Hacker News Reader

> During a single reproduction, there are tons of genetic and epigenetic variables that could be selected for, which determine different things. Why would a specific feature in a specific part of the body, that provides merely a minor improvement at any given time, come to completely replace all other genotype expressions?

As far as I understand it, sexual reproduction with its gene shuffling allows all features to evolve simultaneously. So the genes for the anorectal angle are constantly selected across the whole population in parallel with every other gene.

Have you ever seen the mathematics on this? When so much is being selected "at the same time" it seems like most of the time it would never get selected in the wash. After all, selection works on an individual organism, and it's binary: either the organism survives, reproduces and passes on its genes, or not (we are not talking about DNA swapping between microorganisms here). This binary signal is supposed to be the "vote" for whether a particular combination of genes is more fit or not. But you can't just say "it's all possible given a really long time", you'd have to actually quantify what the chances are of say, the mechanism a single vote selecting simultaneously 1000 different traits, would increase the prevalence of all the "beneficial" traits. Do you have enough resolution to decide e.g. whether the bone density, color of skin or the length of legs was the factor that led to survival?

I am especially not sure why it would select the traits -- like a colorectal angle -- that would make an individual simply more comfortable rather than make a tangible difference to survival.

If we took the predictions of evolutionary theory seriously, then their contrapositive seriously too. In particular why would gay people exist if genetically homosexuality confers a much lower genetic fitness? Why do we say homosexuality is primarily genetic if evolution is true?

> After all, selection works on an individual organism, and it's binary: either the organism survives, reproduces and passes on its genes, or not

Not quite. Selection isn't a 'did reproduce' / 'did not reproduce' binary - there's lots of in-between.

1. At the macro level, any given trait doesn't exist in a single organism, it exists in multiple organisms. Even assuming that an organism reproduces only once, or not at all, among a large population, the individual binary gets washed into a larger, fuzzy, aggregate.

2. A "binary" seems to assume that an organism can reproduce only once - many organisms reproduce multiple times.

3. A "binary" seems to assume that once reproduction happens, the vote is cast and the trait has (at least in aggregate), "won". Merely reproducing isn't good enough, though - most complex organisms have to continue to rear their young after birth.

Yes, obviously the binary decisions are being aggregated, but at the end of the day, everything is filtered through these binary decisions. It's not like "I survived and reproduced and it was all because of X". It's just a binary outcome every time an animal reproduces. It's a vote for a certain combination of traits out of myriads, and now you have to show mathematically that this sort of selection can lead to evolution.

And also keep in mind, the "beneficial" mutation step happens in only one animal, that had the mutation when the genes combined to form its genetic expression. And you need several of these steps. It's not like all of a sudden a bunch of proto-birds started to develop wings. Just one animal developed a proto-wing trait, and then reproduced along with 99.99% of others who didn't. But then, the proto-wing happened to be so beneficial despite not conferring an ability to fly, that the proto birds with the proto wings simply died less than the other 99.99%, and started growing in prevalence. And all this still assumes that, at every step from non-wings to wings, the improvements were beneficial for genetic fitness. Now, what if they weren't? How can evolution wait until a string of mutations detrimental to fitness takes place? Wouldn't that make the population with the proto-wings actually shrink -- when it was already 1 animal or so, it might actually be eliminated if we assume proto-wings were genetically disadvantageous before becoming wings?

Again, there is no math in these explanations. Just descriptions and handwaving. Where is the actual math backing these claims and the explanation of the mathematical results! How does it actually work out? It clearly doesn't seem like it should!

> Why do we say homosexuality is primarily genetic if evolution is true?

In my opinion it must be to help the fitness of the community they live in. The communities with low likelihood to spawn a gay person are (were?) much more likely to go extinct. It can also be a basic property (side effect?) of the sexual appeal arms race.

There is also the case of suicides, when they happen to young people that hasn't reproduced yet. It can also be argued its for the benefit of the community.

So the thesis is that the primary force in evolution is not individual's fitness but it is the community's fitness.

Meh, it is not necessary to postulate that every part of behavior to which genetics contributes is a thing which has been selected for.

