2017/05/31

What is "societal discrimination"?


I shall not today attempt further to define the kinds of material I understand to be embraced within that shorthand description [hard-core pornography]; and perhaps I could never succeed in intelligibly doing so. But I know it when I see it, and the motion picture involved in this case is not that.
Concurring, Potter Stewart, Jacobellis v. Ohio, 378 U.S. 184 (1964).

American jurisprudence has strayed in response to demands from some social justice advocates who perceive an undesirable situation and attribute it to discrimination, racism, or other deplorable practice. However, what has too often occurred is the logical fallacy, post hoc ergo propter hoc. (The result was caused by a preceding event.)

The usual reasoning is that the result could not have occurred unless there was improper discrimination, and that therefore public policy must intervene against such discrimination. However, there can be many causes of undesirable outcomes, some of which may involve some kinds of discrimination, but not always improper discrimination.

We can propose an alternative definition: (Improper) discrimination is treating an individual as though his attributes were those common to his group. It is a failure to treat individuals as individuals, but rather to aggregate individuals into their group.

That brings us to having to confront something that most people don't want to accept or think about. More than a century of intelligence testing on people of many races, ethnicities, and nationalities yields some dismal results for those who want to believe than there are no differences in innate abilities. Tests have been criticized as biased or unsound, but tests have been modified to answer those criticisms. The results persist. American Whites get an average score of 103, Jews get 113, American Blacks get 85, and Hispanics get 89. This has been investigated by Black economist Thomas Sowell, who argues that while it is wrong to treat all members of those groups as though they had those average values. it does cast doubt on whether the low average performance of members of those in fields like education and employment is the result of societal discrimination. If they are in fact innately inferior, and genetic studies indicate that 50-80% of such differences are genetic, than that changes what we should be doing about such disparities.

I invite readers to put aside their reflex rejection of such results and seriously consider what we do if they turn out to be valid. Note that the above paragraph is presented as conjectural, not as an assertion, and it should not be taken as my position on the issues.

The key point here is to propose a better definition of "societal discrimination" than that which has become established in American jurisprudence today, where judges sometimes find that disparate outcomes are the result of societal discrimination, without evidence. That is not a proper basis for judicial intervention. The proposed definition is intended to work regardless of what might be the causes of any traits that might become stably manifest in a group, whether genetic, "environmental" which may include "discrimination", or whatever. It also doesn't matter what the "group" might be, or how it is determined. It works for any group, however defined.

Those who argue against genetic influence on intelligence as its manifestation stabilizes make a key mistake: they presume that there are only two general causative factors, genetic and environmental. In fact there is a third: emergence, referring to the fact that complex living systems are self-organizing, or chaotic, systems, in which small perturbations can have large consequences, and continue to have large consequences as the system develops.

But part of the mistake is to lump emergent development into environment, or "nurture", as though it were something that could be managed purposely. It is inherently unmanageable.

A second mistake is to treat genomes (genotypes) as more deterministic than they are. Genes influence the probabilities of phenotypes, but do not determine them. A typical human genome is thought to contain about 30,000 genes. The information contained in those genes is about eleven orders of magnitude less than enough to determine all the phenotypes in detail. But all the information that can be passed through the senses to influence development is also many orders of magnitude less than what it would take to determine human behavior in detail. Since information cannot be created, that means something else is shaping the details. That something is self-organization, but unmanageable self-organization. It is not susceptible to deliberate intervention.

A third mistake is to imagine that societal interventions, individual or collective, are or can be more effective than they usually are, especially after some period of development when the things we want to change have become stably manifest. At some point the undesirable traits become difficult or impossible to reverse. Once they stabilize they become a reality that has to be confronted, regardless of the causes of the differences. This especially valid for cognitive development, for malleable youthful brains do not remain as malleable beyond a certain age.

Especially illuminating to the question of the extent to which genes are determinative of cognitive capacity are the way genes largely determine species, each of which has a characteristic range of cognitive capacities. Some species even have their own versions of "societal discrimination", but in most it is difficult to identify any systematic effect of the distribution of cognitive capacities. For that purpose it is useful too compare humans with their two closest relatives, chimpanzees and bonobos.

