Do you actually know anyone who's been to prison? What's your source for this?

I do know someone (I'm not going to disclose any details to protect his privacy). But I don't think he'd agree with you on that.

Well, it's seems to be on the increase:

https://www.statista.com/statistics/220920/number-of-state-p...

And given that there are ~ 1.8 million people in prison in the US, you have 1 in 15000 chances to be murdered in prison per year.

For context, your chances are 1 in 20000 in the general US population [1].

[1] From the reported 5 per 100,000 people murder rate, which includes the 0,5% of the population in prison (though the latter is such a small percentage that doesn't really change the general result).

That’s showing 120 murders across just sate prisons. “ 1,291,000 people were in state prison, 631,000 in local jails, 226,000 in federal prisons, 44,000 in youth correctional facilities, 42,000 in immigration detention camps, 22,000 in involuntary commitment, 11,000 in territorial prisons, 2,500 in Indian Country jails, and 1,300 in United States military prisons.[8]” https://en.wikipedia.org/wiki/United_States_incarceration_ra...

So it’s 1 in 10,000 per year aka a 1 in 500 shot of being murdered in a 20 year prison sentence which seems dangerous but still unlikely.

Averages like that, including for the US population, are meaningless. Your chances if you live in Englewood (South Side Chicago) are dramatically higher, while if you're in rural Utah, somewhat less. Same for prisons.

Only for prisons (and jails) you don't get to pick where you'll end up.

And I wouldn't expect prisons/jails to have the same (or comparable) variety in murder rates as there is between say rural Utah and South Side Chicago.

So the argument is kind of moot.

I don't know what you'd find compelling, then. You quoted two articles about jails. Jails are different from prisons.

A prison is a jail. While I would like to say deaths are less likely in say a city or county jail, the data is not differentiating. County "jails" are included in most prison data. One of the links is literally about "jails", so the pedantry is not useful.

>One of the links is literally about "jails", so the pedantry is not useful.

Both links are about jails, when the discussion was about prisons. Their pedantry was completely on point.

And while prisons and jails are similar enough that you can group data from them to get a view of the entire legally detained population, that doesn't make statements about one entirely transferable to the other. They're different institutions run by different people and housing different (though overlapping) populations.

So which case are you arguing?

1) Prisons and jails are horrible places, and inmates are not given enough protection against violence from other inmates, or the guards. Murder statistics are horrifying.

-- or --

2) There are no safe prisons, anywhere.

If it's (1) you'll get no argument from me.

That does not seem to be true based on what people who were in jails/prisons for non-violent offenders say.