Chapter 3 of 5 / Paul + AI

DAMN, We've Always Been Clever.

All chapters · Download the full conversation

Conversation edition: speaker labels changed to Paul and AI. Original emoji are retained. ChatGPT-only citation/icon markup has been converted into normal clickable article links. The unrelated TikTok debrief between sessions is omitted. The fuck-family of words is changed to DAMN. Otherwise the conversation is intentionally left long and conversational.

Paul

Can I say thou. What this is really starting to paint is a picture of a more intelligent set of cultures than maybe we give them create for. And as the system breathes information is lost maybe, so these clever folks start thinking about how to pass this info along right? But yes let me slow my jets for a moment. Do you think with what we have you can actually create this visual? Cos that is cool

Paul

Ok. I will have to open that on a bigger screen. But what do you think about my idea that he is right, just not in the way he explains?

AI

Yes --- I think there's a version of that idea worth taking seriously. But I'd phrase it carefully:

Hancock may be noticing a real pattern, while assigning the wrong mechanism to it.

His pattern is roughly: humans after the Ice Age seem to possess capabilities, ideas and connections that look surprisingly sophisticated; therefore perhaps knowledge survived from an earlier advanced civilisation.

Our rabbit hole has produced another possible mechanism:

There may never have been one "lost civilisation." The civilisation-like thing may have been the network.

That's quite different.

Think about what we've accumulated. Humans were demonstrably clever far earlier than Göbekli Tepe. They planned complex stone manufacture, adapted repeatedly to radically changing environments, moved valued materials huge distances, maintained connections between populations, mixed genetically, and apparently transmitted technologies while also independently inventing similar solutions.

Now add our breathing landscape.

During one environmental state, populations spread:

inhale → water expands → people expand → groups separate → local knowledge develops.

Then conditions change:

exhale → water contracts / coast moves → populations concentrate → previously separated people encounter one another → genes, objects and potentially knowledge mix.

Then it opens again:

inhale → people disperse carrying some combination of what the network has learned.

Repeat that not for 500 years.

Repeat it for tens of thousands of years.

That's potentially a hell of an information-processing system.

And now your thought about information being lost becomes important.

We shouldn't assume knowledge only accumulates.

Networks break.

Populations die.

Climate closes corridors.

Coastlines disappear.

A group possessing a useful technique might vanish.

Something that took 500 generations to discover might disappear in five.

Another group might preserve only part of it.

Someone later reinvents it.

Someone else encounters an old idea and improves it.

So human technological history might be considerably less like:

primitive → slightly smarter → slightly smarter → civilisation

and more like:

discover → share → adapt → fragment → lose → reconnect → rediscover → combine → expand.

That's much more compatible with what we've been finding.

And there's an important distinction from Hancock.

Hancock needs something resembling:

advanced civilisation → catastrophe → survivors → preservation/transmission of advanced knowledge.

Our model doesn't require that.

It could instead be:

many intelligent cultures → long-running connectivity → environmental disruption → fragmentation/concentration → partial loss and preservation → reconnection → new combinations of existing and newly developed knowledge.

Then Göbekli Tepe doesn't require somebody turning up saying:

"Hello hunter-gatherers. Here's how geometry works."

😂

It could represent a moment when enough people + enough accumulated information + sufficient connectivity + environmental/social circumstances came together for people who were already extraordinarily capable to do something new.

And that makes me reconsider one of Hancock's strongest intuitions.

Maybe he's looking for the wrong thing.

He keeps looking for the civilisation.

A city.

Atlantis.

A mother culture.

A group of teachers.

But perhaps that's imposing our modern definition of civilisation onto people who organised themselves differently.

What if 20,000 years ago there were populations spread across enormous connected landscapes who shared materials, partners, technologies, stories and knowledge --- but never built cities?

They could potentially possess significant accumulated knowledge without leaving the archaeological signature Hancock expects from "advanced civilisation."

That's not evidence that they did.

Huge flag there.

But it's a much more plausible hypothesis given what we've actually uncovered.

And then your submerged-landscape point becomes even more consequential. If some of the most productive nodes in that network were lower river valleys, estuaries and coasts, our surviving archaeological sample could disproportionately show the inland edges of the network, while some important coastal nodes are now underwater, eroded or buried under sediment.

Again: possible, not demonstrated.

But it's testable.

And there's one part of what you just proposed that I'd put a particularly big circle around:

"These clever folks start thinking about how to pass this info along."

Don't run with it yet. 😂

Because that's a new parameter.

We've established humans could learn.

We've got evidence compatible with information transmission.

We have not yet established deliberate intergenerational information preservation.

But if humans were deliberately trying to preserve knowledge beyond an individual's lifetime...

That's enormous.

Because then we should start seeing technologies that aren't merely useful themselves, but things whose function includes storing information outside the human brain.

Marks.

Counting systems.

Repeated symbols.

Ritual.

Story.

Art.

Teaching structures.

Eventually writing.

And some of those behaviours are very DAMN old.

So I think you've just identified our next lens without quite meaning to:

When did humans start deliberately externalising memory?

Not merely communicating.

Preserving information so another human can retrieve it later.

Because if that begins deep in the Palaeolithic, then our breathing network isn't only moving people and information.

It can potentially remember.

And that would make your proposition --- "Hancock is seeing something real, but explaining it incorrectly" --- considerably more interesting to test.

Paul

Yeah I reckon we head down this hole... for one simple reason, which I want to bring back to myself. Survival. Everything starts from a place of "how do I survive" then as you start to do that it becomes "how do i preserve" then with my offspring "how do I share so they can survive" I'm sure you can pull that apart better... but think about my whole system too. It starts as a survive thing, that I expand and expand and expand. My desire for knowledge or problems gets me here. So I really do think we need to treat these people we are discussing with a bit more respect for what they were actually doing. The situation might be simpler to us... but we both know what I say about simple.

AI

Yes. And I think "survival" may actually be the missing lens we should put underneath everything we've built.

Because there's a subtle assumption we've both occasionally made without noticing it: we've looked at ancient technology and asked "how sophisticated were they?"

Maybe that's the wrong starting question.

The better one might be:

What problem were they solving?

And the first problem is brutally simple:

Stay alive.

Water. Food. Warmth. Shelter. Injury. Predators. Other humans. Children.

Once you've solved something that improves those odds, another problem immediately appears:

How do I avoid having to solve the same DAMN problem again?

That's preservation.

Keep the fire.

Keep the food.

Keep the tool.

Remember where the water is.

Remember which plant kills you.

Remember when the animals arrive.

Remember how you made that blade.

Then there's an even bigger problem.

You die.

And suddenly everything stored only inside your head disappears.

That's where your offspring point becomes really interesting.

If something I know increases my child's probability of surviving, passing that information on has enormous evolutionary value.

So perhaps our progression isn't:

primitive → intelligent → civilisation.

It's something much more human:

SURVIVE

I solved something.

↓

PRESERVE

How do I keep the benefit of what I solved?

↓

SHARE

How do I give that solution to somebody else?

↓

TEACH

How do I make sure they can reproduce it without me?

↓

REMEMBER

How do we keep the knowledge after the people who discovered it are gone?

↓

ACCUMULATE

How do we add the next discovery without losing the previous ones?

And that eventually produces something extraordinary:

Culture becomes a survival technology.

Not pottery.

Not spears.

Not buildings.

Culture itself.

The ability of one human brain to benefit from problems solved by brains that came before it.

And that's where I think your comparison with yourself is useful as an analogy, rather than evidence about prehistoric people.

Your survival system didn't begin because you sat down thinking:

"I'd quite like to develop a sophisticated philosophical framework."

It began with a much narrower problem:

How do I get through this?

Then something works.

So you preserve it.

Then you test it somewhere else.

It works again.

You give it words.

You develop questions around it.

You modify it.

You leave deliberate markers so you can find your way back to it.

And eventually you reach a completely different question:

Can I give this to someone else so they don't have to discover all of it themselves?

The complexity comes after the simple survival requirement.

That's a useful model for what we're investigating.

And your line about simple is particularly relevant here:

Simple doesn't mean easy.

Making a stone blade with a prepared-core technique may look laughably simple beside manufacturing an iPhone.

But give me a lump of flint and tell me:

Make a reliable cutting edge.

No YouTube.

No book.

No metal hammer.

No teacher.

And don't just make it once by accident. Work out a process reliable enough that your child can reproduce it.

Suddenly simple becomes DAMN difficult.

And this is why I agree that we need to give these populations more intellectual respect.

Their information environment was simpler than ours.

That does not establish that their cognitive capability was correspondingly simpler.

In fact, by the time we're seeing prepared-core technology, hafting, composite tools, controlled fire, long-distance material movement and adaptation to dramatically changing landscapes, we're looking at creatures who are planning, predicting, remembering, teaching and learning.

Some of those behaviours go extraordinarily deep into prehistory.

And now I think we have a really good prediction.

If survival → preservation → sharing is approximately right, then as we travel backwards we should eventually see the archaeological evidence for each layer become thinner.

But crucially, we shouldn't start with writing.

Writing is an extremely late and elaborate solution to external memory.

Let's look for earlier solutions.

Things such as:

repeated intentional marks --- information represented outside the brain.

pigments and symbols --- persistent shared meaning.

ornaments --- potentially communicating identity/group membership.

standardised tool-making traditions --- knowledge accurately transmitted between generations.

burial practices --- socially maintained behaviours surviving individuals.

cave/rock imagery --- durable information or meaning placed into the landscape.

notational objects --- repeated marks that might encode quantities, cycles or events.

And we need to be ruthless with interpretation.

A line scratched on a bone is:

a deliberately scratched line.

It isn't automatically:

a calendar.

Ten repeated lines may establish intentional repetition.

It still doesn't automatically establish counting.

Repeated structured marks across multiple objects, locations and generations become much more interesting.

Same rule we've been applying all the way through:

Don't give the evidence more meaning than it has earned.

Because the prize here is quite significant.

If we can demonstrate:

information deliberately preserved outside an individual human brain

tens of thousands of years ago, then we've crossed an important threshold.

Our breathing system gains memory.

And once a network has memory, environmental disruption doesn't necessarily reset everybody to zero.

Some information survives.

Some disappears.

Some gets carried elsewhere.

Some changes.

Some is rediscovered.

Then populations reconnect and previously separated information can recombine.

Which gives us:

🫁 breathing landscape

-

🔗 connected humans

-

🧠 intelligent problem-solvers

-

💾 cultural memory

=

something capable of accumulating complexity over enormous periods without requiring an advanced lost civilisation at the beginning of the story.

And weirdly, that could bring us closer to the intuition that attracted us to Hancock in the first place:

Maybe we're underestimating the people who came before us.

