I'd be curious about whether that general category of crime has trended up or down over time- I remember reading that improved medical technology has turned a lot of stabbings/shootings that would have been fatal (i.e, homicides) fifty years ago into assaults instead.
Simon. Problem is that States all have their own sets of laws—and definitions of crimes. What’s aggravated here, is not somewhere else. Some jurisdictions define simple assaults vs assaults with “bodily contact”. Here there is no contact with an LEO in the line of duty that is not considered “aggravated” (by law). The difference? 7 years vs 14 years. About all you can say for certain is that adding “aggravated” to the front of the charge is bad. How bad? What State? Hence I’m quite sympathetic to criminals copping pleas to just about anything to reduce such charges. The stakes are just too high and charges often are ridiculously run up by DA’s wanting to produce quick plea bargains and close cases.
It's also the modern view - what country locks up large amounts of men for fist-fighting when both sides were relatively evenly matched and willing to escalate? Maybe an aggressor goes to jail for a couple nights if-and-only-if the loser presses charges or the place where the fight happened was entirely inappropriate, like a doctors office or a school board meeting or something.
The vast majority of people in prison, serving sentences of a year or longer, are in for assaults with weapons, or if "fist fighting" were using their fists on children, the elderly, women, etc..
Good stuff. I wonder what you think of the better arguments I've seen on incarceration causing crime? They either focus on first time offenders or shortening sentences for lower / medium risk offenders. I'll try to dig up the research later, but from what I remember reading there seems to be at least some quality research in this area. That said I've never seen good arguments of good data to support the idea that repeat offenders shouldn't be locked up.
Also, in case anyone is interested, I put together some graphics that compare homicide to incarceration rate across US states and globally. The scatter plots are much like the scatter in this article, but also includes states and most countries
Interesting what would be example of a crime committed by a first time offender for whom we believe doesn’t warrant (lengthy) imprisonment and/or presents a low risk of reoffending. What groups would you categorize in the low/medium category
From what I've read and seen in the data it's the reoffending that is really predictive and not so much the crime itself. With only a couple of exceptions, the various types of violent crimes all have the rate of rearrest (for any violent crime) post-release, and those who committed murder might even have a lower rate than others.
This makes a little more sense when you realize how random and spur-of-the-moment a lot of crime is. A lot of shootings and stabbings start as random fights or situations where one guy bumps into another's girlfriend. And the difference between a murder and an assault might just be an inch right or left in a bullet's trajectory.
So the main idea is how many times have you been arrested. A guy who was arrested for purse snatching 10 times probably isn't going to change, but a guy with a clean record who gets drunk and steals his ex-girlfriend's car maybe just made that one mistake.
The other thing I've read is that some justice systems have become more efficient and improved outcomes by making probation and parole less onerous, but I don't know as much about that. I guess the idea is that parole / probation is the worst of both worlds: it's shitty enough that it's hard to go straight but there's enough freedom that it's easy to get into trouble. So the upshot would be either keep people in longer or take them off probation / parole earlier, but don't keep them in this in-between limbo.
I thought the central salient fact of American crime writ large was the incredible criminality of the country's black population in relation to other racial/ethnic groups. Targeting the (relatively) small number of VERY criminal blacks in the major American cities would do much to lower the overall criminality of the USA, and would deliver welcome relief to those subject to the predation of those criminals. Alas, big city mayors seem to turn a blind eye to the rampant criminality in their cities and refuse to target the small number who make life so miserable for so many.
This is a very regional trend and is most true of populations of downwardly mobile American Descendants of Slavery. You don't really see the same trend in areas where most Black people are either domestic or international immigrants:
To back that up with some data, the homicide rate for Black people in Missouri has averaged of 60 per 100,000 from 2018-2023, around 10x the average for all Americans, while in Rhode Island, where most Black people are immigrants, it was only 7.7
In general the Black homicide victim rate is very high across the Midwest, Deep South and Appalachia, and although not what I'd call good, at least much much lower in the Northeast and Mountain West, which are places that either attract upwardly mobile migrants and immigrants.
So many, like the 3 or 4 that have a Black census category and then track that kind of data? Probably the UK, Canada and...?
But even that, is extremely regional, because the gap between Black and other races in the UK and Canada is pretty modest. From what I remember the homicide rate is like double the overall population's homicide rate, which is a very small number to begin with. But in the US it ranges from a low of double, to a high of maybe 20 times, depending on the region. That kind of variation is the very definition of regional, and it's generally not observed in the other census categories of the USA:
> From what I remember the homicide rate is like double the overall population's homicide rate
That's likely an underestimate, if you are aware of the situation in London. The black homicide victimisation rate is 4.22 higher for the UK [1] and 4.26 higher in Canada [2] compared to the general population (both numbers from 2021).
