deliberate: grounding the exception in Congress's enactment rather than in the Constitution gave the Court a way to preserve it without confronting the harder question of whether federal courts can voluntarily relinquish jurisdiction that Article III extends to them.

Ankenbrandt also narrowed the exception considerably. The plaintiff in that case brought a tort claim, grounded in diversity, against her former husband and his companion for child abuse. The Court held the DRE did not bar the suit. The exception applies only to cases where a federal court is asked to issue a divorce, alimony, or child custody decree, not to every dispute that arises in a domestic context. That limitation matters. A great deal of family-adjacent federal litigation, including civil rights claims by parents against state officials and custody-linked tort actions, sits outside the exception as Ankenbrandt drew it.

But Ankenbrandt's most significant feature is what it tacitly conceded. By relocating the DRE from the Constitution to a statute, the Court acknowledged that it could not defend the doctrine on constitutional grounds. The opinion carries an implicit message: if the diversity statute does not actually codify the exception, Congress can say so. If the statute's scope has been misread, Congress can clarify it. The Court was extending an invitation without quite saying so. The invitation has gone unanswered for three decades. And the Court, having declined to constitutionally validate the doctrine it was preserving, and having shifted formal responsibility to a Congress that did not act, simply continued enforcing the rule. It created the problem, declined to defend it, passed the responsibility, and continued as before. That is not judicial restraint in any coherent sense. It is a parent denying a relationship the DNA already established.

Two practical complications compound the problem. The first is that lower courts have not applied Ankenbrandt's "decree" limitation consistently. Some circuits read the case narrowly, holding the DRE bars only cases where a federal court is asked to issue a divorce, alimony, or child custody decree. Others apply it to any case that arises from a domestic relations context, even where no decree is sought and the claim is purely constitutional. The result is a doctrine whose scope varies by circuit, creating the anomaly that identical civil rights claims arising from identical family court proceedings may be heard in federal court in one jurisdiction and dismissed in another. That inconsistency is not an argument for expanding the DRE. It is evidence that the doctrine was never principled enough to produce consistent results.

The second complication is the efficiency objection: federal courts are already overloaded, domestic relations cases are high-volume and fact-intensive, and the DRE usefully channels family disputes to state courts better suited to handle them. The objection misunderstands the argument.

Nothing here suggests federal courts should become general courts of appeal from state family court proceedings or should issue custody and divorce decrees. Ankenbrandt is correct that those functions belong in state court. The argument is narrower: when a family court proceeding violates a constitutional right, the parent who seeks federal review of that constitutional violation should not be turned away because the underlying facts arose in a domestic context. A court that reviews whether a custody proceeding denied due process is not issuing a custody decree. It is doing what Article III courts exist to do.

The deeper version of this objection is worth acknowledging directly: state courts are not merely the default forum for family law; they are, in many respects, the better forum. They develop genuine expertise in the fact-intensive, ongoing, and highly individualized disputes that family law generates. They understand local norms, have mechanisms for supervision and modification of custody arrangements, and are embedded in the network of family services and counseling resources that complex family disputes require. That competence is real and the argument for it is serious. But competence is not jurisdiction, and institutional expertise does not authorize the elimination of constitutional rights. A family court judge who is highly expert in custody determinations and who nevertheless denies a parent due process in the course of a proceeding has violated the Constitution. The existence of state court expertise does not make that violation less real, and it does not answer the question of which court gets to say so. The DRE does not route constitutional claims to a more expert tribunal. It eliminates federal review entirely. Those are not the same thing.

The Constitutional Parent

The Ankenbrandt solution is more elegant than it is sound. Locating the DRE in the diversity statute resolves the constitutional question only by assuming the answer. If Congress did not actually exclude domestic relations disputes from § 1332 when it enacted the statute, then the Court's construction amounts to judicial amendment of a federal law, and the exception remains a creature of the judiciary rather than of Congress.

Article III is the relevant constraint in either direction. The constitutional text extends the judicial power of the United States to "all Cases, in Law and Equity, arising under this Constitution, the Laws of the United States," to controversies "between Citizens of different States," and to several other categories. The word "all" in the arising-under clause is not decorative. Federal question jurisdiction, as a matter of Article III ceiling, reaches every case arising under federal law, domestic relations character notwithstanding.

