Hacker News Reader

Comment by thyrsus | Hacker News Reader

Good science accepts all non-fraudulent evidence. In the case of evolution, we're dealing with large, highly variable populations. Thus, no individual observation will falsify the theory. Evolution is almost tautological in predicting that traits which increase the likelihood of reproduction in a population will become more common in that population. To falsify evolution, you would have to observe a population in which traits that increased the likelihood of reproduction became less common.

There is an important subtlety with social groups: a trait that decreases the likelihood of an individual's reproduction may increase the likelihood of the group's reproduction. For instance, worker ants have near zero probability of reproducing, yet they highly increase the likelihood of the group's (colony's) reproduction.

View on HN

I think saying "evolution theory" is very vague. There is the theory of common descent. That is what is being supported here: http://www.talkorigins.org/faqs/comdesc/

But claiming that mutation and natural selection alone explains ALL speciation and ALL behavior that we observe today is just a simplistic position, almost an article of faith. How do you know there is no other factor involved? How do you know no other process WAS involved in the past?

We don't. So far as I know, we have never observed macro evolution taking place in recorded history (although I would be happy to be shown wrong on this point), so what actually causes speciation is just speculation at this point.

Common descent is one thing. "Evolution" is a loaded term. What does it mean?

It's a bit like the people who claim that the Bing Bang Theory describes an explosion in the first seconds of the universe. Actually, the BBT just extrapolates backwards from what we observe today -- which is that the visible universe is expanding, and there is microwave radiation at its edges. How do we know there wasn't some other process earlier in the real universe, which would make our extrapolation completely erroneous? How do we know that the universe isn't infinite beyond our visible universe? Every year objects disappear further beyond the light horizon, due to the expansion of space. Who knows hoc much lies beyond that? And yet people talk about "the universe had a beginning" ... how do they KNOW?

Okay I just wanted to throw a crazy idea at the end: Maybe our universe is like a reverse black hole, with a singularity at the edges instead of its interior. Meaning that just as gravity increases as you approach the singularity in a black hole, so the universe expands faster the bigger a "sphere" you take in it.

No on many counts. Evolution theory is an umbrella term for a number of predictions including common descent, mutation, natural selection which is sufficient in theory and observation to create genetic and then physical changes in a population.

We have observed macro evolution countless times in bacteria and viruses. It's not even difficult.

I'm not sure why you'd call Evolution a loaded term in a scientific domain. It's clearly loaded in politics these days, but it's a very clear and celebrated theory with sharp and validated predictions in scientific domains.

You're somewhat right in the statement that BBT extrapolates from data observable today, but due to the nature of the universe, data observable today provides evidence that is not fully localized in time. We believe that the universe is infinite beyond our light cone because if we make the minimal possible assumptions on the net geometry of the universe you end up with three choices, each having to do with the sum of the angles in a very large triangle. We measured this triangle (see WMAP) and found the angles to be very nearly 180 degrees corresponding to a flat, infinite universe. So long as we have no compelling reason to invent more exotic geometry (which if GR holds would require some rather incredible momentum-energy that we haven't observed --- all possible, but very, very exotic) then this is a safe prediction.

BBT theory doesn't even describe an explosion anywhere. There wasn't something to explode in to. It instead suggests that the invariant metric has been expanding throughout the lifetime of the universe making things grow further and further apart. It's unclear that there wasn't more exotic behavior in there again, but without evidence of it, the best prediction is the simplest prediction.

I could go further into how we KNOW things and why that's not really a great way to think about it, but this is already grown long. But if you want your crazy idea to be meaningful at all, please specify it formally enough so that it predicts something about the shape or behavior of the universe different from current theories and then see if that prediction holds. Without at least claiming a prediction, it's difficult to even begin to understand what a crazy idea might mean.

1) re evolution: when I say it's a "loaded" term, it's because it is sometimes by people who seem to make the additional assumption that mutation and natural selection are sufficient to produce the results we see today. I think that assumption is unfounded and seems to be more like a wish to keep things simple. I don't for example see how gay people could be genetically predisposed to be gay, since they would have a reproductive fitness level far lower than others. But that's not even the biggest thing. Little mechanisms which could hardly make a difference in survival one way or the other are doing an amazing job. Maybe it's sexual selection. But where is the evidence that all that we see around us was produced with simple mutation and natural selection?

Regarding bacteria and viruses: that's micro-evolution, isn't it? Why is it macroevolution?

2) Regarding the BBT: so why do people think there was a "big bang" and talk about the "first few seconds of the universe" if the universe is probably infinite, and it is only the visible universe that we see expanding?

3) I don't know, I'm not a physicist... seemed like an intriguing way to think about the expansion though

Genetic algorithms are a pretty simple to replicate demonstration that mutation and selection together constitute a sufficient condition for descent with modification.

Evolution doesn't claim to explain all changes, just the "force" which causes populations to grow to resemble the units of the population which are best selected for in later generations. It, for instance, doesn't explain eye color either; there isn't sufficient reproductive pressure to affect those aleles. Eye color requires understanding of DNA and the structure of our genetics to begin to make sense of.

