September 1, 2012

A New La Griffe essay!

For the first time since 2008, La Griffe du Lion has posted a new essay. It's called Crime and the Hispanic Effect. He builds a regression model for predicting crime rates in cities and finds its largely driven by the percent black. Percent Hispanic doesn't much matter, one way or another.

134 comments:

Anonymous said...

Does he deal with the fact that Hispanic criminals often get counted as white?

Anonymous said...

La Griffe has to be Unz.

Anonymous said...

Nevermind. It's right there in the first paragraph i.e. the "Hispanic effect"...

Anonymous said...

FBI statics count hispanics as whites.

Anonymous said...

This would help explain why the elite are so intent on replacing the black underclass with an Hispanic one.

Simon in London said...

"La Griffe has to be Unz"

Yeah, my first thought on reading this was "Unz vindicated!" >:)

Silver said...

Does he deal with the fact that Hispanic criminals often get counted as white?

Haha, gotta love in the invincible ignorance bound to show up on a thread like this. READ THE ESSAY. It's all covered, I promise. (Read Jarod Taylor's recent review of Jonathan Haidt's latest book too. Funny how so much of the logic there applies to WNs themselves.)

I can understand (and, hell, even side with) whites who do dislike being race-replaced by latinos, but a lot of the viciousness towards latinos you see in WN cirles is based on some serious distortion and wild exaggeration of reality. Is a bit of proportion and moderation more likely to be fatal to your cause than helpful? Worth thinking about.

LemmusLemmus said...

It is possible that I am overlooking something very important about the methodology, but it appears that the author has never heard of the problem known in the social sciences since 1929 and called "the ecological fallacy" since 1958. In short, even if we knew the conditional statistical association between %Hispanic and crime rates at the city level to be causal, we still should not conclude anything about Hispanic involvement in crime. Likewise, if antisemitism were particularly high in cities with large Jewish populations, you wouldn't conclude that Jews are particularly antisemitic. As far as I can see, this is not a useful contribution.

Anonymous said...

I wonder if you could do something similar for GDP. Or IQ. Crime isn't the only problem with a flood of illegal third world hispanics in your country.

fnn said...

No one could 'splain the following in a previous thread.

Aren't criminal gangs a major driver for the incidence of violent crime?

As everyone knows, there are many large Hispanic gangs like MS-13. OTOH, white gangs have virtually disappeared. The Westies are long gone, the Chicago Gaylords retired from activity a generation ago and the various "mafias" have been decimated by the Feds and the remaining members have a median age of 70.

So who's producing all this non-hispanic white violent crime?

fnn said...

OTOH, Unz is great on foreign policy issues:
MEXICAN DRUG CARTELS FIGHT TURF BATTLES IN CHICAGO
(...)
Daily turf battles over drugs and distribution, he said, are turning parts of this Midwest city into a Mexican border town.

“One of the hardest jobs I’ve had in the past couple of years is to convince our law enforcement partners that we need an enforcement mentality as if we’re on the border,” Riley said.

Miles away, Riley says, Mexican cartels have a significant influence in Chicago’s gang violence problem.

“Let’s take the gloves off on that,” he said. “We know that the majority of the drugs here in Chicago, cartels are responsible for. We know that the majority of the murders are gang related. So it is very clear to see the connection and the role.”

As it stands now, at least three major Mexican cartels are battling over control of billions of dollars of marijuana, cocaine and — increasingly — heroin in this city. That includes the ultra-violent Zetas and the powerful Sinaloa cartel, run by its shadowy leader Joaquin “El Chapo” Guzman.

(...)

As if Chicago didn't have enough problems.

B322 said...

I found a similar curiosity using only homicides and reported state-level census data. (Census data sort-of breaks Hispanics out from other groups.)

Blacks and homicide are more strongly correlated than Blacks + Hispanics and homicide. If you apply a factor of 0.315* to the Hispanic fraction, then add it to the Black fraction, you get the maximum correlation -- 0.816.

(0.315 is just made up, but it is pretty consistent with the ratios presented by this table, with Black incarcerations rates over 10 and Hispanics of various nationalities varying between 0.68 and 5.06.)

Trouble is, I took stats so long ago that I don't know if that extra step has any statistical validity. I just chose it to maximize the correlation. Call it the NAM Differential Homicidality Factor. It's sort of like when political scientists multiply unemployment b X and inflation by Y and try to predict who will win the Presidential election.

Anyway, homicide data aren't everything, but at least using census data gets around FBI "Hispanics are White" shenanigans.

Anonymous said...

He builds a regression model for predicting crime rates in cities and finds its largely driven by the percent black. Percent Hispanic doesn't much matter, one way or another.

His model fails when applied globally. Here are the most murderous cities in the world:

http://urbantitan.com/10-most-dangerous-cities-in-the-world-in-2011/

Anonymous said...

"This would help explain why the elite are so intent on replacing the black underclass with an Hispanic one"

"Replacing"? We should be so lucky. Try: "supplementing".

Anonymous said...

This certainly jibes with my instincts. I never feel threatened going to the Home Depot, even though I need to run a gauntlet of about 100 or so Hispanic day laborers milling around the parking lot. And I've never seen a YouTube video of a Hispanic flash mob robbing a grocery store or beating up a gringo.

But while Hispanics don't seem to drive up levels of violent crime in neighborhoods, they do increase rates of property crime. If campesinos can't find work they still gotta eat so they often resort to residential burglary and auto theft.

Anonymous said...

Thanks for the heads-up! Can't wait to read it.

Johnny said...

I do wish we'd drop Hispanic in favor of more accurate descriptions. A blue eyed Argintine and a Mexican hill-person who doesn't speak Spanish are the same ethnicity?

Here in Miami the differences between the groups are obvious. Cubans are normal white people, mostly, but with a brash disposition. The Argentines embarrass us all. The we have the Puerto Ricans, Dominicans, Mexicans.. Out in Doral the Columbians and Venezuelans have their own little pudgy world, but are all mostly white and don't seem to cause problems.

It'd be best to use strict racial categories.

eah said...

"Unz vindicated!"

I guess this must be sort of a case of 'lies' ala 'lies, damned lies, and statistics' (I mean, I wouldn't say the conclusion is worse than 'damned lies').

Having lived in CA for almost thirty years, nothing is clearer than that the quality of an area -- from general prosperity to schools to whether or not part of it is a suburban slum (aka a 'barrio') to (yes) how much crime -- is directly related to the density of Hispanics.

But they are clearly preferred to Blacks in those regards. In that they also contribute to the demographic destruction of a majority white America they are not preferred at all.

stari_momak said...

Rather than Unz being La Griffe, I believe that the suggestion on a previous thread is correct, and the La Griffe is parodying Unz. In Unz's articles under his own name, he hasn't displayed anywhere near the statistical chops of La Griffe, and moreover most of the numbers work seems to have been done by Razib Khan. Further, the new La Griffe seems to be replete with some obvious errors and clues. He notes that you shouldn't believe the stuff you can see with your own eyes, like incarceration, he only deals with big cities (is the white population of cities representative?), he counts crimes versus criminals, but crimes are often really punished (see the US's increase in the incarceration rate), so it number of crimes doesn't really reflect the 'criminality' exactly or its cost to society (prisons cost money). He ignores possible systematic differences between cities in reporting of crimes (altercations between hipsters in Portland bars might well be more likely be brought to the attention of authorities more so than similar events in 'social clubs' in El Paso). Finally, assuming all this stuff doesn't matter, the 14% higher crime rate for 'Hispanics' versus whites La Griffe derives is not anywhere near the white-black differential, it is not nothing either. It means 14% more violence, 14% more costs of prisons, etc, as 'hispanics' replace whites.

Anonymous said...

Not only does FBI count hispanics as whites, my local police departments do the same.

Many police departments have online call reports, arrest records, etc. Verrrry interesting. The vast majority of crime victims seem to be black women.

Anonymous said...

This is just the ecological fallacy. Your methodology in your Mapinator article was much more sound.

His results depend on the assumption that blacks in South Central LA have the same propensity toward criminality as blacks in Beverly Hills. Not likely.

Charlesz Martel said...

We're not replacing an underclass, we're adding to it with another underclass.
Hispanics-real hispanics, not "White Hispanics"- aren't nearly as violent as Blacks, but they do lower the quality of life for Whites in other ways. I lived in Miami in the 70's, 80's, and 90's- ask any refugee from there. As they say:
"Do you know the best thing about Miami? It's so near the Untied States!"

Hispanics are not very civic-minded, institutionalize corruption, don't care about public property (trash everywhere), and smart enough to rig the system in their favor and play the AA or race card. They are not a benefit to a White, Euro-derived nation.
True Spanish, OTOH, are basically like most White Euros.

Anonymous said...


Yeah, my first thought on reading this was "Unz vindicated!" >:)


As I understand Unz's argument, it is that Hispanic (or perhaps Mestizo) IQ is converging on the IQ of the white population.

It is not clear to me that Unz is arguing that they are less crime-prone than blacks.

I think that the Union of Nationalist Zapatistas is a jab in Unz's eye, but I could be wrong.

medvedev said...

LOL. "UNZ convention" - hilarious!

Hapalong Cassidy said...