The problem for the DRE is that federal courts have no constitutional authority to decline jurisdiction that Article III and a valid congressional grant together confer. That principle runs through the case law with some consistency. Marshall in Cohens v. Virginia, 19 U.S. (6 Wheat.)

264 (1821), was emphatic: a court cannot decline to exercise jurisdiction any more than it can arrogate jurisdiction it does not have.

A defender of the DRE will reach immediately for the abstention doctrines, and that response deserves a direct answer. Younger, Burford, Colorado River, and Pullman abstention all permit federal courts to defer or decline in the exercise of otherwise-valid jurisdiction under particular conditions. If those doctrines survive Article III scrutiny, why not the DRE? Because each of them has a defined trigger, a bounded scope, and a foundation in either established equity practice or explicit Supreme Court precedent. A litigant can challenge whether the conditions for abstention are met. Appellate courts review those determinations. The DRE has none of that structure.

Pullman, from Railroad Commission of Texas v. Pullman Co., 312 U.S. 496 (1941), is temporary: the federal court stays the case while state courts resolve an ambiguous question of state law, then the federal proceeding resumes. It is a tool of judicial economy, not a permanent exclusion.

Burford, from Burford v. Sun Oil Co., 319 U.S. 315 (1943), requires a unified state regulatory framework with specialized expertise whose coherence federal piecemeal intervention would disrupt. It does not reach general state court jurisdiction over any subject. Colorado River, from Colorado River Water Conservation District v. United States, 424 U.S. 800 (1976), applies to concurrent proceedings and requires case-specific balancing. The Court said plainly that the circumstances permitting dismissal in favor of a parallel state proceeding are "considerably more limited" than those permitting a stay. Younger, from Younger v. Harris, 401 U.S. 37 (1971), requires an ongoing state criminal proceeding, adequate state remedies, and no bad faith or harassment. The Supreme Court in Sprint Communications, Inc. v. Jacobs, 571 U.S. 69 (2013), curtailed its scope to three specific categories of state proceeding.

The DRE is categorical, not conditional. It is permanent, not a stay. It is unreviewable at the threshold. And its only stated rationale is that family law belongs to the states, which is a policy preference, not a jurisdictional rule. Calling it a member of the abstention family does not make it one.

A related doctrine deserves direct attention. The Rooker-Feldman doctrine, derived from Rooker v.

Fidelity Trust Co., 263 U.S. 413 (1923), and District of Columbia Court of Appeals v. Feldman, 460 U.S. 462 (1983), bars lower federal courts from reviewing state court judgments. It operates alongside the DRE in the family court context, and critics of the argument advanced here sometimes conflate the two. They are different in important ways. Rooker-Feldman applies only to parties who lost in state court and are effectively asking a federal court to reverse or void that judgment. It does not apply to independent claims, brought under Section 1983, alleging that a state court actor violated a constitutional right in the course of proceedings. A parent who alleges that a family court judge penalized religious practice in a custody determination, fabricated the basis for an arrest warrant, or denied due process through a rigged proceeding is not asking a federal court to reverse the custody order. She is asking the federal court to adjudicate a constitutional claim against a state actor. The Supreme Court has confirmed this distinction: Rooker-Feldman does not bar federal claims that are independent of, rather than inextricably intertwined with, the state court judgment. The DRE does bar those claims, in the family court context, on grounds that are constitutionally indefensible for all the reasons this article has set out.

The two doctrines are not the same obstacle, and eliminating the DRE's application to civil rights claims does not require disturbing Rooker-Feldman.

When the Parent Refuses to Come Home

The hardest version of the problem arises when the DRE is used to bar claims brought under 42 U.S.C. § 1983 or the Constitution directly. Federal civil rights are not a matter of state law. A parent who alleges that a state family court judge violated the First Amendment by penalizing religious practice, or violated the Fourteenth Amendment's due process guarantee through a custody proceeding infected by fabricated evidence, is raising a claim that Congress has specifically authorized federal courts to adjudicate. Section 1983 exists precisely because Congress determined, in the wake of Reconstruction, that state courts could not be trusted as the exclusive forum for constitutional claims against state actors.

