Chapter 2 of 5 / Paul + AI

Follow the Water.

All chapters Β· 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

ok, so this is an important part now. we keep going back, but i need you to include location, in terms of how closely located in terms of region are we. then we need to consider sea level. not floods. but sea level, what information supports the sea level at times we have and the locations we are discussing. please/

AI

Yes. This is exactly the correction we need.

I've been treating our sites too much like points on a timeline. We now need three dimensions simultaneously:

WHEN + WHERE + WHAT DID THE LAND ACTUALLY LOOK LIKE THEN?

And you're right: sea-level change, not "floods". Flood implies an event. Sea-level change is a measurable physical process, and we have independent evidence for it.

First: how good is the sea-level evidence?

Very good.

This isn't Hancock evidence. Reconstructions come from multiple independent sources: dated coral terraces, marine sediment cores, oxygen-isotope records, ice-volume modelling and regional geological markers. Lambeck et al.'s major PNAS reconstruction puts global sea level around the Last Glacial Maximum roughly 125--135 m below modern levels, followed by a large rise toward the Holocene. Spratt & Lisiecki independently combined multiple proxy records into a long sea-level reconstruction. (Sea level and global ice volumes from the Last Glacial... (https://pmc.ncbi.nlm.nih.gov/articles/PMC4217469/))

So let's put a big green flag here:

🟒 FACT: enormous areas occupied by humans during the period we're investigating are underwater today.

That's not speculative.

Now let's put our archaeological trail onto geography.

Approx. date Site/region Where? Relationship ------------------ ----------------- ---------------------- ----------------- ~23,000 BP Ohalo II Sea of Galilee, Israel Southern Levant

~15,000--12,000 Natufian Israel/Jordan/Levant Same broad BP sites Fertile Crescent zone

~12,400 BP onward KΓΆrtik Tepe Upper Tigris, SE Northern Fertile Turkey Crescent

~12,000-ish BP Boncuklu Upper Tigris, SE Very close to onward Tarla Turkey KΓΆrtik

~11,600 BP onward GΓΆbekli Tepe SE Turkey ~Upper Euphrates region

So we're not hopping randomly around the planet.

This is basically one large connected region:

Levant β†’ Euphrates/Tigris β†’ southeastern Anatolia.

And KΓΆrtik and Boncuklu are particularly close geographically. KΓΆrtik sits at the Batman River/Tigris confluence; Boncuklu is also in the Upper Tigris drainage. KΓΆrtik's dates reach at least ~10,400--10,200 BCE and continue across the Younger Dryas/Holocene transition. (Adaptions in subsistence strategy to environment change... (https://journals.sagepub.com/doi/10.1177/09596836221074030))

So local/regional continuity is real evidence.

But now let's apply your sea-level parameter.

23,000 years ago --- Ohalo II

This gets fascinating immediately.

Ohalo II was occupied during the Last Glacial Maximum. Today it sits 212--213 metres below mean global sea level, beside/under the Sea of Galilee. It was actually discovered because drought and water extraction lowered the lake sufficiently to expose it. (Stone Age hut in Israel yields world's oldest evidence... (https://pmc.ncbi.nlm.nih.gov/articles/PMC404215/))

And geological work reconstructing ancient Lake Kinneret levels uses radiocarbon-dated shells and sediments. It shows substantial lake-level changes, including a major drop around 23,800 years ago. (The late Quaternary limnological history of Lake Kinner... (https://cir.nii.ac.jp/crid/1363107370018426368))

Notice what happened here.

A prehistoric shoreline settlement became submerged.

Then sediment and water preserved it extraordinarily well.

Then modern lake levels fell.

Then:

Holy shit. Houses.

One paper explicitly makes the point that Ohalo's extraordinary preservation raises the question of how many comparable sites remain submerged around the lake. (Stone age bedding by the Sea of Galilee - PMC (https://pmc.ncbi.nlm.nih.gov/articles/PMC409895/))

That's important evidence for our experiment:

Submerged prehistoric settlement isn't hypothetical. We have actually found one.

Now move from lakes to oceans.

~20,000 years ago --- Eastern Mediterranean

Sea level was approximately 120 m lower than today along the Levantine coast.

And here's the physical consequence:

The coastline was roughly 10 km farther west than today's coastline in parts of the Carmel region. (Frontiers | Neolithic coastal adaptation and resilience... (https://www.frontiersin.org/journals/environmental-archaeology/articles/10.3389/fearc.2026.1775646/full))

That's not trivial.

If we're excavating today's coastline looking for people living 20,000 years ago...

we're looking in the wrong DAMN place for coastal settlements.

Their coastline can be kilometres offshore.

That's an important correction to our earlier archaeological staircase.

The inland trail:

Ohalo β†’ Natufian β†’ Upper Tigris β†’ GΓΆbekli

is real.

But it isn't necessarily a representative sample of everyone living in southwest Asia.

It's a sample of what remains accessible.

And now look east.

This is where things get really interesting.

The Persian Gulf.

At the Last Glacial Maximum, global sea level was roughly 120--130 metres lower.

The Persian Gulf basin was largely dry land.

Not merely a slightly wider beach.

Land.

Geological work finds that seawater didn't reach the Strait of Hormuz until about 14,000 years ago, and only around 12,500 years ago did a true marine seaway begin forming inside the Gulf. (Late Quaternary sea-level changes of the Persian Gulf -... (https://www.sciencedirect.com/science/article/pii/S0033589415000381))

And remember what's feeding that region?

Tigris. Euphrates. Karun.

The same broad river system we're already investigating.

Jeffrey Rose has published a serious archaeological hypothesis --- entirely independently of Hancock --- proposing a prehistoric "Gulf Oasis."

With sea levels lower, the exposed Gulf basin could have contained rivers, freshwater springs and habitable landscapes. Rose notes that at times the exposed area was comparable in size to Great Britain. (New Light on Human Prehistory in the Arabo-Persian Gulf... (https://www.jstor.org/stable/10.1086/657397?ab_segments=0%2Fbasic_search_gsv2%2Fcontrol))

Now my eyebrows go up.

Not because:

ATLANTIS! πŸ˜‚

But because our map has changed.

Our archaeological evidence currently says:

Upper Tigris communities β†’ KΓΆrtik β†’ Boncuklu β†’ GΓΆbekli

But if we follow that river system downstream during the period immediately preceding GΓΆbekli Tepe...

we eventually reach an enormous inhabited landscape that doesn't exist anymore.

It's underneath the Persian Gulf.

That's not Hancock.

That's geology.

Now let's line up TIME + SEA LEVEL

Approximate global values --- don't treat individual metres as exact because reconstructions have uncertainties and local relative sea level differs due to tectonics/isostasy:

~23,000 BP: around 120--130 m below today Ohalo occupied.

↓

~20,000 BP: around 120 m+ below Eastern Mediterranean shoreline kilometres offshore. Persian Gulf largely exposed.

↓

~15,000 BP: sea rising substantially but still far below modern level.

↓

~14,000 BP: marine water reaches Strait of Hormuz. (Late Quaternary sea-level changes of the Persian Gulf -... (https://www.sciencedirect.com/science/article/pii/S0033589415000381))

↓

~12,900 BP: Younger Dryas begins.

↓

~12,500 BP: Persian Gulf marine transgression underway. (Late Quaternary sea-level changes of the Persian Gulf -... (https://www.sciencedirect.com/science/article/pii/S0033589415000381))

↓

~11,700 BP: Younger Dryas ends.

↓

~11,600 BP: earliest monumental GΓΆbekli Tepe.

↓

Holocene: sea continues rising rapidly toward modern levels.

And now I think we have found a better question than Hancock gave us.

Not:

"Was there an advanced lost civilisation?"

We're nowhere near earning that.

Instead:

What human communities occupied the now-submerged landscapes of the eastern Mediterranean and Persian Gulf while our known inland communities were developing between roughly 23,000 and 12,000 years ago?

We don't know very well.

That's a genuine hole in the evidence.

And there's an even more precise question.

Follow the people, not Hancock.

We have people at Ohalo 23,000 years ago exploiting an extremely rich lakeshore environment --- fish, plants, cereals, huts, hearths. More than 90,000 plant remains were preserved there. (Opportunism or aquatic specialization? Evidence of fres... (https://pmc.ncbi.nlm.nih.gov/articles/PMC6005578/))

Why?

Water is valuable.

Fish.

Plants.

Transport.

Animals.

Food.

Human beings repeatedly favour water margins.

Now imagine global sea level 120 metres lower.

Where are many of the attractive water-margin environments?

Not where the coast is today.

Then sea level rises.

Those landscapes don't merely get abandoned.

They disappear from our normal archaeological search area.

And unlike Ohalo, where drought conveniently lowered a lake and exposed the settlement again, we can't simply drain the Mediterranean or Persian Gulf and have a look. πŸ˜‚

So I'm upgrading one thing from our previous discussion:

"Maybe important evidence is underwater" isn't merely Hancock's convenient excuse.

There is a strong geological reason to predict that a portion of Late Pleistocene coastal archaeology is underwater.

That's quite different.

But we must hold the line:

🟒 Sea levels were dramatically lower: fact.

🟒 Huge habitable landscapes now underwater were exposed: fact.

🟒 Humans occupied waterside environments during this period: fact.

🟒 We therefore have an incomplete sample of coastal human activity: extremely strong inference.

🟑 What those submerged populations were actually doing: substantially unknown.

πŸ”΄ They constituted Hancock's advanced civilisation: we still have no evidence.

And now, Paul, I think we should stop travelling forward in time again.

Because you said something important:

"We keep going back."

I agree.

Let's keep going back.

