Mensch tracht, un Gott lacht

Wednesday, June 27, 2018

The Post "demolishes" a Trump immigration claim Trump never made


“If you get people asking the wrong questions, you don’t have to worry about the answers” --Hunter S. Thompson  

I see the Washington Post's "Wonkblog" has set out to "demolish" the Trump position that immigration is associated with crime. Right from the start, "demolition" seems a curious goal for a "wonk," by the name of Christopher Ingraham, whose interest is supposed to be about facts and figures. Yet Mr. Ingraham don't even begin with numbers. What instead is his purely factual starting point? It is Trump's "notion that immigrants here illegally commit more crime." This "notion," of course, is the Post's headline, not Trump's actual words. What is the source of the Post's claim about Trump's "notion"? It is the White House's "alarmist accounts" and "exaggerated [] threat[s]," including about MS-13. 

If you're like to accept the Post's framing of the question then you don't even need to hear the answer. 

Anyway, the Post wonk sets out to "demolish" Trump's supposed "notion" by positing immigrants commit far fewer crimes than native-born citizens. 

Pardon me, that is not quite right. The precise conclusion of the Cato study covered in the piece is that fewer immigrants were "convicted" of crimes in 2015 in Texas. Apparently researchers looked at data submitted by Texas prosecutors about their criminal cases. But do we know how many total victims there were? Do we know how many charges were successfully prosecuted against each criminal? Do we know how many victims each crime created? The wonkperson's reporting does not indicate. 

Obviously, innocent citizens have as much to fear from a single criminal who perpetrates multiple or repeated crimes than multiple offenders who perpetrate a single crime on a single occasion. Moreover, certain groups can be particularly victimized: Brussels Jews had their Purim festival canceled in 2016 because of security concerns. Such measures might help keep immigrant crime stats down, but where on the ledgers are these special costs entered? Without it, there is no way to "demolish" citizens' fears, or Trump's prescriptions for them. It seems "demolition" may need to wait. 

And what do we know of the number of immigrant crimes that are reported -- or underreported? I see this question has been vexing research of the issue for many years, confounding attempts to understand the issue. The MS-13 gang began as an outgrowth of underreported immigrant crime: Los Angeles immigrants from El Salvador in the 1980s formed the gang to protect themselves from fellow-immigrant predators there. And how have law enforcement attitudes changed since then? Just a New Year's celebration ago the world saw that a mass sexual assault could be covered up by every organ of the state. If that's progress, say the half of America that elected Trump, include us out. 

Despite the sympathy of the host country's leadership to immigrant communities, PolitiFact tells us our "overall understanding of immigrants and crime 'remains confused.'" We still don't know how often immigrants fail to report crime, or why. PolitiFact also tells us "incarceration rates are a poor way to measure links between immigrants and crime." 

The sappers may need to resolve this internecine feud before they get to work "demolishing" the President's claims. 

Back to the Post piece. Immigrants, whether documented or no, according to studies that apparently focus on the number of criminals rather than the number of crimes -- and which apparently do not account for underreporting -- are actually more law-abiding than native-born citizens. Is this meant to be any more surprising than the well-accepted assumption that airline passengers are more law-abiding than the general population? And yet to my knowledge the Wonkblog has never undertook to demolish policies for airline security or chastised "alarmist accounts" of terrorists' attempts -- successful and otherwise -- to hijack and blowup planes. To the contrary, it is firmly accepted that criminals who get on airplanes are punching above their weight. Why isn't it even considered remotely plausible, let alone accepted, that foreign criminals who cross our border are punching above their weight? On what purely scientific basis is immigrant crime deemed to be no more frightful or alarming than native-born crime? 

How do we check the fact-checkers' values? 

Perhaps it is plausible that, if we assume the immigrant is law-abiding, the immigrant will remain law-abiding: why make the arduous journey only to piss it away on a life of crime? But it is equally plausible that, if we assume the immigrant is, say, an MS-13 gang member at the time of entry, we might likewise presume the immigrant will remain a criminal menace once here. MS-13 produces many more victims, and with much greater brutality, than the average native-born criminal does. This would suggest the conclusion it may well be reasonable to conduct some sorting before allowing in an undifferentiated mass of people, even assuming they are on average a more law-abiding bunch than the current cohort. We have no less need to sort immigrants than we do airplane passengers. Assuming the average decency of a cohort of people does not excuse us from preventing the great risks that may be created by a small subset. 

So the most we can conclude so far from the Post piece is that an effective open-borders policy does not create a greater number of criminals, even if it tells us nothing about the total effects these criminals are having on Americans. 

And what about the immigrants' children? PolitiFact also tells us crime does spike among the children of immigrants. A state official concurs: 

""We're talking about third- and fourth-generation [immigrants]; these youngsters are born in Belgium, even their fathers and mothers are born in Belgium, and still they are open for these kind of messages. This is not normal -- in the U.S., the second generation was the President; here, the fourth generation is an IS fighter -- so that is really something we have to work on."" 

Do you object to my citing the Belgian Interior Minister? Why? Do you assume immigrants are unequal in their capacity to become good citizens? Do you assume that countries are unequal in their ability to assimilate immigrants? One of the greatest experimental scientists of Francis Bacon's time, William Harvey, said science was more than merely making observations -- "the vital factor of judgment [is] about what to observe and what to pay attention to." Are the Washington Post or other mainstream news sources providing reporting that meaningfully probes these basic assumptions underlying immigration? Or do you think they are leaving out something important? And do you think this omission is sending readers to other sources, including President Trump, to support these plausible and existentially relevant assumptions? 

1 comment:

Tom Van Dyke said...

"Pardon me, that is not quite right. The precise conclusion of the Cato study covered in the piece is that fewer immigrants were "convicted" of crimes in 2015 in Texas. Apparently researchers looked at data submitted by Texas prosecutors about their criminal cases. But do we know how many total victims there were? Do we know how many charges were successfully prosecuted against each criminal? Do we know how many victims each crime created? The wonkperson's reporting does not indicate. "

Your primary challenge here holds. The best crime statistics are based on the reports by the victims, not arrests or convictions. But how can a crime victim know whether the perp was an immigrant, and especially whether the perp was a legal or illegal immigrant!??

Trump simply cannot be refuted by the available statistics.