Racial Behavior

[ApostateAbe]

ApostateAbe, in a slightly different society, you could prove the validity of horoscopes with the very same logic you're using here to argue for a racial component to behaviour.

OK, feel free to explain further. How is the intermediate value of the evidence for within-group heritability of violent behavior in favor of the hypothesis of between-group heritability of violent behavior much like an argument concerning horoscopes? Or feel free to stop that line of argument because I figure it would be too muddled.

 

[Jokodo]

ApostateAbe, in a slightly different society, you could prove the validity of horoscopes with the very same logic you're using here to argue for a racial component to behaviour.

OK, feel free to explain further. How is the intermediate value of the evidence for within-group heritability of violent behavior in favor of the hypothesis of between-group heritability of violent behavior much like an argument concerning horoscopes? Or feel free to stop that line of argument because I figure it would be too muddled.

Within group heritability is of no value, intermediate or otherwise, in favor of the hypothesis of between-group heritability, except in the very limited sense that by showing that there is a genetical component to whatever aspect you're studying, so if divide the population into subgroups by whatever criteria you choose (including "race"), the subpopulations are, in all likelihood, not going to be exactly equal as far as those genetic components are concerned. Even if, at one point in time, the subpopulations were exactly equal, that would cease to be the case as soon as one white person with below-average aggressive potential died without a corresponding number of black people dying.

What it does not allow us is to conclude that the observed inter-group variance is in any way reflective of that genetic component. Given that (a) we know that environment plays a strong role, and (b) we know that the environments in which blacks and whites grow up (in the US and globally) greatly differ on average, an empirical observation of, say, blacks being on average more prone to violent behaviour is of zero informative value regarding the direction in which the groups differ as far as the genetics are concerned, let alone the size of the inter-group difference in genotypes.

If we had a society where Sunday was considered so holy that even hospitals closed down entirely, we'd have good evidence that the calendar (genetics in this analogy) effects probability of surviving infancy (if you're unlucky enough that there are complications during your birth and it's a Sunday, chances are that you and maybe your mother won't survive), and that effect can be demonstrated to be independent of nutrition. If that society is also poor enough for seasonal variability to effect the calories available to the average citizen, which month of the year you're born in would also have an effect on your chances to survive infancy. What you're saying is no different from saying "we know that the day-to-day effect cannot be explained with nutrition alone, therefore we can presume that there is a strong non-nutrition based component to the month-to-month effect even if we all agree that nutrition does play a role".

 

[ronburgundy]

EXACTLY!! That is why those twin studies provide zero evidence related to a genetic contribution to between race differences. Which means there is no evidence of a genetic contribution to racial differences. Sure, it's "possible", which in science means nothing. In contrast, the OP presents an account that have massive evidence to support it to the point that it is not just plausible, but probable such that it would be surprising if those historical factors were not a contributor and would call into question highly validated theories of violent behavior more generally.
There is massive evidence that both being the target of and a witness to abuse and violence has a causal impact on one's own threshold for violence. There is also evidence beyond any doubt that American blacks suffered constant abuse and violence at the hands of whites for many generations and up until recently, and on a scale that no other ethnic sub-group in the US compares to (not even Native Americans who were mostly just slaughtered rather than abused). These two well established facts by themselves strongly predict that the abused history of the black community have made members of that community more likely on average to employ physical violence in their interactions, independent of the many other factors (whether genetic or environmental) that impact the variability in violence among individuals .

Taking the OP's theory seriously gives us plenty of implications to consider and plenty of the taboos you are fond of, without bringing purely speculative non-evidence supported possibilities that genetics create different violent tendencies between races. See my prior post on those controversial implications. In short, if the OP is correct, then those historical abuses predict that blacks should be more aggressive and thus more criminal on average leading to more incarceration and more violent conflicts with police that might end in their death. IOW, the observed outcome disparities are predicted and accounted for, even if current cops and the legal system are in no way racist or differential in their treatment of blacks. This means that no aggregate disparities in those outcomes can be viewed as evidence of racism or unfairness by current cops.

