-
Chimpanzee Tool Traditions: Different Populations, Different Technologies
In the Goualougo Triangle of the Republic of Congo, chimpanzees harvest termites using a two-tool system. First, they manufacture a thick, sturdy puncturing stick from a specific plant species and drive it into the soil to breach the outer wall of a subterranean termite nest. Then they switch to a separate fishing probe—thinner, more flexible, often with the tip deliberately frayed by pulling it through their teeth to create a brush-like end—and insert it through the access tunnel they’ve just made. Termites bite the frayed fibers, the chimpanzee withdraws the probe, and eats them off the bristled end. The whole operation requires selecting the right raw materials, manufacturing two distinct tools in the correct sequence, and knowing how to modify one of them to improve its efficiency. It is, by any reasonable definition, a technology.
Nine hundred kilometers east, at Gombe in Tanzania, chimpanzees also fish for termites. But they use a single tool—a simple probe stripped of leaves, inserted directly into exposed holes in the mound—and they don’t puncture, don’t use tool sets, and don’t fringe the tips. Same species. Same resource. Same basic objective. Completely different technique. And when researchers compared these two populations in a study published in PNAS, they found that in Goualougo—where the task is more complex—mothers were significantly more likely to actively share tools with their offspring and facilitate learning, something that barely happens at Gombe, where the task is simple enough that young chimps figure it out by watching.
That’s not instinct varying by region. That’s culture. And the chimpanzee evidence for culture is, at this point, about as close to settled as anything in behavioral ecology gets.
The 1999 paper that changed the field
The landmark study was Andrew Whiten’s 1999 paper in Nature, which synthesized data from the seven longest-running chimpanzee field sites across Africa—151 combined years of observation. The analysis identified 39 distinct behavioral patterns, including tool use, grooming styles, and courtship displays, that were customary or habitual in some communities but entirely absent in others, even when ecological and genetic explanations had been ruled out. The behavioral repertoire of each community was itself distinctive—not just a few isolated differences but a combinatorial profile of dozens of traditions that, taken together, made each population culturally unique.
That’s worth sitting with. Before 1999, “culture” in the biological sciences was essentially reserved for humans. Other animals had “traditions” or “behavioral variation” or, if you were being generous, “proto-culture.” Whiten’s paper didn’t just add chimpanzees to the list of species with cultural variation—it showed that the scope and combinatorial complexity of that variation was without parallel in any non-human species. Not one tradition. Not three. Thirty-nine, distributed across communities in patterns that looked less like random variation and more like the kind of between-group differences you’d see comparing human societies.
The paper used what’s called the “method of exclusion”—if a behavior is present in one community and absent in a neighboring community with access to the same raw materials, the same prey species, and similar genetic backgrounds, and if that behavior is transmitted socially rather than reinvented independently, then the most parsimonious explanation is cultural transmission. It’s not a perfect methodology—proving a negative (that ecology doesn’t explain the difference) is always harder than proving a positive—but it was rigorous enough to shift the consensus.
The tool traditions themselves
The catalog of chimpanzee tool behaviors now documented across Africa is staggeringly diverse, and the geographic specificity of individual techniques is what makes the cultural interpretation so compelling.
Nut cracking with stone or wooden hammers is practiced by chimpanzee populations in West Africa—in Côte d’Ivoire, Guinea, Liberia, and Sierra Leone—but is completely absent in East and Central African populations, despite the availability of suitable nuts and hard surfaces. The Taï Forest chimpanzees in Côte d’Ivoire select stone hammers of appropriate weight for the hardness of the nut species being cracked, transport hammers to anvil sites they remember from previous visits (sometimes carrying them over a kilometer through the forest), and teach the technique to juveniles through years of observation and practice. Young chimps at Taï spend roughly four to five years learning to crack Coula nuts and seven years to crack the harder Panda nuts. That’s an apprenticeship, not a light-switch moment.
Ant dipping—using a stick to harvest driver ants or safari ants from their nests—varies in technique across populations in ways that map onto geography rather than ecology. Some populations use short sticks and eat the ants directly off the tool with their lips. Others use long sticks, wait for a mass of ants to swarm up the tool, then sweep them off with a single hand motion into their mouth. The technique choice correlates with community membership, not with the ant species or the physical properties of the nest. When researchers at the Taï site compared neighboring communities separated by only a few kilometers, they found that the communities used different lengths of ant-dipping tools and different techniques—despite inhabiting functionally identical habitat with the same ant species available. The difference was social, not ecological.
Spear hunting is one of the most striking recent discoveries. At Fongoli in southeastern Senegal, Jill Pruetz documented chimpanzees fashioning wooden spears from branches—stripping side branches, sharpening the tip with their teeth—and thrusting them into tree cavities to stab bushbabies, small nocturnal primates that shelter in hollow branches during the day. This has not been observed at any other site. It’s a behavior that involves tool manufacture, planning (they modify the spear before approaching the tree, not after), and lethal predatory intent. Female and juvenile chimps at Fongoli do this more frequently than adult males, which inverts the usual pattern of male-dominated hunting in chimpanzees and suggests the spear technique may be an equalizer—a technology that compensates for the strength advantage that adult males have in manual capture.
Honey dipping—using sticks to extract honey from beehives—shows variation across the entire species range. Central African populations at some sites use complex multi-tool sets (a pounder to break open the hive, a collector to extract the honey), while West African populations that eat honey frequently often don’t use tools at all. A 2021 study in Guinea-Bissau documented honey-dipping tools for the first time in the westernmost chimpanzee populations, adding new data points to a behavioral map that’s still being filled in. The variation suggests that honey-dipping technology has been invented and elaborated independently in different populations, rather than spreading from a single origin—which is convergent cultural evolution, a concept that makes the human parallel even more interesting.
