Chicken and the Egg, for 5-year-olds | Hacker News Reader

Chicken and the Egg, for 5-year-olds

1 pointtrashtester12 comments

My 5-year old daughter asked me a few days back, how the fist human came to be.

I gave the Darwinian answer: "She (aribrary gender) was born from an ape mother."

Today, she came back, after obviously having thought about this, and asked: "Who was the father of the children of the first human?"

Now this was more difficult. We could just say that anyone born to her species born after day X were human, and those before were apes. But if we chose to use genotypes or phenotypes to determine this, "she" may very well have mated with someone just short of meeting the criteria, and may also have had children that qualified as "apes".

What do I answer the 5-year-old?

View on HN

Ah. Yeah, that's a tricky one.... The trouble is that it all boils down to the concept being 'turtles all the way down'when followed to it's logical conclusion.

Think of it this way - what created the big bang and what was there first (since things cannot form if there is 'nothing' so it must therefore be 'something' (for whatever any definition of "something"). But then - where did the 'something' come from.

Good luck with explaining that to your 5 y.o. and I'm glad that she is curious enough to ponder such things in an analytical way!

Btw - If you figure it out, let us know ..... because after over 50 decades of life I still, on idle days, ponder the 'What created the 'something' from the original 'nothing'.

In other words – If it is indeed ‘turtles all the way down’ then… what made the original turtle? And what made the 'thing' that made the turtle

It's a problem of classification. Imagine you have one of those machines with a red faucet with hot water and a blue faucet with cold water. Then you can classify water as hot or cold and be happy. Until you discover that the temperature of water can be in between. Anyway, if one of your coworkers are going to the machine, you can ask for a glass of cold/hot water and your coworker will understand. It's a useful classification.

Something similar happens with Australopithecus vs Homo. The discussion is easier because Australopithecus are dead. The latest Australopithecus were very similar to early Homos. The threshold is arbitrary, like the threshold between hot and cold water. (Apparently transitions of species are very fast, so it's difficult to find fossils of the intermediate steps. I'm not sure if this is confirmed.)

So if you take a time machine and go to see the transition, you will find some individuals that are difficult to classify. You will not be able to be sure which one is a Australopithecus and which one is a Homo. Perhaps ir you kill them and cut them in pieces, you can apply some arbitrary threshold like 537.2345 cm^3 of brain, but they will not be happy. Moreover, they will not be able to be sure which one of them is a Australopithecus and which one is a Homo, assuming you can explain the arbitrary threshold to them.

I guess it depends on how you build the phylogenetic trees at the individual level.

1) The fist Homo mated with a Australopithecus and some of the children were probably Australopithecus and some Homos. I guess two Homos can mate and have a children that is an Australopithecus. If you mark Homos with a magic invisible painting, they will be mixed with the community and be treated as a normal member of he community and mate with whoever they can. After a while, most will be Homos, but the few remainder Australopithecus will also be just normal members of the community and mate with whoever they can.

Note that in some parts Australopithecus lived for a longer time, until the Homos from other areas arrived. In the early year we probably ignored the difference and mated. Later when the difference were bigger we probably killed them or forced them to go away.

2) You go and find the first Homo, and define that all the offspring are Homo and it doesn't matter with who they mate. So the first one must mate with an Australopithecus and repeat that during a few generations until the they have cousins far enough to mate. Note that they will not realize the transition. It's just a different way to use the magic invisible painting, but they will interact exactly like in the other method.

species A becomes subdivided into 2 variants Ax, Ay those being the same species, can mate (think for instance, citrus variants)

but somehow they get separated, and start evolving slowly different, and here's the crux, at some point the genetic changes no longer allow viable reproduction should they mix again, at that point they have evolved into separate species

it's not that Humans (Ax) came from ape (Ay) but rather they evolved out from some common ancestor (that may or may not still exists today)

when they say "humans descend from apes" does not necessarily mean from actual current modern apes

and the flaw in the child's reasoning is that presupposes a linear change "from this point on, we have humans, before, we have apes)

that's enough simplification for a 5 year old?

> it's not that Humans (Ax) came from ape (Ay) but rather they evolved out from some common ancestor (that may or may not still exists today)

She knows this already, I did show here pictures of Australopithecus(ape) and Homo Habilis (human). She knows that the mother of the first human was something like an Australopithecus, while the first human was something like Home Habilis.

The problem is a logical one. Granted that one Homo Habilis was born from one Australopithecus, who would she mate with to have offspring? Would she commit bestiality with some Australopithecus of her tribe, or did in fact TWO such homo habilis suddenly appear in that tribe in the same generation? (And become Adam and Eve, or something.) The latter would, by common sense, seem to be extremely unlikely.