A custody order that restricts where a parent may worship, or conditions visitation on discontinuing religious instruction, implicates free exercise. A proceeding that penalizes a parent's political speech, associations, or public advocacy offends the First Amendment regardless of the domestic framing. A state court that credits perjured testimony because it aligns with a preferred outcome, refuses to allow cross-examination of a guardian ad litem, or issues an arrest warrant on a fabricated invoice is not conducting a domestic relations matter; it is conducting a proceeding that denies due process. A parent removed from a child's life based on race, religion, or national origin has a Fourteenth Amendment equal protection claim that does not become inaccessible simply because it arose in family court. And a parent who loses custody because the presiding judge disapproves of his politics, his religion, or his willingness to fight has a claim that goes to the constitutional core of what federal civil rights law exists to protect. These are not exotic theories. They are the ordinary application of constitutional guarantees to a forum that, by design and tradition, operates with less transparency than almost any other court in the American system.

The argument that a determined litigant can always work through state courts and eventually reach the Supreme Court on certiorari misunderstands how the DRE actually operates. When a district court dismisses a civil rights claim on DRE grounds, what the litigant loses is not just the ruling. It is the record. No factual development. No evidentiary hearing. No merits briefing. No judicial analysis of whether the constitutional violation is made out. What goes up on appeal is a dismissal order, reviewed under the same doctrine. What goes to the Supreme Court, if it ever gets there, is a case that looks on its face like a domestic relations dispute. The Court has revisited Ankenbrandt precisely never in thirty years. The theoretical path to Supreme Court correction exists. The practical path does not.

The state court route is not a remedy either. Family court proceedings are reviewed deferentially.

The record is shaped by state procedure and state evidentiary rules. By the time a constitutional question emerges cleanly enough from that record to support a federal petition, years have passed, the custodial status quo has hardened, and the original injury may be permanent. The Supreme Court cannot restore years of a parent's absence from a child's life. For the parent navigating this without resources, state appellate review that is structurally sympathetic to the court below is not merely a ceiling. It is a sealed room.

That problem is worst in the most insular jurisdictions, where the distance between a trial judge's predispositions and the last available state appellate check is very short. Family courts operate in private. Records are sealed. Proceedings are shielded from the press and the public in ways that trial courts in almost every other context are not. A judge who wants to reach a particular result has more room to do so here than nearly anywhere else in the system. The DRE ensures that room is never reduced by the prospect of federal scrutiny.

Federal review is not a guarantee of success. A court confronting a Section 1983 claim from a family court proceeding can examine the merits, find the constitutional violation is not made out, conclude that qualified immunity applies, or determine that adequate state remedies exist. The question is not whether federal courts should automatically reverse state family court decisions. It is whether they should be allowed to look. A court that examines a civil rights claim on its merits and rejects it has done its job. A court that dismisses the claim before reaching the merits because the facts arose in a domestic relations context has not. The first is adjudication. The second is foreclosure.

The DRE, as applied to civil rights claims, produces only the second result. It does not weigh the merits and find them lacking. It does not ask whether the constitutional violation is made out. It asks only whether the underlying facts carry the scent of domestic relations, and if they do, it closes the courthouse door before the analysis begins. That is precisely what Section 1983 was designed to prevent: a system in which a litigant's access to federal adjudication of federal rights depends on whether the state actor who violated those rights happened to do so in a context the federal courts find uncomfortable. The DRE makes the family court the one forum in American life where a state actor can, with reasonable confidence, infringe constitutional rights and face no federal scrutiny whatsoever. That is not deference. It is immunity by venue.

The Parent in the Room

The premise underlying the DRE is that domestic relations constitute a coherent, categorical domain of state sovereignty into which federal courts do not venture. That premise is false. The federal courts have always been the constitutional parent of American family life, and the DRE does not keep them away. It determines, on an ad hoc basis, whose family gets the benefit of their presence.

Start with marriage itself. Obergefell v. Hodges, 576 U.S. 644 (2015), and United States v.

Windsor, 570 U.S. 744 (2013), did not touch on marriage tangentially. They defined its constitutional contours, overrode state licensing law, and compelled states to recognize unions they had affirmatively prohibited. If the domestic relations domain is categorically reserved to state authority, those decisions should not exist. The DRE has never been reconciled with them, because no reconciliation is available.