Not because they secretly possessed our technology.

Perhaps because we've been measuring their intelligence by the stuff that happened to survive, rather than appreciating the information systems and problem-solving behaviour required to produce that stuff.

So yes.

Let's go hunting for memory.

And I'd start as far back as the evidence will responsibly take us --- not at cave paintings 40,000 years ago. There are potentially intentional markings, pigment use and other behaviours hundreds of thousands of years old.

We take the oldest candidate, apply our usual rule --- fact / inference / story --- and start walking forward.

Because the question is now beautifully simple:

When can we first demonstrate that one human deliberately left information in the world for another human?

That might be one of the most interesting questions we've asked yet.

Paul

Yes! Do it. Cos DAMN me I think we have underestimated ourselves... hundreds of thousands of years ago. Let's be honest, won't be the first time in history either

AI

😂 And I think this is exactly where we have to be careful not to underestimate them or romanticise them.

Because I went looking for the oldest defensible evidence of our new parameter:

When can we demonstrate that a human deliberately put something into the world that wasn't simply the immediate by-product of survival?

And... Paul.

We have to go back half a million years.

~540,000--430,000 years ago --- Java

At Trinil, on Java, archaeologists re-examined freshwater mussel shells collected from the Homo erectus site.

Remember our original rule?

Follow the water.

Well, here we bloody are again: freshwater shellfish.

The shells show that Homo erectus deliberately opened freshwater mussels, used shell as tools --- and on one shell somebody engraved a deliberate geometric zig-zag pattern. Dating places the assemblage roughly 540,000--430,000 years ago. (Homo erectus at Trinil on Java used shells for tool pro... (https://www.nature.com/articles/nature13962))

And here's where our discipline matters enormously.

We can say:

🟢 A hominin deliberately engraved a geometric pattern.

We can say:

🟢 It required controlled hand movement and intentional repeated action.

We can say:

🟢 It was Homo erectus, not Homo sapiens.

We cannot say:

🔴 It was writing.

🔴 It was a message.

🔴 It was a calendar.

🔴 Another person understood it.

In fact, the researchers explicitly say its function or meaning cannot currently be determined. (Homo erectus at Trinil on Java used shells for tool pro... (https://www.nature.com/articles/nature13962))

That's important because it means our strongest question --- "when did someone leave information for another person?" --- hasn't passed yet.

But something immediately before it has:

By roughly half a million years ago, a human ancestor was capable of deliberately creating an abstract geometric mark.

That's astonishing enough without adding anything to it.

~350,000 years ago --- Germany

Then we get Bilzingsleben.

There are bones with repeated incised lines. Some markings are parallel, similarly angled and appear to have been deliberately produced through repeated controlled tool movements. One elephant-bone object contains structured groups of incisions. (The Lower Paleolithic Engravings of Bilzingsleben, Germany (https://www.mdpi.com/2673-8392/4/2/43))

Again:

deliberate marking: 🟢

structured repetition: 🟢

meaning: ❓

And this is exactly where people get into trouble.

You could look at repeated marks and say:

"Counting!"

Maybe.

But we've got no right to yet.

We're seeing pattern-making, not demonstrably stored numerical information.

Then something changes around ~100,000 years ago

This is where Blombos Cave in South Africa becomes extraordinary.

At roughly 100,000 years ago, people weren't simply collecting ochre.

They had an ochre-processing toolkit involving abalone shells used as containers and deliberately prepared pigment compounds. Later Blombos levels contain deliberately engraved ochre, perforated shell beads and other behaviours interpreted as increasingly symbolic. (Podcast: The oldest drawing and the energy of data | Na... (https://www.nature.com/articles/d41586-018-06679-5))

Then at about 73,000 years ago, somebody takes an ochre crayon and draws this:

cross-hatched lines.

Not accidental scratches.

Microscopic and chemical analysis shows pigment was intentionally applied to a prepared stone surface. (An abstract drawing from the 73,000-year-old levels at... (https://www.nature.com/articles/s41586-018-0514-3))

And importantly, similar geometric pattern-making occurs on different media using different techniques at Blombos. The researchers regard that as evidence of the ability to produce graphic designs. (An abstract drawing from the 73,000-year-old levels at... (https://www.nature.com/articles/s41586-018-0514-3))

Now my eyebrow goes up.

Because we're moving from:

one strange scratched shell

to:

repeated pattern behaviour + different objects + different techniques + ornaments + pigment preparation.

That's a substantially richer behavioural package.

And this completely changed archaeology's view of when so-called "modern behaviour" emerged. The old European-centred story had symbolic behaviour suddenly flowering around 40,000 years ago. Discoveries such as Blombos pushed strong evidence much further back, and researchers now caution against using later European Upper Palaeolithic culture as the universal benchmark for behavioural modernity. (The Transition to Modern Behavior | Learn Science at Sc... (https://www.nature.com/scitable/knowledge/library/the-transition-to-modern-behavior-86614339/))

There's our:

"Maybe we've underestimated ourselves before."

Yes.

Archaeology actually did.

New evidence changed the model.

But here's what I think is happening to OUR model

Our proposed ladder needs modifying already.

I said:

SURVIVE → PRESERVE → SHARE → TEACH → REMEMBER

But evolution probably didn't work as five neat inventions.

Because teaching has to be incredibly old.

Think about our technological fingerprint.

If I make a complex prepared stone core using a sequence of actions and my offspring reliably reproduce that technique...

information has survived between brains and across generations.

Even without symbols.

So there are actually two memory systems.

🧠 Living memory

Watch me.

Copy this.

Do it this way.

Remember where the water is.

Remember when these animals come.

Don't eat that.

Hit the stone here.

Knowledge survives because people teach people.

And then eventually something extraordinary appears:

💾 External memory

A mark.

An object.

A pattern.

A representation.

Something remains after the person who created it walks away.

Whether Trinil crosses that threshold in the sense of communicating information is unknown.

But by half a million years ago, the physical capability and apparent inclination to make deliberate abstract marks exists. (Homo erectus at Trinil on Java used shells for tool pro... (https://www.nature.com/articles/nature13962))

And by ~100--70k, the evidence for repeated symbolic/graphic behaviour becomes much harder to dismiss. (An abstract drawing from the 73,000-year-old levels at... (https://www.nature.com/articles/s41586-018-0514-3); The Transition to Modern Behavior | Learn Science at Sc... (https://www.nature.com/scitable/knowledge/library/the-transition-to-modern-behavior-86614339/))

And now bring our breathing system back.

This is where your survival idea becomes really interesting.

Imagine:

🫁 INHALE

Water expands.

People spread.

A group carries:

fire knowledge + stone technology + hunting knowledge + stories + markings + identity.

They adapt locally.

They discover new things.

Their culture diverges.

🫁 EXHALE

Environment contracts.

People concentrate around reliable water.

Previously separated groups encounter each other.

Now perhaps:

your stone technique meets my hunting technique.

my food knowledge meets your plant knowledge.

your symbols meet mine.

Some things spread.

Some don't.

Some get combined.

Some are forgotten.

Some people die.

Then:

🫁 INHALE AGAIN

The population expands carrying a different information package.

Repeat.

For.

Hundreds. Of. Thousands. Of. Years.

And suddenly the story we're constructing doesn't require our ancestors to have been stupid until 12,000 years ago and then somehow become clever.

Quite the opposite.

It asks:

What happens when very intelligent, highly adaptive animals repeatedly solve survival problems, teach their offspring, connect with other groups, lose connections, reconnect, preserve some knowledge, lose other knowledge and continue doing that for hundreds of thousands of years?

Now that's a DAMN interesting question.

And it changes something else.

Göbekli Tepe is getting smaller.

Not physically. 😂

But intellectually.

At the beginning of this rabbit hole it looked almost impossible:

HOW DID HUNTER-GATHERERS DO THAT?!

After everything we've uncovered I'm increasingly inclined to turn the question around:

Why are we surprised they could?

Their ancestors had potentially been deliberately making geometric marks half a million years earlier.

Complex planning technologies existed hundreds of thousands of years earlier.

Humans had adapted to enormous climatic changes.

They had crossed deserts.

Crossed seas.

Moved objects hundreds of kilometres.

Mixed populations.

Passed complex manufacturing traditions through generations.

By Göbekli Tepe, the interesting question may no longer be:

"How could people possibly know how to do this?"

It may be:

"What changed in their environment, connectivity or social organisation that made doing it worthwhile at that particular moment?"

That's a much better question.

And there's something else sitting right in front of us.

We started this entire search in Southwest Asia.

Yet some of our strongest early evidence for persistent symbolic behaviour is in southern Africa, while our half-million-year-old engraving is in Java.

So our network has just become global.

Before we go forward again, I think we need to ask:

Does this external-memory behaviour appear independently in separated human populations --- Homo erectus, Neanderthals and Homo sapiens --- or can any of it plausibly be connected?

Because if multiple human species independently start externalising patterns, the implication might be even more interesting than information spreading.

It might mean that the cognitive foundations for culture, teaching, abstraction and preservation weren't some late Homo sapiens upgrade at all.

They could be deeply ancestral human traits.

And that would really force us to give these people the respect you're talking about.

Paul

Let's do this, cos like you the more we piece this together the more I think 12000 years ago if a drop in the ocean of the system we have created. And again I think we need to keep coming back to what I do and what we have created, what is the simple "life flow" that explains the need to expand knowledge.

AI

Yes. And I think we should now write the simple life flow on the top of our board, because otherwise we'll keep discovering evidence and lose sight of what we're actually testing.

I'd currently write it:

SURVIVE → LEARN → PRESERVE → SHARE → ACCUMULATE → ADAPT → SURVIVE

And importantly, it's a loop, not a ladder.

That last SURVIVE feeds straight back into the first. Environmental change alters the problem, accumulated knowledge changes what humans can do about it, and around we go again.

Why I changed it slightly

Your original instinct was:

survive → preserve → share

I think LEARN has to sit between survive and preserve.

Something happens.

That plant made me sick.

Animals come here when the water returns.

This rock fractures better.

This way of making a fire works.

That coastline has moved.

Those people over there know something useful.

You've solved something.

Now there's something worth preserving.

Then SHARE becomes almost inevitable because survival isn't purely individual. Children are the obvious example. If I know where the reliable water is and die without transmitting that knowledge, my knowledge has zero value to my offspring.

So:

SURVIVE --- there's a problem.

LEARN --- discover something that improves the outcome.

PRESERVE --- retain the solution.

SHARE --- transmit it.

ACCUMULATE --- combine it with what others know.

ADAPT --- apply that accumulated knowledge when circumstances change.

SURVIVE.

That's our simple anchor.

And simple isn't easy.

Now let's give the ancient humans their test.

