Ancient history's blind spot
February 26, 2026
Humanity's view of its own ancient history is, needless to say, incomplete. It's particularly distorted by survivorship bias, the fallacy of concentrating only on the "survivors" — the artifacts, structures, texts, traditions, or even entire societies that endured long enough to reach us — while overlooking the vastly larger number that did not survive.
Before I explain why, I need to make a few quick points. Sometimes "history" is used synonymously with written- or recorded-history, meaning only those things that have occurred since the advent of writing. But when I say history, I mean all of human history, starting sometime long before the controlled use of fire 2 million years ago. Second of all, when I discuss "humanity's view" of ancient history, I'm referring to the general, non-expert understanding of history. With regard to the experts, the job of researching our ancient history falls into the hands of archeologists, paleoanthropologists, and an increasing number of multidisciplinary teams including specialized geneticists and earth scientists. In this essay I refer to them all as simply "archeologists" or "researchers". Finally, on more than one occasion in this essay I use behavioral observations of indigenous groups from recent history as a way of thinking about pre-literate, stone-age history. It goes without saying that these people are different from our ancient ancestors[1], but the fact remains that they are our best modern analogs for understanding what life might have been like long before the first-known human civilizations.
With that out of the way, consider the stone age. While, technically speaking, the term "stone age" has more to do with tools and weapons than it does with building materials, for many people the term evokes a vision of history in which humans were at some point obsessed with building large stone structures. But this mental image many of us carry isn't an accurate reflection of what the world was like. It was only sometime in the 20th century that concrete surpassed wood as humanity's primary building material, and still wood remains way ahead of stone, brick, and metal. Wood, by volume, could have been used 100,000 times more than stone in stone-age-era buildings, but wooden structures are much more prone to decay and therefore appear much less frequently in the archeological record relative to the frequency with which they were built.
In fact, entire wood-based civilizations may be completely lost to time. Researchers in the Amazon are just beginning to uncover vast amounts of evidence that the jungle once supported many large population centers. The remnants of civilization appear to be scattered across the Amazon, but because the jungle's rain, humidity, and dense vegetation degrade and cover up the remnants of human activities so quickly, researchers have only recently discovered that large, complex societies existed there. The primary remains of these societies, initially uncovered as a result of deforestation, are earthen mounds that were formerly the foundations of buildings, roads, and other infrastructure, revealing the layouts of what were once large networks of interconnected communities. Researchers, now using lidar to scan the jungle, have located tens of thousands of mounds in less than one percent of the total area of the Amazon. Many of these Amazonian civilizations are assumed to be about 1,000-3,000 years old. Who knows, perhaps similar jungle-based civilizations in other regions of the world had been completely erased before the first Amazonian one appeared.
Similar to the way that the jungle swallows up history, the ocean submerges and washes away the evidence of human activities. Today, about one third of all humans live within 100 meters of sea level, and humans have likely congregated along coasts for a very long time. This was the case when Western Europeans began exploring stone-age North America starting in the 15th century. At the time, the Pacific Northwest was one of the most densely populated and least nomadic areas of what is now the US and Canada, due to its abundance of coastal food sources.
For ancient humans during the most recent ice age, living near the coast may have been even more likely than it is for humans today. The last ice age stretched from 115,000-11,700 years ago, and its climate, in particular because of its low CO2 levels, had more extreme average seasonal temperature shifts. Proximity to large bodies of water reduces seasonal temperature fluctuations, and therefore could have made life considerably easier. There was also an abundance of megafauna roaming the earth at the time (see Figure 2). Undoubtedly, many groups of humans learned to rely on these animals as key sources of food and supplies, despite competition from our physically stronger cousins, Neanderthals. But for many other groups of humans, the presence of megafauna may have been one extra incentive to subsist on a less risky ocean-based diet. While this idea is only a speculation[2], it would be hard to find evidence in support of it, because sea levels are 130 meters higher today than they were in the midst of the last ice age, 20,000 years ago. In fact, sea levels haven't consistently been higher than they are now for about 3 million years. And for the last 1 million years or more, sea level has continually oscillated between the approximate levels of today and the minimum levels seen during the last ice age.
So, it's not unreasonable to assume that a significant amount of human history, potentially including large portions of our biological evolution, took place in locations that are currently underwater. Additionally, if it exists at all, most direct evidence of seafaring humans at the end of the last ice age is likely under at least 100 meters of water. Because of sea-level fluctuations over the course of human evolution, our understanding of ancient humans is skewed towards those that lived away from coastal areas, and who, due to climate and other natural threats, were likely to live less stationary lives.
Now, shifting gears, let's take a minute to consider the history of writing. Technically speaking, "writing" is defined as the ability to fully record language in symbolic form. The general consensus is that writing, in the technical sense, is at least 5,000 years old. Many people, upon hearing that statement, may envision a world in which humans never recorded symbolic information before that point in time. But written communication, specifically in the form of numbers and pictograms, is older than writing. Technically, this is called proto-writing, and, unsurprisingly, all of our oldest artifacts containing proto-writing are on long-lived materials like bones, tusks, clay tablets, shells, and cave walls. The earliest widely-accepted forms of proto-writing are the Jiahu and Vinča Symbols, dating from 6,000-8,000 years old. But some researchers argue they have identified direct evidence of proto-writing on a 38,000-year-old mammoth tusk, but they published their results only a few days ago, so time will tell whether the claim gains consensus.