If a particular gene causes good things 95% of the time (based on other interactions or developmental factors or whatever) and catastrophic failures like heart disease or cancer or crippling depression 5% of the time (and it represents a local maxima, with no simple improvements possible), then it can easily come to dominate a gene-pool, so long as the 95%-benefit outweighs the 5%-failure. It doesn't mean the 5% failure has been "selected for", it just means that it wasn't worth weeding out.

Moths circle lights because of genetic factors which influence their tiny moth brain's development. It doesn't mean that circling-lights was a selected feature.

Yeah, that's definitely a possibility. What you describe is more or less the basic property/side effect I was talking about, but much better explained.

You're betting evolution is not that good. One can also bet it better than you can possibly imagine. I don't think you can be sure either way at this point, but your argument seems to be the most compelling at this point!

So what are the genetic advantages of being gay? And how come nature hasn't figured out a way to make these advantages without also producing gay siblings, which would seem to bring down the overall fitness of the genes?

Since the first comment didn't work, I'll try going into more detail.

Our genetic code doesn't include statements like "if (is_male) { try_to_have_sex_with_ladies(); } else { try_to_have_sex_with_dudes(); }".

Our genetic code produces proteins of different shapes and those shapes influence the development of the structure of our brains and their function and that brain generates feels of various kinds that lead to attractions of various kinds.

Evolution works by changing the coding of the genetic code which changes the shapes or amounts of proteins produced which changes the development... a local maximum happens when the final system produced (our feels of attraction) would be negatively affected by any individual mutation on the genetic code. There may be better feels-of-attraction systems you could build from scratch that would propagate the species better than ours, but you can't turn ours into this better system without making it worse for awhile.

Nature can't make arbitrary changes; it can only make incremental changes. And consequently, we frequently get stuck at local maximums rather than global maximums.

Birth order has been correlated with the statistical likelihood of an individual being gay (i.e., the more older siblings you have, the more likely you are to be gay). This could definitely pose an evolutionary benefit to a stable population that lacks the natural resources to support a sudden boom of growth. Case in point being pandas which I just watched a show about last night that put forth pandas' low reproductive rate as ensuring low competition for their precious bamboo.

See that just further goes to show my point. Where is the evidence of these assertions? Mainstream evolutionists have said that "group selection has been disproven" except possibly kin selection. So your example clashes with what they've said.

https://en.wikipedia.org/wiki/Group_selection#Good_of_the_sp...

I personally think the benefit of the community is definitely one of the factors. And I think the evolutionists have it wrong. But why does the public just accept whatever they say? It hardly sounds like a settled theory.

What you linked says "On the other hand, kin selection is accepted as an explanation of altruistic behavior.[11]"

So, an extended family with a few homosexual kin might have greater reproductive success with their offspring having more caretakers.

Traits aren't evenly distributed throughout the population. Natural selection doesn't operate on the individual level, but at the _population_ level.

In other words, there is no "wash"-- genes aren't all shaken up until they're randomly distributed.

I agree that a lot of "evolutionary arguments" that make their way as far as popular culture are indeed post-hoc "just-so stories".

However...

>Consider just how much relative advantage with survival and replication any given feature must have in order to come to dominate ALL the other features that are competing in the population

Answer: very little! This is the advantage of sexual reproduction (or at least gene transfer) - it allows us to do our evolving "in parallel". A single "feature" might only convey a tiny advantage, and you'd think it'd be swamped by all the thousands of other features - but the "experiment" is repeated zillions of times, enough to bring it out of the statistical noise. It doesn't have to "make the difference" between reproductive success and failure in any particular instance. It also bears repeating that most "features" don't spring out fully formed - most "features" only convey a tiny advantage!

Sorry if this isn't a very good explanation. You've hit upon a key aspect of evolution, and I think you're right that there isn't a good, popular layperson's explanation of the mechanism that actually tackles the statistical aspect instead of glossing over it.

I want to reply in a very strongly worded way to this point.