Immigration law provides the same problem at greater volume. Federal courts and administrative tribunals make findings about the validity of marriages, parental fitness, and family arrangements every day. The Board of Immigration Appeals and the circuit courts review those determinations on the merits. The federal interest in immigration jurisdiction apparently overrides the same state primacy the DRE treats as inviolable, and no one argues that immigration courts are constitutionally prohibited from examining a marriage.

Federal courts adjudicate international child custody disputes directly under the International Child Abduction Remedies Act, 22 U.S.C. §§ 9001-9011, implementing the Hague Convention.

District courts order the return of children across international borders, evaluate habitual residence, and assess grave-risk claims. Stripped of the international framing, they are resolving custody disputes. The DRE draws no principled line between that and a domestic custody case raising a federal constitutional claim.

ERISA produces some of the most extensive federal adjudication of marriage and divorce consequences in the country. Federal courts determine who qualifies as a spouse, whether a Qualified Domestic Relations Order validly divides a pension, whether a beneficiary designation survives divorce, and whether a federal plan document preempts a state court's property division.

Egelhoff v. Egelhoff, 532 U.S. 141 (2001), held that ERISA preempts state laws passing benefits to a former spouse after divorce. A federal statute controlled the financial consequences of an ended marriage. The Court had no difficulty with the jurisdiction.

Bankruptcy courts may be the single most active forum for family law adjudication outside of state courts. Whether a debt arises from a divorce decree, whether it constitutes a domestic support obligation entitled to priority under 11 U.S.C. § 507(a)(1), and whether it survives discharge under § 523(a)(5) or § 523(a)(15) all require bankruptcy judges to examine divorce agreements and interpret separation terms. These are among the most frequently contested issues in consumer bankruptcy. No one suggests the bankruptcy court lacks jurisdiction to resolve them.

The Indian Child Welfare Act, 25 U.S.C. §§ 1901-1963, goes further still. It gives federal law affirmative primacy over state court custody proceedings involving Native American children, requires compliance with federal placement preferences and evidentiary standards, and subjects state court custody orders to federal judicial review. Haaland v. Brackeen, 599 U.S. 255 (2023), litigated ICWA's constitutional scope before the Supreme Court. Federal adjudication of child custody, in that context, is not merely permitted but mandated by Congress, and the mandate is enforceable against state courts.

Section 1983 litigation over parental rights completes the picture. Federal courts have developed substantial doctrine on the Fourteenth Amendment liberty interest in family integrity. Cases involving state removal of children, see Tenenbaum v. Williams, 193 F.3d 581 (2d Cir. 1999), improper investigations by child protective services, and retaliation against parents who resist family court proceedings are regularly litigated in federal court. The circuits have defined what process is constitutionally due before a state may separate a parent from a child. Applied consistently, the DRE would sweep most of that doctrine away, because it arises from facts generated in the domestic relations context.

The Supreme Court itself has reached directly into custody and parental rights proceedings when constitutional stakes were visible. In Palmore v. Sidoti, 466 U.S. 429 (1984), Chief Justice Burger's unanimous opinion noted at the outset that a state court custody judgment "is not ordinarily a likely candidate for review by this Court," then reversed a Florida custody determination that had transferred a child from her mother to her father because the mother had remarried a Black man. The Equal Protection Clause applied in the custody context without hesitation. The DRE was not an obstacle. In Santosky v. Kramer, 455 U.S. 745 (1982), the Supreme Court imposed a federal due process standard, clear and convincing evidence, on New York Family Court proceedings to terminate parental rights, holding that the state's preponderance standard was constitutionally insufficient. Federal constitutional law governed the evidentiary standard in a state family court proceeding. In Troxel v. Granville, 530 U.S. 57 (2000), the Court struck down a Washington grandparent visitation statute as applied, holding it violated parents' substantive due process right to make child-rearing decisions. The foundational parental liberty cases, Meyer v. Nebraska, 262 U.S. 390 (1923), and Pierce v. Society of Sisters, 268 U.S. 510 (1925), established the constitutional liberty interest in family autonomy that the Court has been applying in family court contexts ever since. Stanley v. Illinois, 405 U.S. 645 (1972), struck down a state statute that automatically transferred an unwed father's children to state custody without any hearing. The federal courts had no difficulty reaching the constitutional question.

