Friday, January 25, 2019

Biological Materialism

Let's continue on with what was discussed in Subprime Science. (This post is longer than normal, and takes on some technical arguments at a high level.) A couple takeaways from that piece relating to evolutionary studies were:
  1. Like physics, it shares an infatuation with randomness.
  2. Many of the fundamental mechanisms are assumed but have never been demonstrated in a lab.
DNA is commonly depicted as an algorithm that fully directs the development, maintenance, and operation of all an organism's biological functions. Most associate the nucleus as the cell's brain. In actuality, DNA has just a single role: it encodes proteins. It is a reference book that is consulted as necessary when a particular protein is needed. Change the gene's allele and you get a different resulting protein. An example of this is the genetic determination of hair color. The generated proteins are functionally similar, but not identical. In that fashion, the genome is assumed to account for the entirety of one's inherited traits. The problem is that there aren't nearly enough genes.

The Human Genome Project was a global scientific collaboration undertaken to fully map the human genome. It was lauded as a monumental human achievement; more significant than landing a man on the moon. But - as often happens - improvements in technology only illustrated how little we know. Early estimates for the number of genes were as high as seven million. Even after the turn of the century, molecular biologists were insisting that the human genome must consist of upwards of 100,000. The results of the Human Genome Project indicate around 19,000 genes. That's not a lot of variables. For comparison, the Toyota Camry electronic throttle control system had 11,000 global variables. Granted, that was some nightmare code, but still a human is vastly more complex than throttle control in a low-end family sedan. Traits are often expressed by multiple genes. If it takes a few genes to determine hair color, and a few to determine eye color, and a few to determine height, and IQ, and susceptibility to heart disease, or alcoholism, or schizophrenia, or each of the 96 sexual orientations, or....well, you can see that 19,000 doesn't go very far. And those are just the traits of human variance. There would need to be genes to describe how to build out the complex brain with its 100 billion or so neurons arranged in the proper manner, with all the intricate structures, as well as all the other organs and tissues. How do 19,000 genes describe a kidney, or where to put it?

Surprisingly, the human genome is smaller than that of a nematode - a microscopic worm. Rice has more than double our gene allotment. From the first article,
The fact that the human genome is so parsimonious raises an interesting question. What exactly is it about the human genome that gives rise to our staggering complexity, in the brain for example, compared to other animals such as monkeys, worms or even water fleas?

A good answer to that question will win prizes!
In modern science, it's all about the prizes. Well, I'll field a good answer to that question: the human genome doesn't give rise to our staggering complexity. Simple enough. (I accept checks and all forms of flattery.)

It is now well established that there is biological inheritance that does not come directly from genes. One scientist showed that if rats are trained to fear a certain chemical odor (by administering electric shocks), then that fear is measurable in two generations of offspring. That is quite a challenge to genetic determinism. The strange ability of organisms to pass specific information to offspring constitutes a field known as epigenetics. Scientists can't explain epigenetics, but assume that somehow regular genetic expression is being modified, and offer some theories to how that may occur. As in astrophysics, the response to results that challenge basic assumptions is to stick with those assumptions and throw in as much complexity as needed to handwave the annoying questions away.

A rejection of biological materialism does not refute the theory of evolution. Darwin himself was not a biological materialist, having written about a hundred years before James Watson - who is currently being de-platformed by the rabid left for insisting that genetics affect human intelligence traits - discovered DNA. Take the example of the giraffe. Under materialism, the long neck is explained by random mutations and gene combinations. One day, a gene randomly mutates to express a slightly longer neck, giving that giraffe a slight survival advantage. Over many generations, this process results in the extremely long necks we see today. Darwin assumed that the longer neck somehow resulted from the animal straining to reach leaves. The natural selection is still the same - animals with the adaptation get a survival advantage.