Why this isn’t just “smart animals being smart”
The critical distinction between culture and individual intelligence is social transmission—behaviors that are learned from others, maintained within a group over generations, and resistant to disruption by individual innovation. If every chimpanzee independently figured out how to crack nuts, that would be intelligence. What makes it culture is that the technique is passed from mother to offspring through years of observation and practice, maintained within populations even when individuals migrate between groups, and differentiated between neighboring communities in ways that can’t be explained by the environment.
The conformity data is particularly striking. In a series of experiments by Andrew Whiten and colleagues, researchers introduced two different techniques for solving the same food-extraction problem into separate captive groups by training a single high-ranking female in each group. The technique spread through each group via social learning. But here’s the part that matters: some individuals in each group independently discovered the alternative technique—the one that had been seeded in the other group—and then abandoned it in favor of the locally dominant method. They conformed. They had a working solution, discovered a different working solution, and reverted to the one everyone else was using. That’s not problem-solving. That’s peer pressure. That’s culture.
In the wild, the same conformity pattern has been documented with migrating females. When a female chimpanzee transfers from one community to another—which is the normal dispersal pattern—she adopts the tool traditions of her new community, even if she was proficient in a different technique at her birth community. William McGrew, one of the founding figures of chimpanzee cultural primatology, pointed out the thought experiment: imagine a female from Gombe transferring to Goualougo. If she persisted in fishing for termites the Gombe way—single probe, no puncturing stick—she’d fail, because the Goualougo termite nests require the puncturing step she never learned. She’d have to adopt the local technology or go hungry. The technology is the community’s intellectual property, and you either learn it or you don’t eat.
What this tells us about early human culture
The reason chimpanzee tool traditions matter beyond primatology is that chimpanzees are one of our two closest living relatives (bonobos being the other), and the last common ancestor we shared lived roughly six to seven million years ago. The cultural capacity documented in living chimpanzees—multiple traditions, conformity bias, social transmission across generations, geographically specific tool-use techniques—represents either a shared ancestral trait or an independently evolved one. Either way, the implication is that the cognitive and social infrastructure for culture was present in the hominin lineage long before stone tools show up in the archaeological record around 3.3 million years ago.
The emerging field of “primate archaeology” is making this connection explicit. Researchers are applying the same archaeological methods used to study early human tool sites—analyzing raw material selection, tool morphology, wear patterns, and spatial distribution of discarded tools—to chimpanzee termite-fishing sites and nut-cracking stations. The Kasekela and Mitumba communities at Gombe, separated by just a few kilometers, produce termite-fishing tools that are measurably different in length and width, made from different selections of raw materials, even though both communities have access to the same plant species. That’s the kind of between-population variation in material culture that, if it showed up in a 2-million-year-old hominin site, would be published in Nature and generate a press cycle about “the origins of technology.”
It’s already happening. It’s just happening in chimpanzees, which makes it less glamorous and more informative.
We cover chimpanzee tool traditions—alongside whale dialects, corvid problem-solving, fish social learning, and the full breadth of non-human cultural transmission—across our Animal Culture & Knowledge course. If the spear-hunting bushbaby story made you rethink what “culture” means, the course goes considerably deeper.
-
Humanoid Robots in 2026: Who’s Building Them and How Close Are They to Useful?
Sometime in late 2021, Elon Musk stood on a stage and announced that Tesla would build a humanoid robot. Then a person in a spandex bodysuit walked out and did a little dance. This was the prototype. Five years later, Tesla has converted its Model S and Model X production lines at the Fremont factory to manufacture Optimus Gen 3, committed $20 billion in capital expenditure for 2026, and Musk has projected annual production of one million units. On the Q4 2025 earnings call, however, Musk acknowledged that the robots currently deployed in Tesla’s own factories are not doing “useful work”—they are learning and collecting data. Which is a very expensive way of saying they’re interns.
This is the state of humanoid robotics in March 2026: more money, more companies, more capability demonstrations, and more press releases than at any point in the history of the field—and approximately zero humanoid robots doing anything that a $40,000 industrial robot arm couldn’t do better, faster, and without falling over. The technology is real. The progress is significant. The gap between the demo reel and the work site remains enormous. And the number of people who seem to understand this gap is significantly smaller than the number of people writing breathless headlines about it.
Here’s the actual landscape, company by company.
Tesla Optimus
Optimus is the most visible humanoid robot project in the world, which is what happens when the world’s most famous billionaire decides it’s going to be “more significant than the car business.” The Gen 3 iteration—revealed in early 2026—features 22-degree-of-freedom hands with 50 actuators, which is a legitimate engineering achievement. The hands can pick up an egg without crushing it, which sounds trivial until you consider how many sensors, control loops, and force-feedback calculations are running in real time to make that possible. The body remains the Gen 2 platform: 173 centimeters tall, 57 kilograms, with 28 degrees of freedom and a claimed 24-hour battery life.
Tesla’s strategy is to treat Optimus the way it treated electric vehicles—leverage vertical integration, AI infrastructure from the Full Self-Driving program, and automotive-scale manufacturing to drive the unit cost below $30,000. At volumes exceeding one million units per year, Musk claims production costs could drop below $20,000. For context, Boston Dynamics’ Atlas has never been available for commercial purchase and estimates for comparable platforms from other companies run $100,000 to $150,000.
The skepticism is warranted though. Musk’s timeline projections across every company he runs have a well-documented relationship with reality best described as “directionally correct but temporally delusional.” Full Self-Driving was supposed to be feature-complete in 2019. The Cybertruck was announced in 2019 and began deliveries in late 2023. Robotaxi service was supposed to launch in 2020. External customer deliveries of Optimus are now targeted for late 2026, which—adjusted for Musk Time—probably means sometime in 2028. Robotics pioneer Rodney Brooks has been publicly and specifically skeptical, and his track record on predicting the gap between robotics demos and robotics products is considerably better than Musk’s track record on predicting his own timelines.
