← Back to archive

Game: Police brutality, find the fallacies.

Historical archive only. New interaction is disabled.

Original Question

(This was in response to George Floyds/ Derek Chauvin verdict.)


P 1: Shares a Fox News meme that says police killed a total of 52 black persons, 3 unarmed, and 109 white persons, 5 unarmed so far in 2021. 


P 2: So Fox News made a cherry piking fallacy of a cherry piked tiny sample size of only 3 months and they also did a statical fallacy by not doing per capita.
Here is a large sample size, not a random cherry piked 3 months. This is from 2015 to 2021 which is the last 7 years to get a larger picture of our systems and its per capita. You need per capita to find out if there is actually any discrepancies. No unarmed person should be killed by state sanctioned power, and that power should be held accountable like the rest of us, not just for justice but to stop the potential abuse of power.


P 3:  So, if a 110lb 5’2” female police office is attacked by a 300lb 6’4” male you do not think she should have the right to use all force necessary to protect herself and life?

Comments on Question

Oh, and also in response to the Daunte Wright shooting as well. 

Dr. Bo has hit on the key problem with this line of reasoning, especially as seen in the media these days: "lying with statistics." I think it was Mark Twain who said there are three kinds of untruths: lies, damned lies and statistics. That's worth remembering when considering any argument made with statistics, especially in the context of a politicized topic.


"Lying" may be too strong of a word, though. To be charitable, I think it's more an echo-chamber of people experiencing extreme confirmation bias. It's not necessarily cherry picking. It's that disconfirming data doesn't even make it through the narrative filters.

Answers

5

Biased sample fallacy?


They are working on the audience not knowing that black people do not make up 33% of the US population, which it would need to be for the same to not be biased.  There may be two white people killed by the police for every black person, but white people here are the overwhelming majority of the population. 


argument by fast talking  also seems to apply, the arguer is relying upon the people not having done the research to refute their implied claim.

The question is confusing, but it sounds to me like the OP is focusing on the relative statistics: X number of black people killed by police compared to Y number of white people killed...BUT that's misleading because there are far more white people than black people in the U.S.


If that's the crux of the question, then check out the Base Rate Fallacy.

These aren't premises, so we can't treat this like an argument. However, each "premise" can be an argument in itself. I will address them individually.



Shares a Fox News meme that says police killed a total of 52 black persons, 3 unarmed, and 109 white persons, 5 unarmed so far in 2021. 



There is not enough data here to determine that this isn't about race, if that was Fox's goal. It is not about raw numbers; but percentage/per capita. But there is more to this. I won't rehash this as I have written about it extensively in my book Uncomfortable Ideas (touched on below). Bottom line, most conclusions drawn from this would be a non sequitur .



So Fox News made a cherry piking fallacy of a cherry piked tiny sample size of only 3 months and they also did a statical [sic] fallacy by not doing per capita.



I wouldn't call cherry picking on this. The data is actually pretty representative. It is just "lying with statistics" or using actual data to spin a misleading narrative.



Here is a large sample size, not a random cherry piked 3 months.



One can argue this is still cherry picked, if we argue the other data was cherry picked. To avoid accusations of cherry picking, all data should be presented, or at least a summary of it.



You need per capita to find out if there is actually any discrepancies.



But discrepancies doesn't uncover the problem. It is not just the racial makeup of the whole population, but the population that is likely to interact with the police. But it gets more complicated than that—do the police interact with some races more than others just because of their race, or do some races break the law more warranting more interaction with the police, or a combination?



No unarmed person should be killed by state sanctioned power, and that power should be held accountable like the rest of us, not just for justice but to stop the potential abuse of power.



This is an ideological position that hasn't been supported.



So, if a 110lb 5’2” female police office is attacked by a 300lb 6’4” male you do not think she should have the right to use all force necessary to protect herself and life?



I think this is a legitimate question to follow up the ideological position stated.


 

There's a reason that schools offer semester-long courses in statistics, and that reason is this: statistics are complicated and easy to misunderstand. So, I agree that the sample is too small (only 3 months). Furthermore, after only 3 months, many police departments have not completed their reports, which often take quite a bit of time. I also agree that we need to consider the proportions of the population. E.g., the analysis cited above has validity only if black and white people are equal in the population. We all know that's untrue. We'd also need to know much more about the specific incidents; simply knowing the numbers is a starting point for the argument, not its conclusion. 

I'd like to submit an answer to the "spot the fallacy" game to which you are alluding: oversimplified cause fallacy.


P1. X can contribute to a Black civilian being shot and killed by police.
P2. Many recent cases of police shootings are ultimately about X.
C. Therefore, this recent spate of Black civilians being shot by police can be stopped by solving for X.


This fallacy and its cousin, causal reductionism, are typical of the arguments we are hearing these days.


For the Left, X is "systemic racism." Solving for it involves anti-racism training at least and defunding/dismantling the police at most.


For the Right, X is "fatherlessness." That is, the high rate of single-mother households in Black communities.  Solving for it involves Black leaders somehow persuading members of Black communities to get married before having children.


The point is that everyone tries to reduce the problem to a single, simple cause. But like all issues, this one is quite complex. Even in your writeup, there are multiple issues to think about:



  • Can statistics really tell us anything at all about the underlying causes of this problem when the individual cases are so varied? For instance, does an officer mistaking her pistol for a Taser and shooting a civilian have anything in common with an officer shooting someone who was about to stab another civilian? Do either of those cases have anything to do with an officer kneeling on a civilian to the point where he suffocated him to death?

  • Is it true that "no unarmed person should be killed" by the police? What if, as a recent example demonstrates, that unarmed person is about to get into a car and drive in a reckless manner that is likely to endanger the lives of others? What if the person is armed, but only with a knife?

  • Is "power" really a thing that can be held responsible? If so, how would that look? How would it be done fairly with regard to the people involved? Who would be deemed "power" and why/why not? For instance, should the governor of a state be held accountable if a state police officer kills an unarmed Black civilian? What if the governor is a Black politician who was elected because of his desire to reduce police shootings of Black civilians?

  • Is it really the job of a police officer to use force "to protect herself and life"? Or is her primary job to protect the lives and safety of civilians, even sacrificing her own life or safety in order to do so? We think of the job of fire rescue in the latter way. Why not police officers? (Related question: Should police officers even be armed? Or perhaps armed only with non-lethal weapons?)


 


 

Book

Want the full book?

Get the complete guide to logical fallacies by Bo Bennett.

Buy the Book

Master Logical Fallacies Online

Take the Virversity course and sharpen your reasoning skills with structured lessons.

View Online Course