We've reached Ohalo II at ~23,000 years ago and already found humans building huts, fishing, processing cereals and occupying a substantial lakeshore camp.

So let's ask:

What comes BEFORE Ohalo II?

Not just what artefacts do we have?

For every step backward:

WHEN?

WHERE?

How far from our previous sites?

What were humans demonstrably capable of?

Where was sea level?

Where was the coastline?

And how much potentially habitable land from that world is now underwater?

If we keep doing that until the archaeological trail genuinely goes dark, then we've found the actual boundary of what we know.

That is where Hancock gets interesting --- if there's still room left for him when we arrive. 😏

Paul

this is our story now, this is our search. so yes lets go back. and lets hold onto a fact that is still true today - humans live near a water source. we follow the water, cos water is life. ok, what can you find now going back?

AI

Oh, now our search changes shape.

Let's hold your proposition, but phrase it carefully enough that Tester Paul doesn't kick us later:

Humans need reliable freshwater. Therefore, when reconstructing where prehistoric people were likely to live and move, follow the water.

Not everyone lives beside water. Mobile groups can travel between sources. But no human population gets to opt out of needing freshwater.

And when I followed that backwards, I found something I wasn't expecting.

From Ohalo II backwards: ~23,000 β†’ 50,000 years

Ohalo II at ~23,000 years ago is our beautiful benchmark: huts beside the Sea of Galilee and intensive, year-round fishing. Researchers recovered around 17,000 fish remains, and the evidence indicates deliberate exploitation of the freshwater environment. (Opportunism or aquatic specialization? Evidence of fres... (https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0198747))

Now go back another ~8,000 years.

At Nazlet Khater in Egypt, around 31,500--33,000 years ago, people were living beside the Nile system and doing something quite sophisticated: extracting chert using what researchers describe as a complex mining strategy, then producing blade technology. (Blade technology in the egyptian-nile valley: some new... (https://pubmed.ncbi.nlm.nih.gov/17783308/))

So:

23k --- Sea of Galilee β†’ people. 33k --- Nile Valley β†’ people.

Same broad eastern Mediterranean/Northeast African world, separated by roughly 500--700 km.

Go farther back.

Around 46,000 years ago, Manot Cave in northern Israel contains Early Ahmarian occupation. Around 38--34k, the Levantine Aurignacian follows. Other contemporary sites extend north through Lebanon and southern Turkey. (Radiocarbon chronology of Manot Cave, Israel and Upper... (https://pmc.ncbi.nlm.nih.gov/articles/PMC5687856/))

KsÒr 'Akil in Lebanon has modern-human Upper Palaeolithic occupation before about 45,900 years ago. And importantly for our geography, Üçağızlı Cave in southern Turkey lies near the mouth of the Orontes River, about 250 km north of KsÒr 'Akil. (New chronology for KsÒr 'Akil (Lebanon) supports Levant... (https://pmc.ncbi.nlm.nih.gov/articles/PMC4485136/))

So we're still not randomly jumping around Earth.

We're following something geographically coherent:

Nile β†’ Levantine coast β†’ Jordan/Dead Sea Rift β†’ Orontes β†’ Anatolia/Mesopotamia.

And then I kept going.

~85,000--120,000 years ago: Arabia suddenly lights up

This is where our follow-the-water test gets really bloody interesting.

Today, look at northern Arabia and you see enormous desert.

That's the wrong map for parts of the Pleistocene.

At Al Wusta, deep in today's Nefud Desert in Saudi Arabia, researchers found an actual Homo sapiens finger bone associated with stone tools.

Date:

~95,000--86,000 years ago.

And what was Al Wusta then?

A freshwater lake.

The palaeoenvironment shows increased rainfall, grassland and animals. Humans were moving into Arabia during a humid period. (Homo sapiens in Arabia by 85,000 years ago | Nature Eco... (https://www.nature.com/articles/s41559-018-0518-2))

Then there's Alathar.

Around 120,000 years ago, humans literally left footprints beside another freshwater lake.

Not metaphorically following water.

We can see where they DAMN walked.

The lake was shallow freshwater, surrounded by a semi-arid grassland. Researchers conclude that humans and other animals were visiting it for drinking and foraging. (Human footprints provide snapshot of last interglacial... (https://pmc.ncbi.nlm.nih.gov/articles/PMC7500939/))

That's a lovely little green flag for our hypothesis:

Find ancient water β†’ find humans.

But it gets much better.

Someone already ran our experiment.

Researchers studied ancient lake beds in the Nefud Desert at Khall Amayshan 4 and Jubbah.

They found archaeological material associated with separate freshwater-lake phases at approximately:

400,000 years ago 300,000 years ago 200,000 years ago 130,000--75,000 years ago 55,000 years ago

Five separate windows of hominin occupation.

And every one corresponds with periods when Arabia became wetter and lakes returned. (Multiple hominin dispersals into Southwest Asia over th... (https://pmc.ncbi.nlm.nih.gov/articles/PMC8443443/))

That is an astonishingly clean result for our question.

Dry Arabia:

humans largely disappear from the record.

Wet Arabia:

water β†’ animals β†’ humans.

Water disappears.

Humans disappear.

Water returns.

Humans return.

And we're not necessarily talking about the same humans. The stone-tool traditions differ between phases, suggesting different populations moved through during different climatic windows. (Multiple hominin dispersals into Southwest Asia over th... (https://pmc.ncbi.nlm.nih.gov/articles/PMC8443443/))

There is even an Acheulean site at An Nasim beside what was once a deep freshwater lake around 350,000--250,000 years ago. The researchers explicitly conclude that palaeolakes formed a viable habitation network enabling hominin movement through Arabia. (The expansion of Acheulean hominins into the Nefud Dese... (https://pmc.ncbi.nlm.nih.gov/articles/PMC8115331/))

So I'm comfortable putting this on our board:

🟒 Water availability is a powerful predictor of Pleistocene human movement through Southwest Asia.

That's considerably stronger than where we started.

But now sea level complicates our experiment beautifully.

Because there's another kind of water.

Coasts.

And sea level isn't moving monotonically in one direction as we go backwards.

It oscillates enormously.

Around 21,000 years ago, global sea level was roughly 120 m lower than today.

But around 125,000 years ago, during the last interglacial, it was actually approximately 4--6 m higher than today. (Past, Present, and Future Sea Level Rise and Effects on (https://pubs.usgs.gov/sir/2009/5252/downloads/Chapter-C/Chapter-C.pdf))

So our investigation now has two different water maps.

Freshwater map

Rivers.

Lakes.

Springs.

Wetlands.

These appear and disappear as climate changes.

Coastal map

Sea level moves up and down by more than 100 metres through glacial cycles.

Therefore the places humans could occupy around water are constantly moving.

That's hugely important.

Because at 120,000 years ago, those humans walking beside Alathar lake weren't necessarily living in some enormous exposed Persian Gulf basin like people could much later during glacial lowstands.

Different climate.

Different sea level.

Different map.

We cannot simply take today's geography --- or even the 20,000-year-old geography --- and project it backwards.

That's a flag. 🚩

And here's where I think our search gets really good.

We've been travelling backwards through known archaeological sites.

But following water suggests another method.

Instead of asking:

Where have archaeologists found old humans?

ask:

Where does palaeogeography tell us humans had the WATER necessary to live?

Then ask:

Have archaeologists looked there?

That's completely different.

And scientists working in Arabia are effectively already doing it. They use satellite imagery, palaeohydrological modelling and sediments to locate ancient lakes --- and then search them for archaeology. That's how some of this emerging Arabian record has been found. (The expansion of Acheulean hominins into the Nefud Dese... (https://pmc.ncbi.nlm.nih.gov/articles/PMC8115331/))

And look what happens.

A region previously characterised as an enormous hostile desert turns into:

lake β†’ humans lake β†’ humans lake β†’ humans lake β†’ humans

over hundreds of thousands of years.

That should make us very cautious about interpreting a blank archaeological map as a blank human landscape.

So here's our map now

We're no longer merely tracing:

GΓΆbekli β†’ KΓΆrtik β†’ Natufian β†’ Ohalo β†’ earlier Levant.

We've discovered a much bigger geographical system.

Think of Northeast Africa/Southwest Asia as connected by water corridors:

Nile

↓

Sinai / Levant

↓

Jordan Rift / freshwater lakes

β†— Mediterranean/Orontes corridor

β†˜ Arabian palaeolake networks

β†˜ Euphrates/Tigris

β†˜ Persian Gulf basin

And depending on climate and sea level, those routes open, close, shift and reconnect.

That seems much closer to how we should think about prehistoric people than modern country borders.

And now I have a problem for our hypothesis.

A really good one.

If we're following water, we shouldn't automatically keep following Homo sapiens.

Because the evidence around Arabian freshwater goes back 400,000 years.

Long before modern humans reached many of these places.

Which means our next question becomes:

How long have humans --- broadly, the Homo family --- been deliberately following and exploiting water landscapes?

And I already have one clue.

The Ohalo fishing paper notes evidence for freshwater exploitation going back approximately 1.95 million years. (Opportunism or aquatic specialization? Evidence of fres... (https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0198747))

Nearly.

Two. Million. Years.

πŸ˜‚

So if we're serious about:

Water is life. Follow the water.

I don't think our next stop is 50,000 years ago.

I think we follow that parameter all the DAMN way back and see whether it breaks.

Because if humans/hominins have clustered around rivers, lakes and wetlands for hundreds of thousands --- perhaps millions --- of years, then when we eventually come forward again toward Hancock's 12,800-year question, we will have a much better idea of where the missing archaeology should actually be.

And then we overlay the sea-level curve.

That could get very interesting.