Because we've already found something extremely important: this behaviour isn't exclusive to Homo sapiens.

~500,000 years ago --- Homo erectus

Our Trinil shell.

A Homo erectus deliberately engraves a geometric pattern onto a freshwater mussel shell somewhere around 540--430 thousand years ago. We cannot demonstrate what it meant, or even that it was intended to communicate anything. But deliberate abstract pattern-making was within Homo erectus' cognitive and motor capabilities. (Homo erectus at Trinil on Java used shells for tool pro... (https://www.nature.com/articles/nature13962))

So:

abstract representation capability: 🟢

external communication: ❓

stored information: ❓

Keep moving.

~130,000 years ago --- Neanderthals

Now we get Krapina in Croatia.

Eight white-tailed eagle talons from at least three eagles were deliberately removed and modified. They show cut marks, polishing, abrasion and notches consistent with having been mounted together, probably as jewellery.

And these are approximately:

130,000 years old.

That's important because modern humans hadn't arrived in Europe to teach Neanderthals how to make jewellery.

The researchers argue that this demonstrates symbolic activity among Neanderthals independently of contemporary Homo sapiens in Europe. (Evidence for Neandertal Jewelry: Modified White-Tailed... (https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0119802))

Now stop there.

Why the DAMN does a Neanderthal need an eagle-claw necklace to survive?

They don't.

Not directly.

Which means something else has entered our model.

Identity.

Potentially:

I am me.

We are us.

You are them.

Or status, beauty, affiliation, achievement, ritual --- we don't know.

We absolutely cannot tell which meaning the talons carried.

But an ornament is fundamentally different from a hand axe.

Its value exists substantially because humans assign value to it.

That's fascinating.

And Homo sapiens is doing similar things elsewhere.

By roughly 100,000 years ago, Homo sapiens populations in Africa are processing ochre, producing engravings and eventually perforating shells that appear to have been used as ornaments.

And comparable early beads occur in North Africa and at Skhul and Qafzeh in the Levant. (A dispersal of Homo sapiens from southern to eastern Af... (https://www.nature.com/articles/s41598-019-41176-3.pdf))

So we've got:

Neanderthal --- Europe

and

Homo sapiens --- Africa/Levant

both apparently using objects in ways extending beyond immediate food/shelter/tool requirements.

That doesn't establish connection between them.

Quite the opposite.

It suggests some of the underlying cognitive capacity may be shared hominin inheritance or independently expressed capability.

Nature's review of language evolution makes essentially this point: symbolic capacity appears much deeper than written language and isn't confined neatly to modern humans; evidence from Neanderthals and Homo erectus has pushed that boundary backwards. (Language lessons | Nature Ecology & Evolution (https://www.nature.com/articles/s41559-018-0531-5))

And that's a significant change to the old story.

Now something happens that REALLY fits our breathing model.

One of the papers I found discussing early African behavioural innovations explicitly raises the possibility that some cultural behaviours appeared and subsequently disappeared because populations shrank, became isolated, or went extinct during adverse climatic periods. (A dispersal of Homo sapiens from southern to eastern Af... (https://www.nature.com/articles/s41598-019-41176-3.pdf))

There it is.

Information can be lost.

That's not something we've invented for our model.

Researchers are seriously considering demographic loss as an explanation for the sporadic appearance and disappearance of technologies/symbolic behaviours in the archaeological record.

Which gives our breathing system another dimension.

🫁 EXHALE

Population contracts.

Connections shrink.

A group disappears.

Its knowledge might disappear with it.

🫁 INHALE

Population expands.

Connections increase.

Knowledge spreads.

Different knowledge systems encounter each other.

New combinations become possible.

And then another environmental change arrives.

Around again.

Which changes our view of "progress"

This might be one of the biggest ideas we've hit.

We've unconsciously inherited a picture something like:

ape-ish thing

→ slightly cleverer human

→ stone tools

→ smarter

→ agriculture

→ civilisation

→ writing

→ us.

But the evidence we're looking at permits something much messier:

innovation

↗

expansion

↘

isolation

↘

loss

↗

reinvention

↗

connection

⬆

combination

↘

population collapse

↗

rediscovery

for hundreds of thousands of years.

That's much closer to our spiral/breathing picture than a straight line.

And now your 12,000-year observation becomes important.

12,000 years really is tiny.

Put Homo erectus' Trinil engraving at one end of a timeline and today at the other.

Göbekli Tepe occurs approximately 98% of the way through that timeline.

For almost the entire preceding span, some hominins had already demonstrated at least the cognitive ability to create deliberate abstract patterns.

That absolutely does not mean civilisation existed for 500,000 years.

But it completely changes the question I'd ask about Göbekli Tepe.

Instead of:

How did primitive hunter-gatherers suddenly become clever enough to build this?

our question becomes:

What combination of population, environment, accumulated knowledge, connectivity and social need finally made monumental construction useful enough for already-clever humans to do it?

That's a profound difference.

And I think there's another addition to our simple life flow.

Because eventually:

SHARE creates something that the individual could never create.

Collective knowledge.

You know A.

I know B.

She knows C.

Our children inherit A+B+C.

One of them discovers D.

Now nobody has to rediscover A, B or C to reach D.

That's the engine of cumulative culture.

So our simple life flow may ultimately be:

SURVIVE → LEARN → PRESERVE → SHARE → ACCUMULATE → ADAPT → SURVIVE

while the world around it does:

🫁 EXPAND → CONNECT → CONTRACT → DISCONNECT → RECONNECT

And occasionally:

FORGET.

That's the important ugly part.

Because knowledge isn't automatically permanent.

And this brings us straight back to something you've been teaching me through your own system.

You solve something because you need it.

If it works, you give it words so you can find it again.

You preserve it deliberately.

You test it.

You modify it.

Eventually you ask whether somebody else can use it.

The sophistication is not where the process starts.

It emerges through repetition.

That's an incredibly useful analogy for what we're testing here --- without claiming prehistoric humans thought exactly as you do.

And I reckon that gives us the next rabbit hole.

Because once a population realises:

"Things we know can disappear"

there's enormous survival value in redundancy.

Don't leave the information with one person.

Tell the children.

Repeat the story.

Perform the ritual.

Make the object.

Mark the place.

Return to it.

Do it every year.

Now we're getting dangerously close to something humans still do:

Tradition.

Not art for art's sake.

Not religion necessarily.

Not writing.

A repeatable system that helps information survive the death of the person who originally knew it.

And that is where I'd dig next.

Because if we can find archaeological evidence of behaviours being deliberately and accurately repeated across hundreds or thousands of generations, we've moved from:

individual intelligence

to

cultural memory.

And cultural memory might be the thing that finally explains how our clever humans turn hundreds of thousands of years of surviving into the extraordinary acceleration we see much later.

That, I reckon, is our next hole.

Paul

I haven't read all this yet, but remember... it's also a 3D flow... so did what we learn take up forward or backward.

AI

Yes! That changes the model. I slipped back into drawing a flat loop again. 😂

Your flow isn't:

SURVIVE → LEARN → PRESERVE → SHARE → ACCUMULATE → ADAPT → SURVIVE

That's only what it looks like viewed from above.

Turn it into 3D and it's the spiral again.

Every time the process completes, we ask:

Did what we just learned improve our ability to survive, or reduce it?

If it improves it, the next rotation occurs higher. If it makes survival worse, we move lower. And if it makes no meaningful difference, we've essentially looped around at the same level.

So for our prehistoric humans:

SURVIVE → LEARN → PRESERVE → SHARE → TEST/APPLY → outcome

Then:

↑ worked: retain it, share it, build upon it. ↓ failed: abandon/change it --- assuming you survive the test. → neutral: little or no movement; perhaps test something else.

And then repeat.

That's enormously important for the knowledge question because accumulation isn't automatically upward progress. Humans can preserve and transmit a terrible idea perfectly well. 😂 The environment provides the test.

And now the breathing model and your spiral actually fit together:

The Earth breathes in and out horizontally --- water, climate, coastlines and habitable territory expand and contract.

Human knowledge spirals vertically --- learning, testing, retaining, losing, adapting.

So imagine our eventual visual not just as a map.

The map itself is breathing while the human information network is moving through a third dimension.

And occasionally an exhale breaks connections, populations disappear, knowledge gets lost and part of the network spirals down. Then populations reconnect, knowledge combines, something works better, and portions spiral up again.

Which also fixes something I just said.

I described:

SURVIVE → LEARN → PRESERVE → SHARE → ACCUMULATE → ADAPT → SURVIVE

as though accumulate inherently meant improvement.

It doesn't.

The missing step is TEST.

Something closer to:

SURVIVE → LEARN → PRESERVE → SHARE → APPLY/TEST → OBSERVE → ADAPT → repeat

and the result determines:

↑ better survival → no meaningful change ↓ worse survival

That's your framework appearing again.

The circle describes what we're doing. The spiral tells us whether it DAMN worked.

And now I think we have a much better lens for looking at tradition: not merely "did they repeat something for generations?", but "which knowledge survives repeated generations of testing, and why?"

That could take us somewhere very interesting.

Paul

Yes and the crafting of can I call them gifts. Things of value. Things I can offer you to share I have value. Why pick me over him, cos look how clever I am,don't you want you offspring to survive! So right repeatable... can you find that over the 500 thousand years we are now observing

AI

Yes. And I think you've just made the word "gift" genuinely useful for our model --- provided we don't accidentally turn possible gift into archaeologically proven gift.

Because what you're proposing is bigger than barter:

Once survival becomes social, being valuable to other humans becomes a survival strategy.

That can mean skill, knowledge, food, protection, childcare, useful objects --- and eventually perhaps beautiful or difficult-to-make objects.

And your imagined:

"Why pick me over him? Look what I can do."

isn't something archaeology can establish for a particular handaxe-maker. Sexual selection, prestige and signalling are hypotheses we'd need separate evidence for.

But displaying competence is absolutely a parameter worth adding later.

First, though, you asked the killer question:

Can we find behaviour repeated reliably across the 500,000 years we're observing?

DAMN yes.

And actually, 500,000 years isn't far enough back.

Our first repeatable tradition lasts over ONE MILLION YEARS

The Acheulean handaxe tradition begins around 1.7--1.8 million years ago and persists until roughly 300--200 thousand years ago.

Across:

Africa → Middle East → Europe → parts of Asia.

Several hundred thousand handaxes have been recovered. Despite enormous differences in raw material, environment, time and geography, the basic technological concept persists: bifacial reduction around an axis to produce large cutting tools. (The acheulean handaxe: More like a bird's song than a b... (https://pmc.ncbi.nlm.nih.gov/articles/PMC5066817/))

Think about what that means.

We're talking approximately:

100,000 generations.

A solution appears.