What Tesla does have is scale. No other humanoid robotics company has access to automotive-grade manufacturing infrastructure, a battery supply chain, an AI training cluster processing real-world data from millions of vehicles, and the capital to sustain years of losses while iterating. The question isn’t whether Tesla can build an impressive robot—they already have. The question is whether they can build one that does productive work reliably, in unstructured environments, without a team of engineers babysitting it. That’s a fundamentally different problem than building a robot that looks good on stage.
Figure AI
Figure is the startup that’s moved fastest from announcement to deployment. Founded in 2022, backed by Microsoft, OpenAI, Jeff Bezos, and NVIDIA—a funding roster that reads like a fantasy football lineup for the tech apocalypse—Figure has its Figure 02 robots actively deployed at BMW’s Spartanburg, South Carolina manufacturing facility. Not in a lab. Not in a demo. In a factory, doing factory work, alongside human workers.
Figure 02 stands about 5 feet 6 inches, weighs 70 kilograms, and can lift roughly 20 kilograms. The robot uses a proprietary AI system called Helix that integrates vision, language, and tactile feedback—it can look at an object, understand a verbal instruction about what to do with it, adjust its grip based on the material’s properties, and execute the task. The OpenAI partnership gives Figure access to frontier language model capabilities for natural-language task understanding, which is a meaningfully different approach than Tesla’s end-to-end neural network strategy derived from autonomous driving.
The BMW pilot is the real differentiator. While Tesla’s robots are collecting data inside Tesla’s own factories, Figure’s robots are performing actual tasks in a customer’s factory. That’s the gap between R&D and product, and as of March 2026, Figure is further across that gap than anyone else except arguably Agility Robotics. Figure has also announced Figure 03, targeting consumer home use by late 2026, which—if it ships on time—would make it one of the first humanoid robots marketed for residential deployment.
Boston Dynamics Atlas
Atlas is the robot everyone pictures when they hear “humanoid robot”—the one doing backflips, navigating obstacle courses, and recovering from shoves with unsettling grace. For over a decade, Atlas was the benchmark for what was mechanically possible in bipedal locomotion. Then in April 2024, Boston Dynamics retired the hydraulic Atlas and unveiled an all-electric version designed specifically for commercial applications.
Electric Atlas is targeting industrial deployment in 2026, starting with Hyundai—Boston Dynamics’ parent company—at its Georgia manufacturing facility. Estimated pricing is in the $140,000 to $150,000 range, which positions it as a premium platform for high-value manufacturing tasks rather than a mass-market product. The robot’s athletic capability remains unmatched—nobody else is doing the dynamic whole-body movements that Atlas demonstrates—but the question is whether that athleticism translates into useful work or is primarily an engineering flex.
Boston Dynamics’ history suggests caution. Spot, their quadruped robot, took years to go from viral video sensation to commercial product, and its actual use cases—construction site monitoring, industrial inspection, remote sensing—are narrower than the YouTube highlight reel implied. Atlas could follow the same trajectory: deeply impressive technically, deployed in specific high-value niches, but not the general-purpose factory worker that the hype suggests.
The Chinese contingent
This is the part of the landscape that gets insufficient attention in Western media. China’s humanoid robotics companies—Unitree, Agibot, Leju Robotics, XPENG’s IRON program—are shipping volume. Unitree’s G1 is projected to move 10,000 to 20,000 units in 2026, which would make it the highest-volume humanoid robot producer in the world by a significant margin. China holds an estimated 85 to 90 percent of global humanoid robot shipments.
Unitree’s strategy is the Chinese manufacturing playbook applied to robotics: aggressive pricing, rapid iteration, and a willingness to ship products that are good enough for specific tasks rather than waiting for general-purpose perfection. The G1 is priced aggressively enough that the cost-benefit calculation works for warehouse and logistics applications where you’re replacing a human worker doing repetitive material handling. It’s not elegant. It’s not doing backflips. It’s moving boxes, and it’s doing it at a price point that makes the economics work.
At the AW 2026 conference in Seoul in March, China’s major robotics firms demonstrated live commercial-ready systems with explicit scaling roadmaps. This isn’t the R&D phase anymore. It’s the deployment phase, and it’s happening faster in Shenzhen than in Fremont.
What “useful” actually means
The fundamental problem with humanoid robots in 2026 isn’t locomotion, manipulation, or AI. Those are all improving rapidly. The fundamental problem is that the world wasn’t built for robots—it was built for humans, by humans, with human proportions and human capabilities assumed at every level of design, from doorknob height to shelf spacing to the assumption that whoever is operating a forklift has the judgment to not drive it into a wall.
A humanoid form factor theoretically solves this by being the right shape to operate in human environments without requiring those environments to be redesigned. That’s the pitch: you don’t need to retrofit your factory for robots if the robot is shaped like the worker it’s replacing. In practice, the pitch breaks down because “shaped like a human” and “capable like a human” are separated by decades of unsolved problems in manipulation, balance recovery, spatial reasoning, and the kind of contextual judgment that lets a warehouse worker notice that a shelf is about to collapse before it actually does.
The robots that are doing useful work right now—and there are some—are doing it in structured environments with limited task variety. BMW’s Figure robots are handling specific components on specific production lines. Amazon’s warehouses use mobile robots extensively, but they’re wheeled platforms, not humanoids. The humanoid form factor adds cost and complexity that is only justified if the robot needs to navigate stairs, use human tools, or operate in environments that can’t be modified. For the vast majority of current automation needs, a robot arm bolted to the floor or a wheeled platform following a painted line on the floor is cheaper, more reliable, and less likely to fall over.
The honest forecast: by the end of 2026, there will be thousands of humanoid robots operating in controlled industrial settings worldwide—predominantly in Chinese factories and warehouses, with smaller deployments at Hyundai, BMW, and Tesla’s own facilities. They will be doing structured, repetitive tasks. They will require significant human oversight. They will not be folding your laundry, mowing your lawn, or serving you coffee. The gap between “a robot that can pick up an egg on camera” and “a robot you’d trust to unload a dishwasher without supervision” is wider than any press release suggests, and closing it is measured in years, not months.
