Inductivist

It's the 10th anniversary of 10,000 Year Explosion, a fantastic book by Gregory Cochran and the late Henry Harpending.  You MUST read it if you haven't.  I command you.

The book is chock full of provocative hypotheses about humans.  Let's test just one of them with General Social Survey (GSS) data.  The authors explain that hunter-gatherers routinely share resources, in part because it is difficult for mobile people to hold on to things.  It's also advantageous to cement positive ties and to make others feel obligated to you by being generous with them.

By contrast, farming selects for people who are good at holding on to possessions: seed grain, breeding stock, land, etc. A farmer who gives away everything will starve.

So Cochran and Harpending predict that ethnic groups lacking deep histories of agriculture will tend to be quicker to share with others. In the US context, this would be American Indians and blacks.

The GSS asked respondents: "People help other people in ways besides giving money, time, or other things to organized groups. Sometimes people help needy people directly. During the past 12 months, did you or members of your family or household give money, food, or clothing to any of the following types of people: The homeless or street-people."

Here are the percentages who answered yes:

Percent giving to the homeless in the past year

American Indian  55.4
Blacks  45.7
Irish  44.0
East Asian  41.7

Total sample 40.2

Mexican  39.4
Italian  37.7
French  37.0
Russian  36.8
German  34.2
English/Welsh  33.1
Scottish  30.4
Asian Indian  28.6
Norwegian  20.0

Even though American Indians and blacks are poor groups, they are more likely to give to the homeless.  Groups with long agricultural histories are less likely to give, even if they are wealthy like Asian Indians.

And if you argue that giving to the homeless is largely an urban thing, American Indians tend to live in rural areas but are at the top of the list, while urban Asian Indians are at the bottom.