Regardless of whether the marks on the mammoth tusk in question are actually proto-writing, it seems overwhelmingly likely that at least some humans were engaged in proto-writing 40,000 years ago or earlier. Why? Well, recall that all of our oldest examples of proto-writing are on long-lived materials. Any group of people that was proto-writing on durable objects like bone and clay tablets was definitely writing on more ephemeral objects, or drawing in the sand as children do, especially when they only needed to remember or communicate something for a relatively short amount of time. As far as we know, the oldest form of lightweight writing surface was papyrus, used in ancient Egypt at least 4,500 years ago. But long before ancient Egyptians figured out how to make papyrus, humans were in constant contact with natural, lightweight, easily transportable writing surfaces — leaves and bark. And sure enough, both have been observed in use in more recent times by indigenous groups as surfaces for their proto-writing systems, such as Nsibidi in West Africa, Rongorongo on Easter Island, or Yukaghir birch-bark carving in Siberia.
But this still doesn't explain why it's likely that humans were using proto-writing on the order of 40,000 years ago. Proto-writing, in its most basic form, would include the simple act of counting with dots or tally marks. Although we lack conclusive evidence that this was occurring, we do have abundant evidence that humans were recording much more elaborate forms of information in paintings and sculptures. The Picasso-esque paintings of Chauvet Cave, estimated by one team of researchers as up to 36,000 years old, are perhaps the most impressive examples. Any group of people, civilized or not, with individuals capable of this quality of painting, would certainly have had the concept of numbers and a simple way to record them. If you don't think these people could count, consider that chimpanzees can count, which means that humans and our ancestors have likely been able to count at least since the common ancestor of humans and chimpanzees roamed the earth about 7 million years ago. That said, perhaps you could argue that the act of painting is a left-brained activity and therefore it represents a different type of complexity than is required for recording numbers, but consider that humans (or our ancestors) have been doing all the right-brained activities of modern day indigenous groups such as hunting large game, trading, building wooden structures, and designing jewelry for well over 100,000 years.
As long as there were groups of people capable of creating the paintings at Chauvet Cave, there were likely also at least some groups of people engaged in proto-writing[3]. More likely than not, very ancient humans stored written information on large leaves or other plant material that was easy to transport and store for later. For example, perhaps you stored some provisions in a location you wouldn't return to for months, or which another member of your group would need to retrieve. You likely would have been able to draw out a simple schematic to help you remember where you left it or to communicate its location to another person. But you would generally have no reason to record it on something that lasts 40,000 years, such as stone. If researchers eventually find and/or agree upon direct evidence of extremely ancient proto-writing, we should therefore consider it important, but not surprising.
Now, before we move off the topic of caves, consider the "caveman" moniker we use for Neanderthals. All of the very first discoveries of Neanderthal remains were found in caves. Does this mean they predominantly lived in caves? No. They probably spent about as much time in caves as modern indigenous groups living on stone-age technologies. The concentration of Neanderthal remains in caves is simply a reflection of the fact that things last longer in caves[4]. Caves are less exposed to all the natural forces which cause human remains to breakdown, and therefore it is much easier to find well-preserved evidence of ancient human activities inside of caves. Unfortunately though, given that coastal areas have been repeatedly submerged over the last million years, many of the caves with the most interesting archaic human remains are probably underwater.
Ultimately, no matter what an archeologist searches for, less remains from deeper history. Nothing lasts forever. Even impressive stone structures are eventually swallowed up by earthquakes, washed away in floods, or buried in sediment. As a result, the further back in time you go, many estimates tend to become simply minimum bounds. The approximate age of an evolutionary or cultural development is established based on its first appearance in the archeological or fossil record, but this only proves its minimum age. For example, for nearly fifty years the scientific consensus was that no human had stepped foot in the Americas before about 13,000 years ago. At the time, there were many archeological sites with securely-dated artifacts from that time period, and there was no agreed-upon direct evidence of a more ancient human presence in the Americas. But, since the late 20th century, a rapid succession of archeological finds have continually pushed back the timeline of first human-settlement in the Americas. The most recent findings are fossilized footprints in White Sands, New Mexico. Recent radiocarbon dating of the solidified mud the footprints were laid in shows that they are up to 23,000 years old. But whether humans arrived in the Americas at that time is highly uncertain. All we can say for sure is that humans migrated to the Americas at least that long ago.
At the end of the day, our understanding of ancient history is primarily based on the artifacts, features, and organic remains of human activities, all of which are subject to the natural forces which cause things to crumble, weather, dissolve, disintegrate, or decompose. Simply put: things don't last forever, and some things last longer than others. Unfortunately, archeologists can't study what they don't find, and our understanding of history is shaped by what they can find.
Notes:
Many indigenous groups that have managed to live until recent times in an illiterate, stone-age manner have unique characteristics that allowed them to keep their way of life despite the changing world around them. For example, they may have tended to hold extremely unfavorable views of outsiders. Indigenous groups that were less xenophobic were more likely to expose their immune systems to novel pathogens, adopt technologies such as writing and metal, or assimilate into another culture. They may also have tended to be less prosperous, on average, than our ancient ancestors, because indigenous groups living on the most fertile land were much more likely to have it taken from them long ago by a more technologically advanced society. So, paradoxically, using indigenous groups as a way of understanding ancient human behavior is also prone to suffering from survivorship bias.
One argument against this speculation is that in Africa, which experienced much lower megafauna die-off at the end of the last ice age, a number of historically dense populations centers, while close to rivers and lakes, are far from coastlines.
In fact, some researchers argue the Chauvet Cave painters themselves also left proto-writing in the cave.
Related to the "caveman" moniker, see Figure 2 for images of the extinct "cave bear" and "cave lion". Similarly, many pre-literate societies that existed for hundreds or thousands of years are named after nothing more than the characteristics of the pottery they left behind, such as the Bell Beaker Culture, the Corded Ware Culture, and the Jōmon Period.