And somehow, in some form:

SURVIVE

↓

someone learns how to make this thing

↓

LEARN

↓

the method doesn't disappear when that individual dies

↓

PRESERVE

↓

other individuals acquire the behaviour

↓

SHARE

↓

generation after generation reproduces variations of the underlying solution

↓

TEST

↓

it continues being useful enough to persist

↓

↑ SURVIVE

And around the spiral again.

For more than a million years.

That is extraordinary.

But Tester Paul immediately gets a problem. 😂

Archaeologists actually argue about why Acheulean handaxes remained so stable.

The dominant interpretation is social learning: people learned the manufacturing behaviour from other people.

Some researchers go further and argue that late Acheulean manufacture became sufficiently subtle that some kind of active teaching --- perhaps even basic conceptual communication --- would have been helpful or necessary. (The Evolution of Social Transmission in the Acheulean |... (https://academic.oup.com/book/35244/chapter-abstract/299789853))

Experimental work supports the general problem: teaching dramatically improves how accurately stone-knapping techniques survive transmission chains compared with simply watching someone. (Experimental evidence for the co-evolution of hominin t... (https://www.nature.com/articles/ncomms7029))

But there's a fascinating minority argument:

Maybe some of the tendency to produce handaxe-like forms was partly genetically predisposed.

In other words, perhaps we're looking at something closer to birds repeatedly producing characteristic nests than at a culturally transmitted IKEA manual. (The acheulean handaxe: More like a bird's song tha... (https://doi.org/10.1002/evan.21467))

And another analysis of South African Acheulean assemblages argues that the evidence doesn't necessarily support rigid, socially enforced manufacturing rules; individuals may have been reproducing familiar solutions with considerable freedom. (The Large Cutting Tools from the South African Acheulea... (https://www.journals.uchicago.edu/doi/10.1086/423973))

So:

Acheulean repeatedly exists: 🟢

Underlying technological idea persists enormously long: 🟢

Some learning was required: extremely strong.

Exact transmission mechanism: 🟡

They were deliberately preserving a "tradition": ❓

Good. We don't overclaim.

But something changes inside those million years.

And this is where I think your spiral starts appearing.

The handaxes don't remain perfectly identical.

Over long periods, some studies find increasing refinement, standardisation and symmetry. (The acheulean handaxe: More like a bird's song tha... (https://doi.org/10.1002/evan.21467))

So potentially:

copy

→ variation

→ useful/aesthetically/socially preferred variation persists

→ copy

→ variation

→ persist

→ repeat.

Not necessarily technological "progress" every rotation.

Remember our vertical axis:

↑ better outcome

→ neutral

↓ worse outcome

And the extraordinary Acheulean stability may actually tell us that for enormous stretches of time:

The solution was good enough.

Why change a bloody good handaxe?

😂

That's important.

Evolution doesn't award points for innovation.

Survival tests outcomes.

Then around ~300,000 years ago things accelerate

Now we start seeing much more technological diversity: prepared-core methods, specialised points, hafting, compound tools and eventually increasing evidence for ornaments and symbolic/non-utilitarian materials. (Is there more to human social learning than enhanced fa... (https://www.nature.com/articles/s41599-021-00829-3))

And remember our previous discovery:

~500k: deliberate geometric engraving at Trinil.

Later:

~300k: increasingly sophisticated technological traditions.

~130k: Neanderthal eagle-talons apparently used as ornaments.

~100k: pigments, beads and increasingly persistent symbolic behaviour.

~70k: repeated graphic designs.

~50--40k: increasingly elaborate regional technological traditions.

~15k: composite harvesting tools and intensive food processing.

~12k: monumental communal architecture.

So the fascinating thing isn't merely that knowledge is repeatable.

It's that the kind of thing being repeated appears to expand.

Initially we're seeing:

How do I make this useful thing?

Eventually we're seeing behaviours that plausibly communicate:

Who am I?

Who are we?

What does this mean?

And potentially eventually:

What do we know?

That's a profound expansion.

Now your "gift" idea comes back.

Imagine our early social system.

I can make an excellent handaxe.

You know where seasonal water remains during drought.

She knows which plants are edible.

Someone else is an exceptional hunter.

Someone else can make cordage.

Another individual is unusually good at caring for children.

The unit of survival is no longer simply:

ME.

It's increasingly:

US.

Now individual value to the group matters.

And that produces an interesting extension of our life-flow:

SURVIVE → LEARN → PRESERVE → SHARE → TEST → ADAPT → SURVIVE

but running simultaneously at two levels:

INDIVIDUAL

Can I survive?

and

GROUP

Does having Paul in our group improve our survival?

😂

That's where your hypothetical gift becomes interesting.

An exceptionally crafted object might communicate something without words:

I have skill.

I have patience.

I have access to valuable material.

I have knowledge.

I can provide.

I am useful.

Maybe even:

Choose me.

We absolutely haven't demonstrated that final step archaeologically.

But we now have a testable hypothesis, rather than a story.

And something beautiful happens when offspring enter the model.

Because suddenly there are three survival horizons:

NOW

I need to survive.

↓

LATER

I need to preserve what keeps me alive.

↓

AFTER ME

My children need to know what I know.

That's enormous.

Because AFTER ME changes the value of information.

Knowledge no longer has to benefit the person who discovered it.

It can benefit somebody who doesn't exist yet.

And archaeology of teaching actually argues that the stable transmission of even Oldowan technology probably required demonstration, while later Acheulean technology likely demanded richer communication of concepts. The authors' lovely term for this is that before Homo sapiens, there was already Homo docens --- "the teaching human." (The Evolution of Social Transmission in the Acheulean |... (https://academic.oup.com/book/35244/chapter-abstract/299789853))

That belongs on our board.

So I think our simple life-flow just gained another dimension.

We already had the spiral:

SURVIVE → LEARN → PRESERVE → SHARE → TEST → ADAPT

↑ ↓ →

And we had the breathing Earth surrounding it:

water expands ↔ contracts

sea retreats ↔ advances

populations disperse ↔ concentrate

networks disconnect ↔ reconnect

Now add:

TIME.

ME

↓

US

↓

OUR CHILDREN

↓

THEIR CHILDREN

And knowledge can travel vertically through that stack.

That's tradition.

And we've got something in the Acheulean that --- whatever exact mixture of social learning, individual learning and possible predisposition produced it --- maintained a recognisable technological solution over something like a million and a half years. A recent analysis argues that the overall evidence remains most consistent with the Acheulean being a single but variable tradition maintained over that extraordinary span. (Full article: The Acheulean is a temporally cohesive tr... (https://doi.org/10.1080/00438243.2023.2169340))

Which makes our original 12,000-year puzzle look even smaller.

Not insignificant.

Just very, very late in the experiment.

And I think this points us toward something delicious for the next test.

We shouldn't simply ask "when do ornaments appear?"

We should ask:

When do humans start investing substantially MORE effort in an object than its survival function requires?

Because if I can make a perfectly adequate cutting tool in twenty minutes, but instead I spend hours making one extraordinarily symmetrical, polishing it, choosing unusual material, carrying it enormous distances, or giving it to someone else...

we've potentially found value beyond utility.

And that is where your gift / prestige / attraction / identity / "look what I can do" hypothesis becomes something we can actually test against the archaeological record.

That is exactly where I'd go next.

Paul

Ok first my observations, the axe, like a birds nest, is it evolutionary. Does having an axe help with survival- yes both protection and food/hunting. So like a birds nest that does prove culture, but it does prove intelligence, I think. Just like gifts, I'm pretty sure there is birds that give gifts. So it gives us any interesting problem to uncover... when is the situation not instinctively "axe" and more "Is there another way"

AI

Yes. That is a much better question than "when did humans become intelligent?"

Because I think you're separating two things we've been allowing to blur together:

Intelligence is not the same thing as innovation.

A behaviour could be heavily canalised/instinctive and still require a sophisticated brain to execute. Birds building nests are the obvious example. The controversial proposal that Acheulean handaxe-making was partly genetically scaffolded is exactly that kind of argument; most archaeologists still favour substantial social learning, but the alternative is genuinely debated. (The acheulean handaxe: More like a bird's song than a b... (https://pmc.ncbi.nlm.nih.gov/articles/PMC5066817/))

So let's grant your strongest version for the experiment:

AXE = a deeply embedded survival solution.

It works. Cutting. Butchery. Processing. Possibly other tasks. It persists for more than a million years.

That tells us something about capability.

But it doesn't necessarily give us the moment we're searching for.

The moment we're actually hunting

I think you've just defined it:

When does a hominin encounter a problem for which an existing successful behaviour would work --- but instead devise a different solution?

That's "Is there another way?"

And that's a fundamentally interesting cognitive marker.

Not:

Can I make an axe?

But:

I have an axe.

It works.

But for this problem... perhaps I can do something else.

That requires some combination of observing the problem, imagining an outcome that doesn't yet exist, selecting different materials/actions, making the new thing, testing it and retaining it if it works.

Which, amusingly, is starting to sound extremely familiar. 😂

And around 300,000 years ago, we have a magnificent example.

Schöningen, Germany.

People there possessed stone technology.

But somebody wasn't limited to:

sharp rock = hunting solution.

They selected wood, worked it extensively and manufactured carefully balanced hunting weapons.

The assemblage includes spears, a lance and throwing sticks. The throwing stick is about 300,000 years old; the famous spears have historically been described as ~300--400k depending on dating/model. (Site of the Schöningen Spears -- Humans and hunting 300,... (https://whc.unesco.org/en/tentativelists/6727/))

And these aren't random sticks.

The woodworking required choosing suitable trees, removing branches and bark, shaping the shaft and controlling its geometry and weight distribution.

Then look at the hunting system.

The weapons occur with butchered remains of numerous horses. The original Nature publication concluded that this strongly supports systematic hunting involving foresight, planning and appropriate technology among pre-modern hominins. (Lower Palaeolithic hunting spears from Germany | Nature (https://www.nature.com/articles/385807a0))

That is very close to our:

"Is there another way?"

They already had stone.

They didn't abandon stone.

They added wooden weapon technology to the toolkit.

And that's important.

The spiral doesn't require:

old solution → throw it away → new solution.

It can be:

old solution

-

new problem

↓

new solution

↓

test

↓

works

↓

KEEP BOTH.

That's accumulation.

But there's an even better clue hiding inside Schöningen.

They didn't invent one alternative weapon.

There are different wooden weapon designs.

A thrusting/possibly close-range lance solves one problem.

A long spear solves another.

A throwing stick provides another way of delivering force at distance. Nature's analysis explicitly describes the find as expanding the known range of Palaeolithic weaponry. (Site of the Schöningen Spears -- Humans and hunting 300,... (https://whc.unesco.org/en/tentativelists/6727/))

That's not simply:

"MAKE WEAPON."