The federal courts are the parent in the room. The DRE does not change that. It determines, on grounds that have never been principled, whose constitutional rights get the benefit of the parent's attention. The families who can frame their claim around marriage equality, immigration status, pension benefits, or tribal membership get federal review. The parent whose family court proceeding violated the Constitution in ways that do not carry a sufficiently salient federal hook gets the DRE. The difference between those two outcomes is not a function of constitutional principle. It is a function of which family presented a sufficiently salient constitutional stake.

Habit Is Not Parentage

Long-standing judicial practice carries real weight in constitutional interpretation. It informs what the founders understood, how institutions have functioned, and what disruption a change in doctrine might cause. The argument here is narrower: practice was never meant to operate as a mechanism by which courts could permanently override a constitutional grant of jurisdiction without legislative authorization. The Constitution assigns the power to define the scope of federal jurisdiction to Congress, within the limits Article III sets. Courts interpret what Congress has done.

They do not fill gaps by inventing permanent exclusions that Congress never enacted. The DRE is what happens when that line is crossed. A nineteenth-century dictum became practice.

Practice became doctrine. Doctrine became, in effect, a permanent jurisdictional policy that no individual litigant can dislodge. The Judicial Conference, through its role in shaping how courts understand their own jurisdiction, has institutionalized that policy in ways that are largely invisible. A body that is not itself a court, not subject to Senate confirmation in its policymaking capacity, and not directly accountable to Congress has effectively shaped which categories of claim the federal courts will hear. That is jurisdictional lawmaking outside the constitutional apparatus designed to govern it.

I work on this problem in a different domain. In AI governance, the central question is what happens when a system accumulates authority over decisions it was never authorized to make, and whether accountability mechanisms are adequate to check it. The answer, in almost every case examined seriously, is that they are not. The DRE is an older instance of the same structural problem: authority exercised without authorization, hardened into settled practice, and insulated from scrutiny by the very institutions that benefit from the arrangement. Habit is not parentage.

Custom can inform interpretation. It cannot replace it.

The Institutions That Could Move This Question

Congress will not fix this. Family law is among the most politically sensitive subjects in American public life, and congressional action on federal court jurisdiction is especially difficult when the subject matter invites accusations of intrusion on state domestic authority. Congress could amend the diversity statute to remove any ambiguity about its application to civil rights claims arising in domestic contexts, or clarify the scope of Section 1983's reach in family court proceedings. It has not done so, and there is no realistic prospect that it will. The constitutional structure that created the problem, an unelected judiciary generating jurisdictional doctrine without legislative authorization, is also the structure most resistant to the kind of broad political coalition that produces congressional action.

The Federalist Society has done more than any organization in the past four decades to embed originalism and textualism in the federal judiciary. Those commitments, taken seriously, produce a specific conclusion about the DRE: it is indefensible. The Framers did not carve domestic relations out of federal jurisdiction. The text does not support it. The diversity statute as originally enacted did not exclude it. What exists is a judicially invented exception, created without constitutional authority, sustained by repetition, and dressed in the language of federalism to give it an appearance of principle that the historical record does not support. An originalist who follows the method where it leads has no principled basis for preserving the DRE, regardless of any policy sympathy for state authority over family matters. The Society's forums, publications, and networks are among the most direct channels through which that conclusion could reach the judges who would have to act on it.

The Judicial Conference of the United States has more direct administrative influence over this question than any other institution. As the policy-making body for the federal courts, chaired by the Chief Justice and composed of the chief judges of the circuits and district courts, the Conference shapes how courts understand their own jurisdiction through guidelines, reports to Congress, and the work of its committees. The Conference has not treated the DRE as a problem requiring attention. That choice is itself consequential. A body with this degree of practical authority over federal jurisdictional norms has more capacity to engage this question than any litigant or legislature, and has not yet been called upon to do so.

Whether that posture is sustainable is now an active legal question. America First Legal Foundation v. Roberts, No. 1:25-cv-01232 (D.D.C.), illustrates the structural problem directly: AFLF sued the Chief Justice as Presiding Officer of the Judicial Conference under FOIA, arguing the Conference and Administrative Office are executive agencies subject to disclosure requirements. The case was dismissed by Judge Trevor McFadden, who had disclosed membership on a Judicial Conference committee appointed by the Chief Justice, the named defendant, but remained on the case regardless. The appeal is pending in the D.C. Circuit. The pattern is the same one the DRE presents: a body exercising policymaking authority over federal courts, deciding questions about its own character and reach, insulated from the accountability mechanisms that govern other lawmaking.