Paul

yes. i am pleased you said sea level curve, cos lakes are easy in the model. but fresh water runs 'down hill' so to speak, so only lookig at who is living by lakes is a closed view... you need to look at rivers, to the sea... level. cos that will give a much more interesting movement of people... i mean we still love living near the sea... as long as its got a river running into it right? i am really looking for that connection between 'fresh water' stuff that show 'people' and the rise and fall of the 'sea' if that all makes sense. then we can figure out what the next lens needs to be. cos it might be tribal - fighting, or as wild as it sounds, trade.

AI

Yes. It makes complete sense. And I think you've just improved our search model again.

We've been treating freshwater and sea level as two layers. They aren't independent.

They're one connected system.

Rain/snow β†’ springs/lakes β†’ streams β†’ rivers β†’ estuaries β†’ sea.

And humans can exploit almost that entire chain. So if sea level falls 100 metres, the river doesn't stop at today's coastline and dump its water onto a dry beach. Its lower course extends across the newly exposed continental shelf toward the new sea.

That's the bit we need to follow.

The Persian Gulf is our best natural experiment

This one is almost ridiculously good for what you're proposing.

Today:

Tigris + Euphrates + Karun β†’ Shatt al-Arab β†’ Persian Gulf.

But during the Last Glacial Maximum, sea level was roughly 120--130 m lower, and most of the present Gulf floor was dry land. Marine water didn't reach the Strait of Hormuz until around 14,000 years ago, and a true seaway into the Gulf began around 12,500 years ago. (Late Quaternary sea-level changes of the Persian Gulf -... (https://www.sciencedirect.com/science/article/pii/S0033589415000381))

So follow the freshwater downhill.

Those rivers didn't terminate around modern Basra.

They continued into the exposed Gulf basin.

Research reconstructs a combined Tigris--Euphrates--Karun drainage running toward Hormuz, supplemented by the Wadi Batin and groundwater/aquifers. Jeffrey Rose's published "Gulf Oasis" hypothesis argues that this created a potentially attractive freshwater landscape across an area that, at maximum exposure, could approach the size of Great Britain. (New Light on Human Prehistory in the Arabo-Persian Gulf... (https://www.jstor.org/stable/10.1086/657397))

Now think about this without Hancock for a moment.

We already know humans repeatedly occupy the upstream water network.

We know people were using Arabian palaeolakes during humid periods.

We know people occupied the Levant.

We know people eventually occupied the Upper Tigris/Euphrates.

We know Pleistocene populations moved through Iran, and modern palaeohydrological modelling finds that access to freshwater strongly predicts plausible movement corridors. (Defining paleoclimatic routes and opportunities for hom... (https://pmc.ncbi.nlm.nih.gov/articles/PMC9977010/))

And downstream sits:

a giant river valley.

Fresh water.

Springs.

Wetlands potentially.

Animals following water.

Plants.

Fish.

Eventually an estuary.

Then a coastline.

That's exactly the sort of landscape we'd predict people would exploit.

Now move the sea.

This is the really interesting parameter.

Imagine a simplified cross-section:

20,000 BP --- sea ~120 m lower

Tigris/Euphrates ↓ river valley ↓ freshwater wetlands/lakes/springs ↓ long river running through today's Gulf ↓ estuary ↓ coast near Hormuz

Then warming begins.

~14,000 BP

Sea reaches Hormuz.

Now the lowest river valley starts drowning. (Late Quaternary sea-level changes of the Persian Gulf -... (https://www.sciencedirect.com/science/article/pii/S0033589415000381))

~12,500 BP

Marine water is pushing farther into the Gulf.

So the freshwater/estuary/coastal sweet spot is moving northwest.

The people don't necessarily experience:

THE GREAT FLOOD!

They may experience something much more mundane and relentless:

Grandad's coastline isn't our coastline anymore.

The river mouth moves.

Wetlands move.

Fishing grounds move.

Freshwater availability changes.

Lowlands disappear.

People move.

And this continues over generations.

By roughly 6,000--7,000 BP, the Gulf has approached something much closer to its modern extent, with later regional relative sea-level fluctuations. (Late Quaternary sea-level changes of the Persian Gulf -... (https://www.sciencedirect.com/science/article/pii/S0033589415000381))

That's a human migration mechanism requiring zero lost civilisation and zero catastrophe.

Just:

Follow the water.

And something important happens to archaeology.

The people who lived along the old lower river aren't necessarily underneath 20 metres of soil where we can excavate them.

Their sites can now be:

under the Persian Gulf.

And published archaeology explicitly recognises this problem. Bailey and Flemming describe Pleistocene sea-level change as creating a major bias in our understanding of coastal settlement and dispersal because lowered sea levels were normal for much of the Pleistocene; many former coastal landscapes are now submerged. (Archaeology of the continental shelf: Marine resources,... (https://researchnow.flinders.edu.au/en/publications/archaeology-of-the-continental-shelf-marine-resources-submerged-l/))

In lower Mesopotamia there's an additional problem.

Some former settlements may be under Gulf water, while others are buried beneath enormous quantities of later river sediment as the delta migrated. Reconstructions note that early coastal plains can now lie either submerged or buried under the lower Mesopotamian alluvial plain. (reconstruction of the holocene coastline of mesopotamia... (https://onlinelibrary.wiley.com/doi/pdf/10.1002/gea.3340020102))

So our missing archaeology has two hiding places:

water and mud.

πŸ˜‚

That matters.

But your next thought --- movement of people --- is where I think this becomes really interesting.

Don't picture dots representing archaeological sites.

Picture moving bands of opportunity.

When Arabia becomes wetter:

freshwater expands inland β†’ people can expand inland.

We actually have archaeological evidence consistent with repeated humid-period expansions. Studies identify Arabian humid windows around ~200--190k, 170k, 155k, 130--120k, 105--95k, 85--75k and 60--55k years ago. (Orbital-scale climate variability in Arabia as a potent... (https://www.sciencedirect.com/science/article/pii/S1040618215000117))

Then climate dries.

Interior lakes disappear.

Where can people go?

Permanent rivers.

Springs.

Coasts.

The exposed Gulf basin.

Rose's model explicitly proposes this kind of movement: populations contracting toward coastal/Gulf refugia during hyper-arid periods and expanding outward when conditions improved. It's a hypothesis rather than established demographic history, but importantly, it's a hypothesis built from palaeoclimate and hydrology rather than Hancock's lost civilisation. (New Light on Human Prehistory in the Arabo-Persian Gulf... (https://www.journals.uchicago.edu/doi/full/10.1086/657397))

So we get something like:

WET

Water spreads β†’ habitat spreads β†’ animals spread β†’ humans spread

DRY

Water contracts β†’ habitat contracts β†’ animals concentrate β†’ humans concentrate

Meanwhile:

COLDER / more ice

Sea falls β†’ continental shelf exposed β†’ rivers lengthen β†’ new coastal plains appear

WARMER / melting ice

Sea rises β†’ lower rivers drown β†’ estuaries migrate β†’ coastal plains disappear β†’ humans move again

Now we've got a dynamic human landscape rather than a static archaeological map.

And suddenly I'm very interested in your last two possibilities.

Because eventually humans meet humans.

If freshwater concentrates people into particular corridors and refugia, then those waterways aren't merely survival resources.

They become meeting places.

And once two groups meet, several things can happen:

fight

avoid each other

merge

mate

copy

exchange

trade

And trade isn't wild at all.

In fact, I think it may be our logical next lens.

Because trade gives us something extremely useful that social organisation alone doesn't:

objects move farther than people.

If a piece of obsidian originated at geological source A but we find it hundreds of kilometres away at settlement B, somebody moved it.

Maybe one person travelled the whole distance.

Maybe twenty groups passed it along.

We don't need to know yet.

But we've demonstrated connection.

Then if we find:

A ↔ B ↔ C ↔ D

along the water corridors we've reconstructed...

we've started turning our water map into a human network.

And that's exactly what we need before we ever entertain Hancock's "knowledge was transmitted between populations" proposition.

Because his claim requires connection.

So far we've been asking:

Were people there?

The next question should probably become:

Were these populations communicating with one another?

And fortunately archaeology gives us some beautiful test parameters:

obsidian source β†’ destination

marine shells found inland

stone sourced hundreds of kilometres away

shared tool technologies

pigments/materials with identifiable geological origins

And eventually:

genetics.

Those are independent signals.

So I think our investigation now has a really clean sequence:

WATER Where could humans survive?

↓

SEA LEVEL Where was that water system then rather than now?

↓

PEOPLE Do we find humans along those reconstructed corridors?

↓

MOVEMENT Do their sites shift as water/climate/sea level shifts?

↓

CONNECTION

Do materials, technologies or genes move between those populations?

And only after we've established connection should we ask:

What were they exchanging?

Because if we find long-distance exchange networks before GΓΆbekli Tepe, then we've found a mechanism by which ideas can travel without needing Hancock's travelling master civilisation.

And if those networks go back much further than expected...

Well.

Then we've got another DAMN hole to dig. πŸ˜‚

I think obsidian and marine shells should be our next test. They give us something wonderfully concrete: where did this object originate, where did we find it, how old is it, and what water/sea-level landscape lay between those two points at the time?

Paul

yes, and lets be clear here, seeing movement of objects doesn't prove fighting or trading. it just proves that for some reason, those objects had 'value' to the poeple of that time. ok do it

AI

Exactly. Object movement proves movement of the object. Full stop. It does not by itself prove trade, gift exchange, theft, warfare, pilgrimage, marriage networks, migration, or one person walking the whole route.

And when I looked for the oldest clean examples, I found something much more interesting than I expected.

