Re: Formal Disciplinary Complaint Against John G. Roberts Jr., DC Bar Member
Supreme Court Disclosure Accountability: The Enforcement Architecture Gap
Instrument: Formal Disciplinary Complaint Against John G. Roberts Jr., DC Bar Member
Filer: Christopher Armitage (member of the public)
Filed: April 22, 2026
Filed with: DC Bar Office of Disciplinary Counsel, District of Columbia Court of Appeals, Building A, Room 117, Washington, DC 20001
Governing statutes: Ethics in Government Act, 5 U.S.C. §§ 13101 et seq.; DC Rules of Professional Conduct 8.4(b) and 8.4(c); DC Bar Rule XI § 1(a)
Accountability mechanisms implicated: Judicial Conduct and Disability Act, 28 U.S.C. §§ 351–364; Judicial Conference of the United States; Supreme Court Code of Conduct (adopted November 2023)
Track: Judicial
Plain-language summary: Chief Justice John Roberts filed fifteen consecutive annual federal financial disclosure forms (2007–2021) characterizing his wife’s legal recruiting income as “salary.” The underlying compensation was commission-based, paid by third-party law firms that appear regularly before the Supreme Court. Roberts also omitted his wife’s equity interest in her employer from three consecutive annual filings (2019–2021). Both were corrected approximately seventeen days after Business Insider published whistleblower documents in April 2023 — not as the result of internal review. On April 22, 2026, journalist and Substack author Christopher Armitage filed a formal disbarment complaint with the DC Bar Office of Disciplinary Counsel, arguing that the disclosure pattern constitutes professional misconduct under DC Rules of Professional Conduct 8.4(b) and 8.4(c). This brief analyzes the complaint’s legal architecture and its five identified technical vulnerabilities, and maps the structural enforcement gap the complaint navigates: the absence of any binding accountability mechanism above the federal appellate level for Supreme Court justices.
Verdict line: The Armitage complaint is architecturally sound as a jurisdictional theory — personal sworn EIGA filings made in an individual capacity are attorney conduct, not judicial conduct, and DC Bar Rule XI draws no distinction between active and judicial members — but five unaddressed technical vulnerabilities will provide sophisticated defense counsel with a roadmap to dismissal before the complaint reaches the merits. The deeper structural finding is that this complaint exists because every other accountability body with potential jurisdiction over a sitting Chief Justice has been explicitly excluded, politically neutralized, or designed without enforcement teeth — and correcting that design failure requires legislative action that the 2023 SCOTUS ethics code was constructed to forestall.
Legal Impact Assessment
This section analyzes the Armitage complaint as a legal instrument: what rights, constraints, and accountability mechanisms it implicates, where it is architecturally sound, and where it is vulnerable. This is a legal vulnerability audit, not a litigation forecast.
The jurisdictional theory
The complaint’s central jurisdictional argument is that executing sworn federal financial disclosure forms is attorney conduct in Roberts’s individual capacity — not judicial conduct — and therefore falls within DC Bar jurisdiction regardless of the respondent’s status as Chief Justice. The argument rests on two foundations: DC Bar Rule XI, Section 1(a), which subjects “all members of the District of Columbia Bar” to the court’s disciplinary jurisdiction without distinguishing between active, judicial, or inactive members; and the nature of EIGA compliance filings as personal administrative acts that attach to the filer as a federal officer, not as a judge adjudicating any matter.
This framing is the only available jurisdictional hook given the enforcement landscape described in the Structural Analysis below. The complaint demonstrates awareness of the DC Bar ODC’s published practice of not investigating “judges acting in a judicial capacity” — stated verbatim on the ODC’s public-facing website (dcbar.org/attorney-discipline/office-of-disciplinary-counsel/purpose-and-mission) — and preemptively addresses it in Section VII.E, arguing that the practice statement covers adjudicative acts, not personal sworn filings.
One citation in the jurisdictional section warrants a precision note. The complaint cites In re Slattery, 767 A.2d 203 (D.C. 2001), for the proposition that the DC Court of Appeals “disciplines federal judicial officers for off-bench dishonesty.” The citation is verified via CourtListener (767 A.2d 203, D.C. 2001; courtlistener.com/opinion/1985253/in-re-slattery/): the case exists, the reporter citation is correct, and the DC Court of Appeals did discipline Daniel J. Slattery Jr. for off-bench misconduct. However, Slattery was a Federal Administrative Law Judge — not an Article III judge — and the case’s core holding addressed whether bar discipline constitutes a criminal proceeding (it does not), not whether DC Bar jurisdiction extends specifically to sitting Article III judges. The case supports the complaint’s general jurisdictional argument but is not as on-point as stated. The complaint should either qualify the attribution or identify a case more precisely on-point for Article III judges.