In 2016, scientists completed a giraffe genome project. They found 70 genes they believed were associated with long neck development. They intend to test the results by modifying mice with the giraffe versions to produce long-necked mice. My prediction: it will never happen. None of the genes are encoding neck length. They are encoding proteins that the giraffe needs to incorporate into its metabolism. There are genes correlated with giraffe necks because giraffes have structural differences which require protein compositions which differ from, say, mice. Geneticists must routinely fall prey to the correlation == causation fallacy. It's sort of like determining that batteries are what make a car go, because all cars that go have batteries, and all cars that don't have batteries, don't go. We understand that batteries don't drive cars, but are still a necessary component. (Teslas excluded.) Similarly, a giraffe without the appropriate proteins cannot support a long neck. That doesn't mean the genes "drive" long necks, only that they are necessary.

Some famous research done in Soviet Russia on Siberian foxes showed that selecting for a single trait - friendliness - caused a whole slew of changes to emerge besides behavior. The shape and color of the animals changed, as well as their vocalizations, and reproductive behaviors. Generally, they became more neotenous (puppy-like), and cuter. These secondary adaptations were noted within the first few generations of the project. The broad effects of selecting for a single trait were surprising to geneticists.

Recently, researchers at the University of Illinois analyzed the fox genome to determine the genetic causes of domestication.
They honed in on a single gene, known as SorCS1, which is involved in synapse formation, functioning, and plasticity, as well as additional functions. Although it had never before been known to contribute to social behavior, SorCS1 was clearly associated with a very specific behavior in foxes.
From an earlier article,
"Our analysis revealed that the differences between tame and aggressive foxes may lie in cells in the anterior pituitary gland, which can change their shapes to communicate with one another about when it's time to release stress hormones," Hekman said. "Their pituitary glands may produce the same amount of stress hormones but be less efficient at getting those hormones into the bloodstream."
The researchers found a strong correlation of the SorCS1 gene with the domesticated population of foxes. However, that does not explain the broad-spectrum changes within that group. They address the matter in their article in Nature.
Changes in physiology, morphology and reproduction have also been observed over the course of fox domestication. These by-products of selection for behavior could be caused by several mechanisms including pleiotropy, hitchhiking, random fixation, trade-offs between different biological systems and targeting of genes that have a broad effect on the genome, for example DNA methylation.
That seems a bit speculative for an article in a premiere journal. (Nature is a politicized entity, so their professionalism should always suspect.) The first possible explanation, pleiotropy - the influence of a gene on multiple traits - could be tested by running a similar experiment, but only selecting for one of the related traits, like fur color, with no human interactions. After a few generations, the experiment population should show increased friendliness if pleiotropy is the cause. (Which it isn't.) Hitchhiking requires the genes for the various traits to reside on the same chromosomes. The researchers should be able to to investigate that with their genome data. Random fixation - the reduction of a population to a single allele of a gene - is possible, but in this case the odds of random fixation for all in the various traits, in ways that show gradual change over many generations, is not plausible.

I'd suspect that the interaction with humans itself has a significant impact on the evolution of the foxes. An experiment that duplicates the fox breeding but adds two new test groups - one that is selected for friendliness but has no interactions with humans, and one that encounters heavy human interaction but selection is random - would likely yield results just as surprising to geneticists as the original research was. Another experiment could be done more easily with mice. Half of the mice must always reach out to grab their food, half the mice do not. Within each population, half should be selected for arm length, and the other randomly. This should allow a quantifiable comparison between Darwin's understanding of evolution and the modern random genetics model.

The researchers found a strong correlation between cells that influence the release of stress hormones from the pituitary gland and the domesticated fox genomes. It seems perfectly natural that it would be so. The selection process should give preference to those with a lower stress responses. The correlation with the single gene may be a matter of the random fixation they referred to. There is no other indication that it affects behavior outside of this one study of a particular lineage of foxes. Of course, a single gene doesn't account for the broad and gradual changes. They admit as much themselves.
"We think this gene makes foxes more tame, but we don't want to overemphasize it—tameness isn't associated with a single gene. The picture is definitely more complex," Kukekova says.
When results don't match expectations, always assume more complexity. What other genes are associated, then? They did the statistical analysis. Only one gene jumped out at them. I'd wager that the one gene is the entirety of the genetic selection that occurred, and all other adaptions occurred outside of genotype changes. The article is titled Sequenced fox genome hints at genetic basis of behavior, but the results actually demonstrate the opposite conclusion. The sequenced fox genome hints at an extremely limited genetic basis of behavior (if any) and none for the associated physiological changes. All conclusions can be twisted into support of the prevailing theory if the theory is considered to be infallible.