We cover the full history, engineering, and trajectory of humanoid robots and autonomous drones—from Boston Dynamics’ earliest prototypes to the Chinese manufacturing surge—across our Humanoid Robots & Drones course. If the gap between demo and deployment is the part you want to understand, that’s the deep dive.
-
What Are Fortean Phenomena? A Serious Guide to the World’s Strangest Unexplained Events
In 1919, a largely unknown writer named Charles Hoy Fort published a book called The Book of the Damned. The “damned” in question were not people. They were facts—data points collected over decades from scientific journals, newspapers, and maritime logs that the scientific establishment of the day had either ignored, dismissed, or quietly buried because they didn’t fit any accepted theory. Rains of frogs. Falls of red liquid from clear skies. Unidentified lights tracked by multiple observers. Objects appearing and disappearing. Spontaneous fires. Animals found in places they had no biological business being. Fort spent thirty years in the New York Public Library and the British Museum reading room, copying these reports onto thousands of index cards kept in shoeboxes, assembling what amounted to an enormous filing cabinet of things that weren’t supposed to happen but apparently did anyway.
Fort didn’t claim these events were supernatural. He didn’t build a theory of the paranormal. He didn’t start a religion or declare that aliens were responsible. What he did—and this is the part that gets lost in a century of people projecting their own agendas onto his work—was point at the data and say: science claims to have a comprehensive model of how the world works, and here are several thousand documented instances where that model doesn’t account for what was observed. He called this data “damned” because it had been excluded from polite scientific conversation, not because it was demonic. The exclusion was the point. As the writer Colin Wilson later summarized Fort’s operating principle: “People with a psychological need to believe in marvels are no more prejudiced and gullible than people with a psychological need not to believe in marvels.”
That’s Forteanism in one sentence. And it’s a more intellectually rigorous position than it gets credit for.
What actually counts as Fortean
The term “Fortean phenomena” now functions as a catch-all for anomalous events that sit outside the boundaries of currently accepted scientific explanation. The taxonomy—developed by Fort and expanded by researchers since—covers a genuinely enormous range of stuff, and the breadth is part of the point. Fort didn’t specialize. He collected everything.
The major categories, roughly organized:
Anomalous falls from the sky. This was Fort’s bread and butter—his books are packed with documented reports of things falling from the atmosphere that have no obvious atmospheric origin. Rains of fish, frogs, tadpoles, insects, larvae, worms, mussels, snails, and even snakes. Falls of ice blocks, stones, and chunks of calcium. Falls of red, black, or yellow rain. Falls of sulphur, hay, and unidentifiable organic matter. These aren’t all ancient. Fish falls are still reported regularly—hundreds of small fish raining onto the town of Texarkana, Texas in 2021 made national news. The standard meteorological explanation is waterspouts picking up aquatic organisms and depositing them miles inland, which accounts for some cases convincingly and others not at all, particularly when the species involved don’t inhabit any nearby body of water.
Unidentified aerial phenomena. Fort cataloged unexplained lights and objects in the sky decades before Kenneth Arnold’s 1947 sighting kicked off the modern UFO era. His records include reports from ship captains, astronomers, military personnel, and weather observers describing luminous objects, formations of lights, and structured craft-like things visible for extended periods. Fort invented the word “teleportation.” He also proposed, with characteristic deadpan, that Earth might be the property of some unknown intelligence—”I think we’re property,” he wrote in The Book of the Damned—though whether he meant this literally or as a satirical provocation aimed at scientific arrogance is a debate that Forteans have been having for a hundred years.
Cryptozoology. Reports of animals that haven’t been formally identified by science—Bigfoot, the Loch Ness Monster, the Chupacabra, Mokele-mbembe, the Yeti—fall under the Fortean umbrella, though Fort himself was less interested in specific creatures than in the pattern of scientific dismissal that preceded eventual discovery. More on this below.
Spontaneous human combustion. Documented cases—some backed by coroner reports and forensic investigation—of human bodies found almost completely incinerated in circumstances where the surrounding environment showed minimal fire damage. The wick effect hypothesis (where body fat acts as fuel after ignition from an external source like a cigarette) explains some cases. Others remain genuinely perplexing.
Anomalous animal behavior. Mass die-offs, mass strandings, animals appearing far outside their known range, coordinated behaviors that defy current ethological models. Fort was particularly interested in cases where conventional explanations required more assumptions than the anomaly itself.
Earth mysteries. Unexplained sounds (the Taos Hum, the Bristol Hum, the “Bloop” recorded by NOAA hydrophones in 1997), earthquake lights, ball lightning, crop circles (most of which are obviously human-made, though the plasma vortex hypothesis for the handful that aren’t has some interesting physics behind it), and geomagnetic anomalies.
The surprisingly rigorous intellectual tradition
Here’s where Fortean phenomena get genuinely interesting from an epistemological standpoint, and where the field diverges sharply from the conspiracy-theory adjacent content it’s often lumped in with.
Fort’s actual intellectual contribution wasn’t collecting weird stories. It was developing a framework for thinking about how science handles outlier data. His argument—stripped of the deliberately provocative style—was that the scientific establishment has a systematic bias toward excluding observations that don’t fit existing theoretical models, and that this exclusion is driven not by the data itself but by the social and institutional structures of science. Papers that report anomalous findings are harder to publish. Careers are not built on documenting things you can’t explain. Grant funding does not flow toward investigating phenomena that might turn out to be measurement error. The incentive structure of professional science is optimized for extending existing paradigms, not for cataloging their failures.
This is not a crackpot position. Thomas Kuhn made essentially the same argument in The Structure of Scientific Revolutions in 1962—that normal science operates by suppressing anomalies until enough of them accumulate to trigger a paradigm shift—and nobody called Kuhn a crackpot. He got tenure at MIT. Fort made the same observation forty years earlier, in a more entertaining and less academic style, and was written off as an eccentric. The difference was packaging, not substance.