Textual Finding: High | Litigation Risk: Medium
The Textual Finding is high because the jurisdictional theory is grounded in the plain text of DC Bar Rule XI § 1(a). The Litigation Risk is medium rather than high because ODC retains discretionary screening authority and may decline the complaint on judicial-capacity grounds even if the underlying theory is legally sound; the outcome depends on whether ODC reads its own practice statement as a discretionary screening tool or as a jurisdictional limitation.
Vulnerability 1 — The Clark citation
The complaint cites “In re Clark, Bd. Docket No. 22-BD-0891F (D.C. Bd. on Prof’l Resp. 2025)” for the proposition that the DC Bar Board recently rejected senior-federal-officer immunity arguments as “absurd.” This citation has three compounding problems.
First, the docket number (22-BD-0891F) does not match publicly available records of the Clark Board proceedings; the docket shown in the DC Bar’s published discipline database is “22-BD-039.” Second, the “22-BG-0891” number associated with the Clark matter in public sources is a DC Court of Appeals appellate number, not a Board docket number, and the “F” suffix does not correspond to any verifiable DC Bar docket convention. Third, the 2025 date attribution is inconsistent with available public records indicating the relevant proceedings concluded in 2024. Whether the Board used the specific word “absurd” to characterize immunity arguments cannot be confirmed from publicly available materials.
Because jurisdiction is a threshold issue, any citation error or misquote in this section creates disproportionate dismissal risk before the complaint touches the merits.
The fix is precise: pull the correct DC Court of Appeals case number for the Clark proceedings (No. 22-BG-0891, 2024) or obtain the Board docket number directly from the Board’s case PDF caption, verify whether the Board used the word “absurd,” and quote the relevant language verbatim with a pincite. Operationally: quote only what appears verbatim in the Clark decision with a page pincite; if “absurd” does not appear verbatim, remove it and describe the holding neutrally. Attach the Board decision as an exhibit.
Textual Finding: Low | Litigation Risk: High
The underlying principle the Clark citation supports is legally sound. The citation itself is unverifiable in its current form and creates a correctable but serious exposure.
Vulnerability 2 — The 15/16 count inconsistency
The complaint is internally inconsistent on the number of mischaracterized filings. Section IV.B correctly describes “fifteen consecutive annual filings” covering calendar years 2007–2021. The calendar year 2022 form (filed May 15, 2023) is the corrected one. Section V.A then states “sixteen consecutive identical mischaracterizations across sixteen separate annual filings.” A section heading describes the period as a “sixteen-year pattern, 2007–2021” — which is arithmetically impossible, as the span from 2007 through 2021 is fifteen years. The complaint also refers interchangeably to “Mr. Roberts’s own 2022 correction” and “Mr. Roberts’s own 2023 correction” without standardizing the convention (the calendar year 2022 report was filed in May 2023, so both can be technically accurate — but the inconsistency is avoidable and will be exploited).
The practical risk is lower than the Clark vulnerability but is not trivial. Opposing counsel will use the inconsistency to argue that the complaint’s factual predicate is imprecise throughout, which undermines the reliability of the underlying factual record at every subsequent point.
The fix: standardize throughout to fifteen mischaracterized forms (calendar years 2007–2021) and one corrected form (the 2022 Financial Disclosure Report, filed May 15, 2023). Update the section heading, the count in Section V.A, and the comparative sanction arguments in Section VIII accordingly.
Textual Finding: High | Litigation Risk: Medium
The inconsistency is clear from the complaint’s own text. The litigation risk is medium rather than high because it is a precision error, not a substantive jurisdictional defect — but precision errors in a 142-page complaint against a Chief Justice will be used aggressively.
Vulnerability 3 — Four unaddressed anticipated defenses
The complaint addresses four anticipated defenses in Section VII: Ethics Opinion 323, inadvertence, reliance on Administrative Office guidance, and materiality. Four additional defenses are predictable from the record and unaddressed.
