Find judges by district, circuit, and case type
Move beyond name-only lookup. Search federal and state judges by the venue and matter type that actually matters to strategy.
GavelSight gives attorneys a paid, no-freemium judge intelligence stack: searchable federal judge profiles, an opinion-grounded research chat that answers in plain English with citations to the actual rulings, and an AI Paralegal — research memos, an authority finder, and a cite-checker that flags any citation not found in the corpus. Every answer is grounded in primary federal-court records. Structured ruling analytics are expanding as our corpus completes — we show you exactly what's covered, never a black box.
Below is an unedited GavelSight research answer about Judge David L. Bazelon (D.C. Cir., 820 indexed opinions). Question and answer are real — no script, no curation, no fabricated citations.
Bottom line: Of the three excerpts provided, only one directly addresses the insanity defense, and the corpus here is limited. Counsel should treat the following as a partial picture grounded in these specific opinions.
The insanity-defense opinion. The most on-point excerpt is United States v. Wright opinion #3, where Bazelon (writing for the majority) addressed whether a trial judge abused his discretion by declining to raise an insanity defense sua sponte over the defendant's objection. The defendant, charged with destroying government property (a model of the Capitol), had committed the act in response to what he described as religious "inspiration" to warn of "God's impending judgment." Notably, a St. Elizabeths Hospital report — reflecting the consensus of a psychiatrist, two additional psychiatrists, a psychologist, and a social worker — concluded that the defendant was competent to stand trial but had suffered from a mental disease at the time of the offense, diagnosing him with "Schizophrenia, paranoid type." This excerpt shows Bazelon engaging carefully with the tension between substantial psychiatric evidence of mental illness and the defendant's own objection to asserting the defense — framing the legal question as one of the trial judge's discretion opinion #3.
A caveat on the excerpt's limits. The Wright excerpt cuts off before resolving the question (it ends mid-sentence on the competency finding), so it does not state the court's ultimate holding on whether the judge erred. What it does demonstrate is Bazelon's methodical attention to the full mental-health record — emphasizing the multidisciplinary composition of the evaluating team and the specific diagnosis — and his recognition that competency to stand trial is analytically distinct from mental state at the time of the offense opinion #3.
The remaining excerpts do not address mental health. The other two opinions are off-topic for this question. opinion #1 is a Fourth Amendment dissent on whether police had probable cause for a public-drinking arrest, stressing the gravity of arrest as an intrusion on personal liberty. opinion #2 is a separation-of-powers opinion concerning the Senate Select Committee's effort to enforce a subpoena for the Nixon tapes. Neither bears on insanity or mental-health considerations opinion #1opinion #2.
Recommendation: For a reliable account of Bazelon's well-known jurisprudence on the insanity defense and the role of psychiatry in criminal law, you'll want to pull additional opinions — this set yields only one relevant data point, and even that excerpt is incomplete on the holding.
Move beyond name-only lookup. Search federal and state judges by the venue and matter type that actually matters to strategy.
Judge Chat grounds every answer in the judge's actual rulings and links the citations. No black-box bravado, no fabricated cases — if the corpus doesn't cover it, we say so.
Research answers and reports are saved to account history and exportable for real law-firm workflows.
The current local seed still includes prototype tendency rows. GavelSight now surfaces that caveat plainly and is wired to the live federal opinion ingest pipeline, but you should treat seeded-demo analytics as prototype intelligence until live federal-record coverage is synced.