There is a huge disconnect between the assertions made by proponents (like Dawkins) of evolutionary explanations, that "evolution has been proven LIKE GRAVITY at this point" and the lack of widely accessible explanation of how the most obvious statistical questions are settled in favor of the theory actually working. I don't even mean comfortably working, like "it should have taken 3 trillion years but we got lucky". I mean like, how does a binary decision select for things that are not even correlated to survival or reproduction but merely comfort? And even more astonishingly, how is it able to take such long (1000s of generations) speculative paths, without any apparent benefit until the basic version of the thing is fully formed? Combine that with the contradictory empirical evidence of punctuated equilibria broken by genetic explosions and I'm not so sure the theory is correct. Mutation and natural selection seem insufficient to explain the features we see.

But the disconnect between the dearth of actual answers (as available to someone with a mathematics background like me) with the insistence that evolution "is as sure as gravity" makes me skeptical. I find the situation very close to bullshit that is politically motivated and propagated through systematic bullying.

I should point out that this is not for religious reasons but merely because the theory seems like one of those early simple theories (like the humors or luminiferous ether) that always felt off and were too simple.

PS: just to be clear, I am not challenging the theory of common descent, but rather the proposed explanation for evolution as being completely a product of random mutation and natural selection -- evolution is a vague term which adds to the equivocation problem and is further evidence that there is strong political pressure in the scientific establishment to prop up the theory.

I find this an interesting line of argument, and I wouldn't be at all surprised to eventually find out that the current model is missing a lot of important "second-order" effects that contribute to the convergence towards good designs for life. I would be particularly interested to see if epigenetics play a larger role than we previously thought, because this would provide for a feedback mechanism.

However, I wouldn't discount the importance of _sexual_ fitness being a part of the natural selection picture. In some sense, you have not just nature judging your genes, but an intelligent creature that can make judgments about all sorts of traits that are not critical for survival. Just by being very picky, we all contribute extreme selective pressure. I would also imagine there is somewhat of a runaway effects in this regard. If you confer a slightly crappy gene to your offspring, that will, on average, cause them to find a mate with slightly crappier genes themselves. Go a few generations down that road, and you might finally terminate.

Just some food for thought. Obviously your central point that you would need some statistical backing to buy what I've written is well taken.

> I mean like, how does a binary decision select for things that are not even correlated to survival or reproduction but merely comfort?

If you're referring to

> For example, lets say the human anorectal angle was specifically "evolved" for pooping while squatting. Is it a "just so" story or did the exact anorectal angle contribute SO MUCH to survival or reprosuction that it became the standard for bipedal humans?

I don't think you need to assume that the particular angle and arrangement of organs there was evolved "for comfort", any more than we need to assume that fingers evolved for efficiently typing on keyboards.

The simplest explanation is that the particular angle is just one that happened to work - and then, given the way our bodies work, we try to find the most comfortable posture to poop in. It could have easily been the case that any other angle would also have been the default, and maybe we'd be most comfortable pooping while standing, sitting, or some other posture.

Well, not everything that is successfully modeled (sorry, but I won't go into arguing the meaning of "understood") can be described in a simple way. Obligatory xkcd: https://xkcd.com/1489/

There are answer for people with a mathematical background. They aren't very simple, but they can be verified and simulated til you get tired. The fact is that from postulating a population where individuals die, with hereditary genetics, and a cross-over style sexual reproduction, and a changing landscape, Math goes to show that you'll have all those features you are uncomfortable with.

Evolution works on very long time scales. Say a few people have a gene that gives a 0.01% increase in survival so the next generation there are 1.0001 times as many copies statistically expected. Over 100,000 generations the gene would statistically expand something like 22,000 fold. And over 1m generations 1.0001^(1e6) = 2e43 meaning most of the population would have it. I think most popular books skip math lessons - they are not so popular.

What we found is actually the opposites evolution can and does work on very short time scales exceptionally well favorable genetic traits can propagate throughout the entire population in only a handful of generation (in the best case even under 1 generation).