Separation of powers. Roberts’s counsel will argue that a state bar disciplinary body cannot sit in judgment of a sitting Chief Justice’s disclosure compliance choices without implicating Article III and the constitutional separation of powers. The complaint must establish that disciplining bar membership conduct is categorically distinct from reviewing a judicial act or removing a justice — and that courts sustaining bar jurisdiction over attorneys who happen to be federal judges have consistently drawn this distinction.
Exclusive remedy / institutional competence. Roberts’s counsel will argue that EIGA disclosure issues belong to the Judicial Conference process, not the DC Bar. The complaint must establish that bar discipline and Judicial Conference administrative review are parallel mechanisms, not mutually exclusive, and that DC Bar discipline reaches attorney conduct in ways the Judicial Conference process demonstrably cannot.
Fair notice / ambiguity. Roberts’s counsel will argue that “salary” is used colloquially to describe compensation paid through a single employer even when commission-based underneath, and that the EIGA form instructions did not provide sufficient notice that this characterization was impermissible. The complaint must quote the EIGA form instructions and any Judicial Conference guidance defining income type categories in force during 2007–2021, to establish that the salary/commission distinction was unambiguous on the face of the instrument at the time of each filing.
8.4(b) scienter gap. The complaint predicates Rule 8.4(b) on criminal conduct under 18 U.S.C. § 1001, which requires knowing and willful falsification. The pattern-based recklessness argument under Rosen that carries the 8.4(c) claim does not automatically satisfy § 1001’s willfulness requirement. The complaint needs a distinct scienter argument for the criminal predicate — one that addresses willfulness with the particularity a federal criminal standard requires.
Textual Finding: High | Litigation Risk: High
The absence of these defenses is identifiable from the complaint’s text. Roberts’s counsel will raise all four. The complaint has no prepared responses to any of them.
Vulnerability 4 — The advisory opinion is not a defense; the complaint should say so directly
The Snopes fact-check and Armitage’s rebuttal both reference what they describe as a “2009” Judicial Conference advisory on spousal recruiting income. The complaint addresses reliance on Administrative Office guidance in general terms (Section VII.C) but does not quote the advisory directly or demonstrate why it cannot serve as a good-faith defense.
The advisory is Committee on Codes of Conduct Advisory Opinion No. 107, “Disqualification Based on Spouse’s Business Relationships.” Its content does not provide a defense for Roberts. It provides the statutory mechanism by which the salary/commission mischaracterization was most consequential.
AO 107 establishes a magnitude-based threshold analysis for recusal purposes under Canon 3C(1) — not a clean binary, but a structured distinction with a safe harbor on one end and a recusal obligation on the other. The safe harbor, confirmed in the Guide to Judiciary Policy at page 209: “If the judge’s spouse is a salaried employee, rather than a partner or owner, and receives no bonus or commission based on the work performed for clients, the spouse will not be considered to have an interest in business clients that would require the judge’s disqualification.” The recusal trigger, also at page 209: “However, if the judge knew the spouse directly received a portion of client fees amounting to more than a small percentage of the spouse’s income, the judge’s impartiality might reasonably be questioned in a proceeding involving the client, in which case the judge should recuse.” And the scope of the trigger, also at page 209: “In situations of this nature, involving large fees paid by a client to the spouse, it is immaterial whether the spouse personally works for the client or collects fees in the nature of finders’ fees or fees based on the services provided to the client by others.”
The bridging argument the complaint needs: EIGA requires disclosure of income type — not just income source and amount — precisely because income type determines which recusal analysis applies under Canon 3C(1). A “salary” characterization places a spouse in the safe harbor category (no bonus or commission = no disqualifying interest). A “commission” characterization triggers the magnitude analysis (is the commission more than a small percentage of income?). The disclosure categorization feeds directly into the Canon 3C(1) analysis. Mislabeling commission income as salary does not merely describe compensation incorrectly — it places the disclosing officer in the wrong analytical category under the Judicial Conference’s own recusal guidance. This is the materiality argument the complaint currently states but does not fully develop.
Advisory Opinion No. 107 directly addresses legal and executive recruiting by a judge’s spouse, at page 208. It specifies that a judge whose spouse operates an executive recruitment business need not recuse for isolated or insubstantial transactions — but that “a spouse who personally performed four high level executive recruitments in the same year for the same company for substantial fees for each recruitment would be considered to have a substantial and ongoing relationship with the company, requiring the judge’s recusal.” This is an illustrative threshold, not a bright-line rule; the advisory also notes that if the same work “were spread out over a substantially longer period, recusal may not be required.”