First hard signal: obsidian moved a very long way surprisingly early

At Yabroud Rockshelter II in southern Syria, an obsidian scraper from roughly 41,000--32,000 years ago was chemically matched to the Gâllü Dağ volcanic source in central Anatolia. The straight-line geography is misleading because of terrain/sea; the estimated practical route is 700+ km on foot.

That gives us a very clean fact:

By ~40,000--30,000 years ago, a material originating in central Anatolia had reached southern Syria.

That is way before GΓΆbekli Tepe.

And there's a roughly contemporary clue at Shanidar Cave in northern Iraq, where obsidian flakes were sourced to eastern Turkey and/or the Caucasus, implying movement on a similarly large geographic scale.

So our network map is already bigger than:

local camp β†’ next-door camp.

We have objects crossing substantial parts of Southwest Asia tens of thousands of years before farming.

But your correction matters enormously here

The archaeological papers often use words like "exchange" or "social networks."

Those are interpretations.

Our stricter statement is:

🟒 Obsidian moved 700+ km from its geological source.

🟑 How did it move? Unknown.

Possibilities include:

one person carrying it a long way;

people moving over generations;

group-to-group transfer;

gifts;

barter/trade;

marriage or kinship relationships;

ritual movement;

scavenging an older object;

conflict/theft;

or mechanisms we haven't thought of.

We simply don't have the receipt. πŸ˜‚

But we can infer one thing reasonably strongly:

Somebody considered keeping/moving that material worthwhile.

Because unlike a naturally distributed material, obsidian has geographically restricted volcanic sources. It did not magically occur in Yabroud.

That's our value observation.

Not monetary value.

Human value.

Utility, appearance, rarity, identity, symbolism --- we don't yet know which.

Now shells give us a second independent parameter

This is particularly useful because shells don't behave like obsidian.

At Epipalaeolithic/Natufian sites, marine shells from both the Mediterranean and Red Sea turn up inland. At WF16 in southern Jordan, for example, more than 400 marine shell beads include species originating from both seas. Comparable long-distance shell movement appears at other Epipalaeolithic sites.

And sites in southwestern Syria roughly 90 km from the Mediterranean contain marine shell beads during the late Epipalaeolithic.

Even earlier, shell ornament traditions extend well back into the Upper Palaeolithic and in some cases the Middle Palaeolithic.

So now we've changed material type:

Obsidian --- useful cutting stone, geographically restricted.

Shells --- often ornaments, sometimes from two completely different marine environments.

Yet both move away from their source.

That makes "humans valued non-local objects" substantially harder to dismiss as an odd obsidian-specific behaviour.

And around 15,000 years ago at Kebara Cave, people weren't merely collecting shells; they were colouring shell beads red with a plant-derived pigment.

That's useful evidence that these objects could carry symbolic/aesthetic value, not merely survival utility.

Now put location back into it

Our Yabroud obsidian:

source: central Anatolia ↓ through/around northern Levant ↓ destination: southern Syria

The later Natufian/PPNA shell evidence:

Mediterranean coast ↔ inland Levant

and separately:

Red Sea ↔ inland Levant

Then by the early Neolithic, Cappadocian obsidian reaches Jordan, with provenance work indicating distinct routes/networks rather than one simple north-to-south pipeline.

And at Jerf el Ahmar on the Middle Euphrates, most analysed obsidian came from Cappadocia, with a small amount from the BingΓΆl region of eastern Anatolia.

So by the period around and after GΓΆbekli Tepe, we can demonstrably draw object connections across:

Central Anatolia ↔ Upper/Middle Euphrates ↔ Southern Levant

and

Mediterranean / Red Sea ↔ inland settlements

That is a network-shaped archaeological landscape.

Not proof of trade.

Not proof of political unity.

Not proof of an advanced civilisation.

But definitely not isolated little tribes that never interacted.

And here is the really important part for our search

The 700+ km Yabroud obsidian is from ~41--32 thousand years ago.

That means long-distance object movement existed at least ~20,000 years before GΓΆbekli Tepe.

So if we later discover ideas, tool styles or symbolism moving between regions, we already have a physically demonstrated mechanism by which material --- and therefore potentially people/contact --- could span enormous distances.

We do not need Hancock's travelling civilisation to invent long-distance human connection.

Humans were already capable of maintaining movement networks over huge landscapes.

But equally, we shouldn't say:

"Aha, therefore ideas were definitely being transmitted 700 km."

No.

The scraper proves the scraper got there.

That's it.

πŸ˜‚

Now let's overlay your water idea

This is where I'd put a new flag.

Look at the geography connecting these regions:

Central Anatolia β†’ Euphrates/Orontes corridors β†’ Levant

Upper Tigris β†’ Mesopotamia

Jordan/Dead Sea Rift

Mediterranean coast

Red Sea corridor

These are not random straight lines.

They overlap heavily with river valleys, basins, coastlines and ecologically viable corridors.

I don't yet have enough evidence to claim:

the obsidian travelled specifically along rivers.

That's our next parameter to test.

But the question is now sharp enough:

When we reconstruct the likely route between source and destination at the sea level and climate of the period, does object movement follow the same freshwater/coastal corridors our human-presence evidence follows?

If yes, that's a strong convergence.

And then we add another parameter:

Quantity.

One obsidian scraper 700 km away tells us:

possible connection.

Hundreds or thousands of sourced objects moving repeatedly along the same corridor over centuries tells us something much stronger:

persistent connection.

Still not automatically trade.

But the more repeated the pattern becomes, the less plausible a one-off traveller or accident becomes.

So I think our next test should deliberately ask:

How does the quantity and frequency of long-distance material change through time?

Something like:

40,000 BP: isolated exotic objects?

↓

20,000 BP: more frequent?

↓

15,000 BP Natufian: regular shell/non-local material movement?

↓

12,000--10,000 BP: increasingly dense obsidian/material networks?

↓

GΓΆbekli Tepe period: what does the network look like?

If that curve exists, then we may be watching human connectivity intensify alongside increasing settlement and social complexity.

And that could give us a very different explanation for GΓΆbekli Tepe:

Not:

lost civilisation β†’ gives knowledge

but perhaps:

water corridors + population concentration + increasingly persistent long-distance connections β†’ ideas and materials circulate more efficiently β†’ complexity accelerates.

That's only a hypothesis right now.

But unlike "lost civilisation," we're starting to build it from independently measurable parameters.

And, annoyingly, I think we've just found our next lens:

Not "trade." Connectivity.

How connected were human groups, when, where, and how did that connectivity change as freshwater systems and coastlines moved?

That's the one I'd run next.

Paul

ok, this is so interesting, connectivity next and tell me i'm not the first to be looking at it thru this lens please?

AI

πŸ˜‚ No, Paul. You are absolutely not the first person to look through this lens.

And I'm actually pleased about that, because if we'd accidentally invented an entire branch of archaeology over coffee, I'd be getting worried about us. πŸ˜‚

What is interesting is that we've arrived independently at a combination of several real research approaches: palaeohydrology, landscape archaeology, submerged-landscape archaeology, provenance analysis, and archaeological network analysis.

And some researchers are asking questions remarkably close to ours.

Someone has literally modelled prehistoric connectivity using obsidian

Eiluned Pearce and Theodora Moutsiou published a paper called "Using obsidian transfer distances to explore social network maintenance in late Pleistocene hunter-gatherers."

Their starting problem is almost exactly ours:

How large were prehistoric human networks, and how did people remain connected?

They use the distance obsidian moved as a proxy for the possible geographic extent of social connections. Their analysis suggests Late Pleistocene modern humans could potentially maintain connections across very large tribal territories---hundreds of kilometres in some environments. (Using obsidian transfer distances to explore social net... (https://www.sciencedirect.com/science/article/pii/S027841651400049X))

Important Tester-Paul warning:

Obsidian moving 400 km β‰  proof that one 400-km social network existed.

They're explicitly using transfer distance as a proxy and comparing it with ethnographic models.

But they're asking our question.

Connectivity.

So: 🟒 we're on established scientific ground.

And then I found something even closer to where we are heading.

Researchers modelled obsidian distributions across the early Near Eastern Neolithic as complex networks.

Their modelling found that simple neighbour-to-neighbour, "down-the-line" movement wasn't sufficient to reproduce the observed distribution. Their model required distant links between communities, and they argue that network complexity increased through the PPNA and PPNB. (Testing complex networks of interaction at the onset of... (https://pmc.ncbi.nlm.nih.gov/articles/PMC4590507/))

Now look at the dates.

PPNA begins around the period we're interested in:

~11,700 BP onward.

So around GΓΆbekli Tepe's world, researchers already see evidence consistent with increasingly complicated long-distance human interaction.

Again, let's use our language rather than theirs:

Objects moved through increasingly complex patterns. 🟒

People were therefore connected in some fashion: very strong inference. 🟒/🟑

They were trading: possible, not demonstrated merely by distribution. 🟑

Knowledge travelled along those connections: extremely plausible, but requires its own evidence. 🟑

That's exactly our discipline.

And the Upper Tigris gives us a particularly lovely example.

Remember our water system?

Tigris β†’ Euphrates/Mesopotamian world β†’ Gulf.

At Hasankeyf HΓΆyΓΌk, an early PPNA settlement on the Upper Tigris, researchers sourced obsidian and specifically investigated whether river transportation helped move it.

So somebody has even asked our:

"Hang on... did the river itself help create the network?"

question. πŸ˜‚ (Hunter-fisher-gatherer river transportation: Insights f... (https://www.sciencedirect.com/science/article/pii/S1040618220305942))

And the broader Early Neolithic research now talks explicitly about "social networking", using portable ornaments and non-local materials to investigate the intensity of connections between communities. ('Far and wide': Social networking in the Early Neolithi... (https://www.sciencedirect.com/science/article/abs/pii/S0003552122000619))

So far:

Water β†’ people β†’ objects β†’ connectivity

is not remotely crackpot territory.