The anthropologist Roger Wescott coined the term “anomalistics” in 1973 to describe the interdisciplinary study of scientific anomalies—essentially a formalized version of what Fort had been doing since the 1890s. The field has since developed genuine methodological rigor. The Fortean Times, published since 1973, combines humor and skepticism with original research. The Society for Scientific Exploration publishes peer-reviewed work on anomalous phenomena. Modern Fortean researchers use satellite imagery, digital archives, eDNA sampling, acoustic analysis, and the same statistical tools as any other field to investigate claims.
The phenomena science denied until it couldn’t
The strongest argument for taking Fortean data seriously isn’t the data that’s still unexplained. It’s the data that was once “Fortean” and is now just science.
Ball lightning was reported for centuries—glowing spheres of light appearing during thunderstorms, passing through walls, hovering for seconds before vanishing. Scientists dismissed the reports as hallucinations, optical illusions, or misidentified St. Elmo’s fire until laboratory-produced ball lightning was achieved in 2006 by researchers in Tel Aviv and again by a team in China in 2012 using microwave discharge. It’s now an accepted atmospheric phenomenon with multiple competing physical models.
Rogue waves—walls of ocean water two to three times the height of surrounding waves, appearing without warning—were considered sailor folklore until the Draupner wave was measured by instruments on the Draupner oil platform in the North Sea on January 1, 1995. The wave was 25.6 meters high in a sea state of 12-meter significant wave height. It was real, it was measured, and it immediately invalidated the standard statistical models for ocean wave height distribution. Sailors had been reporting these waves for centuries. Oceanographers had been explaining to them that such waves were statistically impossible.
Meteorites. Before 1803, the idea that rocks fell from the sky was considered superstitious nonsense by the scientific establishment. The French Academy of Sciences had formally dismissed the possibility. Then, on April 26, 1803, roughly 3,000 stones fell on the town of L’Aigle in Normandy, witnessed by the entire town and investigated by physicist Jean-Baptiste Biot, who confirmed the fall. The scientific consensus flipped overnight. Rocks from space had been “Fortean” the day before and were geology the day after.
Continental drift. Alfred Wegener proposed in 1912 that the continents had once been joined and had drifted apart. He was ridiculed for decades—the mechanism he proposed was wrong, and geologists couldn’t accept the conclusion without an acceptable mechanism. Plate tectonics wasn’t established until the 1960s. Wegener was right about the observation and wrong about the explanation, and mainstream science rejected the observation because it didn’t like the explanation. Fort would have had a field day.
The pattern is consistent: observation precedes explanation, sometimes by centuries, and during the gap, anyone who takes the observation seriously is treated as a crank. Fort’s entire body of work is essentially a catalog of phenomena sitting in that gap—things that have been observed repeatedly but not yet explained to anyone’s satisfaction.
Why rational people engage with this
The most common misconception about Fortean phenomena is that interest in them requires credulity—that you have to “believe” in Bigfoot or UFOs or spontaneous combustion to find the field worthwhile. Fort himself would have rejected this framing completely. He wasn’t a believer. He was, if anything, a professional skeptic—skeptical of claims of the anomalous, skeptical of claims of normalcy, and especially skeptical of anyone who claimed to have a complete model of reality. His position was that the only honest intellectual posture in the face of anomalous data is to document it, resist the temptation to explain it away prematurely, and maintain a suspension of judgment that he called “intermediatism”—the idea that nothing is entirely real or entirely unreal, and that all knowledge is transitional.
That’s a position that would fit comfortably in any philosophy of science seminar. It just happened to come wrapped in stories about rains of frogs and mysterious lights over the Atlantic, which made it easy to dismiss.
The Fortean approach doesn’t require you to believe anything. It requires you to take observation seriously even when the observation is inconvenient, to resist the reflexive urge to explain away data that doesn’t fit your model, and to recognize that the history of science is littered with phenomena that were “impossible” until they were measured, at which point they became textbook material and everyone pretended they’d never doubted them.
That’s not credulity. That’s intellectual honesty with better source material than most people expect.
We cover the full landscape of Fortean phenomena—anomalous events, cryptozoology, legendary conspiracies, and the science of why humans believe what they believe—across our Fortean Phenomena & Anomalistics course. If the rogue wave story made you want to know what else science got wrong before it got it right, that’s where to start.
-
Forbidden Zones: The Places on Earth You’re Legally Not Allowed to Visit
There is an island in the Bay of Bengal, roughly the size of Manhattan, where the Indian Navy maintains a permanent patrol to ensure that nobody gets within five kilometers of the shore. The island has no roads, no ports, no cell towers, no modern infrastructure of any kind. It is home to the Sentinelese—an indigenous people who have lived there in voluntary isolation for an estimated 60,000 years and have, with remarkable consistency, attempted to kill anyone who approaches. The Indian government has responded to this situation not by trying to change the Sentinelese’s mind but by making it illegal for you to try. Approaching North Sentinel Island is a criminal act under the Andaman and Nicobar Islands Protection of Aboriginal Tribes Regulation of 1956, and the Indian government has publicly stated it will not prosecute the Sentinelese for killing trespassers.
That is not a hypothetical. In 2018, American missionary John Allen Chau paid local fishermen to take him to the island, where he hoped to convert the tribe to Christianity. He made it ashore. The Sentinelese killed him with bows and arrows. In March 2025, an American YouTuber named Mykhailo Viktorovych Polyakov landed on the beach, blew a whistle, left a can of Diet Coke and a coconut as “offerings,” filmed the whole thing, and left. He was arrested two days later. Police noted he was “lucky he did not make contact otherwise he would have met the same fate.” Survival International, the NGO that advocates for uncontacted peoples, called his actions “reckless and idiotic,” which feels like an understatement for a man who traveled to the one place on Earth where the locals have a documented track record of murdering visitors so he could leave a Diet Coke on the beach and post it online.