Also you should really not underestimate epigenetics (which we only now beginning to understand) which play even a more vital role than random mutation as epigenetics can selectively play with gene expressions in every offspring (as well as any thing which has already been born) based on both environmental factors, the "genetic" memory of the parents, and the current experience of the mother as far as mammals go.

It seems that every part of the tree of life, from bacteria to complex life forms (what's more interesting is that mammals/marsupials, fish, birds, etc. each have their own epigentic mechanisms) have it's own internal genetic restructuring mechanism which is really not dependent on random mutations being pitted against natural selection.

Just because many people make up evolutionary just so stories doesn't mean that there can not also be valid evolutionary explanations.

If you just say "anorectal angle evolved so that people poop squatting" it is a just so story. To make it solid you have to show the context, history and so on. For example, why is it an advantage to be able to hold your poop in?

A lot of things evolved out of serendipity and are not perfect, like those arteries going up and down the whole long neck of giraffes to bridge a short distance. But there are evolutionary explanations for it.

As for features related to survival: I think phases with more and less pressure alternate. If you made a breakthrough, say you discovered skeletons, or invented the iPhone, for a while you may have such a big advantage that you can play around with lots of fancy variants. Eventually the rest of the world catches up, pressure gets higher and the less efficient variants get weeded out.

Just having plausible-sounding evolutionary explanations isn't enough, though. You also need a way to falsify the theory. There is a good essay about this called Science as Falsification, by Karl Popper:

http://www.stephenjaygould.org/ctrl/popper_falsification.htm...

Otherwise how is it different from Marxism or Freud's theories? I can always concoct an explanation in the context of my theory. That isn't science by itself. It's demagoguery.

You make a model, then you make predictions. If they fail, your model was wrong. You can do that in evolution theory, even though you are mostly looking at history. Maybe you predict that all birds should have tails at least 30'' long, then you look around and see that they don't. You can also look at the way genes spread - I suppose you can compare the observed spread of a gene to the spread that would be expected if it were random.

Nevertheless, once you see something unexplained (like zebra stripe), I suppose you start with creating theories, then you try to verify them. Isn't the article here doing exactly that?

Yes and if the a particular theory of evolution were only 5 years old that would be fine. But several classical formulations like the https://en.m.wikipedia.org/wiki/Modern_evolutionary_synthesi... have been around for decades and various proponents around the world yell that they are "as proven as gravity". A strange position to take when there was little falsification. Your examples above serve to illustrate the point - they are just illustrations of how falsification might look like if it were done. If the prediction and falsification framework was robust by now, you wouldn't have to come up with made-up examples, there would have been tons of them like there are for other, widely tested theories, even in biology

I am not a biologist, so you shouldn't draw any conclusions from me not knowing good examples or the state of verification.

Your Wikipedia link seems to mention evidence for the "modern evolutionary synthesis", are you sure that there has been no falsification? Also, I have heard that name for the first time today.

Isn't most of 'anthropology' just invention of plausible sounding stories with no way to verify or falsify ever as well?

If you mean to say that society is a construct, well, yes. You learn that in the 101 course.

That does not lead to a "this is bullshit" conclusion, unfortunately, because we act on the stories we tell ourselves.

There are different levels of "construct" though. There are historical documents from the middle ages, from Rome, from ancient Egypt, etc. that can be checked to a certain degree. And there are stories that are just made out of thin air about how humans supposedly lived 50k years ago. This is not all the same.

There's a concept in electrical engineering and signal processing called superposition. It basically means that a linear circuit will process the sum of a bunch of signals the same way it would process them one at a time (in very rough terms, to the best of my recollection, etc.).

Maybe something similar applies to selective pressures; at a statistical, population-wide scale, the result of optimizing thousands of traits at the same time is mathematically equivalent to optimizing one at a time?

> All these "evolutionary arguments" -- this isn't real science. It's just a bunch of "just so" stories that are tested in some of the flimsiest ways

That's pretty much the process of generating a scientific hypothesis, so that description itself isn't the not-science part. The not-science part is presenting such stories as conclusions and leaving them set.

That is true but not for species that reproduce sexually. This is precisely why sexual reproduction is so successful.