Microevolution and macroevolution are the same thing at different (orthogonally arbitrary) scales. The appearance of a significant difference occurs under strong selective pressure or differential selective pressure leading to punctuated equilibria or genetic splitting, but really it's just a matter of sharp changes of a smooth parameter instead of actual categorical differences.

---

Because there were presumably first few seconds of the universe. The cosmic microwave background radiation is a picture of these first few seconds when everything was closer together and thus interacted strongly! The resolution to this conundrum turns out to not be tied to the finiteness of the universe but instead the astounding fact that if you fix two objects in space the distance between them will still grow (very slowly) with time.

---

I'm not a physicist either, so I avoid seriously proposing strange theories of cosmology. It's gotta be pretty equivalent to having a manager ask for some new feature assuming it'll be trivial to build.

I gave up arguing with creationists long ago, but on the off chance you actually do want to improve your understanding: http://en.wikipedia.org/wiki/Macroevolution#Misuse

Yeah I read that already, but what I said doesn't necessarily imply creationism. What arguments do you have to show that everything we see today came about SOLELY through mutation and natural selection?

It's the true form (generalization) of the old "new proposals require evidence" catchall. The space of possible ways things could happen and be in accordance with our ignorance is enormous. If you want the best chances of being right, you keep you guesses maximally simple!

If you create a more complex model that includes more factors than mutation and natural selection and it is able to predict reality better than one that only includes those factors then you win. As it turns out, people have by using more complete understanding of the actual mechanics of human genetics, but a rough and basically accurate picture can be painted without invoking those mechanisms.

That "mostly accurate" picture is where most people stop when they argue "for evolution". It's a fantastic rule of thumb.

What makes it fantastic, though? What is the hypothesis / theory exactly that they are espousing, when they talk about evolutionary psychology for example?

Isn't it the fact that our behaviors all have an evolutionary basis, including e.g. the desire for a human male to sleep after sex? Each of these gets its specially concocted rendition of the narrator's fantasy -- e.g. "so that the women could get away more often after copulating".

I find this to be completely unscientific and wonder what falsifiable predictions these theories make. I mean the ones that imply that everything can be explained simply in terms of "evolution" by mutation and natural selection.

Nobody ever claims that evolution explains everything. The claim is something closer to "Of all theories of equivalent or lesser complexity as evolution, no other one yet discovered has better predictive power".

I'm not sure of my scientific history here, but I believe one extraordinary falsifiable prediction drawn from evolution in its earlier days is that there should be some common structure of generational information between almost all living things, a prediction which the discovery of DNA clearly validated.

Unless you have some truly spectacular competing theory, with plenty of hard evidence to back it up, there is only one conceivable reason for your attempted nitpicks at evolutionary biology and that is religious conviction. I'm pretty sure you don't have the former, so changing away from this very repetitive channel..

I don't know why you'd think religion is the only conceivable reason. I think both religious people and others start to see enemies where there aren't any (I guess it's an evolutionary psychology instinct, right? Because the human brain evolved to find tigers better, at the expense of false positives, etc.)

To be honest I didn't remember a lot of my gripes off the bat when I wrote it, but a big one is the gratuitous use of evolutionary psychology by all armchair theorists and their book writing moms these days. How is that religious? They take any sort of human behavior and start speculating evolutionary reasons for it purely in terms of reproductive fitness, because in essence they are saying it all boils down to mutation and natural selection + it must be super simple. WHY?

Science doesn't work this way. "Unless you have some truly spectacular competing theory, I'm just gonna use empty speculation to explain why people have non-advantageous trait X, because we know that mutation and natural selection take place". It works by making falsifiable predictions. What falsifiable predictions do armchair speculators about tiny human behavior make? Do they test them?

http://www.youtube.com/watch?v=zv7b-KPg9hY :P

Here is one that I remember was posted by a friend of mine on facebook. Note the empty speculation at the end. Really, narrator? Just because you have a British accent we should implicitly accept that assertion?

This is a shortsighted and reactionary post: DH1 or DH2 at best [1]. I understand your frustration, but if you only have frustration to contribute then HN prefers silence. It's the path to maintaining a respectful community.

It turns out, unsurprisingly, that all human phenotypes cannot be explained by naïve evolutionary biology but instead require more genetics and biochemistry to describe. Consider the specifics of eye color: evolution does not specify this feature, so a theory which goes above evolution (not contrary to what EGreg is asking for) is necessary to predict this phenotype. We're just more complicated than that.

This is a rebuttal to using pure evolution to model the shape of human physiology that doesn't introduce religion.

[1] http://www.paulgraham.com/disagree.html

You're right, I shouldn't have taken the bait. Entirely deserving of the downvotes ...

Ever heard the reproduction rates across the western world are dropping off the charts? Oh, did I just falsify evolution?

Reply on news.ycombinator.com