I know of conflicts that never reach formal grievance procedures: people simply lack the emotional, financial or institutional resources to risk retaliation. A culture of silence isn’t just a series of individual choices. It’s a product of unequal power.

Universities often reproduce rigid hierarchies, sometimes even authoritarian practices. Having experienced academia in both Russia and the US, I see major differences — but also a shared reluctance to challenge those with greater institutional power.

While some universities make headlines for Nobel Prize laureates, Cornell is making international headlines for the Jane Doe case. I spoke with @novayagazeta.bsky.social Europe about why this scandal points to problems far beyond one university.

Thank you!

If you want to know more come at APSA's “Hearing Politics: Audio Data, Political Behavior, and Measurement beyond Text” Thursday 2–3:30 Hynes 105

The broader lesson is about more than Russia: as political science uses more audio-as-data, survey paradata, and ML-based measures, prediction and measurement are different problems.

Once we audit those channels, the acoustic signal is real - but we don’t find convincing evidence that it measures concealed dissent. A much simpler measure does better: how long someone takes on the sensitive item relative to their own pace on ordinary survey questions.

An acoustic model predicts respondents’ stated war positions surprisingly well out of sample. Tempting conclusion: people concealing opposition sound different. But audio contains words, speakers, interviewers, segmentation, and recording conditions—not just “tone of voice.”