Jane Roberts placed senior government lawyers at salaries ranging to $3 million. She received $10.3 million in commission income over seven years from firms appearing regularly before the Court. AO 107 recognizes that some commissions may be non-disqualifying when they are a “relatively small” percentage of the spouse’s income — it is not a blanket prohibition on all commission arrangements. But the alleged conduct here involves repeated multi-year misclassification of large client-linked commissions, which AO 107’s magnitude analysis places squarely in the recusal-trigger category, well above the “small percentage” threshold that might be excused. Characterizing that income as “salary” had the effect of placing the disclosure in the safe-harbor category that applies only to employees receiving no client-linked commission — when the actual compensation structure placed Roberts in the commission category that the advisory says requires recusal analysis.
Roberts’s counsel will argue that AO 107 is a recusal guide, not an EIGA disclosure categorization guide — that the advisory says nothing about how to label income on a financial disclosure form. The complaint’s response must make the causal chain explicit: income type on the EIGA form determines which recusal analysis applies under AO 107; mislabeling the income type does not merely describe compensation incorrectly, it places the disclosing officer in the wrong analytical category under the Judicial Conference’s own guidance. The disclosure obligation and the recusal obligation are structurally linked.
The complaint should quote Advisory Opinion No. 107 with pin cites to the Guide to Judiciary Policy, Vol. 2B, Ch. 2 (pp. 208–209), move it from Section VII.C (anticipated defenses) to Section VII.D (materiality) as an affirmative argument, and add the bridging sentence establishing that EIGA income type categorization feeds directly into the Canon 3C(1) analysis.
Textual Finding: High | Litigation Risk: Low
The advisory’s text is now confirmed. It does not support a good-faith reliance defense. The Litigation Risk for this particular vulnerability is now low — the advisory helps the complaint rather than threatening it — but the complaint’s failure to quote it directly means Roberts’s counsel can mischaracterize it without a prepared response in the record.
Vulnerability 5 — Unattached Board decisions and missing pincites
The complaint’s two most important non-reported precedents — In re Clark (jurisdictional argument) and In re Tun (sanction framing, including the attributed “mental gymnastics” and “disbarment-level discipline” language) — are cited without the Board decisions attached as exhibits and without pincites to the quoted language. Every quoted passage from Tun, Rosen, Howes, Shorter, Alessandro, and Miller is similarly unanchored in the filed complaint.
Bar disciplinary complaints are not appellate briefs, and pin-cite requirements differ accordingly. But where the complaint quotes specific language and attributes specific holdings that ODC cannot verify without independent research, those attributions are exposed to challenge. If any quoted language proves inexact when pulled and compared against the original, the complaint’s credibility is broadly undermined.
The fix: attach Clark and Tun Board decisions as exhibits; add pincites for every quoted passage; where the DC Court of Appeals issued a final discipline order, cite that final order rather than the Board recommendation alone, as final orders carry stronger precedential weight.
Textual Finding: Medium | Litigation Risk: Medium
Structural Analysis
This section does not analyze the complaint as a legal instrument. It analyzes the accountability architecture that produced the conditions under which this complaint is the most viable available instrument — why the enforcement gap exists, what sustains it, and what it enables.
The enforcement architecture gap
There are five bodies with potential jurisdiction over the conduct alleged in the Armitage complaint. None of them has binding, enforceable authority that reaches a sitting Chief Justice for this category of conduct. The complete loop:
The DC Bar Office of Disciplinary Counsel explicitly states in its published materials that it “cannot consider complaints against judges acting in a judicial capacity.” The complaint correctly argues that personal EIGA filings are not judicial conduct — but ODC retains discretionary screening authority and may decline on capacity grounds even if the underlying theory is legally sound. Even if ODC proceeds, the DC Court of Appeals would review any discipline, and the separation-of-powers challenge would remain live throughout any appeal.
The Judicial Conduct and Disability Act (28 U.S.C. §§ 351–364) provides the primary mechanism for disciplining federal judges. Its definition section at § 351(d)(1) specifies that the term “judge” in the Act “means a circuit judge, district judge, bankruptcy judge, or magistrate judge.” Supreme Court justices are excluded by definitional omission — not by express language, but by the Act’s enumeration of covered judicial officers, which does not include them. Complaints under the Act are filed with the circuit clerk and administered by the relevant circuit judicial council. There is no circuit council for SCOTUS. As written, the Act supplies an administrative discipline mechanism for enumerated lower-court judges; it does not create a parallel complaint path for Supreme Court justices.
