4 ms·
Your point is very valid considering the caste distribution of Indian Americans. However, The homicide rate of India proper is 2.9 cases per 100,000 population
by cuteboy19 4y ago
Your point is very valid considering the caste distribution of Indian Americans. However,
The homicide rate of India proper is 2.9 cases per 100,000 population. That of Asian Americans is about the same.
The rate for African Americans is around 18
- karmakurtisaani 4y agoI bet you could find a subpopulation of India that is responsible of most of the homicides in the country. I bet you'd also find a lot of poverty, systemic oppression and alienation from society in that group. Then you'd start to perhaps understand the issue more broadly than by the color of skin. However, if you are motivated by ... other things, I'm sure you can just jump to conclusions and tell us all the statistics you have about African Americans.
- coryrc 4y ago> I bet you could find It seems you base your beliefs on things you imagine instead of facts.
- inglor_cz 4y agoYeah, it read like an article of faith rather than an analysis. It simply must be so. There are many examples in history where it was the privileged class/race/ethnicity/group that was the most murderous one. Spartiates, Normans, elites of Renaissance Italy, even Nazis (see: The Night of Long Knives).
- karmakurtisaani 4y agoIn your examples the murdering was done in a systemic fashion, i.e., driven by the state. It is not relevant for the discussion here.
- inglor_cz 4y agoNot necessarily, e.g. the Normans had a well-deserved reputation of being extraordinarily violent on their own. Even Norman bishops were known to travel armed and get into fights. (And the state back then was far too weak to prevent such behavior even if it wanted to.) Also, gangs can be seen as proto-states, and a lot of today's criminal violence is gang-related. There isn't really that much difference between, say, drug cartels and the kind of violent packs of future nobility that raided and terrorized Early Middle Age villages in the post-Roman world.
- karmakurtisaani 4y agoI'm very sorry, but I have difficulties relating this post to the earlier discussion. I'm sure you are factually correct, but the examples start to be so far removed (and in some sense not terribly relevant even as edge cases) from the earlier context that I simply can't add anything without sounding terribly rude.
- karmakurtisaani 4y agoOk, here's a proof for you: Take the sub-population of only consisting of murderers in the country. You will not find a consistent pattern of well doing middle class people.