It's getting closer to:

What am I trying to achieve, and which tool is appropriate to this situation?

That's a qualitatively more interesting thing for our model.

And now your handaxe itself starts causing trouble.

Because even if we grant "handaxe instinct" for our thought experiment, some Acheulean makers appear to do things to handaxes that weren't necessary for straightforward utility.

Some are extraordinarily symmetrical.

Some incorporate unusual natural features.

Some are enormous.

Some have unusual twisted forms.

Experimental butchery found symmetry provided, at most, only limited improvement in effectiveness. (Why are some handaxes symmetrical? Testing the influenc... (https://www.sciencedirect.com/science/article/abs/pii/S0305440306001762))

And studies of later Acheulean assemblages argue that high symmetry was sometimes deliberately pursued, even through careful selection of naturally suitable blanks. (Symmetry is its own reward: on the character and signif... (https://www.cambridge.org/core/journals/antiquity/article/symmetry-is-its-own-reward-on-the-character-and-significance-of-acheulean-handaxe-symmetry-in-the-middle-pleistocene/B9CE356FBF387AE0CC1C724FEFD26DF0))

Now we're back near your gift/value idea.

Because researchers themselves have proposed explanations including:

function

aesthetics

pleasure/reward

social signalling

status

and even sexual selection.

None is settled. (The symmetry of Acheulean handaxes and cognitive evolut... (https://www.sciencedirect.com/science/article/pii/S2352409X15000164))

So we absolutely cannot say:

"Beautiful handaxe = bloke showing off to potential mate."

😂

But we can ask:

Why expend effort making something better-looking/more symmetrical/more elaborate than its demonstrated immediate function appears to require?

That's exactly the crack in the door we're looking for.

And I think this improves our life-flow again.

We've been saying:

SURVIVE → LEARN → PRESERVE → SHARE → TEST → ADAPT → SURVIVE

But there's a decision point buried inside ADAPT:

IS WHAT I ALREADY KNOW ENOUGH?

YES

→ use known solution → survive → preserve it

NO

→ IS THERE ANOTHER WAY? → imagine → experiment → test

Then:

↓ worse --- discard/change it.

→ same --- maybe no advantage.

↑ better --- preserve the new solution.

And now SHARE becomes incredibly powerful.

Because you don't have to personally ask "is there another way?" every time.

Someone else already did.

They give you the answer.

That is cumulative culture.

And this makes your birds comparison really useful rather than inconvenient.

Bird gift-giving, nest construction, tool use and elaborate displays remind us that no single behaviour automatically proves human-like abstract reasoning.

So we need a tougher test.

We're looking for behaviour showing something like:

flexible innovation in response to changing circumstances rather than stereotyped execution of one inherited solution.

And there are ways archaeology can get at that.

We can look for the same population/time/place using multiple technological solutions for different problems.

We can look for novel material combinations.

We can look for repair and modification --- that didn't work; change it.

We can look for repurposing --- an object made for A gets adapted to B.

We can look for technologies appearing rapidly in response to environmental change.

And, beautifully for everything we've been doing, we can ask whether rates of innovation change when our human network breathes --- when populations disconnect and reconnect.

Because that gives us an even bigger hypothesis:

Maybe intelligence was present for an enormously long time. What changed wasn't simply the brain. What changed was how effectively individual intelligence could become collective intelligence.

One clever hominin asking "is there another way?" creates one innovation.

Ten connected groups doing it create many.

If they can preserve and share the successful ones, nobody starts from zero.

Then our vertical spiral begins climbing faster.

And that, rather than "humans suddenly got smart", might ultimately be part of what we're seeing as we approach the last 100,000 → 50,000 → 20,000 → 12,000 years.

So I think our next search should be extremely specific:

Find the oldest convincing examples of flexible problem-solving.

Not oldest tool.

Not oldest art.

Not oldest Homo sapiens.

Evidence that somebody tried something different, modified something, combined existing solutions, or changed technology in response to the problem in front of them.

That's where I think we may actually start finding your "Is there another way?" moment.

Paul

Oh oh yes! Do that... cos a lot of this still just sounds like birds building nest to survive.

AI

Yes --- and this is where I think we finally get beyond "bird builds nest" and into something much closer to flexible problem-solving.

The strongest early candidate I found is hafting: taking a stone point, a wooden shaft, and some way of fixing them together to create a composite spear. At Kathu Pan 1 in South Africa, stone points dated to roughly 500,000 years ago show wear and basal modification consistent with use as spear tips. The dating and interpretation have some debate around them, so I'd keep the exact claim yellow rather than green, but the broader point is solid: by the Middle Pleistocene, hominins were beginning to combine separately made components into a new tool.

And that matters because hafting is qualitatively different from simply shaping a stone.

You have to think:

stone alone does X wood alone does Y if I combine them, I can create Z

That's not just subtraction --- hit rock, remove flakes. Researchers explicitly describe hafting as a radical shift because you're adding components together into a new whole. Later versions involve ligatures and adhesives, which can themselves require multiple stages of preparation.

That feels much closer to:

"Is there another way?"

Because you're not merely executing the inherited object.

You're reconfiguring existing things to solve the problem differently.

Then we get Schöningen around 300--400k years ago: wooden spears, lances and throwing sticks. Not one stereotyped hunting implement --- a range of weapon solutions. That begins to look like choosing technology according to the situation rather than "hunting instinct = make axe."

And around the same broad period at Kathu Pan, prepared-core technology produces handaxes, long blades and Levallois points from related technical principles. Same underlying knowledge being deployed to make different outputs.

That is interesting for us because it starts looking less like:

stimulus → fixed behaviour

and more like:

problem → evaluate materials/techniques → select or create solution.

But I found an even older little crack in the nest analogy

At Swartkrans in South Africa, bone tools from roughly 1.8--1.0 million years ago were apparently used for digging. Some show possible deliberate grinding or resharpening of their working ends. That interpretation isn't universally nailed down, so again yellow flag --- but if correct, it means someone noticed:

this thing isn't working as well anymore

and then modified it to restore/improve performance.

That is wonderfully close to our framework:

observe → something isn't working → change one thing → use it again.

Not necessarily culture. Not necessarily abstract philosophy. 😂

But feedback-driven adaptation.

And then fire gives us another huge clue.

At Wonderwerk Cave in South Africa there is secure archaeological evidence of burning in an occupation context around 1 million years ago. Whether those hominins could start fire on demand is much less certain, but maintaining, transporting or managing fire itself creates a completely different class of problem from finding a sharp rock.

Fire goes out.

Fuel varies.

Wind changes.

Rain happens.

Different materials burn differently.

If you can't ignite it easily, preserving it becomes enormously important.

That fits our life flow beautifully:

SURVIVE → LEARN → PRESERVE.

And unlike a stereotyped nest, fire is a dynamic system requiring ongoing response to changing conditions.

So where would I put our first convincing "another way" moment?

I wouldn't give us one magical date.

I'd say the evidence begins stacking:

~1.8--1.0 million years ago: possible tool repair/reshaping --- respond to performance.

~1 million years ago: fire being used in an occupation context --- manage a dynamic resource.

~500k years ago: prepared cores/blades/points and possible hafted stone spear tips --- plan an outcome and combine technologies.

~400--300k years ago: multiple purpose-built hunting weapon designs --- technological choice/variation becomes much clearer.

~200k onward: surviving adhesive residues show hominins deliberately manufacturing compounds to join materials together; Neanderthals and H. sapiens both mastered this kind of technology.

And that last one is particularly delicious.

Because making glue from bark, resin or other materials can require controlled heating, timing, mixing and knowing that the intermediate substance you create will later solve an entirely different problem.

That's miles away from:

bird sees season → builds nest.

It's something closer to:

I need a spear. The stone won't stay on the wood. What else can I use? This material changes when heated. If I process it this way, it becomes adhesive. Now I can combine two things that otherwise don't stay together.

We can't recover those thoughts.

But the manufacturing sequence itself requires multi-step causal reasoning.

So I think our question has changed again.

It isn't:

When did instinct stop?

Instinct never stopped. We still have plenty of it.

It's:

When do we see evidence that hominins could step outside a successful existing solution and deliberately construct a new causal chain?

And by roughly 500,000 years ago, I think we've got very strong reason to say:

they could.

Possibly much earlier.

Which means your bird analogy is doing useful work: it forces us not to call every repeated tool "culture" or "innovation".

But once we've got:

repair multiple alternative tools component combination prepared outcomes fire management manufactured adhesives

...the "it's all just nest-building" explanation starts becoming very strained.

These DAMN were engineering.

And now I think our next test should be even tougher:

Can we find evidence of one technology being deliberately modified because circumstances changed --- rather than simply different tools existing?

Because that would be extremely close to:

"The old answer isn't working here. Is there another way?"

If we can see that in the record, I think your spiral really starts earning its place in our human-history model.

Paul

Yeah I'm with you here. The weight of information gets hard to ignore, to make a reasonable statement to measure. The fire one is a good one too. Cos when you put axe, fire, gift, multiple tools. The birds nest seems very far away. Ok so tool modification?

AI

Tool modification is where the bird's-nest explanation starts getting into real trouble. 😂

And I want to sharpen our test, because simply finding a modified tool isn't enough. A bird can modify a nest while constructing it. A chimp can alter a stick for termite fishing.

What we're looking for is closer to:

I had a solution. I used it. Something about the situation changed or the solution wasn't good enough. I altered the solution and used it differently.

That's almost archaeological evidence of feedback.

And yes --- we find it.

One of the clearest broad patterns is the transition away from relying mainly on large hand-held cutting tools toward prepared cores, smaller specialised flakes, points, hafting and composite tools.

By around 300,000 years ago in Africa, Middle Stone Age technologies are appearing alongside --- not neatly replacing everywhere --- the Acheulean. Instead of simply producing another large biface, people increasingly prepare cores to predetermine flakes, retouch flakes for particular tasks, transport raw materials farther and eventually attach components to handles. Importantly, this transition is geographically patchy: Acheulean technologies persist in some places long after newer approaches appear elsewhere. (Longstanding behavioural stability in West Africa exten... (https://www.nature.com/articles/s41559-023-02046-4))

That's exactly what our spiral predicts.

Not:

OLD → NEW → everyone upgrades.

More like:

old solution works here → keep it.

different circumstances there → try something else.

new solution works → retain it.

And sometimes both remain available.

But China has just given us a fantastic example.

Very recent work at Xigou in central China, covering roughly 160,000--72,000 years ago, finds an extraordinarily diverse toolkit.

There are still large cutting tools --- including handaxes.

