Telephone Operators Didn't Become Machines in the 1890s

Chatbots erode personhood through repetitive interaction, machine-like communication patterns bleeding into authentic self-expression—or so the worry goes.

The studies are cited, the concern sounds credible, and people nod along. But the researchers doing this work are missing a control that would have immediately settled the matter—and we know this because a field already solved this exact problem a century ago.

In the 1890s and early 1900s, telephone switchboard operators faced nearly identical accusations. Critics claimed that hours spent connecting calls, repeating standardized phrases. Mediating human interaction through machinery would rewire their personalities—the operator would become mechanical herself, hollow, a conduit rather than a person.

What actually happened was far more mundane. Follow-up studies found that operators did develop distinctive communication patterns on the job—curt, efficient, and rhythmic. But these patterns vanished completely outside work. The same operator who was economical with words at the switchboard was verbose at dinner, sarcastic with friends, fully present in contexts where the job didn't demand otherwise. The role had adapted to the environment. The self had not been erased by the tool.

Today's chatbot researchers are conducting their observations without this temporal control. They document how people shift their language around AI—fewer contractions, different pacing, abbreviated emotional content—then conclude the shift represents personality erosion. But they're measuring the operator during her shift and calling the shift the self. The difference that might change this ending is straightforward to track. Watch whether someone who uses chatbots extensively maintains their full communicative range in private conversation, in unstructured writing. In moments where no system is listening. If they do, we're watching adaptation. If they don't, we're watching something new.