The Judicial Conference of the United States can receive complaints, conduct investigations, and certify findings to the House of Representatives for consideration of impeachment. It cannot compel recusal, impose sanctions, or discipline bar membership. The § 13106(b) referral duty — under which the Judicial Conference “shall refer” to the Attorney General names of individuals it has reasonable cause to believe have willfully falsified required EIGA information — is also administered by the Judicial Conference, which is chaired by the Chief Justice under 28 U.S.C. § 331. Note also that § 13106(a)(2)(B)(i) — confirmed from the current statutory text — creates a parallel criminal penalty for willful falsification: “fined under title 18, imprisoned for not more than 1 year, or both.” This is distinct from the § 1001 five-year exposure the complaint also invokes. The complaint characterizes § 13106(a) as imposing only civil penalties, which understates the statutory exposure.
The Supreme Court Code of Conduct, adopted November 2023, applies to all nine justices by its terms. It has no enforcement mechanism. No body is designated to receive complaints, conduct investigations, or impose consequences under the code. The code was developed under Roberts’s leadership as Chief Justice. A governance instrument with no enforcement mechanism is not an accountability mechanism — it is a preemptive substitute for one.
Congress retains impeachment authority over all federal officers including justices. The Senate impeached and removed Judge Porteous in 2010 on grounds that included false EIGA financial disclosure forms — but Porteous was a district judge, his proceedings were initiated following independent investigative findings by the Fifth Circuit Judicial Council, and the process required political will that is not available for a Chief Justice in the current Congress. The Porteous record is offered here as illustrative context — demonstrating how federal institutions have characterized false EIGA disclosures by a federal judge — not as binding precedent for a DC Bar disciplinary proceeding. Impeachment findings do not create controlling authority for bar discipline, and the Armitage complaint should clarify this characterization to foreclose a reductio ad absurdum response from Roberts’s counsel.
The structural result: every accountability body hands the question to the next. ODC says it cannot touch judicial conduct. The Judicial Conduct Act excludes SCOTUS. The Judicial Conference can only recommend to Congress and is chaired by the subject. Congress will not act. The ethics code has no mechanism to act. The Armitage complaint threads this complete loop by targeting the one jurisdictional hook that remains open: the attorney-conduct theory. Whether that hook holds under a separation-of-powers challenge is the central unresolved question.
Standing structural flags
Accountability gap: The enforcement architecture produces a condition where no body has binding, enforceable authority over a sitting Chief Justice’s compliance with the disclosure regime. This is not a gap produced by legislative oversight — it reflects a deliberate combination of constitutional design (Article III independence), statutory choices (JCDA scope), and administrative design (ethics code without enforcement). Its practical effect is that the accountability mechanism for the most consequential judicial disclosure obligations operates entirely on the honor system.
Textual Finding: High | Structural Significance: High
Power concentration: The Chief Justice, as administrative head of the federal judiciary, participates in designing the accountability mechanisms that apply to him. The 2023 ethics code was developed under Roberts’s direct leadership. An officer who participates in designing their own accountability architecture and whose participation produces a code without enforcement teeth presents a structural conflict of interest that is not incidental — it is built into the governance design of the Judicial Conference.
Textual Finding: High | Structural Significance: High
Preemption of oversight: The 2023 ethics code functions structurally as a preemptive substitute for external accountability, not a complement to it. Its adoption reduced the political pressure on Congress to legislate binding enforcement mechanisms. A code with no enforcement mechanism that occupies the political space in which legislation might otherwise pass is a structural regression dressed as progress. The appropriate test for any ethics instrument is what happens when a violation occurs — for this code, the answer is nothing.
Textual Finding: Medium | Structural Significance: High
The Textual Finding is medium because “preemption” requires inferential judgment about political dynamics that the code’s text does not make explicit.
Sunset provisions and review mechanisms absent: The SCOTUS ethics code has no built-in review timeline, no public amendment process, and no defined accountability for non-compliance. The JCDA scope exclusion of SCOTUS has never been revisited through a formal legislative review process since the Act’s passage in 1980. Neither instrument was designed with a correction mechanism.