The legal academy is the most immediate channel. The DRE has attracted scholarly attention, but not commensurate with its constitutional significance. The argument that the DRE's application to civil rights claims is constitutionally indefensible has not received the sustained scholarly development that moves judicial doctrine over time. Judges read law reviews, or more precisely their clerks do, and the pipeline from academic critique to a concurring opinion questioning a precedent to eventual reexamination by a full court is one of the more reliable mechanisms of doctrinal change in American law.

None of these institutions requires a congressional majority or a sympathetic administration to engage the question. Each operates within the legal system itself. The DRE has survived not because the argument for it is strong but because the subject matter is unglamorous and the people most harmed by it are not a politically organized constituency. That is a reason to press the question, not to leave it where it is.

Answer the Question

Ankenbrandt is binding. The statutory reconstruction the Court adopted in 1992 has hardened into doctrine, and the distinction between decree-issuing jurisdiction and civil-rights jurisdiction has not been drawn with precision in the circuits. Litigants invoking the exception against federal civil rights claims continue to find receptive courts.

But the doctrinal case for the exception, examined closely, consists of four elements: nineteenth-century dictum, a statutory construction that the text does not clearly support, a tradition of deference to state family law courts that predates the modern federal civil rights framework, and an understandable but legally insufficient preference among federal judges to stay out of contested family disputes. None of those elements provides a satisfying answer to the Article III objection when the claim before the court is constitutional rather than domestic.

The Court's own equivocation in Ankenbrandt suggests awareness of the problem. Locating the exception in the diversity statute rather than the Constitution was a narrowing move, not a validation. It also carried an implicit message ignored for three decades: if the doctrine is wrong as a matter of statutory construction, Congress can say so. It has not. That silence is not endorsement.

It is inertia. A Congress that amended the diversity statute to clarify its application to civil rights claims arising in domestic contexts, or that expressly confirmed the scope of Section 1983 in family court proceedings, would resolve the question that Ankenbrandt declined to answer. That Congress has not acted in thirty years while federal courts routinely adjudicate marriage, custody, and parental rights under ERISA, bankruptcy, ICWA, and Section 1983 itself is not evidence that the doctrine is correct. It is evidence that no one has been politically inconvenienced enough by it to force the issue.

The doctrinal path for courts that are willing to engage the argument is not complicated. The DRE, as Ankenbrandt drew it, bars federal courts from issuing divorce, alimony, and child custody decrees. It does not, on any principled reading of that decision, bar federal question jurisdiction over constitutional claims arising from family court proceedings. Courts should hold that the DRE does not apply to Section 1983 claims alleging constitutional violations in domestic relations proceedings. That rule does not require overruling Ankenbrandt. It requires reading it as it was written, rather than as it has been applied. It does not authorize federal courts to relitigate custody determinations, displace Rooker-Feldman, or second-guess family court decisions on non-constitutional grounds. It means only that a constitutional claim does not lose its federal forum because the facts arose in a family court. The circuit courts that have extended the DRE beyond its decree-issuing core have done so without constitutional authorization. They should stop.

The question the baby bird asks is simple: are you my mother? The federal courts' record answers it. They prosecuted the marriage of a religious minority. They dissolved a church. They defined who may marry whom. They told states that race cannot determine custody. They set the evidentiary standard for taking children from parents. They declared that Americans have a constitutional right to marry the person they love. At every turn, when the constitutional stake was visible and the political culture was aligned, the answer was yes. I am your mother. I am here.

The Domestic Relations Exception is the federal judiciary's attempt to claim, when the constitutional stakes are less salient, that it was never in the room. The historical record does not support that claim. Neither does the constitutional text. Neither does the century and a half of family law jurisprudence the federal courts have produced.

The bird found its mother. The question has been answered by the courts' own conduct, across every generation of American constitutional history. The exception should be closed. The answer has always been yes.

Michael Simon Baker is a New York attorney and the author of Prophets and Prejudice: Race, Power, and the Mormon Priesthood Ban. This article is for informational purposes only and does not constitute legal advice.