Bonobos differ greatly from chimpanzees in their behavior. Unlike Chimps, they do not become aggressive and ill-behaved after the age of puberty. Therefore they have become a favorite subject for learning studies. They are able to learn to understand spoken English, and to communicate using symbolic keyboards that produce spoken words when pressed, but so far only up about the level of a 2.5-year-old human. They share about 3% of the human genome, or 0.5 megabases, which is more than they share with chimps. Humans have about 14.6 megabases they share with neither of their cousin species. Within that subgenome we can expect to find the genes that most influence human cognitive development. It is a large number, but finite. We can speculate that the number is a few thousand. The genes don't have to determine the wiring of the brain. All they have to do is set up the emergent process which then determines its own structure and behavior. But that process is, in general, unmanageable, and once it stabilizes, either to work well or not, it is likely to be irreversible.

Consider fingerprints. Identical twins both have them, but they are not identical. The details of the fingerprint are the result of emergence. Similarly hearts. Identical twins both have them, but the details of vascularization are not the same. Tissue that is to become a heart becomes one by responding to pressures from adjacent tissue that shape its development. The result is a chaotic system, that beats but not governed by a pacing signal. Similarly brains. Both twins get one, but the details of neuronal net structuring differs, although there can be similarities in the ways each twin leads its life.

Much is made of the capacity of humans, unlike most other species, to choose to overcome their seemingly innate limitations through determination and effort, perhaps with a little encouragement. That works with some, although not with all. There may be genes for that as well. Of course, no one really overcomes innate limitations, only realizes the potential that was there, but the ability of the brain to learn can continue into later years, especially with enough fortitude. It can sometimes even compensate for damage, as from injury or stroke. But only sometimes. Such things cannot be counted on to solve widespread or entrenched disparities of outcomes.

And no amount of determination is going to enable someone of average ability to become a cosmologist. Perhaps a successful lawyer in a small market.

Something also needs to be said about "white privilege". Nothing is said about how it is supposed to work. When I was young almost anyone could get a job within a day or two if he pitched himself properly. There was plenty of work to be done, and if someone had work to be done it didn't matter much who did it. But things have changed.

I am a highly skilled, highly experienced computer professional, yet I have difficulty getting work. The hiring process has become so difficult that about the only way I have gotten work has been by random contact with a hiring manager outside the workplace. That leads to the old saying, "It's not what you know but who you know that counts." Jobs today are surrounded by hordes of gatekeepers, each determined not to hire anyone they find unsuited, or even if they do find someone suited. But of course, if most of the gatekeepers are white, or whatever is the dominant shade in the community in which the job is, then you have to get to know at least one of the gatekeepers, and if that person happens to be white, then one can come away with the impression that "white privilege" has been the deciding factor. Getting past the gatekeepers can be largely a matter of luck. Many a successful businessman, if pressed for the secret of his success, will confess it was mainly a matter of luck. Of course, persistence can overcome bad luck, but it can only overcome so much bad luck.

Most disadvantaged people just don't know how to enter the circles where hiring managers dwell. Part of that is in their appearance and manner of speaking. They don't have a Henry Higgins to coach them. To get a professional position one has to fit the image of a top professional.

Another problem is technological unemployment. Machines are taking or eliminating jobs. Rapidly. That means many middle class persons are going to descend in their economic status. It's going to get rough for everyone, even for the top 1%. Eventually even they will be replaced.

Is "societal discrimination" as important as some think?

There is a widespread belief that disparities of socio-economic condition are entirely the result of discrimination, in the absence of which everyone would be educated, middle class persons. But is that more myth than real? Or could disparities in the treatment of people be more the result of differences in their attitudes or merits?

Obviously there is some discrimination, enough to seize on as a explanation for much of what we observe, but that doesn't make it the explanation for everything. Some disadvantages are a matter of choice, and will not be overcome by "ending discrimination".

It means he's up against middle-class morality for all the time. ...
I ain't pretending to be deserving... no... I'm undeserving, and I mean to go on being undeserving. I like it, and that's the truth. -- Alfred P. Doolittle, My Fair Lady.

______


  1. IQ and Race, Thomas Sowell, 11/26/2012. http://mobile.wnd.com/2012/11/iq-and-race/
  2. Intelligence, New Findings and Theoretical Developments, Nisbett, Blair, Dickens, Flynn, Halpern, Turkheimer, February–March 2012, American Psychologist https://www.apa.org/pubs/journals/releases/amp-67-2-130.pdf
  3. Commentary on the above, by Turkheimer, et. al., Vox Media, May 18, 2017.
  4. The bonobo genome compared with the chimpanzee and human genomes, Kay Prüfer, Kasper Munch, Ines Hellmann, et al., 486/7404, Nature Letters https://www.nature.com/nature/journal/v486/n7404/full/nature11128.html