ronburghundy, I think our ideological biases tend to excessively affect our judgments about what counts as "zero evidence" and what counts as "massive evidence." In this debate, all evidence contributes an intermediate amount of probability toward or away from one position or the other. You think that the within-group heritability of violence counts for "zero evidence" that there would be between-group heritability of violent behavior. But, suppose we we were to find an exactly zero value of within-group heritability of violence. You are likely to claim that this effectively disproves the hypothesis of between-group heritability of violence, and you would be correct!
But, you cannot have both at the same time. Suppose, on the other side of the coin, that within-group violent behavior is 100% heritable. It would leave absolutely no room for the position that between-group violent behavior is merely environmental. The reality as it stands is intermediate. The within-group heritability of violence is 50%. The higher that value is, the higher the probability that the between-group differences are likewise mostly heritable. This is actually a mathematically true statement. When you go from 99% within-group heritability to 100% within-group heritability, the evidence provided for between-group heritability does not suddenly jump from 0% to 100%.

It is a case of only the most extreme values having inferential meaning. IOW, only if either heredity or genetics is ruled out as impossible due to zero variance explained at the individual level of variance does the individual level variance have much implication for group level differences. However, when the variance explained by either type of factor at the individual level is closer to 50%, and the between group variance is only a small fraction of the total variance, then this means that their is almost no difference in the probability of between group differences being 100% environment, 50% environment, or 100% genetic. IOW, the individual level data doesn't change the relative probabilities of the various possible ratios of influence driving the between group differences. Something that leaves all possibilities as similarly plausible is, by definition, not relevant evidence to that question.

Since there is 50% within-group heritability of violent behavior, then the hypothesis of between-group heritability must be confronted as though the evidence counts for something, because it really does, not dismissed as though it counts for absolutely nothing.

IT means the same thing as it would if the withing group heritability were 5% or 95%. In all those cases, the between group variance can still plausibly be 0% to 100% genetic. They only means something relative to 0% or 100% within group heredibility, and this nothing more than that it moves the heredibility hypothesis from impossible to the possible, but has little impact on its actual level of probability.

In contrast the environmental theory follows logically from established facts with mountains of evidence establishing them as extremely probable. Thus, it is extremely probable, while your genetic claim is merely possible.

 

[ApostateAbe]

Jokodo and ronburgundy, I have uploaded an image of a table from Arthur Jensen's book, The g Factor: the Science of Mental Ability, 1998, page 455. The focus of the book would be intelligence differences, but the argument applies to differences in violent behavior. The same calculations were done in Neven Sesardic's, "Philosophy of Science that Ignores Science: Race, IQ and Heritability," 2000, page 589.

WGH is within-group heritability (about 50% in our case), BGH is between-group heritability (you would hold exactly 0, presumably), and sigma-sub-E is the effect of environmental variations between groups, standardized at sigma-sub-E = 1 for one standard deviation of the environmental effect of variations within a group. The table shows that, for your hypothesis to hold, sigma-sub-E must be 1.41, meaning that, whatever environmental variations cause a person to be one standard deviation more violent than the average in his own race (going from the 50th percentile to the 68th percentile), such environmental variations must apply 1.41 times as much between races ON AVERAGE. The greater the within-group heritability, the greater the environmental effects must be to account for the between-group differences. The values in this table apply if the between-group differences are one standard deviation on average, which is true for the American racial IQ gap, and I am unsure of the violence gap, but one standard deviation looks to be a minimum judging from the FBI crime statistics (violent crime among blacks is about triple the rate among whites).

ronburgandy, you claimed "[it] means the same thing as it would if the [within] group heritability were 5% or 95%." I will do the math if we assume a value of 95%, using Jensen's formula.

sigma-sub-E = sqrt((1-BGH)/(1-WGH)) = sqrt((1-0)/(1-0.95)) = 4.47

So, if we assumed a within-group heritability of 95%, the effect of environmental effects between groups would need to be four and a half times as much as the environmental differences that would affect one standard deviation more violent behavior within a group, to keep your hypothesis true. WGH is not actually as high as 0.95, and sigma-sub-E = 1.41 is pushing it but is still plausible, but I think it is time to stop pretending the value of WGH doesn't matter. It matters.

 

[Jokodo]

The greater the within-group heritability, the greater the environmental effects must be to account for the between-group differences.

That's true. What's also true: The greater the within-month calendar effect on infant survival rate, the greater the nutrition effects must be in order to account for the between-month differences. Same logic. It works. Congrats to proving astrology!