North Sentinel Island is probably the most famous forbidden zone in the world, but it’s far from the only place where you are legally, physically, or practically unable to go. The planet is littered with them—restricted by governments, militaries, treaties, geography, or the simple fact that getting there would kill you. And the reasons a place becomes off-limits are often more interesting than the place itself.
Area 51, Nevada, United States
The Groom Lake facility in the Nevada Test and Training Range—universally known as Area 51—is the one everyone knows about and nobody knows anything about, which is the ideal combination for generating decades of conspiracy theories. What we do know: it’s a classified United States Air Force testing facility that has been used since the 1950s to develop experimental and stealth aircraft, including the U-2 spy plane, the SR-71 Blackbird, and the F-117 Nighthawk. What we don’t know: basically everything else. The airspace above it is restricted. The perimeter is monitored by motion sensors, cameras, and security personnel. Warning signs inform you that photography is prohibited and that the use of lethal force is authorized—the kind of signage that really makes you feel welcome.
The facility’s existence wasn’t officially acknowledged by the U.S. government until 2013, when the CIA released declassified documents about the U-2 program. Sixty years of officially pretending the place didn’t exist while simultaneously stationing armed guards around its perimeter is a very specific kind of bureaucratic performance art. The conspiracy theories—alien technology, reverse-engineered spacecraft, extraterrestrial autopsies—are almost certainly nonsense, but the government’s own commitment to secrecy created the perfect conditions for them to thrive. If you act like you’re hiding aliens, people are going to think you’re hiding aliens.
The Korean DMZ
The Demilitarized Zone between North and South Korea is 250 kilometers long and roughly 4 kilometers wide—a buffer strip that has been essentially untouched by human activity since 1953. It is simultaneously one of the most heavily militarized borders on Earth and one of the most pristine wildlife corridors in Asia, because it turns out that landmines and machine gun emplacements are an extremely effective conservation strategy. Red-crowned cranes, Asiatic black bears, and possibly the critically endangered Amur leopard have been spotted in the DMZ, thriving in a landscape that humans have been too afraid to enter for seventy years.
Tourists can technically visit portions of the DMZ through organized tours from the South Korean side—the Joint Security Area at Panmunjom is the one you see in photographs, where soldiers from both sides stare at each other across a concrete line. But the vast majority of the zone is completely off-limits, laced with an estimated two million landmines, and anybody wandering off the designated tour path would be facing a combination of unexploded ordnance and North Korean border guards, which is not a recreational combination.
Poveglia Island, Italy
A small island in the Venetian Lagoon, roughly halfway between Venice and Lido, Poveglia has been closed to the public since 1968 and the Italian government has repeatedly declined to sell or develop it. The island’s history is a greatest hits compilation of everything you’d want in a haunted location. In the late 1700s, it was used as a quarantine station for plague victims—ships arriving in Venice were required to stop there first, and those showing symptoms of the plague were left on the island. By some estimates, over 100,000 people died there during various plague outbreaks, and the island’s soil is reportedly so densely packed with human remains that fishermen in the lagoon occasionally pull up bones in their nets.
In 1922, the island became a psychiatric hospital, which—given the setting—feels like a creative decision that no modern institutional review board would approve. The hospital closed in 1968, and the island has been abandoned since. Italy’s government has periodically entertained bids to develop it, but access remains restricted, and anyone trying to visit without authorization faces fines. The paranormal tourism industry would very much like Poveglia to be open. The Italian government would very much like everyone to stop asking.
Svalbard Global Seed Vault, Norway
Built into the permafrost of a mountain on the island of Spitsbergen in the Svalbard archipelago—roughly 1,300 kilometers from the North Pole—the Global Seed Vault stores duplicate samples of seeds from gene banks worldwide. It’s a backup drive for agriculture. If a crop variety goes extinct due to disease, climate change, war, or catastrophic incompetence, the seeds are here. The vault holds over 1.3 million seed samples from virtually every country on Earth, stored at minus 18 degrees Celsius in a facility designed to remain frozen even if the refrigeration fails, because the permafrost provides passive cooling.
You cannot visit. The vault is not a museum. It opens a few times a year for depositors—organizations that are adding or withdrawing seeds—and for the occasional delegation of dignitaries who want a photo op in front of the world’s most famous door. The entrance, with its illuminated art installation visible against the Arctic landscape, has become one of the most photographed structures in Norway, which is remarkable for what is functionally a very cold filing cabinet.
Heard Island, Australia
A volcanic island in the southern Indian Ocean, roughly 4,000 kilometers southwest of Perth, Heard Island is technically Australian territory but might as well be on another planet. It is one of the most remote places on Earth, dominated by Big Ben—a 2,745-meter active volcano covered in glaciers—and home to colonies of penguins, seals, and seabirds that have never been habituated to humans. The island is a UNESCO World Heritage Site and access requires a permit from the Australian Antarctic Division, which is granted almost exclusively to scientific researchers. The weather is, by all accounts, spectacularly hostile—constant gale-force winds, temperatures rarely above freezing, and seas that make the approach by boat genuinely dangerous.
The restriction isn’t about secrecy or military sensitivity. It’s about preserving one of the last places on Earth where a complete subantarctic ecosystem functions without any human footprint. No introduced species, no infrastructure, no trails, no garbage. The Australian government’s position is essentially: we have one of these and we’re not going to let anyone ruin it.
Lascaux Cave, France
The Lascaux cave paintings in the Dordogne region of southwestern France—created approximately 17,000 years ago—are among the most important archaeological sites on Earth. They depict horses, aurochs, deer, and abstract symbols in pigments that have survived for seventeen millennia. The cave was opened to the public in 1948 and received roughly 1,200 visitors per day through the 1950s. By 1963, the carbon dioxide, heat, and humidity generated by all those human bodies had caused visible damage to the paintings—green algae, white calcite deposits, and black fungal growth on surfaces that had been stable for 170 centuries until tourists showed up.