I know the real science behind it. Clearly, the problem is to think about a species as a category that is invariant across time. The real answer is that "The category 'human' is not well defined over evolutionary time, so your question is absurd."

That is good answer to a 15-year-old. But I was struggling to come up with a good explanation that would extend my methaphor in a way a 5-year-old might understand.

With all due respect. I can’t make sense of this.

Apparently, you have to be 15 to imagine gradual transitions between things (slush must be mind blowing for the little one), but 5 year olds are insisting you give thorough answers to questions about rarely-occurring abstract moral concepts like bestiality?

I know a few five year olds, and I can’t imagine then asking questions like that or caring how sensible the answer is if they did. I guess their parents just haven’t been teaching them the things you’ve been teaching yours?

> Apparently, you have to be 15 to imagine gradual transitions between things (slush must be mind blowing for the little one),

She is unable to grasp time horizons over millions of years, and tell them apart from intuition that at best goes to 3 digits. Visualizing that the transition between Australopithecus and Homo Habilis as incremental steps over hundreds of thousands of years, I presume, is too hard for her (or for my ability to explain). Maybe she can do it before 15, but not yet, I think.

Anyway, fact that we use taxonomies in the first place, is part of the problem. As long as we insist on giving names to ancient life forms, this type of question comes up by itself.

> but 5 year olds are insisting you give thorough answers to questions about rarely-occurring abstract moral concepts like bestiality?

Not explicitly. She is not aware of the concept. I suppose that's what she was asking. But obviously, it is implicit in the question, don't you think? If we insist in defining an individual as belonging to either Australopithecus or Homo, there must have been a time when someone we label as one mated with someone we label as the other.

Obviously, I didn't give her that answer. But I did think it.

> But obviously, it is implicit in the question, don't you think? If we insist in defining an individual as belonging to either Australopithecus or Homo, there must have been a time when someone we label as one mated with someone we label as the other.

It’s implied that there must have been mating, but you’re the one deciding to complicate that with moral language like bestiality. If a horse and donkey, lion and tiger, dog and wolf, or two formally distinguished finch species mate, we just acknowledge it as two compatibly close species mating.

You seem to be projecting your own sense of inviolate categories (maybe you imagine one being divinely ensouled and the other not?) instead of sticking in what the evidence speaks to directly. But that’s a different philosophical dilemma to reconcile.

> maybe you imagine one being divinely ensouled and the other not?

Not at all, I'm a pretty hardcore atheist.

> projecting your own sense of inviolate categories

Now, this is a good point. Even if I do not seen these categories as inviolate myself, mating between members of Genus Homo (aka 'Humans') and other animals is a social taboo, and punishable by up to three years in prison where I live.

> If a horse and donkey, lion and tiger, dog and wolf, or two formally distinguished finch species mate, we just acknowledge it as two compatibly close species mating.

All of these examples produce offspring that are not usually able to reproduce in the wolf, except dog and wolf who are both of the species Canis Lupus (with dog being the subspecies Canis Lupus Familiaris).

I don't have a problem with the real science behind all of this. Indeed, I don't even think it makes sense to define a "first human", as every member of her tribe, both before and after should be categorized in the same category as her.

Previously, I've been of the mind that, for a question like "Who cam first, the chicken or the egg?", the answer was obvious. Regardless of your definition of "Chicken", it did come from an egg. So the egg must have come first.

Let's compare this to the difference between the colors "red" or "orange". Anything of wavelength above 780nm is considered red, anything below is considered orange. So let's say a "generation" of colors is 0.00001nm then 780.000005nm would be read and 779.999995 would be orange. Now, using the category definition strictly, we can make this distinction. Even if the bird that lay the first egg that would become a chicken were almost identical to the first chicken, we could chose a defintion that would make the mother belong to the category of not-a-chicken.

But today, after thinking about this, it is a mistake to accept the question as well formulated. The term "chicken" is a social construct. At any point of time, a species can refer to a group of animals that would belong to the same species, had they lived today. Also, they would belong to the same species as their parents and grandparens at least a few generations back, probably hundreds if generations or more.

In other words, my current opinion is that it does not make sense to make a definition of human vs ape very sharply. Instead, we need to accept that a spectrum exists between fully human and fully ape.

Even if we choose to define a typical Homo Habilis as "fully human", there must have been their ancestors must have been near the middle of that spectrum, and even if we COULD define a set of criteria that would sharply split individuals between Homo Habilis and Australopithecus, that over some time period, proto-Homo animals must have been somewhere on that spectrum. (It has to be a spectrum, as just adding more categories would just move the category split somewhere else.)

it's not a "sudden" change there's no "first chicken", no "first human", no "last ape", no "first human that mated with last ape" either

just gradual diverging changes , and then at some point one branch is different enough from other branch that we call them separate species

so the answer to your connundrum is explaining your kid this, not explainign that a human mated with an ape

(ELI5)

So yes, I suppose my error was to assume that regardless of what definition you use for the category "Human", the mother of the first human must have been an ape.