While I'm not the only one to make the observation of the gene shortage problem, the skeptic commentary out there is scant. The most prominent voice has been Paul Ehrlich, a Stanford biologist most noted for his controversial 1968 book The Population Bomb. Ehrlich, observing the ever-decreasing estimated number of genes coming out of the Human Genome Project, opined that there were simply not enough genes to justify evolutionary psychology.
"People don't have enough genes to program all the behaviors some evolutionary psychologists, for example, believe that genes control."
"Evolutionary psychology is dead but doesn't seem to know it yet."
Ehrlich isn't defecting from biological materialism, he's just pointing out that the numbers don't work. He's quite sure that inheritance == genetics, so his conclusion is that evolutionary psychology just doesn't exist. (A ludicrous conclusion.) His detractors are just as bad.
If gene shortage is such a problem, then how can there be so many hundreds of adaptations and features in a body? There are already so many types and locations of bone, hair, sense organ, nerves, hormones, blood vessels etc. If there is no apparent gene shortage for making such variety in a body, then why should there be a problem in making a multitude of mind modules and inherited instincts as well?
There don't seem to be enough genes for evolutionary psychology, but then there don't seem to be enough genes for general biology either...therefore, there must be enough genes for psychology. The basic assumptions can never be questioned. They go on to explain that it's not the genes that determine phenotypes, but the combinations of genes. That's the answer you'd have to give when faced with a glaring gene shortage. Combinatorial explosion will certainly grant you an enormous state space, but that doesn't mean it's correct. What's the evidence? How do different combinations of genes induce complex traits? The majority of animal behaviors are inherited, not learned. What is the combination of proteins that tells a bird how and when to make a nest? Can scientists manipulate those genes to compel the bird to build a bridge instead? If gene combinations drive genetic expression, rather than mere genes, it should have been detected in the statistical analysis of the fox genome. All the genes are still there. What they're dangerously close to outright stating is that the genetic combinations work in mysterious way that can't be measured, and thus can't be disproven. (Making it not a useful theory.) What they are more careful to say is that it's just very complex, and will require many years of heavy funding to suss out all the various details of a theory which is assumed to be true. As in astrophysics, the theories are axiomatic.

This is a field where assumptions are unquestioned and rationale is vague and speculative. I once read a theory that our genes encode fractal patterns that the body follows. This was to account for the proper routing of blood vessels, for one. It's an implicit acceptance of the gene shortage problem, as it's hard to conceive how genes detail the size and locations of the body's countless blood vessels. A fractal pattern would give a concise formula that the body follows - somehow - to get consistent results. In either case, where is the evidence that blood vessels - or any of the body structures - are described by genetics? What genes can be altered to produce three kidneys instead of two? How do I genetically modify my offspring to have flippers instead of hands? (I would do it.)

And that's the first counter anyone would make to my skepticism. "Uh, genetic engineering bro, so you're wrong." But was has been engineered? In the most famous example, engineers swapped a gene in soybeans that encodes an enzyme that is destroyed by Roundup, with an allele from another plant that is resilient. Thus, Roundup Ready soybeans were born. Cool engineering, but not evidence that genes do anything besides encode proteins. Recently, a breed of salmon were engineered that grow to full-size faster. The trick was to inject salmon eggs with genes that encode a growth hormone, compelling the animal to grow faster. The most significant genetic engineering was the transplantation of an entire genome between two similar species of a bacteria. "The entire protein repertoire changed," said the principle investigator. Well, you'd expect that. The cell survived because of the close similarity of the bacteria. It is also conceivable that such a transplant might succeed in closely related animals, like wolves and coyotes, or humans and New Yorkers. (I keed, I keed.) I would predict that there will not be any similar accomplishment between even mildly dissimilar organisms, and I have to note that there seems to be no followup successes now in over a decade. The research was intended towards the goal of creating custom genomes from scratch. Prediction: the venture will fail, dismally, if it hasn't already.