My guess is that Hispanic crime is more a factor of poverty and education than it is for Blacks. If you compare the crime rate of Hispanics to that of Whites with similar income and education levels, there probably isn't much difference. For Blacks, on the other hand, I think studies have shown that Blacks with college degrees have higher crime rates than White high-school dropouts.

B322 said...

Oh! I'm so disappointed. I was sure La Griffe would at least mention that US-born Hispanics are dramatically more crime-prone than foreign-born ones. That fact, established by Rubén, Roberto, Golnaz, and Charlie, is a huge admission against interest by the pro-immigration screw-the-gringos camp. (Unz doesn't care about it those data, of course.)

Dahlia said...

I've heard the rumors about who La Griffe is, but even if I hadn't, he seems like a pretty doctrinaire conservative unlike Mr. Unz. Can't wait to read it. My biggest disappointment in all my years of reading alt-Right blogs was reading Mr. Unz's article, way after the fact because I trusted the alt-right that it was garbage, and realizing he was right and that this information had been out there all along. Especially disheartening was learning that the "Hispanic paradox" not only existed, but that liberals had been obsessed with it.

Anonymous said...

IIRC, Unz has donated to Steve in the past (and perhaps in the present). Maybe Steve should mention this in the interests of full disclose.

Dahlia said...

"Madam President, honored guests, delegates to the national convention of the Union of Nationalist Zapatistas, ladies and gentlemen, I need scarcely tell you how proud I am to be invited to present this evening’s lecture at the UNZ national convention."

ROTFLOL!

Anonymous said...

La Griffe du Lion makes the same mistake that Ron Unz made in his Hispanic crime article a couple of years ago. Both disregard the fact that native-born populations are more crime-prone than immigrants, something which is true regardless of race. Because a large percentage of Hispanics are foreign-born, the Hispanic crime rate estimated by La Griffe is confounded by this immigration effect. This is pertinent because in the future, Hispanic-Americans are increasingly going to be native-born and thus more crime-prone.

Severn said...

What a typically bizarre Unz piece.

It has been conjectured that the contribution of Hispanics to violent crime is on the point of advancing to the standing enjoyed by blacks. This, however, is not confirmed by our evidence, at least in our largest cities. Whoever thinks or has thought this to be so has come to this determination from evidence not directly related to what is happening on the street, but rather from incarceration records, court appearances or sentencing data. When crimes rather than criminals are counted, and the Hispanic effect is appropriately removed, the data show that violent crime rates for Hispanics and non-Hispanic whites, though a bit higher for Hispanics, are in actual fact quite similar. As for blacks, their crime rate remains by any measure uniquely high.



You got that? Yes, if we foolishly look at things like "incarceration records, court appearances or sentencing data" then it sure looks like Hispanics are guilty of a lot of crime. But if we ignore mere criminals and look at crime then Hispanics don't look so bad. In fact if we ignore actual criminals and look instead at "crime" then the Hispanic crime rate looks a lot like the white crime rate - hardly a surprise since the official government stats lump white and Hispanic crime together.

If you look at the "Ten Most Wanted" fugitives in Texas, all but two are Hispanic and none are white. But Texas law enforcement, in its Orwellian commitment to thoughtcrime, claims that six of them are "white"!

Ten Most Wanted.

If Ernesto Alonso Garcia had won an Olympic Gold Medal or a Nobel Price in medicine he'd be touted as a an example of why America needs Hispanic immigrants. Since he's wanted for murder, he's a "white male"!

Anonymous said...

Of course, it was Euler who said on reading a new proof by Newton - " I recognise the lion by its claw!".
Hence, 'Griffe de lion'.

Lisa said...

My little city of Lubbock is already defeated by UNZ's Hispanic effect. This is easily seen by looking at the Lubbock County Jail roster that does use Hispanic to classify criminals. The Hispanic defeat is also seen in our schools' failing test scores.

josh said...

Hispanic crime is better organized than black crime. Less random and angry. Probably less likely to lead to arrest or even being reported by your illegal neighbor or victim. Find the white equivalent of the Nortenos and Suerenos and I might listen.

Peter Frost said...

La Griffe commits the same error that Ron Unz did in his earlier essay: he lumps foreign-born Hispanics in with U.S.-born Hispanics. Since the foreign-born typically have lower rates of criminality, this has the effect of offsetting the higher rates of the U.S.-born.

If you wish to know the long-term effects of immigration, it's better to exclude the foreign-born generation.

Why do the foreign-born have lower crime rates? In part, many of them realize they can be deported for committing a crime. The main reason, however, is that their behavior is more culturally constrained by religious beliefs and by community and family supervision. These cultural constraints dissolve as they assimilate and move into an environment, like that of the U.S, where the individual has much more freedom to act as he/she wishes.

Aaron in Israel said...

This seems like a clever idea. Is anyone qualified to evaluate the argument? I'm not.

One thing seems clear: The p-values and confidence intervals he presents are obviously wrong, because they're conditioned on his proposition, the one he highlights in red. How does one go from the confidence in the intermediate, "red" proposition to confidence levels in his final results?

And by the way, is "regression through the origin" even the proper method to use? I understand that the true regression line must pass through the origin, but does that justify requiring the estimated regression line to do so? I'm nowhere near qualified to answer these questions, only (maybe) to ask them.

Aaron in Israel said...

Re La Griffe's identity: I don't know who he is, but I know who he's not.

He's not Ron Unz. Besides the stylistic differences, why would Unz publish some things under his own name and others, no more controversial, under a pseudonym?

He's not someone with an educational background in statistics, probability, or math. His background is probably in engineering or the physical sciences. That's clear from his earlier articles, where the mathematical "language" that he uses is not that of mathematicians or statisticians, but rather (I think) that of engineers and physical scientists.

Anonymous said...

My biggest disappointment in all my years of reading alt-Right blogs was reading Mr. Unz's article, way after the fact because I trusted the alt-right that it was garbage, and realizing he was right


Which particular Unz article do you think "was right"? His article on how Hispanic crime is really low? That article was and is wrong. Hispanics commit violent crime at rates three times greater than do whites.

Tony said...

That may all be well and good, but how do you explain Stockton CA which is only 10% black and one of the top ten most violent cities in America.

Severn said...

You may have wondered to yourself - Why is he so anxious to discount all the data which we do have on Hispanic crime? Why is the bulk of his article given over to arguing in favor of ignoring all the data which we do have about the race and ethnicity of actual criminals?

Because it allows him to just wave his hand and make all sorts of things like this disappear.

http://www.ag.ca.gov/cjsc/publications/homicide/hm01/ar.pdf

Of 1,754 people arrested for homicide in California in 2001, 22.8% were white, 23.1% were black, and 47.4& were Hispanic.

That does not look very good for the "Hispanic crime looks just like white crime" position.

This "Griffe" character argues in exactly the same fashion as does Unz.

"First of all, let's start by dismissing all the data which contradicts my preferred conclusion. Then we "interpret" the remaining data in my own decidedly non-standard way and from a skewed perspective. If we do these things we discover that - mirabile dictu! - we arrive at precisely the destination I set out for."

Anonymous said...

I wonder how La Griffe du Lion would reconcile apparently grossly outlying data such as the racial composition of LAPD's extensive most-wanted list.

http://www.lapdonline.com/all_most_wanted

Statistical modeling is nice and all, if you've got no alternative, but there is a lot of actual data like this out there.

Severn said...

In 2009, 50.3% of all those arrested for homicide in California were Hispanic.

http://ag.ca.gov/cjsc/publications/homicide/hm09/preface.pdf

Steve Sailer said...

The LAPD Most Wanted list is of people the LAPD can't arrest, typically because they've fled back to the Old Country. Homeboys, in contrast, usually get caught eventually.

IHTG said...

...this is hilarious.

Masterful troll.

Anonymous said...

Assuming the LAPD list is full of people who fled back to their home countries, why are there no Canadians?

Lisa said...

@DirkY Crime is not the only thing is the most troubling. Simple things, like grocery shopping have to be calculated so I won't be in a sea of brown and Spanish conversations.

stari_momak said...

OT--I saw an all white roofing crew in a very expensive SoCal neighborhood today. A glimmer of hope?

Steve Sailer said...

"And I've never seen a YouTube video of a Hispanic flash mob robbing a grocery store or beating up a gringo."

Right. Not many Latino flash mobs, at least on English language YouTube.

Cyprian Korzeniowski said...

I can understand (and, hell, even side with) whites who do dislike being race-replaced by latinos, but a lot of the viciousness towards latinos you see in WN cirles is based on some serious distortion and wild exaggeration of reality. Is a bit of proportion and moderation more likely to be fatal to your cause than helpful? Worth thinking about.

White nationalists rarely bother with the latino poison. Even angry white power skinheads are far more likely to (rightly!) lament the pernicious presence of homeboys and jews than to even bother with the scantest mention of latinos. Latinos are a quasi-invisible serf class.

Steve has posted every now and then on almost-white-could-pass-as-Italians latino leaders. Why have none of them achieved any fame? I suspect it's because, given the media spotlight they would demand higher wages (either via citizenship or some guest worker program) for their clientela, thus making them more trouble than they're worth for your Mitt Romney types who don't want to pay a penny more for their lawn manicures.