As I wrote in the other subthread, is that I should not have made the assumption that it doesn't make sense to make ANY defintion of "Human" sharp enough that a child belongs to the category and not the mother.

In other words, the category "Human" has no sensible definition, and any attempt to make up an arbitrary one leads to such absurdities.

Transferred to the chicken-and-egg problem. I've always held that it is clear that the egg came first, as regardless of your definition of "chicken" it came from an egg. But since I now belive that there cannot be a sensible definition of "chicken", the question turns from having an obvious answer to a question based on an absurd premise.

Thx for pointing this out to me.

First, let's be clear that the "ape mother" answer implies a distinction between humans and other apes, which is not true. We are all born of ape mothers, because humans are apes. Another issue is that "human" can refer to the Homo sapiens species, or it can also refer to any species in the genus Homo, depending on context and preference.

With that out of the way, the biggest problem here is that "species" don't actually exist. There are several different definitions, and they each have arbitrarity and/or striking exceptions. Species are just an intellectual tool humans invented to help us organize our understanding of how different living things are related to each other, but evolution is gradual and messy and doesn't care much for clear separations.

So the answer to "who was the father of the first human mother's children" depends on which definition of "species" and which definition of "human" you want to use. If you go with the common idea of "two things are different species if they can't have fertile offspring together" (and ignore all the asexual reproduction and fertile hybrid exceptions), and "human" meaning just Homo sapiens, then... well, you're screwed, really. Because by that definition, the only possible father of the first human's children would have to be one of their siblings, as they definitionally couldn't reproduce with any other existing creatures.

I'm not ruling out that such inbreeding happened; however... as I said, to use that definition, you need to ignore fertile hybrids, which is a problem because we know Homo sapiens interbred with other Homo species -- we have the DNA in ourselves to prove it. We are all fertile hybrids in that regard. Meaning the actual father would probably have been a pre-sapiens human species. So perhaps you'd prefer to use the morphological definition instead: "humans are the creatures who look most like modern humans". Obviously, that gets messy and arbitrary, too, because different humans look different, and finding a "different enough" threshold is more preference than anything else. But ignoring that problem, the answer becomes the same: "the father of the first modern-looking human was a species that didn't look modern".

Same goes for the genetic definition of species as well; the first human whose DNA was "different enough" for us to call it a modern human had children with someone whose DNA was "different enough" to call it "not a modern human".

In the end, the answer is approximately the same: the father of the first modern human would have been a different hominid species.

> In the end, the answer is approximately the same: the father of the first modern human would have been a different hominid species.

The question was not about the father if the first human (regardless of how you define human, as long as you use a sharp definition). I had already told my daughter that the mother (and implicitly father) of the the first human were belonging to the prior species in the chain. (For me "human" was used as Genus Homo, but it doesn't matter if one uses homo sapiens sapiens instead)

The difficulty of the question came from my daughter asking who would be the father of the CHILDREN of the first human (assuming the first human was female). Either that male would have to still belong to the previous species, or have been born independently of her as human, somewhat after.

Given that the second option is very unlikely. If we set sharp enough criteria for qualifying as human to be able to separate different members of the tribe these individuals lived into either human or non-human, the probability of the first two who qualified would arrive so close in time and independently, is probably near zero.

In other words, we must assume that if we insist on splitting indivuals into the categoris of "human" and "not human" at that specific point in time, most likely the father of the children of the first human must have ben non-human. Also, it would be very likely that even the children may have been non-human (or it may be tha the first "human" didn't have any children at all). In fact, it could be a 100 or a 1000 years before another individual was born that qualified according to whatever definition of "human" we pick.

And this is when I realized that:

> the biggest problem here is that "species" don't actually exist.

Not only is any definition of "human" in some way arbitrary, the very act of creating such a definition leads strange contradictions such as above.

Intuitively, all members of the tribe that brought us the first humans, would have appeared to us as all being EQUALLY human, had we had some way to observe them directly. To draw some red line between individuals of that group makes no sense. (In fact, being in the same tribe, these individuals would be closely related, and so much more similar to each other than two randomly selected modern humans living today would be.)

Also, it doesn't help to just create a new category, since the problems at the boundaries will be precisely the same.

In other words, the only thing that makes sense, is to not assign categories of human vs non-human to these individuals at all. The closest thing that would make sense would be to assign some "degree of human-ness".

Let's say that the first individual with such human-ness would have a human-ness score of more than 50% would have a score of 50.1%, while the average for the tribe would be 49.7% at the time. By thinking of this transition as a real (as in floating point) number, instead of as a boolean, the problem goes away.

Reply on news.ycombinator.com