One more aspect I want to quickly mention before we move on to the social impacts is another domain that falls within biological materialism: consciousness. The standard assumption is that it all happens in the noggin. The first troubles began to appears all the way back in the 1950s, when scientists failed to isolate the locations of specific memories in a frog's brain. They did discover that memories were dulled proportionally to the amount of cortex cut away, wherever it was cut. This indicated to them that the memory was distributed over the entirety of the cortex, giving rise to the holonomic brain theory. There has been virtually no progress on the "hard question" of neuroscience, with all theories more or less amounting to speculation. The core questions of human experience - how do we think? why do we sleep? - remain as mysterious as they've ever been.

Impacts

Some consequences if biological materialism is indeed false.
  • It is the biggest scientific swindle of all time. The other contenders for the title would probably be global warming and the big bang theory. Global warming is less of a swindle because there is widespread skepticism, and there is a degree of truth to the claims. (Carbon dioxide is a minor greenhouse gas, and there was a warming trend at least up to the turn of the century.) The big bang theory is a bigger swindle because it is generally accepted by everyone, although there is a small skeptic community online. Biological materialism is the biggest ruse because it is almost universally accepted, and it is the assumption used for huge amounts of bad investments and research, mostly of public funds. And, as mentioned in Sherlock's Folly, its prominence means no one is looking for the better answer.
  • DNA is not the code of life. It's more like a recipe book. It tells how to make a soup or a cookie, but not how to run the kitchen. Giraffes are tall because giraffes are tall, and the genome instructs how to build the proteins the cells need to function. 
  • Parents don't just pass material genes to their offspring, but also behaviors. Want your kids to turn out good? Do good yourself. This would predict that children of older parents should be more capable and successful, and there is supporting evidence. They have more life experience to pass epigentically to their children. (Their assumed theory is that older women just make better mothers.) PTSD and other mental disorders are also passed down.
  • GMO food isn't that scary. Copying over alleles of an enzyme from another edible plant is nothing to be concerned about. Jacking up growth hormones in salmon? Might be a problem - because of the growth hormone, not the GMO. 
  • Genetically engineered super humans isn't happening. There was a recent buzz that a Chinese scientists had already produced modified humans, stoking all kinds of fears. They are misplaced. Improvements will be minor at best, but probably nil. The best way to create supersmart humans is to breed them, but people tend to get creeped out when you start pining for eugenics.
  • IQs are on the rise, globally. (They're sinking in the west because of immigration, but that's another matter.) People like to credit education, but that's not it. Our public education is crap. IQs are going up because we use our brains more. (And thanks to c-section childbirths.) Even people in everyday jobs are cognitively engaged. Mechanics troubleshoot complex systems every day. Cabbies are constantly strategizing routes in a busy cities. (The most mundane jobs do not give the same benefit, such as picking lettuce all day.) Generally, though, automation is making us smarter, because more people are tasked with big-brain duties. Our intellectual endeavors stretch our brains like a giraffe stretches its neck.
  • Conscious AI isn't happening. There won't be a Kurzweil singularity, which is based on the assumption that consciousness is a material phenomenon. Which is fine, because superintelligent machines would lead naturally to human extinction. (Only liberal philosophers would be happy with that.)
  • The people freezing their severed heads in cryogenic chambers are wasting their money and dignity. The people holding out that technology will one day allow them to upload their consciousness onto a chip and gain immortality are more likely to gain it in a spiritual afterlife.

Conclusion

The natural question to ask is, if genes don't provide a general blueprint or operating instructions, then what does? I have no idea, and that is a perfectly valid answer. I'm only making the case that the gene theory doesn't add up, and hardly anyone seems to be noticing it. Whatever the unknown mechanism is, we'll never find it if the entire scientific community is working under the premise that unknown combinations of genes are a catch-all that can explain away all apparent contradictions.

No comments:

Post a Comment