"Very regional" like in every major city (Providence, Rhode Island not among them) in the United States of America. List all American cities of say over 150,000 population in descending order of criminality (of all kinds, not just homicide). Then show the percentage of the population which is black in each of those cities. Then sit back and think about correlations, causes, and remedies.
Providence, Rhode Island has a population of 150,000. Everybody uses homicide to compare cities and states because it's the only crime that is reported and counted enough to be comparable.
“Compared with other highly developed nations, the United States has a much higher rate of serious crime (e.g., homicide). “
The interesting comparison here would be to take the violent crime rates of our major minority populations and compare those to the minority’s home countries. Once you take a look at that, then it’s an easy process to compare our crime rates—without minority inclusion—to the developed, Western, world. You’ll produce some rather startling findings I’m sure.
Yeap, but also Hispanic. Where I live, it’s majority Hispanic. Crime stat’s are not only elevated, but also of a particular “flavor”. Look to South America and you will see similarities in rate and type of crime.
I personally haven't come across this view expressed very much, but maybe I'm not in the same circles.
The concern I have is how victimless/trivial crimes either accelerate descent into crime or impede rehabilitation.
Maybe a 19 year-old arrested on drug possession later resorts to burglary after their criminal record makes it harder to get a job. Or a drug possession triggers a parole violation on the tale end of their sentence.
I found the per-homicide rate interesting to see, I liked how it helps hone down which parameters are too high, since the final prison population metric is too nebulous.
Most justice systems do take these kinds of things into account. I posted a reply to another commenter with more detail, but in general if you want to go "softer" on crime those are exactly the types of situations (parole and first-time offenders) with some evidence to support the idea that more punishment might make things worse.
Where I might push back is on the victimless / trivial crime part, mainly because there just aren't a lot of people in prison or even jail simply for victimless / trivial crimes. For example, the chart in today's post shows drug possession as 3.2% of the prison population. But the vast majority of those were people who were arrested for more serious crimes (like assault, robbery, carjacking), and pled guilty to drug possession in exchange for dropping the more serious charge.
I can speak to how inaccurate this post is, from personal experience.
I got a 7 year sentence for a nonviolent victimless crime. I have a master's degree and had a Certified Public Accountant license. I harmed no one.
I had no criminal history and had never been arrested or been in any trouble at all. But when prosecutors want to destroy your life, for political reasons, they have the power to do it.
The number of arrests is pretty useless as a measure of "harm caused". Here's some reasons as to why.
I was arrested four separate times for one single "crime". Prosecutors can easily get multiple arrests by adding on more counts or charges at later times. If prosecutors charge you, then add on more charges at later times, and you bond out each time, prosecutors can easily punish you financially by forcing you to pay additional bond each time, while also racking up your number of lifetime "arrests". That's what happened to me. If you looked at my place on that chart, I'd be in the "four" lifetime arrests category. But I only committed one "crime" (which I didn't even commit, but that's a separate topic). I could have easily been in the "one" lifetime arrest category, but the essence of what happened wouldn't be substantially different.
We've all watched multiple videos of someone being arrested simply for mouthing off to a cop, as well. Even if the charges later get dismissed for being excessive, that "arrest" remains. It doesn't mean you're an unrepentant monster.
Even when in prison, most people there weren't really bad people. About 15-20% of people were bad guys, ones who were just racist or evil or terrible people. The other 80-85% were railroaded unfairly, or they made a mistake. They weren't that different from the people on the outside. I know, because I've been there, unlike everyone else commenting and viewing this post. Most people in prison agreed with my 15-20% figure, as well (which, perhaps not coincidentally, is also close to the percentage of prisoners serving life sentences, which is about 15%).
Your figures are also misleading because most people will gloss over the difference between prison and jail. Both have hugely negative effects on people's lives. (In my experience, the conditions in county jail were better than in prison, but still pretty awful. County jail at least had A/C, whereas most prisons in Texas lack A/C. Many, many people die in prisons each year due to the lack of A/C.)
