I think you've hit on a very interesting way of framing the issue. Perhaps it can be explained with (slightly) statistical language: - Taking the self-immolato

I think you've hit on a very interesting way of framing the issue. Perhaps it can be explained with (slightly) statistical language: - Taking the self-immolator's experiences and actions as sample sets, they are different from the 'stereotypical American' by a statistically significant margin. - This does not mean that the cause is that the self-immolator is categorically different from the 'stereotypical American' (because correlation does not necessarily imply causation), in fact, it suggests that the experiences and actions may be related.