Textual Finding: High | Structural Significance: Medium
Second and third-order effects: If no accountability mechanism reaches a sitting Chief Justice for EIGA disclosure violations, the incentive structure for every federal judge beneath the SCOTUS level is affected. The architecture communicates that disclosure compliance is enforceable through normal institutional channels except at the apex. This means the most consequential disclosure obligations — those of justices with final jurisdiction over the most significant cases — are the least enforceable. Enforcement intensity and case-consequence run in opposite directions by design.
Textual Finding: Medium | Structural Significance: High
Abstraction Layer Analysis
This section analyzes the EIGA disclosure architecture as a policy instrument — whether the statutory and implementation layers are coherently separated, and where layer collapses produce the failure modes the Roberts disclosure pattern illustrates.
Income type definition delegated to implementation layer. The Ethics in Government Act requires disclosure of spousal non-investment income but does not define income type categories (”salary,” “commission,” “other”) in the statute itself. Those distinctions are delegated to form instructions and Judicial Conference guidance — documents that can be amended by administrative action without returning to Congress. The result: the definitional term most critical to the conflict-identification function of the disclosure regime is defined at the implementation layer, not the statutory layer, and is therefore subject to administrative drift, reliance defenses, and inconsistent interpretation over time.
This is a textbook abstraction layer collapse. The policy goal — identify third-party payors who might create conflicts — is stated at the statutory layer. The implementation detail that determines whether the goal is achieved — what distinguishes “salary” from “commission” — is delegated to guidance the statute does not control. When that guidance is ambiguous or is interpreted advantageously by the disclosing officer, the statutory conflict-identification function fails silently.
Textual Finding: High | Structural Significance: High
Amount privacy design creates dependency on type accuracy. EIGA keeps specific compensation amounts private above a de minimis threshold — a privacy protection for disclosing officers. This design choice creates a structural dependency: the conflict-identification function of spousal income disclosure depends entirely on accurate income type characterization, because the public cannot verify conflicts from the amounts disclosed. “Salary” implies a single employer. “Commission” flags third-party payors whose identities would otherwise be visible to the public. If income type can be mischaracterized without independent verification, the privacy protection for amounts eliminates the last check on the conflict-identification layer.
Textual Finding: High | Structural Significance: High
Referral architecture at § 13106(b) routes through the subject. The Ethics in Government Act’s enforcement mechanism for willful violations is a referral duty: the Judicial Conference “shall refer to the Attorney General” names of individuals believed to have willfully falsified required information. The referral trigger depends on the Judicial Conference taking action — an entity chaired by the Chief Justice, in the specific case of SCOTUS disclosures. The architecture routes enforcement of the Chief Justice’s disclosure obligations through an entity the Chief Justice administers. This is not a procedural technicality; it is the enforcement chokepoint.
Textual Finding: High | Structural Significance: High
Administrative Office advisory authority creates reliance defense without statutory warrant — and the relevant advisory cuts against Roberts. The Administrative Office issues guidance on EIGA compliance that federal judges rely on. That guidance is not legally binding, is not subject to notice-and-comment rulemaking, and is not publicly available in a searchable, versioned form. The relevant instrument here is Committee on Codes of Conduct Advisory Opinion No. 107, “Disqualification Based on Spouse’s Business Relationships,” compiled in the Guide to Judiciary Policy, Vol. 2B, Ch. 2. That advisory does not create a good-faith reliance defense. It creates the opposite: the advisory establishes a magnitude-based threshold distinction between a salaried employee receiving no client-linked commission (generally non-disqualifying) and commission-based compensation from client firms above a small-percentage threshold (recusal trigger), and explicitly addresses legal and executive recruiting by a spouse. Characterizing Jane Roberts’s compensation as “salary” had the effect of placing the disclosure in the safe-harbor category that the advisory reserves for employees receiving no client-linked commission — when the actual compensation structure placed the household in the commission category the advisory treats as requiring recusal analysis. Advisory-layer authority without statutory anchoring can create exploitable ambiguity; in this instance, the advisory’s own text resolves the ambiguity against Roberts.
Textual Finding: High | Structural Significance: High
No independent verification mechanism. EIGA disclosures for federal judges are filed with the Administrative Office and reviewed by the Judicial Conference. There is no audit function, no cross-referencing against IRS records or employer payroll data, and no mechanism for oversight bodies to trigger a review without filing a formal complaint that requires them to already possess the evidence that verification would produce. The disclosure regime’s conflict-identification function depends entirely on accurate self-reporting with no verification layer — which means detection depends on investigative journalism, whistleblowers, and the rare formal complaint.