Severn said...

a lot of the viciousness towards latinos you see in WN cirles


Can you cite any examples of this supposed "viciousness towards latinos"?

The people who come in for the heaviest criticism are the non-Hispanic enablers of the illegal Hispanic invasion who spin the facts to make Hispanics look good. People like "Griffe".

Severn said...

The LAPD Most Wanted list is of people the LAPD can't arrest, typically because they've fled back to the Old Country.


Sounds like Yet Another Good Argument Against Open Borders to me.

Anonymous said...

For all ages past 20 years, the rate of Hispanic deaths from drunk driving is almost twice that of whites:
www.nhtsa.gov/staticfiles/traffic_tech/TT398.pdf

I wonder how the number of violent deaths compares with deaths from car accidents caused by drunk driving?

This is La Griffe's weakest piece (who is, of course, not Unz - it is utterly ridiculous to suggest that he is). The assumption that Latinos in Philadelphia behave the same as when they are in Phoenix is very questionable. Disregard for the existing differences between foreign- and native-born is inexcusable.

josh said...

btw,

note that the author is "prodigy" as opposed to "mentor".

jody said...

well, la griffe is obviously wrong. i can do the statistics, but i don't have to, having followed this topic for 12 years now.

3 or 4 years ago me and steve took a look at the los angeles violent crime statistics over period of several recent years and found that "white" killers were killing people at about the same rate as "asian" killers were. and if you took out all the people who weren't european in the "white" category (armenians, iranians, various people from the middle east and central asia) and only counted europeans, then their rate fell BELOW the "asian" rate.

i'm talking murders, homicides, and various other ways people were taking out other people. not talking about vehicular homicides or suicides.

mexicans and various central americans were doing most of the killing in LA, then africans. everybody else was way behind, with europeans actually the least violent group, not "asians" (everybody from asia gets stuffed in this category but it's a large swath of humanity).

the same pattern seems to hold in most cities that i've seen. texas cities might be different. then again, an HBD informed perspective on the red state-blue state dichotomy suggests that texas should be a blue state already. but it's a solid red state (for now). so the longstanding mestizo population in texas is somewhat different from other mexicans. mexico itself is in the most violent period in it's history, a 6 year long drug war which has killed more people than were killed in the mexican-american war. mexicans routinely assault US border agents now, and mexican police ambushed american CIA operatives in mexico last week.

the way "hispanic" is used in american english it totally wrong now and almost meaningless. for instance i haven't talked about the various islanders, like puerto ricans, dominicans, haitians, and so forth. they aren't genetically related to mexicans. but these guys just straight up commit more crime than most groups. we don't need statistical analysis to know that.

i've posted before that in 2011, puerto rico had the most violent year in it's history. something like 1130 murders on an island with only 3 million people. and that we know from the US census, that puerto rico itself LOST 82,000 people net between 2000 and 2010. most of them, probably, moving to the US. certainly for jobs, but indisputibly in part to leave the violence of PR behind.

jody said...

definitely anybody who has experienced a mexican invasion of their town over the last 20 years can attest that when the mexicans show up in force, crime rates of all kinds go up. the violent crime rate does go up, as well as various property crimes. it doesn't skyrocket as if the town had suddenly descended into mogadishu, but there's not doubt at all what happens. denying what happens when half the town slowly turns into a mexican slum is intellectually dishonest.

5 out of 10 guys on the FBI's 10 most wanted list are mestizos from mexico, central america, or south america.

james bulger was on there for a long time but he was caught last year. osama bin laden was on there for years but for obvious reasons he's gone. he was replaced by eric toth, a possibly jewish pedophile. that leaves 3 regular euro american murderers, and 1 international jewish type who is a financial fraudster.

Anonymous said...


NYC murder/homicide arrestees
White 6.2%
Black 55.5%
Hispanic 33.8%


Your problem is that you are concentrating on the criminals, not the crime.

Every one of those whites are mass murderers, so their effect is so much greater than a few bumbling blacks or Hispanics.

Anonymous said...

La Griffe Fail. I think La Griffe gets it wrong right off the bat by only looking at the data for the 35 largest US cities. Just a guess, but maybe many old city downtown urban ghettos are much more likely to be heavily Black than Hispanic, and they already have a "history". I challenge La Griffe to look at the Central Valley of California, or even California as a whole.

He (and some posters on the thread) must not live in a place where they live with this stuff. Lies, damned lies, statistics, very confused people, and "why, you must all be racists, I never see bad stuff!" (And just because people see a lot of criminals doesn't mean they're saying, "why, they're all bad people and that's that." Maybe they just see a lot of criminals.) Get out some, ride your public transportation system around the "bad" parts of town. It's cheap. Maybe La Griffe can help out the San Jose police force. I hear they got it under control by flooding the zone.

Beecher Asbury said...

Though unrelated, if you look at La Griffe's argument and that professor from a previous post who argues we shouldn't be deporting immigrants for 'minor' offenses, there seems to be a disconnect. Given the professor was singling out Salvadorans, if hispanic crime rates were really that low, then why all the fuss about deporting them for criminal offenses? There must be a not too insignificant number of hispanic immigrants who are criminal. Otherwise, it wouldn't be an issue.

So whom are we to believe, La Griffe or the professor?

Aaron in Israel said...

There are two issues about the article: (1) Are the results a valid answer to the question they were addressing, i.e., the Hispanic effect on crime? (2) Was the question itself the right question?

Many of the criticisms in the comments confuse those two issues. They're ostensibly addressing issue (1) but are actually relevant only to issue (2).

All the comments about "who are Hispanics" - mestizo vs. other races, domestic vs. foreign born, etc. - are irrelevant to the question of the validity of La Griffe's analysis. He defined "Hispanic" according to self-reporting on census forms, but he could have defined it as "anyone who likes burritos" and the analysis itself would have been just as valid, as long as those people were counted correctly. The distinctions among Hispanics - immigrant/natural citizen, racial, etc. - are completely irrelevant to the quality of the analysis. That's the beauty of the method he's using.

The author does not assume that group crime rates are constant over different cities. He tests that hypothesis and claims to have established it empirically. He then assumes its truth in what follows. I think this is by far the biggest flaw in his analysis: The degree of confidence in that intermediate hypothesis is not reflected in his final confidence levels. This makes his final assertion practically meaningless, because the only valid thing he's left with is point estimates.

Other mistaken criticisms from the comments: "How do you explain [this exceptional case] then?" Answer: it's an exceptional case.

Paraphrasing: "By dismissing contrary data from the start, he just arrives at the destination he set out for." He didn't dismiss data, except for Honolulu. He analyzed a certain type of data, because there's no way to statistically analyze Everything In The World. That's a completely valid approach, and it's clear that he might have arrived at the opposite conclusion if the data had been different. It's also valid to ask, how does the conclusion fit with other evidence, including incarceration rates (answered by the author) and our proverbial lying eyes?

But the "lying eyes" objection cuts both ways. If the author should explain the discrepancy between his conclusion and your lying eyes, then you critics are equally obliged to explain that discrepancy. Saying "his analysis must be wrong then," or "lies, damned lies, and statistics," is no more acceptable than saying "your lying eyes must be wrong."

"[The hypothesis of constant crime rates by group] is demonstrably false." Then why don't you demonstrate that? Of course, any statistical model is "false," in that it doesn't explain 100 percent of the data. The author reports that r = 0.89, which means that his hypothesis explains about 80 percent of the variation in crime rates by city (excluding Honolulu). If he's mistaken in that reasoning, then where is his mistake?

Instead of arguing whether the article is right or wrong, I'm interested in another question, which I have no idea how to answer: Can the obvious error in the analysis regarding confidence levels be fixed? Is there any way to get any valid confidence levels at all for the final point estimates? Maybe if there were someone with lots of resources at his disposal, a millionaire maybe, who was interested in the UNZ presentation...

Anonymous said...

La Griffe had no problem with the young black man I was working with at the time, the one who learned two West African languages and French at home, moved to France deepened French and added German, then moved to America and learned English. The solidly capable engineer and mathematician. As for other blacks maybe he goes crazy and runs away when he sees them how would I know?

Funny happenstance wasn't it, I hadn't been back there for years until just a few days ago. Glad to see you're still writing.

A No Longer Young Man

jqhart said...

So does Mexico freely let African Americans immigrate?

Ex Submarine Officer said...

Hispanics need a couple of generations of trench warefare, tarawa hitting the beaches, Vietnam S&D patrols to weed out their bad boyz.

Then they will be just like the 95% gutless, wimpy, homo-lovin, diversity-lovin white boys that are overpopulating our nation.

Steve Sailer said...

"But the "lying eyes" objection cuts both ways"

Here's a lying eyes question: Where are all the Hispanic flash mob videos?

stari_momak said...

"Where are all the Hispanic flash mobs?"

Well, here's one, if you consider Xavier Cugat 'Hispanic'.

Silver said...

Because it allows him to just wave his hand and make all sorts of things like this disappear.

Severn, you ignore evidence that threatens your beliefs all the time.

Just from that California homicide arrests report 2009 you linked to we can see that gang violence was close to 50% of the known causes of homicide among males (and known causes accounted for some 70% of male homicides). The latino propensity to join gangs is obviously distorting the picture of their "true" degree of innate tendency towards violence.