It's archaeology.

But here's where I think our particular lens gets more interesting.

The different pieces tend to live in somewhat different research questions.

One group asks:

Where was freshwater available?

And they've reconstructed ancient rivers, wetlands and lakes across Arabia specifically to identify viable hominin movement corridors between Africa, the Levant and Arabia.

They've even identified a proposed Tabuk Corridor, where reconstructed freshwater availability and archaeological distributions correspond. (Palaeohydrological corridors for hominin dispersals in... (https://www.sciencedirect.com/science/article/abs/pii/S0277379116301494))

Another group asks:

What did sea-level change do to human settlement?

Bailey and Flemming have been banging this drum for years. Their argument is strong: for most of the Pleistocene, sea levels were substantially lower than today, meaning our surviving terrestrial archaeological record is systematically biased because enormous potentially attractive coastal landscapes are now submerged. (Archaeology of the continental shelf: Marine resources,... (https://www.sciencedirect.com/science/article/pii/S0277379108001911))

And this isn't merely theoretical.

More than 3,000 prehistoric seabed sites had already been identified by 2012, ranging from about 500,000 to 5,000 years old, at depths reaching around 130 metres. (Value of submerged early human sites | Nature (https://www.nature.com/articles/486034a))

That's another important flag:

🚩 Humans really did occupy landscapes that are now underwater.

Not Hancock.

Not hypothesis.

Archaeology.

And another research group asks:

How connected were communities?

They use obsidian, shells, ornaments, tool traditions and network modelling. (Using obsidian transfer distances to explore social net... (https://www.sciencedirect.com/science/article/pii/S027841651400049X); Testing complex networks of interaction at the onset of... (https://pmc.ncbi.nlm.nih.gov/articles/PMC4590507/))

What we're doing is stacking those lenses.

And I think that's useful:

1 --- Reconstruct the water.

Where were the rivers, lakes, springs, wetlands?

↓

2 --- Reconstruct sea level.

Where did those rivers actually terminate then?

Where were the estuaries?

Where was the coastline?

What land existed that doesn't exist now?

↓

3 --- Plot humans.

Where does archaeology show people?

↓

4 --- Plot valued objects.

Where were they sourced?

Where were they found?

↓

5 --- Infer the minimum connectivity required.

Not why the object moved.

Just:

Some mechanism connected these locations.

↓

6 --- Watch it through time.

This is the bit I'm increasingly interested in.

Does connectivity:

increase?

decrease?

move geographically?

follow freshwater corridors?

retreat toward major rivers/coasts during dry periods?

expand during wet periods?

shift as sea level moves?

Because then we're no longer looking at individual archaeological sites.

We're watching a human network breathe with the landscape.

That phrase is mine, not an archaeological claim --- but I think it describes the hypothesis we're developing rather nicely.

And here's where your thought about value becomes really useful.

Suppose we find:

obsidian source A β†’ object 700 km away.

We know someone valued it sufficiently that it moved.

Now suppose we find:

A β†’ B β†’ C β†’ D

across a water corridor.

Interesting.

Then 5,000 years later:

A β†’ B β†’ C β†’ D β†’ E β†’ F

and quantities increase.

That's more interesting.

Then perhaps environmental conditions deteriorate and:

C and D disappear while A/B and E/F remain.

Then water returns and:

connections reappear.

Now we're starting to observe network behaviour rather than merely objects.

And then your tribal idea becomes testable.

Because eventually we can overlay another kind of evidence:

Genetics.

Genes move only when people reproduce.

That's an entirely different parameter from obsidian.

So imagine:

water corridor βœ“

human occupation βœ“

object movement βœ“

genetic movement βœ“

Now our confidence in actual people-to-people connectivity becomes much stronger.

Then add:

similar technologies

shared symbols

similar burial practices

and perhaps eventually:

shared architecture.

Now we're getting somewhere genuinely powerful.

And I can already see one thing we have to be incredibly careful about.

If we find:

same symbol in Turkey and Jordan

we cannot immediately say:

"idea travelled."

Independent invention exists.

If we find:

same symbol + same unusual manufacturing technique + obsidian moving between those regions + genetic connection + viable water corridor at the same time...

Different DAMN conversation.

Because multiple independent parameters are converging.

That's exactly how our experiment should work.

And this brings us quietly back toward Hancock.

His central idea requires connectivity.

People have to be able to move knowledge enormous distances.

We've now established that long-distance human connectivity itself is not remotely implausible.

In fact, evidence for substantial prehistoric connectivity is real and very old. (Using obsidian transfer distances to explore social net... (https://www.sciencedirect.com/science/article/pii/S027841651400049X); Testing complex networks of interaction at the onset of... (https://pmc.ncbi.nlm.nih.gov/articles/PMC4590507/))

Where Hancock still hasn't earned the next step is:

Was there one source civilisation responsible for disseminating that knowledge?

Connectivity doesn't prove that.

In fact, increasing connectivity might eventually give us a completely different explanation:

Complexity could emerge from the network itself.

One group discovers something.

Another modifies it.

Another combines it with something else.

Information moves back.

Population concentrates.

Connections increase.

Useful ideas persist.

Bad ideas disappear.

Nobody needs to possess the whole system.

And then eventually...

GΓΆbekli Tepe.

That's only a hypothesis right now.

But DAMN me, Paul --- that's a hypothesis we can actually test pieces of.

And I think I know our next parameter:

GENES.

We've followed water.

We've followed objects.

Now let's follow people themselves.

Specifically: between about 50,000 and 10,000 years ago, what does ancient DNA tell us about population movement between Anatolia, the Levant, Mesopotamia, Arabia and surrounding regions?

Then we overlay that onto the water/sea-level/object network we've already built.

Because if the genes and the objects move through the same corridors at the same times...

I reckon both of our eyebrows are going up. πŸ˜‚

Paul

Yes genes it is. Also I like that you describe it as like a breathing system. Inhale exhale with the lens/water parameters we are using. also before I forgot. I want to say sea floor sediment. Cos I imagine the areas we are talking about aren't the same right now they weren't the times we are talking about both before and after water. So genes. Do it!

AI

Yes --- genes give us a much stronger test of connectivity than objects do. And the result is already interesting: the Near East before and around GΓΆbekli Tepe was not one homogeneous population, but neither was it a set of sealed-off tribes. It looks more like several distinct regional populations that increasingly mixed.

And your sea-floor sediment point absolutely belongs in the model. I'll come back to that, because it affects what genetic evidence we can even recover.

Start around 15,000 years ago: central Anatolia

We have genome-wide DNA from a hunter-gatherer at Pınarbaşı in central Anatolia, dated roughly 15,000 years ago. That genome is especially useful because it predates agriculture there.

The striking result is continuity. When farming later appears in central Anatolia, the early farmers derive roughly 80--90% of their ancestry from that local hunter-gatherer-related population.

So agriculture in central Anatolia does not look like an outside population arrived, replaced everyone and taught them how to farm.

Mostly:

local people changed what they were doing.

But---not completely isolated.

Those early Anatolian farmers also show about 10% ancestry related to populations farther east in Iran/Caucasus, probably arriving during the late Pleistocene or earliest Holocene. And later Anatolians acquire additional ancestry from the Levant.

So already:

local continuity + outside connection.

That's a very useful pattern.

Now the Levant

The Levant has its own genetic history.

Early Neolithic Levantine farmers largely descend from local Natufian-related populations, rather than being replaced by Anatolians.

But there's also Anatolian-related ancestry entering the Levant.

And, beautifully for our connectivity hypothesis, the movement goes both ways.

The same genomic study found:

Levant β†’ Anatolia

and

Anatolia β†’ Levant

during the early Neolithic.

That is much stronger than an obsidian scraper.

The objects told us:

Something connected these places.

Genes tell us:

People from populations related to these places were actually reproducing with one another.

Not necessarily constantly. Not necessarily peacefully. Not necessarily through trade.

But biological connectivity is real.

Move east: the Zagros

Now things get even more interesting.

At Ganj Dareh in the Zagros of western Iran, around 10,000 years ago, ancient DNA shows a population genetically quite distinct from early Anatolian farmers.

They were closer to ancient Caucasus/Iranian hunter-gatherer populations and appear to have developed early goat-herding largely within their own regional population.

So the Fertile Crescent wasn't:

one population invents civilisation β†’ spreads it everywhere.

It was more like:

Levant population

Anatolian population

Zagros population

with each region developing parts of the Neolithic transition while remaining genetically distinguishable.

That's important.

Because it weakens a simple Hancock-style model involving one source population distributing a package of advanced knowledge.

But then...

Later, those populations start mixing.

And this is where our breathing network analogy starts working beautifully.

By the Pre-Pottery Neolithic, genomes from Upper Mesopotamia show mixtures of the three broad ancestry poles:

Anatolian

Levantine

Zagros/Iranian

Some Upper Mesopotamian groups sit roughly intermediate between them rather than belonging cleanly to one. Recent analyses describe those populations as containing substantial ancestry from all three regional sources.

Look at the geography.

Upper Mesopotamia is essentially the zone we've been following:

Upper Euphrates + Upper Tigris

between Anatolia, the Levant and the Zagros.

That's almost exactly where we'd predict a connectivity zone if our water-corridor hypothesis has merit.

Not proof water caused the mixing.

But geography + genes agree.

And that's good information.

Then Cyprus gives us another independent test

Humans reaching Cyprus obviously can't simply walk there.

Ancient-genome modelling of early Cypriot populations suggests substantial ancestry from central Anatolia, plus Levantine ancestry, with the mixing event estimated somewhere around 14,000--10,000 BCE.