The cave was closed to the public in 1963 and has remained closed since. Even the researchers who monitor the paintings are limited in how often and how long they can enter. The French government built Lascaux II, a precise replica of the two most famous chambers, about 200 meters from the original, and later Lascaux IV, an even more elaborate facsimile. You can visit a copy of the cave. You cannot visit the cave. The originals sit in climate-controlled darkness, preserved by the absence of the very species that created them.
What forbidden zones actually tell us
The through-line across these places—and the dozens of others like them, from Mezhgorye in Russia to the Heard Island volcano to the closed military cities of the former Soviet Union—is that the reasons for restricting access are almost always more revealing than whatever’s behind the fence.
North Sentinel Island is forbidden because of an unresolved collision between indigenous sovereignty and the modern world’s inability to leave anything alone. Area 51 is forbidden because military secrecy, once established, takes on institutional momentum that outlives its original justification. The DMZ is forbidden because two countries that never signed a peace treaty have been in a frozen standoff for seven decades, and the buffer zone between them has accidentally become an ecological paradise. Lascaux is forbidden because we loved something to death and had to wall it off from ourselves. Svalbard is forbidden because someone had the foresight to build a backup plan for civilization and the wisdom to not turn it into a tourist attraction.
These aren’t just places you can’t go. They’re places that tell you something about how humans interact with the things they find most valuable—whether that’s a species, a secret, a painting, or a seed.
We cover forbidden zones, disputed territories, unrecognized states, and places erased from official maps across our Off The Map: A Global Atlas of Non-Existent Places course—a full atlas of the places the world doesn’t quite know what to do with.
-
Directed Energy Weapons Explained: Lasers, Microwaves, and the Future of Missile Defense
For roughly forty years, military laser weapons have been perpetually five years away. Ronald Reagan announced the Strategic Defense Initiative in 1983—popularly known as Star Wars—and promised a constellation of space-based lasers that would zap Soviet ICBMs mid-flight like something out of a movie that, at the time, hadn’t even finished its original trilogy. The orbital lasers never materialized. The Soviet Union collapsed on its own. And directed energy weapons settled into a comfortable holding pattern as the technology that was always about to be revolutionary but never quite ready for Tuesday.
That changed. Not in the dramatic, press-conference way. In the boring way—which, if you’ve been paying attention, is how all the important stuff actually happens. As of March 2026, the U.S. Navy has laser-equipped destroyers deployed to active combat zones. Israel has fielded what it calls the first operational ground-based laser defense system. Australia is building autonomous laser turrets that can destroy 200 drones on a single battery charge. The Pentagon just announced it wants directed energy weapons fielded at scale within 36 months. And the entire push is being driven not by some theoretical breakthrough in physics but by a brutally simple math problem: a $3 million Patriot interceptor missile is an absurd thing to fire at a $30,000 Iranian drone, and you’re going to run out of missiles before they run out of drones.
That cost asymmetry is the whole story. Everything else is engineering.
What directed energy actually means
A directed energy weapon uses concentrated electromagnetic energy—rather than a physical projectile—to damage, disable, or destroy a target. The two categories that matter militarily right now are high-energy lasers (HEL) and high-powered microwaves (HPM). Particle beam weapons exist in theory and in certain classified research programs, but they’re not close to deployment, so we’ll leave them in the filing cabinet.
High-energy lasers work by focusing a beam of coherent light on a target and holding it there long enough to transfer sufficient thermal energy to cause structural failure. That’s the clinical version. The practical version: you point an extremely powerful beam of light at a drone, the beam heats the surface material until something critical—a motor, a wing spar, an electronics housing, a fuel line—melts, burns through, or otherwise ceases to function, and the drone falls out of the sky. The key phrase is “holding it there long enough,” because unlike what the movies suggest, a laser weapon doesn’t vaporize things on contact. It’s more like using a magnifying glass on an ant, except the magnifying glass costs $30 million and the ant is traveling at 200 miles per hour.
High-powered microwave weapons take a different approach. Instead of heating a small spot to destruction, they emit a broad cone of electromagnetic energy that fries the electronics inside a target. No need to melt through a fuselage—just overwhelm the circuit boards, the flight controller, the GPS receiver. The drone doesn’t explode; it just stops being a drone and becomes debris. The advantage: you can hit multiple targets simultaneously if they’re within the emission cone. The disadvantage: the effective range is shorter than lasers, the physics of directed microwave propagation are less forgiving, and the technology is less mature.
What’s actually been deployed
The U.S. Navy has the most operational experience. The AN/SEQ-3 Laser Weapon System (LaWS) was tested aboard the USS Ponce in the Persian Gulf starting in 2014—a 30-kilowatt system that successfully engaged small boats, drones, and other targets. That was the proof of concept. Since then, the Navy has moved to ODIN (Optical Dazzling Interdictor, Navy), a soft-kill laser system that’s been installed on multiple Arleigh Burke-class destroyers. ODIN doesn’t destroy targets—it blinds or dazzles the sensors and optics on incoming drones and missiles, degrading their ability to navigate or track. In February 2026, an ODIN-equipped destroyer was photographed launching Tomahawk missiles during Operation Epic Fury—the U.S. military campaign against Iranian targets—marking what appears to be the first deployment of a shipborne laser system in a major combat operation against a state adversary.
The Navy’s next step is HELIOS (High Energy Laser with Integrated Optical-dazzler and Surveillance), a 60-kilowatt system from Lockheed Martin designed to actually destroy targets rather than just blind them. Beyond that, the SONGBOW program is developing what could become the first 400-kilowatt shipboard laser, combining multiple 50-kilowatt industrial laser units into a single directed beam capable of engaging drone swarms, cruise missiles, and fast-moving threats at considerable range. The Navy’s stated ambition—and I’m quoting senior leadership here—is “a laser on every ship.” Whether that ambition survives contact with procurement budgets and manufacturing constraints is a separate question.