When we look at female homicides, however, a different picture emerges. While black women were almost as overrepresented as black men (22% of all female homicide arrestees, 24% of all male arrestees), hispanic women were arrested at only 1.23 times the rate of white women. Hispanics, however, have a roughly one third greater proportion of their population between the ages of 15-34, the age group in which homicide is vastly more likely (75% of female homicides in California in 2009), so this statistic alone may explain the higher homicide rate among hispanic women.

As for WN viciousness towards hispanics, lol, just log onto any WN forum. You'll see all manner of crude insults flying and no end of bs about how horrible they are. As I said, whites are well within their rights to be upset about what is happening to them, but I don't think it justifies this degree of raw hatred, particularly not when it's based on outrageous distortions of the reality of hispanics' effect on the societies they live in.

Steve Sailer said...

That's an important question: are gangs going out of fashion? It could be. God, I hope so.

Anonymous said...

I hope that we will not wait for the next blogpost from La Griffe for 4 years, like this time.

However, the quality is there and it is nice to know that the bloggist is alive.

Now, waiting for the next Tool album... ;-/

Anonymous said...

I hope that we will not wait for the next blogpost from La Griffe for 4 years, like this time.

However, the quality is there and it is nice to know that the bloggist is alive.

Now, waiting for the next Tool album... ;-/

fnn said...

Here's a lying eyes question: Where are all the Hispanic flash mob videos?

In their own way, I think Hispanic bangers are too "contained" or "dignified" for that sort of thing.

Anonymous said...

"Well, here's one, if you consider Xavier Cugat 'Hispanic'."

How can one not, when that's so much more pronounceable than Xochitl Hinojosa!

Anonymous said...

I guess I feel about Silver the way Silver feels about WNs.

Technically I agree with you, Silver. I'm more on your side than anything else. But you're still emphasizing all the wrong things. Anti-Hispanic bile needs to be toned down a lot, because it's mainly Mexicans, PRs, and Guatemalans who are misbehaving, and then, not as much as the blacks. Anti-Mex and anti-PR bile should be toned down too, because they're just a red cape being flashed at us by the journalists and professors.

Acknowledged.

You come across as believing you're from a more sophisticated part of the Anglosphere, lecturing all the dirty Yanks on how clumsy and backward we are. This is not just an impression, it is the truth. Americans exalt Australians almost as much as we exalt Britons. We know your ed system hasn't been dumbed down as much because there isn't an Afro-Australian or Mexi-Australian lobby lobbying to protect their self-esteems from learning. So, yeah, we're envious.

See the problem? Every American who skips that article and asks about its contents was formed in a school system that objectively supported ignorance by objectively punishing success. You should be counting your lucky stars that your mind wasn't savaged by Prussified American intellectual suicidality. Tone down your contempt and people will start listening. The American right-wing needs the overseas right-wings. Every right-wing should be supporter and mentor--not dominatrix--to every other.

Lisa said...

We can't trust the crime statistics anyway, as proven in the recent Harlingen video

Anonymous said...

"This would help explain why the elite are so intent on replacing the black underclass with an Hispanic one."

They were useful as a weapon against white people but now they've won they want to get rid of them because of the insane levels of violence.

Anonymous said...

"Haha, gotta love in the invincible ignorance bound to show up on a thread like this. READ THE ESSAY. It's all covered"

Erm, Boas, Gould, Lewontin, Unz etc - there's absolutely no reason to take this on trust until it's been checked.

Isn't Griff de Lion the "smart fraction" guy? That in itself should be a red flag.

Lisa said...

My last post should be Hidalgo Co. The dishonest crime stats can likely apply to Harlingen anyway.

Anonymous said...

"Just from that California homicide arrests report 2009 you linked to we can see that gang violence was close to 50% of the known causes of homicide among males (and known causes accounted for some 70% of male homicides). The latino propensity to join gangs is obviously distorting the picture of their "true" degree of innate tendency towards violence. "

LOL. Let's just ignore 50% of 'Hispanic' perpetrated murders because, as we all know, gang-related crimes aren't a sign of criminality -- it's the culture. And let's also take out any cases of spousal murders -- it's the pasiĂłn latina, don't you know. That'll go away when they've been assimilated via 'English for the Children'. Also, let's concentrate on the 1/8 of murders committed by females, too much attention has been paid to male criminality.

Hey, if I can keep going, I'll get 'Hispanic' crime down to Amish levels.

Truth said...

Gangs are fraternal organizations for poor males, why would they go out of style?

Anonymous said...

more balkanization = more gangs

Anonymous said...

...it was Euler who said on reading a new proof by Newton - " I recognise the lion by its claw!"...

Bernoulli, not Euler.

[Leibniz and l'HĂ´pital also solved Bernoulli's challenge.]

Severn said...

Severn, you ignore evidence that threatens your beliefs all the time.


He pontificated pompously, before going on to NOT mention "all the times" when I have ignored evidence which "threatened my beliefs". He did, however, go on to say some remarkably stupid things.


Just from that California homicide arrests report 2009 you linked to we can see that gang violence was close to 50% of the known causes of homicide among males (and known causes accounted for some 70% of male homicides). The latino propensity to join gangs is obviously distorting the picture of their "true" degree of innate tendency towards violence.


That is an almost Unz-like degree of insanity. It's like saying "If we take away the mafia, Italian-Americans are not known for their criminal behavior. Clearly the mafia are distorting the "true" Italian-American crime picture".

Or "The black propensity to commit murder in the course of robbing liquor stores is obviously distorting the picture of their "true" degree of innate tendency towards violence".

You are doing EXACTLY the thing I pointed out was the common Unz tactic - trying to dismiss all the evidence which is not congenial to your preferred result. And trying lamely at that.

Severn said...

The author does not assume that group crime rates are constant over different cities. He tests that hypothesis and claims to have established it empirically.


Then his claim is a lie. Group crime rates are very emphatically not constant across America. White Southerners are notably more violent than are white New Englanders, for instance.

Stupid errors like this are why I think it's an open question whether the piece was intended as a parody.

TGGP said...

Four years ago I discussed a popular candidate for the real identity of La Griffe. The candidate was not Unz.

I liked the piece overall, but I was very disappointed that he said the claim about crime varying by racial/ethnic group and NOT city can be (and is) empirically verified, but it wasn't explicitly done later on.

Perspective said...

Silver Said: "hispanic women were arrested at only 1.23 times the rate of white women. Hispanics, however, have a roughly one third greater proportion of their population between the ages of 15-34, the age group in which homicide is vastly more likely (75% of female homicides in California in 2009), so this statistic alone may explain the higher homicide rate among hispanic women."
It would be interesting to see the homicides rates for Arab/Middle Eastern men and women if such a data set exists somewhere. I would be willing to bet that in macho cultures there will be a larger male-female homicide gap than among a more gender neutral culture.

Also, the Hispanic (Mestizo) propensity for joining gangs could have a cultural as well genetic explanation.

Anonymous said...

"We already knew that cities with large Hispanic populations and low black populations have low reported crime rates. San Diego is the safest large city in the USA, and El Paso is the safest poor city in the USA. Both have many Hispanics and almost no blacks."

Maybe it's the interaction between the two groups. They don't seem to bring out the best in one another. At least that's my impression from news reports of gang shootings.

Anonymous said...

That's an important question: are gangs going out of fashion? It could be. God, I hope so.

Yes, I am sure the males that would have joined gangs in the old days will take up crochet any day now. Or perhaps they will become rocket scientists.

Anonymous said...

I just assumed this article was intended as a satirical illustration of modern government "Garbage In, Garbage Out" statistics.

Ok, I see at the start of his article he says:

"I formulated this proposition:

In any given calendar year, the violent-crime rates of the racial and ethnic groups that inhabit a city are characteristic of the groups, but are independent of the particular city in which the groups reside."


I didn't read much further. This makes no sense to me at all. It's likely there's a large non-linear difference between cities that don't have a large enough 'hispanic' population to have a norteno-sureno problem (or have such a problem for whatever tactical reason) and those that do. Or just between those that just have a 'hispanic' populaton of a sufficent size to feed gang behavior and those that don't.

He's claiming group behavior is independent of the size of the group in a city, independent of the ratio of that group's size to other groups. Do we have reason to believe this?

He's also claiming that 'hispanics' in all cities in the US have the same characteristics. This is manifestly not true. Some cities will be dominated by, say, anti-Castro elite former Cubans (of largely european ancestry), while the 'hispanics' in some cities in California are likely near 100% from 'native' populations in Mexico. The 'hispanic' word/category is so broad it is near meaningless. Thus his meaningless results.

Maybe he should check with police departments of San Jose, California; Salinas, California; Portland, Oregon; and Quantico, Virginia. Perhaps set them straight!

Is this a real La Griffe essay or maybe an imposter?

Steve Sailer said...

La Griffe's shtick has always been extreme reductionism to the point of stylization. It makes for helpful models for seeing the world.

Silver said...

They don't give the demographics for the 75 counties, but assume it roughly reflects the US demographics of the 2000 census.

Totally unrealistic assumption.

I just counted and the population of the top 75 counties is about 112 million, with the top 20 counties containing 50% of that figure.

