It's a bit hard to judge since there are only a few crimes where measurement has been consistent and good, e.g., homicide and violent crime. And there you have to take into account the decline in lethality due to better medical care:
https://twitter.com/RAVerBruggen/status/754446756805509120
https://twitter.com/RAVerBruggen/status/756512858331082752.
I'd like to see the trend in assaults, which would be most relevant to free-range parenting, but haven't been able to find it anywhere.
That's true. However there are estimates that historically, during periods in peace and without severe economic hardship, the murder rates had settled at very low numbers in America (lower than today). In the data I linked it is 1/100,000 in 1800 which is lower than around 4 today, and around 50 in several cities. However, there are problems with the datasets as you mentioned, but some of those problems would make the historical homicide rate even lower than higher (i.e. better survival rate from medical care).
There also variable crime by region in America. Baltimore, St Louis, Detroit and New Orleans have homicide rates around 50. That's higher than any country in Africa or the Middle East.
1 - https://en.wikipedia.org/wiki/List_of_countries_by_intention...
2 - https://ourworldindata.org/wp-content/uploads/2013/03/homici...
3 - https://en.wikipedia.org/wiki/List_of_cities_by_murder_rate
Edit: to below: go click on the links to the datasets I provided you. Look at the datasets. Do you see historical murder rate in the USA around 1 - 5 in the USA, and that it is now over 50 in some parts.
>nor the crime rate in the USA historically compared to now
My thesis was that some parts of the USA have higher crime rates than historical (they do). That was my thesis. An additional thesis is also true: At some points in history the historical murder rate in the USA was lower than it is today.
You need to link to actual data. The data I linked to supports my thesis.
Click on this link and look for yourself. I honestly don't know what your problem is. Are you not clicking the links or being intentionally difficult or something?: https://ourworldindata.org/wp-content/uploads/2013/03/homici...
>nor the crime rate in the USA historically compared to now
My thesis was that some parts of the USA have higher crime rates than historical (they do). That was my thesis. An additional thesis is also true: At some points in history the historical murder rate in the USA was lower than it is today.
You need to link to actual data. The data I linked to supports my thesis.
Click on this link and look for yourself. I honestly don't know what your problem is. Are you not clicking the links or being intentionally difficult or something? https://ourworldindata.org/wp-content/uploads/2013/03/homici...
You are then comparing that to certain city numbers in the USA that are ~50 p100k now. That makes more sense as a thesis and I think I now understand what you are saying.
That said, I think it is an entirely incorrect way to interpret the data, as the aggregate numbers that have sub 4 p100k in the historical record do not break out p100k numbers per city. So you have no idea if St. Louis (for instance) is worse now than it was in the "ourworldindata" data set as it isn't broken out that way.
It is true though that during many years of comparable peace and without economic hardship the rate in the USA had gone down to below what it is today. For example, estimated at 1/100,000 in year 1800.
1) There are cities in the US that currently have intentional murder rates higher than aggregate medieval European ones.
2) There are points in history where the aggregate murder rate is better than the current one.
If I understand those 2 correctly, I'm no longer arguing that the data doesn't agree with that. That was a misunderstanding on my part about what you were suggesting.
Now what I'm saying is that it is largely not germaine to the article as:
A) in 1926 when the article makes its first sample point, the aggregate rate in the US at least, was worse than now and
B) specific city numbers in comparison to aggregate numbers are not revelatory as you have no idea what they were historically (at least from the citation).
The current aggregate number is clearly much lower than the current St. Louis number so if we assume a linear correlation in those numbers then the St. Louis number would be worse in 1926. If we aren't going to assume a linear correlation between the aggregate numbers and the city specific ones, I'm fine with that, but I'd like to know why as none of the data sets imply that city murder rates have changed.
I think we're close to being on the same page. About your objection, my argument actually does not really revolve around the idea that there is some correlation b/w city vs state rates.
I'm just looking at the statistics themselves. The fact is there are several cities in the USA that have higher murder rates than any country in Africa or the Middle East. So if you were randomly placed somewhere at random b/w several US cities and vs Africa or the Middle East, you would live in an area with less murders choosing a place at random in Africa or the Middle East.