Israel deployed Iron Beam in 2025, a ground-based laser system designed to complement the Iron Dome missile defense system by handling the lower end of the threat spectrum—drones, rockets, and mortars that Iron Dome currently intercepts with missiles costing tens of thousands of dollars each. The logic is identical to the Navy’s: use the laser for the cheap threats, save the kinetic interceptors for the expensive ones. Iron Beam reportedly can destroy a target in seconds, and the cost per engagement is, by missile defense standards, effectively negligible—a few dollars’ worth of electricity versus a $50,000 Tamir interceptor.
The U.S. Army has been prototyping aggressively. The Directed Energy Maneuver-Short Range Air Defense system (DE M-SHORAD) mounts a 50-kilowatt laser on a Stryker armored vehicle, and four units have been deployed operationally. The Army has also tested systems ranging from 10-kilowatt palletized units for fixed sites to 300-kilowatt systems designed to engage larger threats like cruise missiles and artillery rockets. Of the 17 directed energy prototypes the Army’s Rapid Capabilities and Critical Technologies Office has developed, 11 have been deployed. The service is now pursuing an Enduring High Energy Laser program—its first program of record for laser weapons, meaning the first one that’s not just a prototype or experiment but an actual acquisition program intended to produce systems at scale.
Australia’s EOS Defense introduced Apollo, a laser system targeting around 150 kilowatts that can provide 360-degree coverage and destroy up to 200 drones on a single battery charge. At a time when everyone is talking about counter-drone capabilities in the abstract, EOS is quoting a per-shot cost that makes conventional ammunition look like a luxury good.
Why “infinite magazine” is more complicated than it sounds
The sales pitch for directed energy weapons always includes some version of “unlimited shots.” And it’s true that a laser doesn’t consume physical ammunition—there’s no magazine to empty, no missile to reload, no shell casing to eject. But “unlimited” has asterisks, and those asterisks are where the engineering actually gets difficult.
Power generation is the first constraint. A 50-kilowatt laser needs roughly 150 kilowatts of electrical input power (lasers are not 100% efficient—current military solid-state lasers operate at roughly 30-40% wall-plug efficiency). A 300-kilowatt laser needs close to a megawatt. Generating that kind of power on a Navy destroyer is one thing—the ship has gas turbines producing tens of megawatts. Generating it on a Stryker armored vehicle or a Joint Light Tactical Vehicle is a completely different engineering problem. The vehicle’s power plant wasn’t designed to run a megawatt-class weapon while also driving, running communications, and keeping the AC on. This is the size, weight, and power problem—SWaP in Pentagon jargon—and it constrains how powerful a laser you can put on which platform.
Thermal management is the second constraint and arguably the harder one. That 60-70% of input energy that doesn’t become laser light becomes heat. A 300-kilowatt laser firing sustained bursts generates enough waste heat that you need an industrial cooling system to keep the optics and gain medium from degrading. On a ship, you can dump heat into seawater. On a ground vehicle in a desert, your options are more limited. The Army has found that the optics on its laser prototypes are one of the highest-failure-rate components—not because the optics are bad, but because thermal cycling degrades precision surfaces over operational use in field conditions.
Atmospheric effects are the third constraint. Lasers travel at the speed of light, which sounds unbeatable until you remember that the light has to travel through atmosphere. Rain, fog, dust, smoke, and humidity all scatter and absorb the beam, reducing the energy that actually reaches the target. A 50-kilowatt laser in clear desert air at 2 kilometers performs very differently from the same laser in North Atlantic fog at 5 kilometers. Microwaves are less affected by weather but more affected by range—the beam spreads, and the energy density drops with distance.
This is why nobody serious is claiming that directed energy replaces conventional weapons. The operational concept is layered defense: lasers and microwaves handle the high-volume, low-cost threat layer—drone swarms, rockets, mortars, loitering munitions—while conventional missiles handle the high-end threats that require a kinetic kill at long range. The laser is the cheaper, faster, deeper-magazine first line. The Patriot battery is still there. It just doesn’t have to waste a $3 million interceptor on something a $10 beam of light can handle.
Where this goes next
The $250 million directed energy R&D funding in the 2025 reconciliation bill, the Army-Navy joint laser program under Trump’s “Golden Dome for America” missile shield concept, the Pentagon’s stated 36-month fielding timeline—all of this points in the same direction. The institutional commitment is real. The technology is mature enough to be useful against today’s primary threat—small drones—and the manufacturing base is what’s lagging.
That manufacturing base problem is real and worth taking seriously. The companies building these systems—Lockheed Martin, Raytheon (now RTX), BlueHalo (now part of AeroVironment), nLight, EOS—have prototype and low-rate production capabilities, not assembly lines. EOS’s new Singapore hub plans to produce five to ten systems per year. That’s artisanal output for a capability the Pentagon wants on every ship and every forward-deployed ground unit. Scaling from “onesies and twosies” (the Army’s actual phrase) to hundreds of units requires a manufacturing ramp that hasn’t started yet. And—because everything connects to everything—the advanced optics and laser components in these systems require rare earth elements and specialty materials with the same supply chain vulnerabilities we keep running into.
The DOD roadmap calls for scaling from current 150-kilowatt-class systems to 500 kilowatts by 2030 with reduced size and weight, and eventually to megawatt-class systems. If those numbers hold, you’re looking at weapons that can engage not just drones and rockets but cruise missiles, ballistic missile warheads, and possibly aircraft at tactically significant ranges. That’s when directed energy stops being a complement to missile defense and starts being the missile defense.
But that’s a big “if,” and the history of directed energy programs is littered with timelines that didn’t survive contact with physics, budgets, or the bureaucratic reality of defense procurement. The technology works. It’s deployed. It’s killing drones in active combat zones right now. The question is whether it can scale fast enough and far enough to matter against the volume of threats that modern warfare is producing—and whether the manufacturing base, the power systems, and the thermal management can keep pace with the ambition.
We cover directed energy weapons—alongside drones, electronic warfare, autonomous systems, and every other technology reshaping how wars are fought—across 36 lectures in our Battlefields of the Future course. The full kill chain, the full technology stack, the full timeline from 2025 to 2125.
