Reporting methodology
Articles 1 and 2 are primary-source led. Their language follows the evidence category supporting each claim.
Evidence categories
- Adjudicated fact
- A final judgment, conviction, guilty plea, official certification, or equivalent authoritative disposition.
- Institutional statement
- Evidence that an institution made a statement, not automatic proof that the statement is true.
- Investigator-recorded allegation
- An account documented by the outside investigators without a finding that the reported conduct occurred.
- Institutional record
- A personnel, student, donor, court, or administrative record described or preserved in the source archive.
- Respondent statement
- An admission, denial, partial acknowledgment, nonresponse, or statement through counsel, preserved with attribution.
- Boho News analysis
- A transparent calculation or inference drawn from identified evidence and labeled as analysis.
Part 2 public-record search
Boho News reviewed 33 surviving Traverse City Record-Eagleindex PDFs and identified 106 actionable date-and-page locators. No underlying newspaper pages were acquired in that phase, so the locators were treated as leads rather than evidence of page contents.
General index years were missing for 1950, 1971, 1977, 1986, 1987, 1988 and 1989. Parts of 1983 through 1985 were only partially or category indexed.
The resulting finding is deliberately bounded: No contemporaneous Interlochen misconduct exposé was identified in the public archives searched. It is not a claim that no public reporting existed.
Source-by-proxy records
For Respondents 14, 31 and 32, Boho News has only the 2026 investigators’ descriptions of underlying personnel files. The files themselves were not obtained. Part 2 therefore attributes those chronologies to the investigators and does not present them as independently verified file contents.
Counting the approximately 70 accounts
The report counts conduct involving each adult separately when one person described multiple adults. Its main decade-organized section contains 61 separately counted firsthand accounts involving 45 respondents. The Respondent 1 section adds seven; the Epstein section adds two. Boho News therefore reconstructs the published total as 61 + 7 + 2 = 70 while retaining the report’s “approximately” qualifier.
This is not a count of 70 proven assaults, 70 victims, 70 survivors, or necessarily 70 different alumni.
Privacy and limits
Boho News preserves the report’s anonymization and does not attempt to identify unnamed alumni or respondents through biographical clues. Descriptions of alleged conduct are limited to the detail needed to explain the record.
Unknown facts remain unknown. The absence of a surviving file does not prove that no report was made, and a reported disclosure does not by itself establish that senior leadership received or understood it.