This truism only helps you as an argument for a genetic component to the between group difference if you have a sound way to estimate the size of the environmental effects, and find that the observed inter-group difference falls clearly outside its range of plausible values. You don't have that. Not even close. Worse, you aren't even trying.

 

[Perspicuo]

Perfectly explained by skin color and other "racial" features being the cues for discrimination. Same horse, same carriage, but in inverse order, so it's still an awkward animal, most certainly a house cat being passed for an odd elephant that licks its ass after eating tuna with trunk and car-sized abdomen notoriously not getting in the way between the prickly tongue and anus.

The heritability studies examine identical twins (and other family pairings) WITHIN a race. Identical twins have the same skin color when born, so skin color or other "racial" features being the cues for discrimination would not seem to be relevant to explaining the within-race heritability of violent behavior. There is room for that as an explanation for why some races are more violent than others, but not as much room as we would like.

"Within a race"? Races don't exist.

 

[Rhea]

Jokodo and ronburgundy, I have uploaded an image of a table from Arthur Jensen's book, The g Factor: the Science of Mental Ability, 1998, page 455. The focus of the book would be intelligence differences, but the argument applies to differences in violent behavior. The same calculations were done in Neven Sesardic's, "Philosophy of Science that Ignores Science: Race, IQ and Heritability," 2000, page 589.

Abe - could you please try to take some time to address the generational violence data that is at the heart of this thread? You jumped right to, "I don't want to discuss your thesis, can't everyone discuss mine?" Which is your sign that you should start a different thread.


This one is about the overwhelming data that shows the effects of childhoods that witness and experience violence and how that translates into the way they raise their children or interact with the world.

You seem to have decided that you can bring up a different topic and it is somehow relevant. It isn't.

 

[ApostateAbe]

Jokodo and ronburgundy, I have uploaded an image of a table from Arthur Jensen's book, The g Factor: the Science of Mental Ability, 1998, page 455. The focus of the book would be intelligence differences, but the argument applies to differences in violent behavior. The same calculations were done in Neven Sesardic's, "Philosophy of Science that Ignores Science: Race, IQ and Heritability," 2000, page 589.

Abe - could you please try to take some time to address the generational violence data that is at the heart of this thread? You jumped right to, "I don't want to discuss your thesis, can't everyone discuss mine?" Which is your sign that you should start a different thread.


This one is about the overwhelming data that shows the effects of childhoods that witness and experience violence and how that translates into the way they raise their children or interact with the world.

You seem to have decided that you can bring up a different topic and it is somehow relevant. It isn't.

Explanations are probable only when they effectively compete with rival explanations of the same data. As you dismissed the genetic explanation in your first post, you brought the argument on the table. I am not changing the subject. And I addressed your explanation in my first post. Cultures change and evolve. The fact that the black culture has NOT changed so much to be less violent seems to make the pure-culture perspective less plausible. How long will the black culture of violence persist, in your opinion?

 

[Rhea]

Explanations are probable only when they effectively compete with rival explanations of the same data. As you dismissed the genetic explanation in your first post, you brought the argument on the table. I am not changing the subject. And I addressed your explanation in my first post. Cultures change and evolve. The fact that the black culture has NOT changed so much to be less violent seems to make the pure-culture perspective less plausible. How long will the black culture of violence persist, in your opinion?

I discarded it because it is well known that the effect of being raised in violence is deleterious for everyone, and I wanted to explore that thesis. You did not explore it, you immediately said, oh that's nothing, it's all genetic, see, the twins.

Your twin study does nothing to determine whether the black twins in separate households were raised in black or white households, or in places with or without systemic racial violence. Even if a black twin in raised in a white household, you have not establish that the child is raised free from the racial oppression that faces black children in America, or the stories they hear from their own household, or even the effect of being raised by someone who was raised in violence/oppression. How does a twin study make one of the twins not experience the oppression that so many people of color face? You'll need to explain how they controlled for that, if you're claiming that they drew a conclusion that takes it into account.

My point here is that it is well known, widely studied and firmly established that experiences of abuse will create increased incidences of perpetration of violence and abuse. That is the cycle that transmits the systemic societal abuse from an oppressive racism into the family and onto the next generation.