Only two of the top 20 counties reflected overall US demographics of 2000. Overall, only 31 of the top 75 counties did -- and that was by being generous and rounding up anything 65% or over.

Combine the two data sets to get an estimate of the racial propensity for producing violent felons.

Relative violent felony charge rate
White non-Hispanic - 0.4
Black - 3.3
Hispanic - 2.4


Naturally, then, these numbers are WAY off.


Silver said...

LOL. Let's just ignore 50% of 'Hispanic' perpetrated murders because, as we all know, gang-related crimes aren't a sign of criminality -- it's the culture. And let's also take out any cases of spousal murders -- it's the pasiĂłn latina, don't you know. That'll go away when they've been assimilated via 'English for the Children'. Also, let's concentrate on the 1/8 of murders committed by females, too much attention has been paid to male criminality.

Hey, if I can keep going, I'll get 'Hispanic' crime down to Amish levels.


Lol, that's very funny.

I didn't suggest "ignoring" anything. The point is to put the data into perspective. If you just want to scream "Hispanics 3 times more violent than whites, yes, THREE TIMES!" a la...well, I won't name any names...then go ahead. But if it's actual understanding you want to cultivate and disseminate then I'm afraid there's no other way but to "look behind the numbers."

Gangs have been a mainstay of US latino culture in large part as a reaction to the hostility latinos experienced. As hostility diminishes you should expect the tendency to join gangs to likewise diminish. That doesn't mean gangs will disappear overnight, only that they'll be running much more on cultural inertia than anything else and in time can be expected -- not at all unreasonably -- to effectively vanish/see numbers severely reduced. Also, gang bangers mostly kill other gang bangers, so there's in which you could see them doing God's own work (with respect to demography).

Looking at female homicide is helpful because it can be used to confirm a genetic racial propensity towards violence (or not). As we see with blacks, both the males and females are vastly more violently inclined than whites. With hispanics this isn't remotely so, which suggests to me that the racial/genetic contribution to elevated latino crime rates is rather small (or, we can hope, even non-existent).


Silver said...

Then his claim is a lie. Group crime rates are very emphatically not constant across America. White Southerners are notably more violent than are white New Englanders, for instance.

It depends on what you mean by "constant."

White homicide rates have typically varied between about 1 and 5 across time and place. Black homicide rates tend to vary from about 10 to 50. Notice something? There's no overlap. Blacks at their most peaceful are still much more violent that whites at their most violent. So it makes sense to speak of a race "constant," even if that constant can vary.

With hispanics, unlike with blacks, you do see substantial overlap, with hispanics at their most peaceful quite often being substantially less violent than whites at their most violent.

As for that this means or implies, that's for thinking people to interpret. But the data is what it is.

Steve Sailer said...

Silver,

Thanks for the female ratio -- most helpful.

Silver said...

You come across as believing you're from a more sophisticated part of the Anglosphere, lecturing all the dirty Yanks on how clumsy and backward we are. This is not just an impression, it is the truth. Americans exalt Australians almost as much as we exalt Britons. We know your ed system hasn't been dumbed down as much because there isn't an Afro-Australian or Mexi-Australian lobby lobbying to protect their self-esteems from learning. So, yeah, we're envious.

See the problem? Every American who skips that article and asks about its contents was formed in a school system that objectively supported ignorance by objectively punishing success. You should be counting your lucky stars that your mind wasn't savaged by Prussified American intellectual suicidality. Tone down your contempt and people will start listening. The American right-wing needs the overseas right-wings. Every right-wing should be supporter and mentor--not dominatrix--to every other.


The thing is, in my own short life I've gone from being convinced that "racism" is totally wrong to grudgingly admitting it may be right to being totally convinced some version of it is the best way forward (and not just for whites). The initiation of this process was a real baptism of fire. It shook me up for a long time, with the result that the "conversion" took much, much longer than necessary. My being contemptuous is just me hitting back at some of the cruder elements whose BS I had to work through on my own.

As someone who always looked up to Americans it was very disheartening to learn how deeply the anti-reality programming reaches in so many of you. And you're right, I do thank my lucky stars I was able to receive the education I did. Get this: I graduated high school in 1994 and don't recall a single lesson specifically about racism/anti-racism (and you can be sure that in those days I would have listened up). We read "To Kill a Mockingbird" in 10th grade, but never really explored the racial angle beyond the obvious common sense conclusion that prejudices can sometimes harm the innocent. Things have certainly changed since then.

Severn said...

It depends on what you mean by "constant."

White homicide rates have typically varied between about 1 and 5 across time and place. Black homicide rates tend to vary from about 10 to 50. Notice something? There's no overlap. Blacks at their most peaceful are still much more violent that whites at their most violent.


By a "constant" I mean a constant. If you don't know what the word means then look it up.

A) What you describe is still not a "constant".

B) What you say still contradicts what Griffe is claiming.



With hispanics, unlike with blacks, you do see substantial overlap, with hispanics at their most peaceful quite often being substantially less violent than whites at their most violent.


Golly! Could that possibly have anything to do with the fact that some "Hispanics" are whites, some "Hispanics" are blacks, and many other "Hispanics" are neither?

Severn said...

Homicide rate by country.

The ten worst.

Honduras
El Salvador
CĂ´te d'Ivoire
Jamaica
Venezuela
Belize
U.S. Virgin Islands
Guatemala
Saint Kitts and Nevis
Zambia

If a "genetic racial propensity towards violence" does exist, then all the evidence says that it's very high in Hispanics.

Next up - Silver will explain how the violence in El Salvador and Honduras is all Reagan's fault.

Severn said...

for WN viciousness towards hispanics, lol, just log onto any WN forum. You'll see all manner of crude insults flying and no end of bs about how horrible they are.



How terribly British and lefty of you. (These days "British" and "lefty" are almost the same thing) Your idea of "raw hatred" and "viciousness towards hispanics" is .... something somebody allegedly said on some "WN forum" which Hispanics don't even read - but somehow these words are responsible for Hispanic violence in the US!

You claim to be a former lefty. Let me suggest that you have not traveled nearly as far from your roots as you think you have.

Peter Frost said...

"Looking at female homicide is helpful because it can be used to confirm a genetic racial propensity towards violence (or not)."

Perhaps I'm stupid, but I'm really having trouble understanding the above argument. A genetic propensity towards violence (with or without respect to race) should involve an interaction with gender. Women, on average, are much less prone to violence.

Anonymous said...

Naturally, then, these numbers are WAY off.

Silver, do you mean way OVER or way UNDER? And, since you went through the numbers, what's your ballpark estimate of the net demographics of those 75 largest counties?

Anonymous said...

It turns out the DOJ in fact did publish demographic data on the 75 largest counties.

They didn't break out % non-Hispanic white, but they do provide Hispanic and black percentages. This numbers are for 1990, which is at the start
of the crime data collection period.

Demographics of the 75 largest counties
White non-Hispanic - NA
Black - 31%
Hispanic - 11%

Again, here are the DOJ figures on violent felony defendants in those counties:

Defendants
White non-Hispanic - 26%
Black - 41%
Hispanic - 30%

Now we can combine the data sets to estimate the racial propensity of blacks versus Hispanics for producing violent felons:

Relative violent felony charge rate
White non-Hispanic - NA
Black - 1.3
Hispanic - 2.7

Even if we accommodated for the rapid growth of the Hispanic population during the 12-year crime data collection period, say by doubling the % of Hispanics in those counties, their violent felony rate would still be higher than that of blacks.

Anonymous said...

"more balkanization = more gangs"

Actually on reflection the inverse of that does make sense if a lot of hispanic crime is a direct function of diversity and immigration.

Immigration, especially illegal, is disproportionately young males and they initially tend to concentrate in the same areas so you get a beachhead of young males who are a minority of the total population and gangs form on both sides as a kind of ethnic militia - so it may be a function of balkanization itself. If the process continues and the original population is gradually ethnically cleansed - partly as a result of the increased violence - and the area gradually becomes mono ethnic again then the impetus for those ethnic militias declines.

So from the average white person's point of view mass young male skewed immigration of poor mexicans means heightened diversity-driven levels of violent crime until the point where the white population has been ethnically cleansed - so lose-lose.

However from TPTB's point of view using poor mexicans to cleanse both white and black people is win-win because they think hispanics are less of an IQ threat to them than white people while at the same time being less violent than black people.

In reality i think they'll find the cartel types will turn out to be a bigger threat because they're more cohesive than white people and smarter than black people but time will tell.

So yes, hispanic crime going down a lot after the original black or white population has been cleansed might make sense after all. Most of my knowledge of those kind of environments is based on *before* the white population is fully cleansed but afterwards - maybe.

This doesn't apply in black areas because black crime is disproportionately much more random and impulsive so it's not really effected by the environment much.

.
"the claim about crime varying by racial/ethnic group and NOT city can be (and is) empirically verified, but it wasn't explicitly done later on."

The important bit of this is the pattern of proportions between different categories of crime. If you make the categories
- direct theft
- direct theft with violence
- fraud type theft
- male-male violence
- male-female violence
then different groups have their own signature based on their combination of IQ, base level of violent traits, levels of impulsivity and also culture.