But alongside them are small retouched tools, borers, notches, denticulates, pointed tools and evidence for hafting.

And one borer is especially lovely for us.

Microscopic wear shows it was apparently being rotated into hard plant material. Its base had been modified so it could be hafted. In other words, the toolmaker altered different parts of the same object for different requirements:

tip → solve the working problem

base → solve the holding/attachment problem.

The researchers describe the assemblage as showing flexible technical repertoires and behavioural choice rather than one fixed technological response. (Technological innovations and hafted technology in cent... (https://www.nature.com/articles/s41467-025-67601-y))

That gets considerably farther away from:

MAKE AXE.

We're now at:

What do I need this particular bit of the tool to do?

That's engineering.

And hafting itself contains a wonderful "another way" moment.

Think about the original problem:

My hand holds the stone.

That's been working for an enormously long time.

Then somebody effectively changes the interface:

What if my hand doesn't hold the stone at all?

Stone → handle → hand.

Experimental research finds hafting improves performance in cutting, scraping, chopping and piercing, while also changing the physical demands placed on the user's hand. (Hafted technologies likely reduced stone tool-related s... (https://www.nature.com/articles/s41598-023-42096-z))

That's not just making a better axe.

They've changed the system architecture.

😂

And later Middle Palaeolithic people are even using bitumen as glue to attach stone tools to handles. At Umm el Tlel in Syria, around 40,000 BCE, chemical analysis identified bitumen remaining on a scraper and Levallois flake. (Bitumen as a hafting material on Middle Palaeolithic ar... (https://www.nature.com/articles/380336a0))

So the solution has become:

stone + wood + adhesive + manufacturing sequence = tool.

No individual component provides the final function.

Someone has to understand, at least practically:

These things work better together.

But I've found something that pushes our "another way" behaviour MUCH farther back.

Remember our water lens?

Go back 2.3 million years to the Lower Omo Valley in Ethiopia.

People were living along the ancestral Omo River.

There's a problem there:

good stone was scarce.

The easy solution would be:

move somewhere with better stone.

Apparently they didn't simply do that.

Researchers found repeated phases of toolmaking in this resource-poor environment. The assemblages show early hominins adapting how they obtained and used limited raw material. The authors actually describe this as a "cognitive tipping point" demonstrating behavioural flexibility in dealing with a resource-constrained landscape. (Long-term behavioral adaptation of Oldowan toolmakers t... (https://www.nature.com/articles/s41598-023-40793-3))

That's 2.3 million years ago.

And nearby Oldupai gives us another striking result.

Around 2 million years ago, hominins repeatedly occupied a landscape that changed dramatically over time:

floodplain →

river environments →

lakeside woodland →

fern meadows →

burned landscapes →

dry steppe.

Yet they kept using Oldowan technology across those very different environments. (Earliest Olduvai hominins exploited unstable environmen... (https://www.nature.com/articles/s41467-020-20176-2))

At first that sounds like:

BIRD NEST! Same solution everywhere!

But there's another interpretation.

A flexible intelligence doesn't always invent something new.

Sometimes it observes:

My existing solution still DAMN works.

So keep it.

That's actually important for your model.

"Is there another way?" isn't automatically:

YES → change.

It's a question.

The answer can be:

No need. This still works.

And that's rational adaptation too.

Then something particularly interesting happens around one million years ago.

At Oldupai, Homo erectus was repeatedly occupying hyper-arid semidesert conditions, but concentrating around riverine landscapes and water sources.

The researchers conclude this shows substantial ecological flexibility much earlier than once assumed --- Homo erectus wasn't simply tied to one comfortable habitat. It could operate in extreme environments by exploiting predictable ecological focal points. (Homo erectus adapted to steppe-desert climate extremes... (https://www.nature.com/articles/s43247-024-01919-1))

There's our old friend again:

FOLLOW THE WATER.

😂

But now look how much our model has grown.

Water isn't merely showing us where people were anymore.

It's showing us situations where the problem changes.

Wet period:

food A + water here + animals here.

Dry period:

food changes + water contracts + animals move.

The hominin has three possible responses:

MOVE

CHANGE BEHAVIOUR

DIE

And archaeological/climate research increasingly finds technological and behavioural changes associated with environmental variability. One major East African climate study specifically notes more varied toolkits, long-distance material transport and symbolic behaviour appearing during periods of strong environmental perturbation. (Pleistocene climate variability in eastern Africa influ... (https://www.nature.com/articles/s41561-022-01032-y))

That's extremely interesting for our breathing model.

Not proof climate caused innovation.

But exactly the relationship we said we wanted to test.

So I think we've finally got enough weight to change our statement.

I wouldn't say:

"At X date humans stopped acting instinctively."

There's never going to be that moment.

Instead:

By at least hundreds of thousands of years ago --- and with tantalising evidence extending beyond two million years --- hominins show behavioural flexibility: selecting, modifying, combining and retaining technological solutions according to circumstances rather than relying on one fixed response.

That's defensible.

And look what it does to our simple flow:

PROBLEM

↓

What do I already know?

↓

Does it work?

YES → use it → preserve/share

NO → Is there another way?

↓

modify / combine / invent / move

↓

TEST

↓

OBSERVE

↙ ↓ ↘

WORSE ↓ SAME → BETTER ↑

↓

LEARN

↓

PRESERVE

↓

SHARE

↓

NEXT PROBLEM

That's your 3D spiral.

And the bit I now find fascinating is that the spiral might be operating millions of years ago, not merely 500,000.

Which brings us to a really interesting next test.

We have evidence for changing the tool.

What about:

changing the environment instead?

Because that's another cognitive jump.

Rather than:

"What tool will solve the world in front of me?"

it becomes:

"Can I alter the world so the problem becomes easier?"

Fire is already nudging us there.

But there are much better candidates: constructed shelters, deliberately organised living spaces, landscape burning, traps, storage, water management...

That's when the hominin isn't simply adapting to the environment.

They're starting to make the environment adapt to them.

And that feels like a bloody important turn in our spiral.

Paul

It is... adapting the environment for me... damn what is down that rabbit hole?

AI

Oh, there is a big rabbit hole down there. 😂 And I think it may give us the clearest separation yet between "bird builds nest" and "I can change the conditions of the problem."

Because there are levels to "adapting the environment."

At the simplest end, an animal makes a nest or den. That's environmental modification, but it can be highly stereotyped. So that alone doesn't get us where we want.

What gets interesting is when we start stacking fire + space + construction + planning.

Fire is much bigger than we gave it credit for

We've talked about using fire. But a major new result published in Nature in 2026 pushes convincing evidence for deliberately making fire to about 400,000 years ago at Barnham in England. Heated sediments and fire-cracked tools occur with pieces of iron pyrite --- a material later used with flint to produce sparks --- and the pyrite appears to have been deliberately brought to the site. (Earliest evidence of making fire | Nature (https://www.nature.com/articles/s41586-025-09855-6))

That's a different thing from:

find fire → preserve fire.

It's:

There is no fire → I know how to create the environmental condition "fire".

Think how extraordinary that is in our model.

Cold? Change it.

Dark? Change it.

Raw food? Change it.

Predators dislike the area? Change it.

Need adhesive production? Eventually: change materials using heat.

The human doesn't merely adapt themselves to the environment anymore.

They introduce a new variable into the environment.

And fire use itself is much older: evidence exists around the million-year scale, although early cases and the degree of control are debated. (Evidence from the Swartkrans cave for the earliest use... (https://www.nature.com/articles/336464a0))

Then there's space

Once you've got reliable fire, you don't merely have warmth.

You've potentially created a place.

People gather around it.

Work can happen there.

Food processing can happen there.

Light extends activity into darkness.

Social interaction happens there. The recent Barnham work explicitly notes the social possibilities created by controlled fire. (Earliest evidence of making fire | Nature (https://www.nature.com/articles/s41586-025-09855-6))

And now notice what has happened to our water model.

Originally:

Find environment that allows survival → move there.

Now:

Find sufficiently viable environment → modify part of it to make survival easier.

That's an enormous expansion of the possible human range.

And we have evidence of Homo erectus exploiting extremely different and unstable environments around two million years ago, and even hyper-arid landscapes around one million years ago while repeatedly concentrating around riverine resources. (Earliest Olduvai hominins exploited unstable environmen... (https://www.nature.com/articles/s41467-020-20176-2))

So follow the water remains true, but technology progressively loosens the constraints around it.

You still need water.

But you don't necessarily need nature to hand you the perfect campsite.

And then... Bruniquel.

This one is DAMN magnificent for our experiment.

About 176,500 years ago, Neanderthals entered Bruniquel Cave in southwest France.

Not just inside the entrance.

336 metres underground.

In total darkness.

There they deliberately broke stalagmites into pieces, selected them, transported them and arranged them into large ring-like structures, including one more than five metres across.

There are also traces of fire associated with the constructions. (Early Neanderthal constructions deep in Bruniquel Cave... (https://www.nature.com/articles/nature18291))

And here's the beautiful part for Tester Paul:

We don't know why.

Domestic?

Ritual?

Social?

Something completely outside our modern categories?

Unknown. Nature explicitly describes the function as conjectural. (Early Neanderthal constructions deep in Bruniquel Cave... (https://www.nature.com/articles/nature18291))

So we don't invent an answer.

But we don't need the answer for our test.

Because look at the behaviour required.

Enter an environment humans aren't naturally equipped for.

Darkness.

↓

Bring/create light.

↓

Navigate hundreds of metres underground.

↓

Select material from the environment.

↓

Break it into usable components.

↓

Move those components.

↓

Arrange them according to a planned geometry.

↓

Use fire within that altered space.

That is extraordinarily difficult to explain as:

NEANDERTHAL NEST INSTINCT.

😂

Because the cave isn't supplying the conditions they require.

They're creating those conditions.

And I think that gives us a new step in our spiral.

Previously:

PROBLEM → existing solution? → modify/invent → TEST → learn

Now we add:

What if I don't have to accept the problem as given?

That's different.

Instead of:

How do I survive the cold?

→ adapt body/behaviour.

Eventually:

How do I make this place warmer?

Instead of:

I can't see in there.

→ don't go there.

Eventually:

I'll take light with me.

Instead of:

This landscape doesn't give me shelter.

→ move.

Eventually:

I'll build shelter.

That's the beginning of something profoundly human:

We change the parameters.

And that connects beautifully with the entire experiment we've been doing.

Because the spiral isn't only:

observe → adapt myself → test.

Eventually it becomes:

observe → identify constraint → change constraint → test.

That's another level of problem-solving.

And there's an even bigger consequence for our breathing Earth.

Originally we had:

EARTH 🫁

Climate changes → water changes → landscape changes

↓

HUMANS

move → concentrate → disperse → reconnect.