When you say "62.5% of all prisoners are in there for violent crimes", many people will probably misinterpret that as "every time someone is sentenced to a prison or jail term, there's a 62.5% chance it is for a violent crime". That is inaccurate. The real figure would be much lower-- per Prisonpolicy.org, when including county jails, only 47% in prison & jail are there for violent crimes. When we consider that nonviolent crimes get lesser sentences on average, so they spend less time in prison, that means that the percentage of people sentenced for violent crimes is even less-- perhaps as low as 30%. That means out of every 100 people who has their lives destroyed by being sent to jail or prison, perhaps only 30 of those were for violent crimes. And most of those nonviolent crimes are victimless crimes, as happened with me.
As to your supposed "dark figure of crime", arguing that prisoners committed other crimes they weren't incarcerated for-- sure. But that's heavily misleading, because so does everyone else!
Ever not reported some taxable income on your return? Like, selling an item to a friend for $50, or not reporting eBay sales, or not reporting cash tips? Nearly everyone has done this in their lifetime. Congrats-- you're a felon, you were just lucky enough to escape being prosecuted!
Ever opened or thrown away mail addressed to anyone else? Felony.
Ever owned six or more dildos in Texas? Felony.
Ever moved drugs out of their prescription container? Felony.
Ever smoked weed? Felony.
Ever looked at porn online? Some of those girls are undoubtedly underage, that's a serious felony.
Ever drank underage? Felony.
Drove while buzzed? Felony.
Every single American has committed tons of felonies in their life (check out the book "Three Felonies a Day" on Amazon). Most people are lucky enough to avoid prosecution for them. I wasn't. That doesn't make me a monster.
There are over 4,000 federal crimes, and thousands more state ones. How many can anyone name? 100, maybe? So how confident are you that you have never commited a crime, when you can only name 1% of crimes that are on the books?
The data also seem to indicate - although not directly - a high rate of recidivism, and that is what I have often heard. If that is the case, how does it compare with other countries?
Officer Harris did her part to put black men in prison for weed. And then kept them past their release dates to exploit their labor. A literal slaver. You voted for her anyway, you didn’t care.
I don't know what the proportion is but i had a female friend who was attacked by the neighborhood bully resulting in a broken leg and cracked ribs. The perp got a suspended sentence.
Wrote a bit on this a month ago, and probably will do a follow up. For the most part, the causes of crime are "drunk, high, or stupid." But most drunk, high, or stupid people don't commit crimes, so that just moves the question again. I take your meaning, but I wouldn't describe the US as having high law enforcement capacity. We are severely under policed.
The anti-mass incarceration policy group you referenced above have a section in that same report that claims prison sentences - long ones in particular - do little to deter future re-offending and might even increase the likelihood that someone will reoffend when they get released.
Is the research in this area one where the findings are mixed/inconclusive; or do you see it as clearly leaning in a particular direction?
It's a highly misleading section. Prisons can reduce crime through several mechanism, and deterrence isn't the main one. That section misleadingly portrays deterrence as the primary goal--it's not.
As I mention in the piece, prisons reduce crime mainly through incapacitation, not deterrence or rehabilitation. Criminals tend to be repeat offenders, and they are incapacitated from doing that while in prison. In other words: prisons reduce crime by preventing the crimes that these offenders would otherwise (had they not been incapacitated) commit over the duration of their imprisonment.
So, yes, longer prison sentences may do little to *deter* people from committing crime. The evidence is mixed on the deterrence effect. But they greatly reduce crime through incapacitation. That is not in dispute.
I saw an interview with a prison official a number of years ago concerning crime and imprisonment and deterrence. One thing struck me from this man as prescient. He remarked, ‘…we simply house these guys away from society until their testosterone levels decrease in their 50’s…’. That basically was his answer to the concept of deterrence and rehabilitation.
I'd like to know more about the "Aggravated or Simple Assault" category in violent crime.
Simple assault can include punching or even shoving.
I hold the old-fashioned view that men should be able to have a fist fight without the law getting involved.
Unfortunately, I'm not aware of any public data with even more fine-grained offense categorization than this.
I'd be curious about whether that general category of crime has trended up or down over time- I remember reading that improved medical technology has turned a lot of stabbings/shootings that would have been fatal (i.e, homicides) fifty years ago into assaults instead.
Simon. Problem is that States all have their own sets of laws—and definitions of crimes. What’s aggravated here, is not somewhere else. Some jurisdictions define simple assaults vs assaults with “bodily contact”. Here there is no contact with an LEO in the line of duty that is not considered “aggravated” (by law). The difference? 7 years vs 14 years. About all you can say for certain is that adding “aggravated” to the front of the charge is bad. How bad? What State? Hence I’m quite sympathetic to criminals copping pleas to just about anything to reduce such charges. The stakes are just too high and charges often are ridiculously run up by DA’s wanting to produce quick plea bargains and close cases.