Racial groupings are obviously broad brush - Chinese descended from people living in a densely populated fertile valley near Beijing for the last 2000 years will be different from Chinese descended from people who've been living up a remote mountain over the same time period - but the broad pattern is correct.

Silver said...

Frost,

Perhaps I'm stupid, but I'm really having trouble understanding the above argument. A genetic propensity towards violence (with or without respect to race) should involve an interaction with gender. Women, on average, are much less prone to violence.

Well, sure. Black women are much less prone to violence than black men. But black women are more prone to violence than white woman while this doesn't seem to be case with hispanic women. Hispanic women's homicide rate is only slightly greater than white women's (and possibly accounted for by age distribution differentials), while black women's is four times as great.

Silver, do you mean way OVER or way UNDER? And, since you went through the numbers, what's your ballpark estimate of the net demographics of those 75 largest counties?

Well, if I'm right, then he is underrating white criminality and overstating hispanic criminality.

As for the demographics, a VERY ballpark estimate would be 50% non-hispanic white. I only counted the counties that approximated 2000 census for the whole USA (ie 'yes' or 'no'), but I wish I took stock of the actual racial proportions, since I did it manually anyway and it wouldn't have required much more effort. Oh well, I can't be bothered doing it again just yet. I'd feel pretty safe saying 50% though, generally greater for the smaller counties than for the top 20 (which are usually 50% or less).



candid_observer said...


Now we can combine the data sets to estimate the racial propensity of blacks versus Hispanics for producing violent felons:

Relative violent felony charge rate
White non-Hispanic - NA
Black - 1.3
Hispanic - 2.7


Assuming you've correctly characterized the data from the DOJ, then: Wow. A little hard to square La Griffe's analysis with these apparent facts -- which seem far more directly derived from pretty hard data than La Griffe's results.

Severn said...

Crime in NYC.

Black New Yorkers are 13 times more likely to be murdered - or arrested for murder - than whites, an NYPD crime analysis shows.

Blacks and Hispanics dominated tallies of both suspects and victims, according to an NYPD racial breakdown of crimes requested by the Daily News.

Among murder arrests, blacks accounted for 64.9%, Hispanics 27.2%, whites 7.3% and Asians less than 1%.

The racial tallies bear little resemblance to the ethnic makeup of New York City: 34.8% white, 27.6% Hispanic and 23.7% black, 2006 Census data shows.



The high black crime rate should be news to nobody. But it seems that there are still people out there who are falling for the Unz schtick about Hispanic crime being very similar to white crime. And all the data says that his schtick is every bit as absurd as the liberal claims about black crime being the same as white crime.

Hispanics have a very high crime rate when compared to whites. There is an abundance of data to demonstrate this, except to those who are willfully blind to it.

Perspective said...

Has any one taken a look at this document: Homicide in California
http://oag.ca.gov/sites/all/files/pdfs/cjsc/publications/homicide/hm10/preface.pdf?

In 2010
48.2% of those arrested were Hispanic
25.3% Black
17.9% White
8.3% were classified as other

In 2010, California was approximately

38% Hispanic
6.6% Black
41% White
15% Asian/Other

Also, regarding the age and sex profile of those arrested, the report noted:

"A greater percentage of white arrestees were female than were Hispanic or black (20.8 vs. 8.7 and 13.3)."

"The largest proportion (39.9 percent) of white arrestees were aged 40 and over, while the largest proportion of Hispanic and black arrestees were aged 18-29 (62.7 and 56.5 percent, respectively)."

As I mentioned before, I wonder if being part of a macho/masculine culture suppresses the female crime rate.

Steve Sailer said...

I believe California Hispanics have a pretty low burglary arrest rate -- that would go along with the Culture of Honor explanation for the high homicide rate.

Steve Sailer said...

When I read through 2700 homicides in L.A. County over three years, I noticed that a lot of the white on white homicides were of the 50-year-old bachelor who lives with his mom shoots her then shoots himself variety.

Steve Sailer said...

It's also useful to adjust for age. Hispanics are younger on average, so, all else being equal, they'd be more crime prone. But one issue is that a fair fraction of the youngish Hispanic men are immigrants who came as young adults. Somebody who comes at age 25 isn't likely to join a youth street gang. The issue of most longrun concern is: what are their sons going to turn out like?

Perspective said...

"I believe California Hispanics have a pretty low burglary arrest rate -- that would go along with the Culture of Honor explanation for the high homicide rate."

In a another report on the same website, among property offenses, the gap between whites and Hispanics is narrowest for burglary and auto theft.
http://oag.ca.gov/sites/all/files/pdfs/cjsc/publications/candd/cd10/preface.pdf?

For violent crimes, the gap is largest for robbery and kidnapping, and narrowest for assault.

Silver said...

Frost,

But why would the gender difference be the same for all human populations?

That's an excellent question and I'm very tempted to say you "got me." But I may save myself yet, so hear me out.

Firstly, let me state that I reject the assumption of equality. Equality may be a fact of human existence or it may not, but we err to assume it axiomatically.

Imagine, though, a world in which inequality was assumed (by researchers). For whatever reason, the researchers felt certain that inequality simply had to be a fact of human existence and they then set about searching for evidence to substantiate their assumption. Imagine further, however, that this evidence was failing to turn up with the regularity they expected. They were counting on differing human heights, for example, but they found racial groups had the same heights. They were expecting to find significant differences in innate intelligence, but the results kept coming in equal. And so on. Might not our researchers eventually see fit to discard or downgrade their assumption of inequality? If they were reasonable men, they probably would.

In today's world, most reasonable investigators allow that a significant proportion of the observed differences between groups are innate. The real debate is to what extent observed differences are innate, and it's rarely clear. While it may be most reasonable to expect an innate crime rate differential between whites and hispanics, (a) it's not clear just how much of an innate difference should be expected, and (b) when evidence turns up not indicative of substantial innate difference, rather more suggestive of substantial equality, it should give us pause.

Arab women, for instance, tend to be less assertive than North American women. In contrast, Arab men tend to be more assertive than North American men. "Don't step on my blue suede shoes!!"

It's interesting how many hereditarians would insist this means Arabs are more assertive than whites, heh. (I disagree, though, that arab women are less assertive. That hasn't been my experience with them at all. They may be submissive to the males of their own family or wider ethno-cultural group, but they are quite assertive in their dealings with males and females other ethno-cultures. White females are the reverse.)

What I find strange in your reasoning is that you present female behavior as being more authentic than male behavior, i.e., men would behave like women were it not for discrimination.

Females are 50% of the species so any attempt to tease out innate racial differences would surely have to consider the ways in which females of various races differ as well as the ways in which males do. Otherwise, we should specify that we are referring to male differences, rather than general racial differences.

Anonymous said...

Then his claim is a lie. Group crime rates are very emphatically not constant across America. White Southerners are notably more violent than are white New Englanders, for instance.

Wonder why such facts are rarely pointed out....

Anonymous said...

Unz was very clear that his analysis didn't apply to the northeast. That stands to reason because the hispanics there are a very different breed, maybe 50% of them being some variety of black.

Unfortunately for you and Unz international data shows that mestizo and amerindian nations with very few blacks such as Honduras, Belize, Guatemala etc are ranked among the top ten homicidal nations on earth.

Northern Mexico which also has few if any blacks is also one of the most violent places around....

Anonymous said...

Silver's not touching that demographic data from the DOJ, is he. It obviously has some sort of problem with Hispanics.

Severn said...

That's an excellent question and I'm very tempted to say you "got me." But I may save myself yet, so hear me out

What follows is a lengthy attempt at obfuscation and rationalization.

most reasonable investigators allow that a significant proportion of the observed differences between groups are innate.


Except, apparently, in the case of Hispanics. In the case of Hispanics we see people who imagine themselves to be "reasonable investigators" tying themselves into knots in an effort to claim that the high Hispanic crime rate is not innate. No, high Hispanic crime is a result of the thoughts and words of the wicked "white nationalists"!


Those of us of a certain age can remember when it was widely accepted that the high black crime rate was also attributable to - you guessed it - white racism.

The left never has any new ideas.

Severn said...

See, this is one reason I find it difficult to trust anything you say. Unz was very clear that his analysis didn't apply to the northeast.


See, this is one reason I find it difficult to trust anything you say. What Unz analysis are you talking about? Are you referring to the Griffe analysis and now claiming that "Griffe" is actually Unz, or are you talking about some other previously unmentioned Unz analysis?


hispanics there are a very different breed, maybe 50% of them being some variety of black. You're surely aware of this, so for you to bang on about NYC is...it's just frustrating, ya know?


If I had not already linked to and mentioned the arrest data for homicide in other places, you'd have a point. But since I did so, you don't. As usual.

Your habit of seeing only what you want to see is frustrating, ya know?

Severn said...

The point of the NYC data on homicides is that it directly contradicts the Griffe piece. So does the California data though not as directly, since Griffe only looks at cities.

If spite of the claims that "we can't know what Hispanic crime looks like because the FBI does not track criminal offenders by their Hispanic status and therefore we have to engage in this obfuscationary analysis", there is a wealth of hard data out there on Hispanic crime from cities and states around the country.

And that hard data directly contradicts the claims by Unz and Griffe with respect to Hispanic crime.