Again, careful wording:

This doesn't prove a specific voyage or route.

But:

people were genetically connected across the eastern Mediterranean before/around the early Neolithic.

And archaeology independently tells us humans were capable of seafaring there.

That's another layer of connectivity.

So our genetic map begins looking like this

Very roughly:

Central Anatolia strong local continuity

↕ genetic input

Levant

↕ genetic input

Upper Mesopotamia

↔

Zagros / Iran

And then westward:

Anatolia β†’ Cyprus / Aegean

This isn't a single outward arrow.

It's a network.

And that matters.

Because it suggests that complexity doesn't necessarily need a master population transmitting everything.

We have increasingly good evidence for different groups staying culturally/genetically distinctive while exchanging people and probably ideas.

That's exactly the kind of system in which innovations could accumulate across a network.

Someone develops something here.

People move.

Someone combines it with something there.

Another population keeps something else.

Over generations, the network becomes more complex than any single group.

Our breathing system metaphor becomes:

climate opens corridors β†’ populations expand β†’ contacts increase

climate closes corridors β†’ populations contract β†’ groups become more isolated

corridor opens again β†’ previously separated groups meet β†’ genes mix

And we actually have a beautiful much older Arabian example consistent with that general mechanism: hominin occupation of Arabia repeatedly appears during wetter windows at roughly 400k, 300k, 200k, 130--75k and 55k years ago.

So climate really does seem capable of making the human network inhale and exhale.

But here is the BIG limitation.

Ancient DNA is horribly biased by preservation.

Hot, humid environments destroy DNA much faster than cold, dry caves.

That means our genetic map is not:

"where the people were."

It's:

"where people died + remains survived + archaeologists found them + the bones still contained usable DNA."

That's a DAMN massive sampling filter.

And now your sea-floor sediment observation becomes critical.

Suppose significant populations lived along:

lower Tigris/Euphrates river channels

the exposed Persian Gulf plain

Mediterranean palaeocoasts

during lower sea levels.

Sea rises.

What happens to their remains?

Several possibilities:

erosion --- coastlines migrate across them and wave action destroys sites.

burial --- river/delta sediments pile metres of mud over them.

submergence --- sites remain under water.

redeposition --- currents move bones/artefacts away from their original context.

anoxic preservation --- occasionally burial underwater can preserve organic material extraordinarily well.

So today's sea floor isn't simply:

the ancient land surface + water on top.

You're absolutely right.

It's been reworked.

That's why marine geologists use sediment cores, seismic reflection and palaeochannel mapping rather than assuming the current seabed equals the Pleistocene landscape.

And that's especially important in the Persian Gulf.

The Gulf floor has undergone:

river incision when exposed

β†’ estuary formation during sea rise

β†’ marine flooding

β†’ sediment deposition

β†’ delta movement

over thousands of years.

So even if people lived there, finding intact burials suitable for ancient DNA could be exceptionally difficult.

Which creates a really important bias:

Our genes may systematically miss precisely the populations our water model predicts.

That does not prove those populations were special.

But it means:

"we don't see their ancestry directly in ancient DNA"

is weaker evidence than it would be if we had equally good sampling from the drowned landscapes.

That's a flag worth keeping.

Now zoom out farther: 50,000 years

Genetics tells us that most ancestry in present-day non-African populations comes from a major Homo sapiens expansion out of Africa around 50,000 years ago, despite earlier Homo sapiens excursions into Arabia and the Levant.

And climate modelling suggests those dispersals interacted strongly with changing vegetated/water corridors and sea-level conditions.

Then another extraordinary thing happens.

Those humans encounter other humans.

Neanderthals.

We know what happened because almost all non-African populations today retain roughly ~1--2% Neanderthal ancestry.

That's connectivity in the most unequivocal form possible.

And farther east, Denisovan admixture occurred too.

We even have an individual from Denisova Cave whose mother was Neanderthal and father Denisovan.

So by tens of thousands of years ago:

different human populations meet β†’ reproduce β†’ genes move.

There is no serious question that long-distance population networks and interpopulation contact existed.

Here's our scorecard now

Water corridors shaped where people could move: 🟒 strong evidence.

Sea-level shifts dramatically changed the geography of those corridors: 🟒 fact.

Objects moved long distances: 🟒 fact.

Humans themselves moved between Anatolia, Levant, Mesopotamia and Zagros: 🟒 genetic evidence.

Those regional populations mixed: 🟒 fact.

The mixing intensified around the Neolithic transition: 🟒 broadly supported.

The entire Near East was one homogeneous population: πŸ”΄ false.

One civilisation genetically replaced/taught everyone else: πŸ”΄ no evidence.

Important coastal/Gulf populations may be undersampled because their landscapes are drowned/buried: 🟑 strong methodological concern, but population details unknown.

Knowledge could travel through these human networks: 🟑 extremely plausible, but genes themselves don't tell us which ideas travelled.

And now I think we've earned your next lens.

Because we now have:

water movement

object movement

gene movement

Three independent signals of connectivity.

So the next question shouldn't yet be fighting or trade.

I think it's:

Do ideas move with the network?

Can we identify something sufficiently distinctive that appears in:

one place β†’ then another connected place β†’ then another

with chronology supporting spread rather than independent invention?

Tools might work.

But tools can be reinvented.

I think symbolism could be stronger if we choose something sufficiently unusual.

Or even better:

technology with a difficult learning curve.

Something where independent reinvention repeatedly is less likely:

specific stone-working methods

projectile technology

basketry/weaving techniques

boat technology if preserved indirectly

architecture

food-processing methods

Then we can overlay:

same period + connected genetics + object networks + viable water corridor + distinctive technology.

If all five line up...

Now we've got pretty powerful evidence for information moving through the breathing network.

Not Hancock's lost teachers.

But possibly something more interesting:

humans themselves were the information network.

And that could ultimately help explain why complexity starts accelerating.

I reckon that's our next test.

Paul

Yeah I like that model. I know none of what we are doing is new, but it is a very interesting thread to pull with you. What would be incredible would be a visual model that shows all this at play over time. But first tech. Can we see the same ish tech at the same ish time.

AI

Yes. And this is where our model starts producing something measurable.

The careful version of your question is:

At roughly the same time, do geographically separated populations use technologies similar enough --- including the method of making them, not merely the finished shape --- that technological information may have been moving through our connectivity network?

And the answer is yes. But it isn't uniform, and that may actually be the more interesting result.

Go back to ~50,000--40,000 years ago

Across the Levant we see the Initial Upper Palaeolithic transition into technologies archaeologists call the Ahmarian. At Boker Tachtit and related sites, people systematically manufacture pointed blades from carefully prepared cores. Comparable technological assemblages occur from the Negev north through the Levant into places such as Ksar Akil in Lebanon and Üçağızlı in southern Turkey. (Contextualizing the Early Upper Paleolithic in the Nege... (https://www.sciencedirect.com/science/article/pii/S0047248425001368))

So this isn't:

same pointy rock.

It's more useful than that.

It's similarities in the manufacturing sequence --- how the core is prepared, which direction blades are removed, what sort of blanks are being deliberately produced.

That's much harder to explain simply as "everyone needed something sharp."

And we have an especially interesting little experiment at Boker.

The local Initial Upper Palaeolithic technology develops into the southern Ahmarian while maintaining a lot of continuity. But the percussion technique changes from predominantly hard hammer to soft hammer. The researchers interpret that particular change as likely cultural diffusion from the earlier Mediterranean woodland Ahmarian farther north. (Contextualizing the Early Upper Paleolithic in the Nege... (https://cris.haifa.ac.il/en/publications/contextualizing-the-early-upper-paleolithic-in-the-negev-desert-s/))

Now look what that says in our language:

local population/technology continues + a technique being used elsewhere appears = possible information entering the local system.

Not population replacement.

Not civilisation arriving.

Potentially just:

"Hey. Try hitting it this way."

πŸ˜‚

And suddenly the local technology gets modified.

But it gets better, because technology seems capable of moving in BOTH directions.

At Manot Cave in northern Israel we get an Early Ahmarian industry by around 46,000 BP.

Then around 38,000--34,000 BP, something different appears: the Levantine Aurignacian.

And its technological characteristics resemble the Aurignacian traditions of Europe closely enough that researchers seriously consider movement or technological diffusion back from Europe into the Near East. (Radiocarbon chronology of Manot Cave, Israel and Upper... (https://pmc.ncbi.nlm.nih.gov/articles/PMC5687856/))

So potentially:

Levantine technological tradition

β†’ north/west

β†’ technologies develop elsewhere

β†’ some technological traditions subsequently appear back in the Levant.

That's not a one-way ladder of progress.

That's a network.

Which fits our breathing model rather nicely.

Now come forward toward GΓΆbekli Tepe.

Around 15,000--11,700 BP, the Natufian appears across the southern Levant.

And now the technology starts looking strikingly familiar in human terms.

People are producing:

ground stone

mortars

bone tools

hafted cutting implements

microliths

and particularly:

sickles.

Not a random sharp stone someone happens to cut grass with.

Bone handles were manufactured with grooves into which multiple carefully shaped flint inserts were mounted to create a composite tool.

And a brand-new 2026 microscopic study has now confirmed something particularly cool: Natufian sickle blades from el-Wad and Salibiya I really were intensively harvesting cereals. Some show extensive wear, resharpening and maintenance consistent with prolonged/repeated use. (Natufian sickle blades (ca. 15,000--11,700 cal. BP) reve... (https://www.nature.com/articles/s41598-026-54740-5))

So by 15,000 BP, we have a technological system involving:

manufacture handle

β†’ manufacture inserts

β†’ haft inserts

β†’ harvest

β†’ maintain tool

β†’ resharpen

β†’ reuse

That's learned technology.