It's also the modern view - what country locks up large amounts of men for fist-fighting when both sides were relatively evenly matched and willing to escalate? Maybe an aggressor goes to jail for a couple nights if-and-only-if the loser presses charges or the place where the fight happened was entirely inappropriate, like a doctors office or a school board meeting or something.
The vast majority of people in prison, serving sentences of a year or longer, are in for assaults with weapons, or if "fist fighting" were using their fists on children, the elderly, women, etc..
Simple assault can also be punching a stranger
Good stuff. I wonder what you think of the better arguments I've seen on incarceration causing crime? They either focus on first time offenders or shortening sentences for lower / medium risk offenders. I'll try to dig up the research later, but from what I remember reading there seems to be at least some quality research in this area. That said I've never seen good arguments of good data to support the idea that repeat offenders shouldn't be locked up.
Also, in case anyone is interested, I put together some graphics that compare homicide to incarceration rate across US states and globally. The scatter plots are much like the scatter in this article, but also includes states and most countries
https://theusaindata.pythonanywhere.com/murder_prison
Interesting what would be example of a crime committed by a first time offender for whom we believe doesn’t warrant (lengthy) imprisonment and/or presents a low risk of reoffending. What groups would you categorize in the low/medium category
From what I've read and seen in the data it's the reoffending that is really predictive and not so much the crime itself. With only a couple of exceptions, the various types of violent crimes all have the rate of rearrest (for any violent crime) post-release, and those who committed murder might even have a lower rate than others.
This makes a little more sense when you realize how random and spur-of-the-moment a lot of crime is. A lot of shootings and stabbings start as random fights or situations where one guy bumps into another's girlfriend. And the difference between a murder and an assault might just be an inch right or left in a bullet's trajectory.
So the main idea is how many times have you been arrested. A guy who was arrested for purse snatching 10 times probably isn't going to change, but a guy with a clean record who gets drunk and steals his ex-girlfriend's car maybe just made that one mistake.
The other thing I've read is that some justice systems have become more efficient and improved outcomes by making probation and parole less onerous, but I don't know as much about that. I guess the idea is that parole / probation is the worst of both worlds: it's shitty enough that it's hard to go straight but there's enough freedom that it's easy to get into trouble. So the upshot would be either keep people in longer or take them off probation / parole earlier, but don't keep them in this in-between limbo.
I thought the central salient fact of American crime writ large was the incredible criminality of the country's black population in relation to other racial/ethnic groups. Targeting the (relatively) small number of VERY criminal blacks in the major American cities would do much to lower the overall criminality of the USA, and would deliver welcome relief to those subject to the predation of those criminals. Alas, big city mayors seem to turn a blind eye to the rampant criminality in their cities and refuse to target the small number who make life so miserable for so many.
This is a very regional trend and is most true of populations of downwardly mobile American Descendants of Slavery. You don't really see the same trend in areas where most Black people are either domestic or international immigrants:
To back that up with some data, the homicide rate for Black people in Missouri has averaged of 60 per 100,000 from 2018-2023, around 10x the average for all Americans, while in Rhode Island, where most Black people are immigrants, it was only 7.7
In general the Black homicide victim rate is very high across the Midwest, Deep South and Appalachia, and although not what I'd call good, at least much much lower in the Northeast and Mountain West, which are places that either attract upwardly mobile migrants and immigrants.
Not regional at all, high black crime rates relative to other ethnic groups can be observed in so many countries...
So many, like the 3 or 4 that have a Black census category and then track that kind of data? Probably the UK, Canada and...?
But even that, is extremely regional, because the gap between Black and other races in the UK and Canada is pretty modest. From what I remember the homicide rate is like double the overall population's homicide rate, which is a very small number to begin with. But in the US it ranges from a low of double, to a high of maybe 20 times, depending on the region. That kind of variation is the very definition of regional, and it's generally not observed in the other census categories of the USA:
https://theusaindata.pythonanywhere.com/regional_gaps
> From what I remember the homicide rate is like double the overall population's homicide rate
That's likely an underestimate, if you are aware of the situation in London. The black homicide victimisation rate is 4.22 higher for the UK [1] and 4.26 higher in Canada [2] compared to the general population (both numbers from 2021).