Anonymous said...

@Silver
"See, this is one reason I find it difficult to trust anything you say."

The same applies to you.


"You don't like hispanics, do you? You didn't like them before you knew a thing about their crime rates and you like them even less now that you do, right?...But I don't believe distorting the truth about them is necessary or even helpful when it comes to such things as ending immigration and white rights/identity."

The truth is there have already been MILLIONS of casualties from mass immigration - millions of people stabbed, shot, raped or killed from the promotion of this policy and the cover-up of the consequences of this policy.

Regardless of the crime *rate* this policy has led to a *vast* amount of *extra* violent crime simply because of the importing of millions of *extra* young men then multiplied by the effects of diversity.

The people who promote this policy and the people who cover it up either directly or through attempted misdirection should be on trial for attempted genocide and accessories to murder.

Severn said...

We don't have national data on the ethnicity of criminals, but we do have it on the ethnicity of victims. And most homicide victims are are killed by members of their own group. The biggest victims of black crime are other black people. With that in mind, look at the national data on homicides.

During 1991--2007, homicide was ranked as one of the top four leading causes of death each year for persons aged 1--40 years living in the United States.


During 1999--2002, among persons aged 10--19 years, the homicide rate for blacks was estimated to be 17.8 per 100,000 population, a rate 10 times that of whites (1.8 per 100,000) and higher than the rates reported for American Indians/Alaska Natives (AI/ANs) (6.0 per 100,000), Asian/Pacific Islanders (A/PIs) (2.9 per 100,000), and Hispanics (8.0 per 100,000).



Note that this is age adjusted data. Some people claim that Hispanic crime only seems high because they are younger on average than the white population. But after adjusting for age, Hispanics are more than four time more likely to be homicide victims than are whites in the same age cohort.

This does not translate directly to "The rate at which Hispanics commit homicide is more than four times that of whites" because the perpetrator is not always from the same racial/ethnic group as the victim. Some of those murdered Hispanics were murdered by blacks,. Or, less likely, by whites or Asians. But the great majority were killed by their fellow Hispanics.

The rate at which members of a group are perpetrators of violent crime is closely and inextricably linked to the rate at which the members of that group are victims of violent crime. Just as we could deduce very high levels of black criminal behavior by looking at the victims of crime even in the absence of federal data breaking out the black/white disparity in felons, so we can make some pretty accurate assessments of national Hispanic criminality by looking at the frequency with which they are crime victims.

And when we do that it turns that, after adjusting for age, Hispanics are far more likely to engage in violent crime than are whites.

Anonymous said...

Silver
"Females are 50% of the species so any attempt to tease out innate racial differences would surely have to consider the ways in which females of various races differ as well as the ways in which males do. Otherwise, we should specify that we are referring to male differences, rather than general racial differences."

Seeing as men are responsible for the vast majority of violent i'd say that would be more a case of attempted misdirection.

Severn said...

That link on national homicide rates didn't work for some reason, so here it is again.

Severn said...

Here is a handy table from the CDC data I linked to before. It breaks out the homicide rate for (non-Hispanic) whites, blacks, native Americans, Asians, and Hispanics - and does so by age cohort.

As before these are murder victims. But as before the victims tell us who the criminals are. The reason why the homicide rate for black men aged 20 to 24 is a sky-high 109.4 per 100,000 is because other young black men are killing them.

With that in mind, the homicide rate of white, Asian and Hispanic males in the 20 to 24 year-old range are 6.3, 8.8, and 35.9 respectively.

Peter Frost said...

"Otherwise, we should specify that we are referring to male differences, rather than general racial differences."

Silver,

Not quite. The differences result from an interaction between race and gender. Actually, it's a three-way interaction between race, gender, and age.

Anonymous said...

To nobody's surprise, Ciudad Juarez I think usually qualifies as the most dangerous city on earth, for instance, '"the most violent zone in the world outside of declared war zones" in 2009' and "As of August 2009, Juárez's murder rate was the highest reported in the world...". Despite all this, Ciudad Juarez is booming. Maybe the violence is going down due to massive Army presence. Of course, El Paso, right across the border, is often held up as an example that large Hispanic cities can have very low crime rates. Go figure. Must be the drug gangs and the drug war. (El Paso is 15% white and 3% black. Of course Fort Bliss, second largest Army base in the US, next door, could have an effect.)

But... has there every been relative peace in Mexico? There's been a lot of wars, revolutions, it almost seems continuous... so many that we don't know their names. Most of the big wars seem to have been out-and-out race wars. What's the modern probability of real revolution in Mexico? Steve, have you ever sketched Mexican history for the more harried among us?

Silver said...

Unfortunately for you and Unz international data shows that mestizo and amerindian nations with very few blacks such as Honduras, Belize, Guatemala etc are ranked among the top ten homicidal nations on earth.

Northern Mexico which also has few if any blacks is also one of the most violent places around....


I'm very well aware of that. But there's a limit to how much can be inferred from it. Otherwise I could say the western United States during the 19th century was one of the most violent places around (white homicide rates in some parts exceeding 100 -- Brazil at its worst) and conclude on this basis that northern europeans are innate savages.

Severn,

In the case of Hispanics we see people who imagine themselves to be "reasonable investigators" tying themselves into knots in an effort to claim that the high Hispanic crime rate is not innate. No, high Hispanic crime is a result of the thoughts and words of the wicked "white nationalists"!

I said "hostility." I didn't claim it flowed only from whites. In many cases it was blacks dishing it out, and in deed as much as in word. And I haven't claimed that this explains everything (nor that it could, in theory, explain everything), only that it's an obvious mitigating factor. I certainly do challenge your claim that hispanic crime rates are innately "high." They may be innately higher than whites' but this doesn't necessarily equate to being "high."

Say the innate rate of white robbery is 75 and the innate rate of hispanic robbery is 225. The hispanic rate is three times higher. Three times! That's a lot. That is front page news, according to you. It's the stuff racial revolutions are made of. But what it means is that in a city composed of 50% whites and 50% hispanics the robbery rate will be 150. And while at a rate of 75, 94% of people who live to age 85 will never be robbed even once, at a rate of 150, 88% of people will never be robbed even once. In other words, the vast, vast majority of people experience no change in victimization, and the effect on whites would be even less, since robbery is often a crime of opportunity and since whites and hispanics self-segregate to an extent the opportunities would be less (provided there's no animus encouraging hispancis to target whites).

Silver said...

See, this is one reason I find it difficult to trust anything you say. What Unz analysis are you talking about? Are you referring to the Griffe analysis and now claiming that "Griffe" is actually Unz, or are you talking about some other previously unmentioned Unz analysis?

Uh, you brought up "the Unz shtick" so I assumed you were talking about Unz's general views, as expressed in the "HisPanic" article and ensuing debate.

The point of the NYC data on homicides is that it directly contradicts the Griffe piece. So does the California data though not as directly, since Griffe only looks at cities.

And the point of citing El Paso and south Texas is that in contradicts your claims of permanently high latino crime.

Silver's not touching that demographic data from the DOJ, is he. It obviously has some sort of problem with Hispanics.

I haven't looked at it yet. And in truth I haven't really wanted to because the first half of the 90s were the peak years for homicide across large swathes of America, which can only provide a seriously distorted picture of what one should expect from a hispanic presence. That doesn't require me to deny an innate role for usually elevated hispanic crime levels.

In view, hispanics have an innately higher "criminal excitability." It seems to take less for them to be prompted to engage in criminal activity. When they're prompted (and the mechanism isn't entirely clear) their crime levels can be astronomical. But when they're not, their crime levels can be very, very low (and for extended periods, decade-long judging by the last 10 years).

Silver said...

Frost,

Not quite. The differences result from an interaction between race and gender. Actually, it's a three-way interaction between race, gender, and age.

Okay, you're the expert. But I'm a political animal and a fact this murky -- a black cheque to claim whatever you want -- is politically useless.

Silver said...

I haven't looked at it yet. And in truth I haven't really wanted to because the first half of the 90s were the peak years for homicide across large swathes of America, which can only provide a seriously distorted picture of what one should expect from a hispanic presence.

Okay, now that I've looked at it I was wrong about the above, since it only covers 2002. I was relying on the poster's characterization of it as 1990-2002, but it's not.

The study actually breaks down felony defendants in 2002 by race for a select number of counties.

In the spreadsheet linked to below I've provided the percentage racial share of whites and hispanics (often an estimate) in 2002 and the percentage share of crimes of defendants of whites and hispanics. The last column attempts to calculate how much more (or, occasionally, less) hispanics are than whites. It divides (A) the hispanic share of defendants by the hispanic share of the population (B) the white share of defendants by the white share of the population, and then divides (A) by (B). For example, for Pima County, AZ it finds that hispanics are 1.73 times as criminal as whites.

File http://depositfiles.com/files/nosqoicpg

Hispanics in the most hispanic parts of the country appear to be about 1.5 to 2 times as criminal as whites. When you adjust for age you should expect this to decrease to about 1.15 to 1.5. That's more criminal, but it's not outrageously more criminal.

Severn said...

the point of citing El Paso and south Texas is that in contradicts your claims of permanently high latino crime.


