Turn out it's the definition of rational as it can be empirically supported. From further up in the thread:
There actually is proper data showing that this kind of thing can actually make a meaningful difference, and surely we're all evidence-driven people here, right?
A 2008 experiment researched teenagers' perception of epilepsy with respect to people-first language. Teenagers from a summer camp were divided into two groups. One group was asked questions using the term "people with epilepsy", and the other group was asked using the term "epileptics", with questions including "Do you think that people with epilepsy/epileptics have more difficulties at school?" and "Do you have prejudice toward people with epilepsy/epileptics?" The study showed that the teenagers had higher "stigma perception" on the Stigma Scale of Epilepsy when hearing the phrase "epileptics" as opposed to "people with epilepsy".
https://onlinelibrary.wiley.com/doi/10.1111/j.1528-1167.2008.01899.x
Not an exact parallel, but the point stands that these kinds of language patters can genuinely influence perception.
Because it leads to measurable difference in empathy response:
https://onlinelibrary.wiley.com/doi/10.1111/j.1528-1167.2008.01899.x
The human brain is weird and wired wrong for modern life. The best way to master our worst impulses is to try and do small things that nudge it in the right direction.
If you've ever done Cognitive Behavioral Therapy, it's a similar idea. Identify behaviours you can practice that help you think the way you want to think.