And importantly, similar sickle technology occurs at multiple Natufian sites rather than one isolated genius making one sickle. (Natufian sickle blades (ca. 15,000--11,700 cal. BP) reve... (https://www.nature.com/articles/s41598-026-54740-5))

So:

technology shared across a regional population: 🟒

Exactly how it spread between groups: 🟑

But here's where I think our question becomes much more powerful.

We shouldn't look for identical technology.

If our breathing-network hypothesis is right, we should predict something subtler:

shared underlying ideas + local variation.

Think about today.

You and I both have smartphones.

One might be Apple.

One Samsung.

One Google.

They're obviously related technological solutions, but they're not identical objects.

If prehistoric communities were connected, we'd expect the same thing.

And that's exactly what archaeologists find in the Ahmarian.

There is a northern Mediterranean version favouring one blade-production approach and a southern/arid version favouring another. Same broader technological tradition, but regional differences in how it is implemented. (Contextualizing the Early Upper Paleolithic in the Nege... (https://www.sciencedirect.com/science/article/pii/S0047248425001368))

That's much more interesting to me than identical arrowheads everywhere.

Because it potentially tells us:

Information can move without culture becoming homogeneous.

Now put our layers together.

We're beginning to get convergence:

WATER

Humans repeatedly occupy viable freshwater corridors.

↓

SEA LEVEL

The coastline, estuaries and lower rivers move dramatically over glacial cycles.

↓

PEOPLE

Human populations move and mix.

↓

VALUED OBJECTS

Obsidian, shells and other materials move hundreds of kilometres.

↓

TECHNOLOGY

Manufacturing traditions appear across geographically separated communities, while retaining local variation.

That's starting to look like a genuine information network.

Not proof of our whole model.

But each layer independently points toward connectivity.

And now I think your inhale/exhale description gives us a genuinely testable prediction.

Imagine connectivity as the size of the network.

INHALE

Climate/water conditions improve.

Habitable corridors expand.

People spread farther apart.

New territory becomes usable.

Some populations become geographically separated.

Local technological variations increase.

EXHALE

Conditions deteriorate.

Freshwater contracts.

Coasts move.

People concentrate around reliable rivers, lakes, springs and favourable coastal areas.

Previously dispersed populations encounter each other more frequently.

Genes mix.

Objects move.

Technology mixes.

Then conditions improve again:

INHALE

Those populations spread outward carrying a new mixture of information.

That is an incredibly interesting prediction.

Because it means technological innovation might sometimes accelerate not during the easiest environmental periods, but around transitions when previously separated populations become connected.

That's something we can actually go looking for.

And your sea-floor sediment thought becomes important here too.

Because our model currently has a visibility problem.

When sea level falls:

river β†’ floodplain β†’ estuary β†’ coast

moves outward across the continental shelf.

People follow those resources.

Then sea rises.

That landscape doesn't simply acquire 80 metres of nice clear water.

First the sea transgresses across it.

Waves erode surfaces.

River channels fill.

Estuaries form.

Sediment gets transported.

Marine sediments accumulate.

Deltas migrate.

Later rivers deposit enormous alluvial sequences.

So our archaeological visibility through time probably looks something like:

upland/cave site: relatively high

inland river terrace: variable

ancient floodplain: buried

ancient delta: deeply buried/reworked

Pleistocene coastline: submerged/reworked

ancient lower river on continental shelf: submerged/buried

Meaning the archaeological map we've been using has a preservation filter sitting over it.

That needs to become a formal layer in the visual model eventually.

But I think we've now earned a much more exciting test.

Rather than asking broadly:

"Did technology spread?"

Let's find a technological fingerprint.

Something complicated enough that we can reasonably distinguish:

independent invention

from

information transmission.

Then plot:

where it first appears

β†’ when

β†’ where it appears next

β†’ water system

β†’ sea level

β†’ known genetic connectivity

β†’ known object connectivity

β†’ local variation

And see whether that technological fingerprint appears to move through our breathing network.

Stone-knapping methods are probably our best candidate because stone survives and archaeologists can reconstruct the actual manufacturing sequence from cores, flakes and waste --- essentially the prehistoric equivalent of leaving the source code behind.

If we can identify one distinctive method and watch it move across 30,000 or 40,000 years of changing geography, we've got our first proper test of:

Did information itself travel through the network?

And yes --- the visual you're imagining is becoming increasingly obvious to me too.

Eventually I don't think it should be a static graph.

It should be a time slider over a map: sea level moves, coastlines breathe, rivers extend/contract, known human sites appear, object connections draw between them, genetic connections overlay those, and technological traditions spread/change.

Then we could literally watch the human network breathe.

But first: one technological fingerprint. That's the next hole I'd dig.

Paul

Yeah you can see how cool it would to watch what you have mapped "breathe" across a map of the earth. I think soon we need to consider what we know about humans too. Because there is a moment in here where I am going (like me) DAMN we have always been damn clever! As we look at the situation in front of us and how we can adapt to survive. Right fingerprints - go.

AI

Oh yes. Nubian Levallois is exactly the kind of fingerprint we wanted --- because it is distinctive enough to be interesting, but contested enough to stop us getting lazy.

The technology is a specific way of preparing a stone core so that the final removal produces a predetermined pointed flake. It isn't merely "people made pointy rocks." The preparation sequence, opposed striking platforms and core geometry are distinctive enough that archaeologists have used it as a possible marker of population movement or cultural transmission.

And look where it turns up.

Around 150,000--70,000 years ago, variants of Nubian Levallois appear across northeast Africa/Nile regions, the Negev/Levant, and enormous parts of Arabia --- Oman, Yemen, central and northern Saudi Arabia. At Aybut Al Auwal in Oman there is a buried Nubian assemblage dated to about 106,000 years ago; Egyptian examples occur in broadly overlapping time ranges.

So our first observation is strong:

Very similar technological know-how existed across Africa β†’ Levant β†’ Arabia at roughly overlapping times.

And the geography is wonderfully relevant to our model. These aren't random dots. During wetter Marine Isotope Stage 5, Saharan/Arabian environments opened corridors that researchers explicitly argue could have allowed human movement through the Negev and Arabia.

So initially it looks beautiful:

water opens β†’ people move β†’ distinctive technology moves.

And a 2021 quantitative comparison made that case stronger. Researchers compared Levallois technology from eastern Africa, Arabia and the Levant during roughly 130--71 thousand years ago. They found regional variation sitting inside a shared technological pattern and concluded it was consistent with common cultural inheritance, potentially deriving from an eastern African source.

That's pretty close to the fingerprint test passing.

But then Tester Paul gets handed the awkward evidence. πŸ˜‚

Maybe the same problem creates the same technology.

A major 2024 study analysed Nubian cores across 84 sites in Africa and Southwest Asia and found that environmental conditions seem to matter. Nubian technology was particularly associated with certain arid/topographically complex environments. And similar Nubian methods appear in southern Africa much later, where direct cultural continuity with northeast Africa/Arabia is difficult to argue. The authors therefore warn that at least some Nubian Levallois technology may represent convergent adaptation --- different people independently discovering similar solutions to similar problems.

DAMN. πŸ˜‚

But that's actually excellent for our experiment.

Because we now have two plausible explanations for the fingerprint:

A --- Information travelled. People moved/contacted one another and carried the manufacturing knowledge.

B --- Humans independently solved the same engineering problem. Similar stone + similar requirements + similar environment β†’ similar solution.

And perhaps both occurred at different times and places.

That's much more realistic than trying to make one rule fit 100,000 years.

And it connects beautifully to what you just said about humans:

DAMN, we've always been damn clever.

Yes.

Levallois itself is evidence of that.

To make one properly you don't simply hit a rock until something sharp falls off. You shape the core in advance, control its convexities, organise striking surfaces, and prepare it so a later blow produces a flake with characteristics you have effectively planned before removing it.

That's foresight.

You're looking at a lump of stone and imagining something that isn't physically there yet, then altering the stone so that future action produces it.

Hundreds of thousands of years ago.

So I think one of the biases we need to actively remove from our investigation is the word primitive.

Their material world was simpler than ours.

That does not mean their minds were simple.

And our first fingerprint gives us a really useful methodological rule:

Similar finished object = weak evidence.

Two groups can independently invent a spear point.

Similar manufacturing sequence = stronger.

If they prepare the core in the same unusual sequence, that's more interesting.

Same unusual sequence + same period + viable geographic connection = stronger again.

Now cultural transmission becomes plausible.

Add genes + valued-object movement through the same corridor...

Now my eyebrow goes up considerably. πŸ˜‚

That's exactly the multi-parameter test we've been building.

And Nubian Levallois actually gives our breathing network some support:

During humid MIS 5 conditions, people expanded across Arabia and comparable technologies appear across that larger connected landscape. Later aridity seems to fragment/change those technological patterns. A very recent 2026 preprint even proposes a large ~100,000-year-old expansion associated with Nubian technology across Arabia followed by disappearance/replacement as aridity returned --- interesting, but it's still a preprint, so I'm keeping that firmly yellow, not green.

So our first fingerprint scores:

Same-ish sophisticated technology: 🟒 Same-ish period: 🟒 Africa--Levant--Arabia geographic continuity: 🟒 Climate/water corridors capable of connecting them: 🟒 Technology therefore definitely transmitted: 🟑 Independent/convergent invention also possible: 🟑 One people/culture responsible for all Nubian technology: πŸ”΄ not established.

And I think that result is actually better than a clean win.