In fact, it does not. Unless you have some actual hard data on crime in south Texas broken out by race/ethnicity which you somehow neglected to mention thus far?

Still waiting for you to notice the data on crime in the Hispanic countries of Latin America which I posted. They don't seem to have gotten your memo about how Hispanics have "very, very low .. innate crime levels"!

But then, your entire intellectual model is built around not seeing all the many, many things which contradict your preferred narrative.

"Don't look at Mexico City, or New York City! Don't look at San Salvador, or Bridgeport, or St Louis, or Tegucigalpa, or Los Angles, or Guatemala City. Don't notice that of the 50 most violent cities in the world, 12 are in Mexico, 14 in Brazil, five in Colombia, 45 in the Americas and 40 in Latin America, or that five of the the most violent cities in the world are in Mexico. No, intelligent and nuanced observers such as I, Silver, know that the correct way to evaluate the Hispanic propensity towards crime is to look at ... El Paso!"

Severn said...

Uh, you brought up "the Unz shtick" so I assumed you were talking about Unz's general views, as expressed in the "HisPanic" article and ensuing debate


I'm starting to wonder if you're another Unz sockpuppet. You certainly mimic the Unz style of argument perfectly.


I certainly do challenge your claim that hispanic crime rates are innately "high." They may be innately higher than whites' but this doesn't necessarily equate to being "high."


Yes, dammit, it does. "High" in this context means "higher than that of the existing Americans", which means "higher than that of white people". "High" means that we are importing more crime when we throw the borders open to Hispanic immigration.


Say the innate rate of white robbery is 75 and the innate rate of hispanic robbery is 225. The hispanic rate is three times higher. Three times! That's a lot. That is front page news, according to you.


According to me? It certainly seems to be news to YOU. Gotta love your fall-back position - "All right, fine, the Hispanic crime rate is three times higher than the white crime rate. But who cares?"

You're supposed to care, you dope, because you've been tying yourself into a pretzel all through this thread in a desperate attempt to show that the Hispanic crime rate is not higher than the white crime rate.

Steve Sailer said...

El Paso has been famous for a low murder rate for a long time. I read an article in Time in the 1970s that attributed it to lithium in the water supply.

Silver said...

"High" in this context means "higher than that of the existing Americans", which means "higher than that of white people".

That's one way to look at it. But I wouldn't characterize hispanic crime in the United States as "high" in any general sort of sense.

Gotta love your fall-back position - "All right, fine, the Hispanic crime rate is three times higher than the white crime rate. But who cares?"

I don't claim it's three times higher. I'd say that's a worst case estimate. While it's most certainly a cause for concern, I don't think it justifies the hysteria you're trying to generate.

You're supposed to care, you dope, because you've been tying yourself into a pretzel all through this thread in a desperate attempt to show that the Hispanic crime rate is not higher than the white crime rate.

That's news to me. I thought I was just trying to put it into a more reasonable context, in particular, one suggesting that it's not necessarily the case that it will forever be higher or forever higher by as much as it currently is.

In fact, it does not. Unless you have some actual hard data on crime in south Texas broken out by race/ethnicity which you somehow neglected to mention thus far?

The south Texas I have in mind are the border cities which are 90%+ latino.

Still waiting for you to notice the data on crime in the Hispanic countries of Latin America which I posted. They don't seem to have gotten your memo about how Hispanics have "very, very low .. innate crime levels"!

My claim is they have the potential for very low crime, not that they always realize that potential -- quite clearly they don't. The reason I haven't wanted to discuss it is the considerable social and developmental differences between the US and latin America. I can accept that a large part of those differences are innate in origin, but it's again a question of just how large a role those innate differences play, and I don't see any good reason to rush to conclusions here.

There are substantial differences between closely related countries there that should give you pause. Consider Chile. That's a mestizo nation if there was ever one. Homicide rate of some 3.5 for a number of consecutive years now. Are Chile's people really so wildly divergent from Colombia's that we must ascribe the mammoth homicide rate differential to the respective people's genetic qualities? What about Nicaragua and Panama (low 10s) and El Salvador and Honduras (50+) over the same time period? Are their populations really so drastically different?

Severn said...

As of mid 2004 - Among males age 25 to 29, 12.6% of blacks were in prison or jail, compared to 3.6% of Hispanics and about 1.7% of whites.

http://proxy.baremetal.com/csdp.org/research/pjim04.pdf

But Silver assured us that after adjusting for age, Hispanics are only 1.15 to 1.5 times more criminal than whites. Then how come they're more than twice as likely to be locked up? Must be "institutional racism" at work.

And the fact is that a lot of Hispanic crime goes unpunished. If we catch an illegal immigrant for some crime less serious than murder or rape, we usually just deport him. In some cases we then re-deport him after he comes back and commits more crime. Such was the case with Jose Lopez Madrigal. Before being arrested for rape in 2010 he had already been arrested and deported nine times, including three times in 1999 alone.

We give illegals such a good deal than even US citizens now claim to be illegals to try to evade jail time. The incarceration rates undoubtedly understate the degree of Hispanic criminal behavior in the US.

Severn said...

While it's most certainly a cause for concern, I don't think it justifies the hysteria you're trying to generate.


The emotive person here is you. The person citing facts and figures here is me.

You still argue like a liberal.


I can accept that a large part of those differences are innate in origin, but it's again a question of just how large a role those innate differences play, and I don't see any good reason to rush to conclusions here.


We are well on our way to transforming the US into a minority white country. The topic of Hispanic migration has been debated extensively for decades now. I'd hate to force to "rush to conclusions here" but you resemble a guy who, in May 1944, announced that he did not want to be rushed to a conclusion on the question of whether America should fight a war with Germany and Japan. You pose as somebody who just wants all the facts, but in reality you're stalling for time - time for more illegals to enter America and create "facts on the ground".


Just a little while ago you were arguing that Hispanics had a low innate tendency towards crime. Now you say that they have the "potential" to have a low innate tendency towards crime.

Of which people could that not be said?

Severn said...

Consider Chile. That's a mestizo nation if there was ever one.


You argue exactly like Ron Unz. Are you sure you're not him?

Chile is not "a mestizo nation if there was ever one".

In 2011, Chile had an estimated population of 17,500,000, of which approximately 9.1 million or 52.7% are of European descent, with mestizos estimated at 44%.[6] Other studies found a white majority measured at 64% to 90% of the Chilean population. Chile's various waves of immigrants consisted Spanish, Italians, Irish, French, Greeks, Germans, English, Scots, Croats, and Palestinian arrivals.

European and, to a lesser extent, Middle Eastern emigration to Chile, chiefly during the second half of the 19th century and throughout the twentieth, was the most important in Latin America after emigrations to the Atlantic Coast of the Southern Cone (that is, to Argentina and southern Brazil)

The Afro-Chilean population has always been tiny, reaching a high of 2,500 people during the colonial period; their current percentage of the population is less than 0.1%.[14] According to the 2002 Census, 4.6% of the Chilean population considered themselves indigenous.


But Columbia ...

According to the 2005 census by the DANE the population of Colombia was composed of these ethnic groups:[3]

58% Mestizo (European and Amerindian).
20% White (European).
14% Mulatto (European and Black/African).
4% Afro-Colombian.
3% Zambo (African and Amerindian).
1% Amerindian.

Anonymous said...

"El Paso has been famous for a low murder rate for a long time. I read an article in Time in the 1970s that attributed it to lithium in the water supply."

El Paso and Ciudad Juarez are essentially the same city with the Rio Grande in the middle. Yet El Paso is tranquilo and Ciudad Juarez is often the literal murder capitol of the world. If it was something in the water you'd expect it might work on both sides of the river... Combined population of the two cities is is about 2.1 million, Ciudad Juarez populaton is about 1.3 million. Something doesn't add up. Maybe the US Federal types really clamped down on El Pase back in WWII and never let up.

Okay, I just learned that the area played an exensive role in the Mexican revolution (with Ciudad Juarez being, as always particularly violent), that Fort Bliss (near El Paso was the base of Pershing's operations chasing Pancho Villa in Mexico before WWII (dare we call it an invasion?)... and that a large number of the Mexican upper class escaped the revolution to El Paso, presumably being grateful for safety in the US:.

A significant Mexican middle class population developed cultural, social, and educational institutions catering to their needs...

The Jesuits also retreated from Mexico to El Paso (they were a particular target).

There are (or were) 100,000 people at Fort Bliss, a "major Cold War base". I'm suspicious that military MPs, FBI agents alert for spying, and an important chunk of the hispanic population being the equivalent of the original Cubans in Florida, that is, ex-Mexican upper elite fleeing the "permanent revolution" and constantly on the alert for trouble, are what keeps El Paso tranquil. I'd be interested in what anyone on the ground had to say...

Severn said...

I seem to recall Steve describing some well-off black neighborhoods in So-Cal. The crime rates of such places is surely much lower than the black norm. But it would be foolish to conclude from this: "There, you see, blacks are not any more prone to violence than whites".

As I pointed out before, the rate of white violence varies within the US, and within Europe.

For ANY large group of people it will always be possible to find outliers on both ends of the crime bell curve. The important question is "Where is the center of the bell curve?"

And for Hispanics, the center of their bell curve for crime is located a long, long way from El Paso.