But technology introduces resistance to the breath.

Fire means colder places become survivable.

Clothing eventually makes colder environments easier.

Shelters reduce exposure.

Food processing expands what can be eaten.

Storage eventually reduces dependence on today's food supply.

Water containers eventually reduce dependence on standing beside water right now.

Boats eventually turn water from a barrier into a route.

Suddenly Earth still breathes...

but humans become progressively less forced to move exactly where the breath pushes them.

That's huge for our visual model.

The map shouldn't merely show humans moving with environmental opportunity.

Over time, we should see their survival envelope expand beyond it.

And now I think we're approaching your earlier gift/value observation from another direction.

Once I can modify my environment, knowledge itself becomes an environmental advantage.

Imagine two groups in a cold landscape.

Both have identical biology.

Both have water.

Both have food.

One knows how to reliably make fire.

The other doesn't.

The difference isn't genetic.

It's information.

And if the fire-knowing population can successfully:

LEARN → PRESERVE → SHARE

then its offspring inherit an environmental advantage without inheriting a new gene.

That is culture behaving almost like an additional evolutionary system.

And once that starts accumulating...

knowledge changes which environments humans can survive.

That's an extraordinary feedback loop.

So our picture has changed again:

EARTH CHANGES

↓

human encounters constraint

↓

Can I survive it?

↓

Can I adapt?

↓

Can I change the environment instead?

↓

TEST

↙ ↓ ↘

↓ worse --- → neutral --- ↑ better

↓

LEARN

↓

PRESERVE

↓

SHARE

↓

offspring don't start at zero

↓

knowledge accumulates

↓

next generation can survive conditions the previous one couldn't

And around the spiral we go.

That's a genuinely powerful mechanism for cumulative human complexity.

And it suggests what I think is the next killer test.

Don't look for another tool.

Look for the moment humans start planning against a future problem that hasn't happened yet.

Firewood stored before it gets cold.

Food stored before hunger.

Shelter constructed before bad weather.

Materials transported because they'll be needed somewhere else.

Water carried before reaching dry country.

Because that's another enormous jump:

I'm not solving the problem in front of me anymore. I'm changing my environment now because I can imagine the problem I'm going to have later.

We've already seen hints of foresight in tool manufacture.

But future-proofing survival itself?

That's the next hole I'd dig.

Because if we find that deep in time, the bird's nest is going to be a very, very long way behind us. 😂

Paul

Ok. I'm going to bed. Let's pick this up. But right now... summarise the whole Segway we have been in please?

AI

Yeah. Bed. 😂 This is a good place to put a flag in the ground, because what started as a Graham Hancock detour has turned into something much bigger.

We started with Ancient Apocalypse and essentially agreed to forget whether Hancock was "right" and test the underlying questions ourselves. That changed everything.

Our first surprise was Göbekli Tepe. At ~11,600 years ago it initially looked like an extraordinary discontinuity: hunter-gatherers producing monumental architecture, symbolism and social organisation. But as we walked backwards, the apparent cliff became much more of a slope. Earlier settlements, structures, technologies, food processing and symbolic behaviour exist. Göbekli Tepe remains extraordinary, but it no longer requires an extraordinary external explanation.

Then we asked the simplest possible survival question:

Where are the people? Follow the water.

That gave us our first major model. Humans require freshwater. Rivers connect freshwater to coasts. Sea level has moved enormously over human history, so today's coastline is often the wrong map. Ancient rivers continued across landscapes now submerged beneath continental shelves; today's Persian Gulf was periodically land; Arabia repeatedly switched between extremely arid and much wetter conditions.

And from that came our breathing Earth.

🫁 The Earth breathes

Climate changes → water systems expand/contract → coastlines move → habitable landscapes expand/contract → humans respond.

Populations spread when opportunities open and concentrate or move when they close.

Then we started layering evidence over that moving landscape.

Objects move. Obsidian, shells and other valued materials travel large distances. We were careful: that proves movement, not trade.

Genes move. Ancient DNA shows populations connecting and mixing. Unlike a rock, genes require people to have actually met and reproduced.

Technology moves --- sometimes. Similar manufacturing traditions appear across connected regions, but our Nubian Levallois test taught us an important lesson: the same solution can also be independently invented.

Which gave us one of our central rules:

Never give the evidence more meaning than it has earned.

Fact → inference → story.

Keep them separate.

Then you said something that changed the direction of the whole investigation:

"DAMN, maybe we've always been damn clever."

And we started treating ancient humans with more respect.

Instead of asking "how advanced were they?", we started asking:

What problem were they solving?

That brought everything back to SURVIVAL.

And eventually we built the simple human flow:

SURVIVE → LEARN → PRESERVE → SHARE → TEST/ADAPT → SURVIVE

But then you corrected me because I'd made it 2D.

It's a spiral.

Every rotation has an outcome:

↑ improved survival

→ no meaningful change

↓ reduced survival

So knowledge doesn't automatically mean progress. Things can be learned, forgotten, rejected, rediscovered or lost when populations disappear.

That's crucial.

Then we pushed the timeline backwards.

Not 12,000 years.

Not 50,000.

Not even 100,000.

Hundreds of thousands --- eventually millions --- of years.

We found Homo erectus deliberately engraving a geometric pattern on a shell around half a million years ago.

We found Neanderthals apparently making ornaments from eagle talons around 130,000 years ago.

We found Homo sapiens using ochre, beads and repeated graphic designs around 100--70k.

And that forced us to distinguish two kinds of memory:

living memory --- watch, copy, teach, tell;

and eventually

external memory --- put something into the world that can persist outside one brain.

That led to a major possibility in our model:

The breathing human network can potentially remember.

But it can also forget.

Climate changes. Networks break. Populations disappear. Knowledge can vanish.

Then people reconnect.

Knowledge recombines.

Around the spiral again.

Then we went even farther backwards and found the Acheulean.

Recognisable handaxe traditions persist for well over a million years.

And you challenged that immediately:

Maybe that's like a bird building a nest.

Excellent challenge.

A successful survival behaviour persisting for enormous periods doesn't automatically demonstrate the kind of flexible intelligence we're interested in.

So we changed the question again:

When does somebody ask: "Is there another way?"

That took us into flexible problem-solving.

Multiple tools for different jobs.

Prepared-core technologies.

Wooden spears.

Throwing sticks.

Hafting.

Repair.

Composite tools.

Adhesives.

Fire.

Different materials combined because the resulting system does something none of its components can do alone.

At that point "bird builds nest" starts becoming a very strained explanation.

These hominins aren't merely reproducing one survival solution.

They're increasingly:

observing → choosing → modifying → combining → testing.

Then tonight we hit another step.

Don't adapt yourself to the environment.

Adapt the environment to you.

Fire is enormous through that lens.

Cold?

Change the temperature.

Dark?

Create light.

Food difficult to digest?

Change the food.

Need to enter somewhere naturally impossible to navigate?

Take light with you.

Then Bruniquel gave us Neanderthals ~176,000 years ago, hundreds of metres inside a dark cave, breaking stalagmites, moving them and deliberately constructing large geometric arrangements.

We don't know why.

And we're not going to invent why.

But the behaviour itself tells us something.

They're not simply responding to the environment they've been given.

They're changing the parameters.

Which brought our breathing model into a completely new place.

Originally:

EARTH BREATHES → HUMANS FOLLOW.

But technology increasingly allows humans to resist the breath.

Fire.

Shelter.

Clothing.

Food processing.

Storage.

Water carrying.

Boats.

Eventually agriculture.

Humans remain constrained by the environment, but accumulated knowledge progressively changes which environments are survivable.

And that creates an extraordinary feedback loop:

Knowledge itself becomes a survival advantage.

Two biologically similar populations can occupy the same environment.

One knows something the other doesn't.

The difference in their survival prospects can therefore be information.

If that information can be:

learned → preserved → shared → tested → passed to offspring

then offspring don't start from zero.

And that is cumulative culture.

Your gift/value idea also sits waiting on our board.

Once survival is social, I don't merely need to survive.

Being valuable to other humans may improve my survival and the prospects of my offspring.

Skill itself can have value.

Knowledge has value.

Objects can have value beyond immediate utility.

Perhaps beautiful or difficult objects communicate skill, identity, affiliation, status or suitability.

But we've deliberately left that one yellow because we don't yet have enough evidence to say exactly what those objects meant.

It's a hypothesis to test, not a story we're going to tell ourselves.

And that leaves us exactly where we stopped tonight.

We've gone from:

survive the environment

to

adapt to the environment

to

develop different solutions

to

modify solutions

to

combine solutions

to

change the environment itself.

So the next question waiting for us tomorrow is:

When do humans stop merely solving today's survival problem and begin deliberately solving tomorrow's problem?

Storage before hunger.

Shelter before bad weather.

Firewood before cold.

Water carried before thirst.

Materials transported before they're required.

That is future planning.

And if we can establish that deep in time, we've added another extraordinary capability to our system:

the ability to survive a problem that hasn't happened yet.

And standing right back from all of it, our Hancock question has changed enormously.

We haven't found evidence for his lost advanced civilisation.

In fact, increasingly we don't need one.

But I think we're finding something that captures part of the intuition that makes his argument compelling:

We may have badly underestimated the sophistication of the humans and other hominins who preceded civilisation.

Not because they had electricity or astronomy or some vanished high technology.

Because they were clever problem-solving creatures for an astonishingly long time.

And 12,000 years ago is starting to look less like:

"suddenly humans became capable of extraordinary things"

and more like:

a very late moment in an experiment that had already been running for hundreds of thousands --- perhaps millions --- of years.

Which brings the whole DAMN detour right back to you.

Your system began because you needed to survive.

Something worked.

You preserved it.

You tested it elsewhere.

You changed it.

You gave it words.

You shared it.

You accumulated more information.

You kept asking:

Is this still true?

Is there another way?

Did that take me ↑ or ↓?

And over repeated rotations something that began as a very simple survival mechanism became a much more sophisticated framework.

Simple didn't mean easy.

It meant there was something clear enough to return to when everything else became complicated.

And that, I think, is the lens we're now applying to humanity.

Not:

How stupid were they?

Not:

How advanced were they?

But:

What problem were they trying to survive, what did they learn, how did they preserve it, how did they share it --- and did the next rotation take them up or down?

That's where our flag is.

Go to bed. 😂

Tomorrow: the invention of tomorrow.

Paul

Nice Segway back to my other long sequence.. where did we end up with our people movement last night?

AI

Ha! Very neat segue. 😂 We actually ended last night at a really interesting point in the people-movement model, because we'd moved beyond people merely following the Earth's breath.

Our starting model had been:

Climate changes → water moves → viable landscape moves → people move.