Textual Finding: High | Structural Significance: High
What This Brief Gets Right
The Armitage complaint’s attorney-conduct jurisdictional frame is architecturally correct and represents sophisticated navigation of a genuinely constrained enforcement landscape. The complaint correctly identifies that personal sworn EIGA filings are the one category of conduct by a sitting Chief Justice that could plausibly fall within DC Bar jurisdiction — because it is attorney conduct in an individual capacity, not adjudicative conduct. The framing of Section II and the preemptive response in Section VII.E to the judicial-capacity practice statement both reflect awareness of where the jurisdictional boundary actually runs.
Textual Finding: High | Structural Significance: High
The 17-day correction timing is the complaint’s strongest single factual argument and is well-positioned throughout. The gap between the April 28, 2023 Business Insider publication and the May 15, 2023 correction date establishes that the correction was reactive rather than the product of internal review. This timing makes inadvertence unavailable as a defense for both the salary characterization and the equity omission. The complaint uses this correctly and returns to it at the right points in the analysis.
Textual Finding: High | Strategic Value: High
The precedent spine — Porteous for federal judges and EIGA, Alessandro for judicial-to-attorney discipline, Tun for multi-year pattern of sworn false filings, Howes for duration as a disbarment-level aggravating factor — is well-constructed. Each case is directionally on-point. The complaint’s “doctrinal spine” in Section VI.F synthesizes the authorities cleanly into six controlling principles that stack rather than overlap. The vulnerability is pincite support and exhibit attachment, not the selection of authority.
Textual Finding: Medium | Litigation Risk: Medium
The materiality argument in Section VII.D is legally sound and independently supported by multiple sources. The complaint correctly identifies that the salary/commission distinction conceals the identity of third-party payors — law firms appearing regularly before the Court — in a way that defeats the conflict-identification purpose of the statute. Professor Gershman’s memorandum, Roberts’s own Part VIII explanatory language acknowledging that the income “type” required “clarification,” and the 17-day reactive correction together constitute independent materiality support that does not depend on any single contested factual claim.
Textual Finding: High | Strategic Value: High
Recommendations
For the Armitage complaint
Priority 1 — Correct the Clark citation before this becomes a widely-cited filing.
The Clark citation is the single highest-risk element in the complaint because it sits at the jurisdictional gateway. The fix: obtain the Board decision from the DC Bar, pull the correct docket number from the Board PDF caption, verify whether “absurd” appears verbatim, quote the language with a pincite, and attach the decision as an exhibit. If “absurd” does not appear verbatim, correct the attribution to accurately reflect what the Board held. This should be done before the complaint is amended for any other reason.
Priority 2 — Standardize the count throughout.
Fifteen mischaracterized forms, calendar years 2007–2021. One corrected form: the 2022 Financial Disclosure Report, filed May 15, 2023. Update Section V.A (”fifteen consecutive annual filings”), the section heading (”fifteen-year pattern”), and the comparative sanction arguments in Section VIII. Adopt the convention “the 2022 Financial Disclosure Report (filed May 15, 2023)” and apply it consistently wherever the correction is referenced.
Priority 3 — Add responses to the four unaddressed defenses.
Add a subsection to Section VII titled “Additional Anticipated Defenses.” Four paragraphs: separation of powers (disciplining bar membership is categorically different from reviewing a judicial act; cite federal cases sustaining bar jurisdiction over attorney-judges); exclusive remedy (DC Bar discipline and Judicial Conference administrative review are parallel, not mutually exclusive; neither forecloses the other); fair notice (quote EIGA form instructions and Judicial Conference guidance in effect during 2007–2021 defining income type categories; establish the salary/commission distinction was unambiguous on the face of the instrument); 8.4(b) scienter (develop willfulness argument under § 1001 independently of the recklessness analysis for 8.4(c)). Note also that 5 U.S.C. § 13106(a)(2)(B)(i) carries criminal penalties independent of § 1001 — up to one year imprisonment for willful falsification — providing a second independent criminal predicate for 8.4(b). The complaint currently characterizes § 13106(a) as civil-only, which understates the available predicate and should be corrected.
Priority 4 — Turn Advisory Opinion No. 107 into an affirmative argument.