How long do I expect it to last? As long as it is normalized and not addressed. As long as the childhoods of people are affected by ongoing abuse. The external nature of racial discrimination (especially such as mob or police abuse) keeps the fuel going. I think generations will tend to diminish the effect, but when an external source adds additional abuse (helplessness, anger, fear) how can it diminish with distance from the worst?

How long will it last? Until we DO something about it? About the abuse from outside of the family that ADDS to and strengthens the residual generational abuse that comes from the parents' and grandparents' experiences of abuse.

It's not about excusing the behavior or coddling it. It's about figuring out the cause so that the most productive and effective measures can be applied to address it. Throwing people who are black in jail at 5x the rate that you do for the SAME crime in white people will NOT diminish this effect.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1494926/

While the association between abuse in childhood and adverse adult health outcomes is well established, this link is infrequently acknowledged in the general medical literature. This paper has 2 purposes: (1) to provide a broad overview of the research on the long-term effects of child abuse on mental and physical health including some of the potential pathways, and (2) to call for collaborative action among clinicians, psychosocial and biomedical researchers, social service agencies, criminal justice systems, insurance companies, and public policy makers to take a comprehensive approach to both preventing and dealing with the sequelae of childhood abuse.

And I would add, acknowledging that racism is an addition to that abuse, it is not always coming from parents or exclusively from parents.

 

[Jolly_Penguin]

This one is about the overwhelming data that shows the effects of childhoods that witness and experience violence and how that translates into the way they raise their children or interact with the world.

You titled the thread "Racial Behaviour" though. This idea of yours goes well beyond race, and may be independent of race in many settings.

 

[Rhea]

This one is about the overwhelming data that shows the effects of childhoods that witness and experience violence and how that translates into the way they raise their children or interact with the world.

You titled the thread "Racial Behaviour" though. This idea of yours goes well beyond race, and may be independent of race in many settings.

True, I did do that. I was starting with people calling violence a racial behavior and me thinking it wasn't, actually, at all. And that my thoughts led to causes that could be connected to possible solutions, and... but, yeah, I did kinda do that, didn't I?

 

[ronburgundy]

ronburgandy, you claimed "[it] means the same thing as it would if the [within] group heritability were 5% or 95%." I will do the math if we assume a value of 95%, using Jensen's formula.

sigma-sub-E = sqrt((1-BGH)/(1-WGH)) = sqrt((1-0)/(1-0.95)) = 4.47

So, if we assumed a within-group heritability of 95%, the effect of environmental effects between groups would need to be four and a half times as much as the environmental differences that would affect one standard deviation more violent behavior within a group, to keep your hypothesis true. WGH is not actually as high as 0.95, and sigma-sub-E = 1.41 is pushing it but is still plausible, but I think it is time to stop pretending the value of WGH doesn't matter. It matters.

And there is absolutely nothing that suggest it is not highly plausible that the influence of the BG enviornmental factors would be many times greater then the same factors on the WG variance. There is absolutely no logical or empirical basis to thing the degree of influence would be the same within versus between groups.
In fact, given the indisputible facts regarding the extreme differences in environment experienced for the different groups, it would extremely implausible for the degree of impact to be remotely similar for WG and BG variance. A 4.5 difference in magnitude of environmental influence is at least if not much more plausible than a zero difference in magnitude. So, yes, the WGH sets the bar for how much environmental influence there needs to be for it to account for BG variance. But the point is that such a degree of influence is no less plausible than a lesser degree of influence. Thus, the WGH studies give us virtually no information about the relative plausibility of the different possible accounts of the BG variance.

Here is an illustrative example. Imagine that research shows us that 40% of the overall population variance in getting lung cancer can be explained by a genetic predisposition.
A group of 100 people are found in a deep dungeon in someone's mansion. They spent their whole lives there under extreme circumstances rarely experienced in the general population. Every surface was covered in asbestos, and there are several other cancer causing agents in the air, plus the people that held them prisoner blew tobacco smoke into the room on a daily basis. So, now it turns out that the number of these people with lung cancer is statistically higher than the general population. What is the more plausible explanation for their higher rate of cancer? IS it that 40% of the extra cases are due to genetic predisposition, thus this group for some reason has a more cancerous genes? Or is it that 100% of the extra cases are due to the extreme environmental circumstances that were known to be in and are known to cause cancer?
The latter is at least if not much more plausible than the former or than any other combo of the two.