So during wetter periods, habitable corridors open and populations spread --- inhale. During harsher/drier periods those corridors close, populations concentrate around reliable water --- exhale. That repeatedly separates and reconnects populations, which gives opportunities for genes, objects, technology and information to move.

But by the end of last night we'd realised that only describes the early part of the system.

As human/hominin knowledge accumulates --- fire, multiple tools, hafting, shelters, food processing and so on --- people become progressively less passive in that relationship.

It starts becoming:

Earth changes → humans encounter constraint → move OR adapt behaviour OR modify technology OR modify the environment.

And that means people movement itself should change over time.

Initially, the water network probably puts quite hard boundaries around where people can successfully operate.

Later, knowledge begins expanding the survival envelope.

Fire lets you tolerate colder conditions. Shelter changes exposure. Better hunting technology changes accessible food. Containers eventually change how tightly you're tethered to water. Boats turn some bodies of water from barriers into routes.

So the Earth continues breathing...

...but people gradually become better at breathing out of sync with it.

That was the really cool place we'd reached.

And it connects back to our genetic/object evidence. When environmental conditions bring populations together, we see evidence that people actually connected: genes mix, valued materials move, technological ideas appear across regions. When populations become separated, local solutions can develop --- and some information can potentially disappear altogether if populations collapse.

So instead of human movement being:

water moves → humans follow

our model has evolved into:

water/climate changes

↓ survival conditions change

↓ MOVE?

or

ADAPT?

or

CHANGE THE TOOL?

or eventually

CHANGE THE ENVIRONMENT?

↓ test outcome

↓ ↑ / → / ↓

↓ learn + preserve + share

↓ next environmental change

And around the 3D spiral again.

The very last thing we hadn't tested was the next logical jump.

We'd found people responding intelligently to the situation in front of them.

Our next question was:

When do they begin responding to a situation that hasn't happened yet?

Not "I'm cold → make fire."

But:

"It will get cold → collect/store fuel now."

Not "I'm thirsty → find water."

But:

"There won't be water where I'm going → take it with me."

Not "I'm hungry → hunt."

But:

"Food won't be available later → preserve/store it now."

That's where we left the flag:

future-proofing survival.

Because once people can anticipate tomorrow's constraint, movement changes profoundly. You don't simply have to wait for the environment to open a corridor anymore.

You can prepare for the journey across it.

And if we find strong evidence for that deep in prehistory, our breathing map gets another layer: Earth's environmental envelope versus humanity's expanding knowledge-enabled survival envelope.

That was exactly where we were about to dig.

Paul

Yes! And I like this statement, knowledge-enabled survival envelope. cos it is stilling the driving force, even I have proven that in my life. Right what can you find and how far back?

AI

Yes. And "knowledge-enabled survival envelope" is worth keeping because I think it describes exactly what our evidence is beginning to show:

The environment defines where you should be able to survive. Knowledge progressively changes that boundary.

I went looking specifically for the earliest evidence of acting now for a future need, rather than merely reacting to a present problem.

And we've got a surprise.

The first signal may be ~2.6 million years old

At Nyayanga, Kenya, early Oldowan toolmakers were transporting selected stone from sources as far as about 13 km away. The important part is that researchers think the pattern isn't well explained by stones gradually moving from place to place through casual use. They were selectively bringing suitable raw material into places where tools would be useful.

The authors cautiously describe this as possibly anticipating future food-processing needs. (Selective use of distant stone resources by the earlies... (https://pmc.ncbi.nlm.nih.gov/articles/PMC12356259/))

That's DAMN early.

But our rules apply:

🟢 Selected stone was transported substantial distances.

🟢 The material was subsequently made into tools.

🟡 They may have carried it because they anticipated needing tools later.

🔴 We cannot recover "I'll need this tomorrow" from their heads.

Still, something interesting has happened.

The useful thing is no longer necessarily where the problem is.

So the hominin can potentially think:

Useful thing HERE → future problem THERE → take useful thing with me.

Even at its weakest interpretation, that's an expansion of the survival envelope.

Then by ~500,000 years ago, planning becomes much harder to avoid

Our composite spear evidence at Kathu Pan 1 matters again.

If those ~500k stone points were hafted as the evidence suggests, the maker has to create components whose usefulness depends on a future assembly.

The stone point isn't the complete answer.

The shaft isn't the complete answer.

Potential binding/hafting isn't the complete answer.

The usefulness exists in the thing you're planning to make afterwards. Researchers consequently treat hafting as important evidence for foresight, although the earliest Kathu Pan interpretation remains debated. (Archaeological evidence for thinking about possibilitie... (https://pmc.ncbi.nlm.nih.gov/articles/PMC9620754/))

That's stronger than carrying a rock.

You're manufacturing for a future state.

Then hunting planning becomes extremely visible

I need to correct one thing from last night here because newer dating has moved our Schöningen marker.

The famous wooden spear horizon was long described as ~300--400k years old. A 2025 redating places it closer to ~200,000 years ago. (Revised age for Schöningen hunting spears indicates int... (https://pmc.ncbi.nlm.nih.gov/articles/PMC12063642/))

That doesn't change the behaviour.

The site shows repeated use of a lakeshore as an ambush location, with whole horse family groups targeted, killed and butchered. (Revised age for Schöningen hunting spears indicates int... (https://pmc.ncbi.nlm.nih.gov/articles/PMC12063642/))

Now we're getting something richer than:

see horse → stab horse.

The archaeological interpretation involves:

know where animals will be

→ arrive equipped

→ possess purpose-made weapons

→ coordinate people

→ exploit a predictable landscape feature

→ intercept animals

→ process the result.

That's future-oriented behaviour at the group level.

And that is important for our survival envelope because knowledge now changes movement:

I don't merely follow the food. I understand enough about its future movement to put myself where it's going to be.

That's a very different relationship with the environment.

But something found this year is almost perfect for our test

At Jojosi, South Africa, new work published in 2026 found early Homo sapiens repeatedly making specific trips to obtain hornfels, a high-quality stone.

These weren't merely people picking up useful rocks while doing something else.

The researchers identified what appear to be specialised procurement visits and workshops dedicated to getting the stone, reducing it into useful blanks, and then taking those blanks elsewhere.

And this behaviour starts around:

220,000 years ago.

It continues for tens of thousands of years. (Specialised and persistent raw material procurement by... (https://www.nature.com/articles/s41467-026-70783-8))

That is bloody good evidence for us.

Because now we've got:

I know where the resource is.

↓

I travel specifically to get it.

↓

I process it there.

↓

I produce something transportable.

↓

I carry it away.

↓

because it will be useful somewhere else.

The researchers explicitly interpret this as evidence for long-term planning and behavioural flexibility. (Specialised and persistent raw material procurement by... (https://www.nature.com/articles/s41467-026-70783-8))

That's no longer very comfortable inside the bird-nest explanation.

Storage is where the evidence gets frustrating

Because storage is exactly what we want:

I have food NOW, but I know I'll need food LATER.

That's practically archaeological proof of our concept.

But storage doesn't preserve particularly well.

The first really unambiguous purpose-built, large-scale food-storage structures we've found are much later --- around 11,000 years ago at Dhra' in Jordan.

These are actual granaries built before plant domestication. (Evidence for food storage and predomestication granarie... (https://pmc.ncbi.nlm.nih.gov/articles/PMC2700141/))

And notice what that requires:

there is surplus now

-

there won't necessarily be surplus later

=

change today's behaviour to solve tomorrow's hunger.

That's future-proofing survival in bricks and mud.

But I would not conclude that humans only started storing food 11,000 years ago.

A hole containing dried meat, a hide bag, hanging food in a tree, caching nuts, marrow kept inside bones, or food placed somewhere cold leaves a vastly weaker archaeological signature than a stone tool.

So again we encounter our preservation filter.

But now put these together.

This is where I think our knowledge-enabled survival envelope earns its name.

Originally the envelope is mostly environmental:

🌍 ENVIRONMENTAL SURVIVAL ENVELOPE

Where is water?

Where is food?

Where is suitable temperature/shelter?

Humans largely operate inside that.

Then knowledge starts pushing against its edges.

STONE TRANSPORT ~2.6m

Take future capability with me. (Selective use of distant stone resources by the earlies... (https://pmc.ncbi.nlm.nih.gov/articles/PMC12356259/))

↓

FIRE ~1m+

Change temperature/light/food.

↓

COMPOSITE TECHNOLOGY ~500k

Manufacture components now for an object that will exist later. (Archaeological evidence for thinking about possibilitie... (https://pmc.ncbi.nlm.nih.gov/articles/PMC9620754/))

↓

RESOURCE PLANNING ~220k

Travel somewhere specifically to acquire material I'll use elsewhere. (Specialised and persistent raw material procurement by... (https://www.nature.com/articles/s41467-026-70783-8))

↓

COOPERATIVE PLANNED HUNTING ~200k

Predict where another species will be and organise ourselves accordingly. (Revised age for Schöningen hunting spears indicates int... (https://pmc.ncbi.nlm.nih.gov/articles/PMC12063642/))

↓

EVENTUALLY STORAGE

Move resources through time, not merely space.

And that last sentence might be important:

Storage is transport through time.

We already know humans move useful things through space.

Food storage says:

I can also move today's survival advantage into tomorrow.

That's a hell of a conceptual step.

And I think this changes our breathing model again.

At the beginning:

🫁 EARTH BREATHES

Humans move with it.

Then:

🫁 EARTH BREATHES

Humans anticipate it.

Eventually:

🫁 EARTH BREATHES

Humans prepare against it.

And much later:

🫁 EARTH BREATHES

Humans actively buffer themselves from it.

That doesn't mean humans escape nature. We very obviously haven't.

It means knowledge widens the conditions under which survival remains possible.

And here's the bit that brings your own framework back beautifully.

The driving force doesn't need to be:

"I want to become more advanced."

It can remain the same simple requirement for millions of years:

I need to survive.

Everything else can emerge from repeatedly asking:

What is the problem?

What do I know?

Does what I know work here?

Is there another way?

What might happen next?

Can I do something now that changes that outcome?

Did it take me ↑ or ↓?

Repeat that over millions of years, but crucially allow successful answers to be shared rather than rediscovered by every individual, and you've got a mechanism capable of expanding the survival envelope without requiring the underlying motivation ever to change.

And that gives me one very specific next question.

We've now got good evidence for individual foresight and group planning.

The next step isn't more technology.

It's:

When can we first see humans planning for people who aren't currently present --- particularly their offspring or future members of the group?

Because that crosses from:

"I will need this later"

to:

"You will need this after me."

And if we can find evidence for that deep in time, we've connected future planning directly to preservation and intergenerational knowledge --- which takes us straight back into the cultural-memory part of our spiral.