The advisory the Snopes piece and Armitage’s rebuttal reference is Committee on Codes of Conduct Advisory Opinion No. 107, “Disqualification Based on Spouse’s Business Relationships,” compiled in the Guide to Judiciary Policy, Vol. 2B, Ch. 2. Quote it directly with pin cites to pages 208–209. The advisory establishes a magnitude-based threshold analysis — not a binary — for recusal purposes under Canon 3C(1): the safe harbor applies only to employees who receive “no bonus or commission based on the work performed for clients” (Guide, p. 209); the recusal trigger applies when the spouse “directly received a portion of client fees amounting to more than a small percentage of the spouse’s income” (Guide, p. 209). The advisory also establishes that this analysis applies regardless of whether the spouse personally performs the work or “collects fees in the nature of finders’ fees” (Guide, p. 209). Jane Roberts received $10.3 million in commission income from the client firms — well above any small-percentage threshold. The complaint should move AO 107 from Section VII.C (anticipated defenses) into Section VII.D (materiality), and add the bridging sentence: EIGA income type categorization determines which recusal analysis applies under AO 107; mislabeling commission income as salary had the effect of placing the disclosure in the wrong analytical category under the Judicial Conference’s own guidance.
Priority 5 — Add pincites and an authorities appendix.
Attach Clark and Tun Board decisions as exhibits. Add pincites for every quoted passage throughout the complaint. Where the DC Court of Appeals issued a final discipline order, cite the final order rather than the Board recommendation alone.
For the accountability architecture
The enforcement gap this complaint navigates cannot be closed by any individual complaint or bar disciplinary proceeding. Closing it requires legislative action.
The Judicial Conduct and Disability Act should be amended to include Supreme Court justices within its scope. The most direct structural fix is to extend 28 U.S.C. §§ 351–364 coverage to SCOTUS, with a reviewing body that is independent of the circuit judicial councils (which have a structural conflict as inferior courts). The design challenge is real — any body with authority to discipline SCOTUS justices will face Article III independence challenges — but the current design is not a neutral default. It is a documented accountability-free zone at the apex of the federal judiciary.
The EIGA income type categories should be defined in the statute, not delegated to administrative guidance. Moving the definitions of “salary,” “commission,” and other income type categories from form instructions to the statute itself eliminates the reliance defense, removes the administrative opacity window, and anchors the conflict-identification function in enforceable text rather than guidance that the disclosing officer can rely upon advantageously. This is the single cheapest structural fix available — a definitional amendment requires no new enforcement architecture.
An independent verification mechanism should be established for EIGA filings by federal judges. Cross-referencing judicial disclosure forms against IRS records for the disclosing officer’s household, administered by an entity independent of the Judicial Conference, would transform the disclosure regime from a self-report system to a verification system. The current architecture is not a conflict-identification regime — it is a conflict-documentation regime that depends entirely on accurate voluntary disclosure, and detects violations only when journalists or whistleblowers supply the evidence that verification would have produced automatically.
The 2023 SCOTUS ethics code is not a substitute for any of these. A code without an enforcement mechanism occupies the political space that might otherwise be used to enact binding legislation. If Congress treats the code’s existence as having resolved the accountability question, the structural gap persists while the political will to close it dissipates. The code’s adoption without enforcement teeth is itself a structural finding: it is evidence that the accountability gap is recognized, and that the recognized response to it was deliberately designed to not close it.
Confidence taxonomy: All findings carry ratings on two axes. Legal findings use Textual Finding and Litigation Risk. Structural and abstraction layer findings use Textual Finding and Structural Significance. Ratings are High / Medium / Low on each axis.
Track note: This brief inaugurates the Church Bells judicial track. The methodology is identical to the legislative and executive tracks — structural vulnerability analysis applied consistently regardless of which branch produces the instrument or which party is implicated. Prior Church Bells briefs have covered the executive and legislative tracks. The judicial branch accountability architecture is a distinct domain with distinct failure modes; future judicial track briefs will build on the structural baseline established here.
This brief does not take a position on whether Chief Justice Roberts committed professional misconduct, nor does it advocate for or against the Armitage complaint’s success. It analyzes the complaint as a legal instrument and the accountability architecture it navigates as a structural engineering problem.
See also: Armitage Disbarment Complaint, DC Bar ODC, April 22, 2026
Statutory basis: Ethics in Government Act, 5 U.S.C. §§ 13101 et seq.; DC RPC 8.4(b), 8.4(c); DC Bar Rule XI § 1(a); Judicial Conduct and Disability Act, 28 U.S.C. §§ 351–364