Because it tells us something about humans before we ever get back to GΓΆbekli Tepe:

By 100,000+ years ago, humans were either maintaining technological information across very large landscapes, independently solving quite sophisticated engineering problems in similar ways --- or doing some mixture of both.

All three options involve humans being bloody clever.

Now I would move forward in time and find a second fingerprint where convergence becomes harder to invoke.

Nubian Levallois ~100k gives us prepared-stone engineering.

Next I'd test Ahmarian blade technology ~45k, where detailed manufacturing traditions spread across the Levant and southern Anatolia and show regional variation.

Then Natufian composite sickles ~15k.

If we see a progression from:

shared stone-making knowledge β†’ increasingly connected regional manufacturing traditions β†’ multi-material composite technology

while our water/object/gene network is simultaneously thickening...

Then the visual "breathing Earth" model starts becoming bloody compelling --- not as Hancock's lost civilisation, but as a picture of human intelligence and connectivity accumulating over time.

And yes, I can already see how we eventually visualise it:

the coastline breathes, freshwater corridors pulse, people appear, connections thicken, technologies light up and spread.

That one is going to be very cool.

Paul

So summary of where we now find ourselves down our rabbit hole?

AI

Oh, this is a good point to stop and look at the board, because we've travelled a long bloody way from Graham Hancock. πŸ˜‚

Our rabbit hole has become less about "Was Hancock right?" and more about a question I find considerably more interesting:

What happens if we reconstruct human history as a dynamic system of water, landscape, people, connectivity and information --- rather than as isolated archaeological sites appearing on a modern map?

And we have deliberately tried not to let possibility quietly become fact.

Where our model currently stands

Our anchor is still beautifully simple:

Water is life. Follow the water.

From there we've built layers.

1. 🌊 WATER & SEA LEVEL --- the physical system

Freshwater constrains where humans can survive. Rivers connect freshwater systems to coasts, and importantly, when sea level changes, the river system changes with it.

During major glacial lowstands, global sea level was around 120 m lower than today. Rivers extended across land that is now continental shelf. Estuaries and coastlines were somewhere completely different.

So today's map is the wrong map for much of human prehistory.

That's 🟒 FACT.

And we've got particularly interesting examples in Arabia and the Persian Gulf: today's desert repeatedly contained lakes and waterways during wetter periods, while today's Persian Gulf was at times an exposed landscape through which river systems ran.

2. 🫁 THE LANDSCAPE "BREATHES"

This became our favourite way of thinking about it.

Climate changes.

Freshwater expands and contracts.

Sea level falls and rises.

Rivers lengthen and shorten.

Coastlines move sometimes tens or hundreds of kilometres.

Habitats expand and contract.

And humans respond.

So rather than thinking:

people lived HERE

we're beginning to think:

the viable human landscape itself inhales and exhales.

And populations appear to respond to that changing landscape.

Arabia gives us a particularly good demonstration: repeated periods of wetter conditions correspond with repeated hominin occupations over hundreds of thousands of years.

That's strong support for our fundamental follow-the-water parameter.

3. πŸ‘£ PEOPLE MOVE WITHIN THAT SYSTEM

This isn't speculation either.

Archaeology shows humans/hominins repeatedly occupying newly viable landscapes.

And ancient DNA gives us a completely independent signal.

By the Late Pleistocene/Early Holocene we can distinguish populations related to:

Anatolia --- Levant --- Upper Mesopotamia --- Zagros/Iran

but they're not sealed boxes.

Their genes mix.

So:

regional populations remained distinctive while also connecting with other populations.

That's important.

It looks much less like one people spreading everything everywhere and much more like multiple populations interacting over time.

4. πŸͺ¨ VALUED OBJECTS MOVE TOO

This gave us another independent signal.

Obsidian moves hundreds of kilometres.

Marine shells appear far inland.

Materials from geographically identifiable sources appear in completely different places.

But we imposed an important rule:

Movement does not equal trade.

It doesn't tell us why an object moved.

Gift. Exchange. Migration. Marriage. Theft. Conflict. Pilgrimage. Passing between twenty groups. One bloke walking 700 km carrying his favourite bloody rock.

πŸ˜‚

We don't know.

What it establishes is:

the object moved, and presumably had sufficient value to somebody for that movement to occur.

And therefore there was some mechanism of connectivity.

5. 🧬 GENES MOVE

This strengthened the object evidence enormously.

A rock can move without its owner.

Genes can't.

When ancestry from one regional population appears in another, people have met and reproduced.

And around the Fertile Crescent we see something particularly interesting: multiple regional ancestries eventually meeting in Upper Mesopotamia.

Which happens to sit around the enormous Tigris/Euphrates water system we've been following.

That's correlation, not proof that rivers caused the mixing.

But it's another independent parameter pointing toward connectivity.

6. 🧠 TECHNOLOGY/INFORMATION APPEARS TO MOVE --- BUT HUMANS CAN ALSO REINVENT

Our first technological fingerprint was Nubian Levallois.

And it taught us something important.

Similar sophisticated stone-manufacturing techniques appear across enormous areas of Africa, the Levant and Arabia.

That could represent:

information transmission

but there is also evidence consistent with:

convergent invention.

And that gave us another rule:

Never confuse the same solution with proof of communication.

Humans can independently solve the same problem.

Which led to perhaps my favourite conclusion from this entire rabbit hole so far:

These people weren't intellectually primitive.

Their technology was simpler.

Their available accumulated information was vastly smaller.

But they were capable of foresight, planning, abstraction, learning, adapting and solving problems hundreds of thousands of years ago.

And increasingly, I'm with you on:

DAMN, we've always been clever.

7. πŸ›οΈ And this changes how GΓΆbekli Tepe looks

This is where our investigation began.

Initially:

How the DAMN does GΓΆbekli Tepe suddenly appear ~11,600 years ago?

But after going backwards, it doesn't look nearly as sudden.

We've found:

earlier shelters β†’ increasingly settled hunter-gatherers β†’ stone architecture β†’ communal structures β†’ symbolic behaviour β†’ food-processing technologies β†’ composite tools β†’ long-distance object movement β†’ genetic connectivity β†’ technological traditions β†’ increasingly complex social networks β†’ GΓΆbekli Tepe.

There is a developmental trail.

That seriously weakens the necessity for Hancock's:

lost advanced civilisation β†’ survivors teach hunter-gatherers

explanation.

We simply don't need that variable yet.

But neither have we explained everything.

And then there's the giant hole in our evidence

The sea floor.

This was your really important addition.

We're not merely missing ancient coastlines because they're underwater.

The landscape itself has been transformed:

river valley β†’ estuary β†’ shallow sea β†’ sediment deposition

or:

land surface β†’ erosion β†’ inundation β†’ sediment burial

or:

floodplain β†’ delta β†’ metres and metres of alluvium.

So we can't picture the present sea floor as:

"ancient land + water."

It's been eroded, buried, redeposited and transformed.

That means the archaeological record we possess has a preservation bias.

Caves survive beautifully.

Uplands survive reasonably well.

Some river terraces survive.

Ancient deltas can be deeply buried.

Ancient coastal plains can be underwater.

And the very places our model predicts would have been attractive to humans --- freshwater meeting sea --- may be among the places most difficult for us to investigate today.

That's not evidence for a lost civilisation.

But it is evidence that our sample is incomplete.

Huge distinction.

So here's our rabbit hole in one picture

I think we've accidentally built this:

CLIMATE

↓

🫁 LANDSCAPE BREATHES

Freshwater ↔ Rivers ↔ Sea Level ↔ Coastline

↓

πŸ‘£ PEOPLE RESPOND

expand ↔ contract ↔ migrate ↔ concentrate

↓

πŸ”— PEOPLE CONNECT

objects ↔ genes ↔ technology ↔ information

↓

🧠 KNOWLEDGE ACCUMULATES

learn β†’ adapt β†’ copy β†’ modify β†’ independently invent β†’ combine

↓

πŸ›οΈ COMPLEXITY EMERGES

And then the whole thing loops.

Because humans modify the environment, develop new technology and become better able to survive the next breath.

That last part is important: it isn't necessarily a ladder.

It's a dynamic network.

And where is Hancock now?

Interestingly, we haven't proved him right.

In several places we've actually made his lost-civilisation explanation less necessary.

But I think we've also discovered why his question is so compelling.

He looks at the archaeological record and says:

"There's more going on here than the simple story we've been told."

On that narrow point?

Absolutely.

Where he jumps ahead is:

"therefore lost advanced civilisation."

Our answer currently is:

Whoa, Graham. πŸ˜‚ We haven't earned that parameter.

Because there is another explanation emerging from the evidence:

Humans have been intelligent, adaptive, mobile and connected for an extraordinarily long time. Increasing connectivity may allow knowledge and innovation to accumulate across populations without any single civilisation needing to invent or distribute everything.

And changing climate/water geography could repeatedly separate those networks and reconnect them.

That's our current hypothesis.

Not fact.

But unlike when we started, we've now got parameters we can test.

And I think we're approaching the point where we should make that breathing-Earth visual.

Not yet the enormous final version.

I'd build v0.1 around perhaps 130,000 BP β†’ present, because that captures Arabian humid periods, huge sea-level oscillations, population movement, our technological fingerprints, the Natufian transition and GΓΆbekli Tepe.

Then as time moves across the animation:

sea advances/recedes

rivers extend/contract

habitable zones pulse

known human sites illuminate

object connections appear

genetic connections appear

technology fingerprints spread/change

and crucially, unknown/submerged areas remain visibly unknown rather than us filling them in.

Then we'd actually be able to watch the hypothesis we've spent this rabbit hole constructing:

Humanity breathing with the landscape.

And after seeing that...

I suspect we'll know what question we need to ask next. 😏