1: https://www.ethnicity-facts-figures.service.gov.uk/crime-justice-and-the-law/crime-and-reoffending/victims-of-homicide/latest/
2: https://www.justice.gc.ca/eng/rp-pr/jr/obpccjs-spnsjpc/index.html
That is only because immigrants are filtered.
"Very regional" like in every major city (Providence, Rhode Island not among them) in the United States of America. List all American cities of say over 150,000 population in descending order of criminality (of all kinds, not just homicide). Then show the percentage of the population which is black in each of those cities. Then sit back and think about correlations, causes, and remedies.
Providence, Rhode Island has a population of 150,000. Everybody uses homicide to compare cities and states because it's the only crime that is reported and counted enough to be comparable.
I can show you the data by state level:
https://theusaindata.pythonanywhere.com/regional_gaps
It's pretty clear that whatever is happening in Chicago and St Louis is quite different than what is happening in NYC.
“Compared with other highly developed nations, the United States has a much higher rate of serious crime (e.g., homicide). “
The interesting comparison here would be to take the violent crime rates of our major minority populations and compare those to the minority’s home countries. Once you take a look at that, then it’s an easy process to compare our crime rates—without minority inclusion—to the developed, Western, world. You’ll produce some rather startling findings I’m sure.
It is ze blacks
Yeap, but also Hispanic. Where I live, it’s majority Hispanic. Crime stat’s are not only elevated, but also of a particular “flavor”. Look to South America and you will see similarities in rate and type of crime.
I personally haven't come across this view expressed very much, but maybe I'm not in the same circles.
The concern I have is how victimless/trivial crimes either accelerate descent into crime or impede rehabilitation.
Maybe a 19 year-old arrested on drug possession later resorts to burglary after their criminal record makes it harder to get a job. Or a drug possession triggers a parole violation on the tale end of their sentence.
I found the per-homicide rate interesting to see, I liked how it helps hone down which parameters are too high, since the final prison population metric is too nebulous.
Most justice systems do take these kinds of things into account. I posted a reply to another commenter with more detail, but in general if you want to go "softer" on crime those are exactly the types of situations (parole and first-time offenders) with some evidence to support the idea that more punishment might make things worse.
Where I might push back is on the victimless / trivial crime part, mainly because there just aren't a lot of people in prison or even jail simply for victimless / trivial crimes. For example, the chart in today's post shows drug possession as 3.2% of the prison population. But the vast majority of those were people who were arrested for more serious crimes (like assault, robbery, carjacking), and pled guilty to drug possession in exchange for dropping the more serious charge.
Yeah, don't punish illegal behavior or you might get more of it.
I can speak to how inaccurate this post is, from personal experience.
I got a 7 year sentence for a nonviolent victimless crime. I have a master's degree and had a Certified Public Accountant license. I harmed no one.
I had no criminal history and had never been arrested or been in any trouble at all. But when prosecutors want to destroy your life, for political reasons, they have the power to do it.
The number of arrests is pretty useless as a measure of "harm caused". Here's some reasons as to why.
I was arrested four separate times for one single "crime". Prosecutors can easily get multiple arrests by adding on more counts or charges at later times. If prosecutors charge you, then add on more charges at later times, and you bond out each time, prosecutors can easily punish you financially by forcing you to pay additional bond each time, while also racking up your number of lifetime "arrests". That's what happened to me. If you looked at my place on that chart, I'd be in the "four" lifetime arrests category. But I only committed one "crime" (which I didn't even commit, but that's a separate topic). I could have easily been in the "one" lifetime arrest category, but the essence of what happened wouldn't be substantially different.
We've all watched multiple videos of someone being arrested simply for mouthing off to a cop, as well. Even if the charges later get dismissed for being excessive, that "arrest" remains. It doesn't mean you're an unrepentant monster.
Even when in prison, most people there weren't really bad people. About 15-20% of people were bad guys, ones who were just racist or evil or terrible people. The other 80-85% were railroaded unfairly, or they made a mistake. They weren't that different from the people on the outside. I know, because I've been there, unlike everyone else commenting and viewing this post. Most people in prison agreed with my 15-20% figure, as well (which, perhaps not coincidentally, is also close to the percentage of prisoners serving life sentences, which is about 15%).
Your figures are also misleading because most people will gloss over the difference between prison and jail. Both have hugely negative effects on people's lives. (In my experience, the conditions in county jail were better than in prison, but still pretty awful. County jail at least had A/C, whereas most prisons in Texas lack A/C. Many, many people die in prisons each year due to the lack of A/C.)
When you say "62.5% of all prisoners are in there for violent crimes", many people will probably misinterpret that as "every time someone is sentenced to a prison or jail term, there's a 62.5% chance it is for a violent crime". That is inaccurate. The real figure would be much lower-- per Prisonpolicy.org, when including county jails, only 47% in prison & jail are there for violent crimes. When we consider that nonviolent crimes get lesser sentences on average, so they spend less time in prison, that means that the percentage of people sentenced for violent crimes is even less-- perhaps as low as 30%. That means out of every 100 people who has their lives destroyed by being sent to jail or prison, perhaps only 30 of those were for violent crimes. And most of those nonviolent crimes are victimless crimes, as happened with me.
As to your supposed "dark figure of crime", arguing that prisoners committed other crimes they weren't incarcerated for-- sure. But that's heavily misleading, because so does everyone else!
Ever not reported some taxable income on your return? Like, selling an item to a friend for $50, or not reporting eBay sales, or not reporting cash tips? Nearly everyone has done this in their lifetime. Congrats-- you're a felon, you were just lucky enough to escape being prosecuted!
Ever opened or thrown away mail addressed to anyone else? Felony.
Ever owned six or more dildos in Texas? Felony.
Ever moved drugs out of their prescription container? Felony.
Ever smoked weed? Felony.
Ever looked at porn online? Some of those girls are undoubtedly underage, that's a serious felony.
Ever drank underage? Felony.
Drove while buzzed? Felony.
Every single American has committed tons of felonies in their life (check out the book "Three Felonies a Day" on Amazon). Most people are lucky enough to avoid prosecution for them. I wasn't. That doesn't make me a monster.
There are over 4,000 federal crimes, and thousands more state ones. How many can anyone name? 100, maybe? So how confident are you that you have never commited a crime, when you can only name 1% of crimes that are on the books?
The data also seem to indicate - although not directly - a high rate of recidivism, and that is what I have often heard. If that is the case, how does it compare with other countries?
You’re clearly a major lying leftoid shit fucker. We all know the reason: niggers.
Try to be honest at least next time.
US crime rates by demographic category are comparable to the rest of the world.
Officer Harris did her part to put black men in prison for weed. And then kept them past their release dates to exploit their labor. A literal slaver. You voted for her anyway, you didn’t care.
I don't know what the proportion is but i had a female friend who was attacked by the neighborhood bully resulting in a broken leg and cracked ribs. The perp got a suspended sentence.
Breakthrough on the cushion: ✓
Breakthrough when the alarm goes off and everything’s already on fire: ❌
Here’s the difference (and how to fix it): https://open.substack.com/pub/nwlehmann90/p/so-youre-a-guru-now-but-you-still
“People enter and exit prisons every day, and those with shorter sentences leave *quicker*.”
Should be “more quickly”.
This is what everyone should know about Obama’s legacy (and subscribe to me)
https://open.substack.com/pub/sergemil/p/how-obama-islamized-america-and-the
Wrote a bit on this a month ago, and probably will do a follow up. For the most part, the causes of crime are "drunk, high, or stupid." But most drunk, high, or stupid people don't commit crimes, so that just moves the question again. I take your meaning, but I wouldn't describe the US as having high law enforcement capacity. We are severely under policed.
https://sabrdance.substack.com/p/life-in-prison
The anti-mass incarceration policy group you referenced above have a section in that same report that claims prison sentences - long ones in particular - do little to deter future re-offending and might even increase the likelihood that someone will reoffend when they get released.
Is the research in this area one where the findings are mixed/inconclusive; or do you see it as clearly leaning in a particular direction?
It's a highly misleading section. Prisons can reduce crime through several mechanism, and deterrence isn't the main one. That section misleadingly portrays deterrence as the primary goal--it's not.
As I mention in the piece, prisons reduce crime mainly through incapacitation, not deterrence or rehabilitation. Criminals tend to be repeat offenders, and they are incapacitated from doing that while in prison. In other words: prisons reduce crime by preventing the crimes that these offenders would otherwise (had they not been incapacitated) commit over the duration of their imprisonment.
So, yes, longer prison sentences may do little to *deter* people from committing crime. The evidence is mixed on the deterrence effect. But they greatly reduce crime through incapacitation. That is not in dispute.
I saw an interview with a prison official a number of years ago concerning crime and imprisonment and deterrence. One thing struck me from this man as prescient. He remarked, ‘…we simply house these guys away from society until their testosterone levels decrease in their 50’s…’. That basically was his answer to the concept of deterrence and rehabilitation.