Current explanations of what makes our species unique and successful focus on factors like our ability for shared intentionality or cooperation or developing agriculture. Here we provide a new, overlooked explanation: what makes us H human is our extraordinary capacity to invent better methods and tools that expand our mind, to test them and to improve them. Our uniqueness comes down to this: we are cumulative method-makers who amplify our own cognition. Humans are not just tool users—we are master method-makers. Our species has mastered this unique survival strategy: building, upgrading and switching between powerful methods to tackle new challenges. This remarkable capacity is not just the driver of our survival but also of scientific progress. At the heart of this capacity is our universal and adaptive toolbox: a set of evolved methodological abilities to observe, experiment and solve problems that we have collectively developed into increasingly powerful methods over time—from rigorous experimentation to statistics. Our method-making capacity is why our species has been so successful in generating knowledge and controlling our environment. While we are not the only tool-using species that communicates and cooperates, we are the only complex method-making species using diverse tools across domains. Only our species can leverage the shared ability for trial and error to develop controlled experiments, or go from the shared sense of quantity to create geometry and statistics, and so on. Ultimately, cumulative method-making has been the key missing link enabling us to understand the world in ways no other species ever could. It characterises our unique mind and endows us with a different relationship to the world than any other animal. With tools, we come up with better ways to represent, test and even simulate the world. Our toolbox gradually set our species on a new method-driven evolutionary path that no other animal had yet taken. It made us increasingly unique. A new kind of species emerged: homo methodologicus, the cumulative method-making human. We diverged from other primates as our capacity to invent methods expanded, together with more complex language, imagination and cooperation. In fact, method-making enabled us to turn language and imagination into something more permanent and shareable—by creating external tools like mathematical symbols, written notation and writing systems. How our powerful methodological mind developed needs to feature at the centre of how we explain our evolutionary history. Our ability to develop an adaptive toolbox represents arguably the most unique feature in understanding the gap between the evolution of the mind of humans and smart non-human animals— an ability that is not just likely or we can imagine, but was necessary to be able to extend our mind beyond theirs. Without it, we could not have ever created science, technology and complex societies. Our cumulative method-making capacity is what distinguishes and defines us as the extraordinary species we have become—extending our mind beyond our DNA (Table 7.1).
knowledge over human history For over a millennium, scholars have been searching—from Aristotle to Popper— for a unified method for how we gain knowledge. The nearest we can get to a single, universal method over history—for what our species has always relied on to uncover knowledge about the world—is our universal methodological 234 T HE ENGINE OF SCIENTIFIC DISCOVERY abilities to observe, solve problems, experiment and develop methods. Our evolution can provide a natural foundation for this understanding. Our ability to observe is fundamental to nearly all knowledge we acquire (Chapter 6). Our mind's methodological capacities are inherited regularities that link our human nature to knowledge. We have had no choice but to use these evolved abilities. This universal toolbox view here is as close as we may get to a natural definition—a universal method—of what generating knowledge has been over our species' history. In contrast, we explored contemporary science in earlier chapters; we reframed the classic scientific method— from our mind's internal processes of observing, hypothesising and experimenting that we have used throughout human history—to the sophisticated external scientific methods and tools we more recently began developing that spark science's major discoveries (Chapter 3). While scholars have asked for millennia how we generate knowledge and discover things, the deeper question is how we even became able to develop knowledge and discoveries at all? No answer was possible as long as we did not understand the evolution of our species and mind. And many did not until the mid-19th century and Darwin's On the Origin of Species (Table 7.1). Science has evolved from our evolved mind. We are not just a unique species, we are methodologically unique. Our large flexible system of methods embodies our intellectual evolution: from observing, solving problems, experimenting and abstracting throughout history to observing methodically, solving problems statistically, controlled experimenting and modelling abstractly in science. Viewing humans as homo methodologicus—using a universal and adaptive toolbox that extends our mind— better explains how we have been directly able to meet our needs, solve problems and develop technological and scientific knowledge—and give rise to complex society and science. Our powerful toolbox explains our success better than commonly focusing on geography, culture or religion as the driving force behind these great human achievements. It is how we even became able to surpass our animal nature and limits. In the final section, we now explore how our extraordinary toolbox has not just shaped our understanding but has shaped our species' remarkable cognitive evolution itself.
drove human evolution What has driven our remarkable evolution to enable us to develop sophisticated methods and technological and scientific knowledge in the first place? How did we get from our early ancestors running in the savannahs and forests using trial and error, to today's scientists running complex experiments using statistics? Scientists have long debated the origins of human intelligence. They propose three main explanations. One view is the general intelligence hypothesis. Over time, the human brain evolved to be bigger and better at processing information than our primate relatives. Our brains are about three times larger than theirs, giving us more memory, faster H thinking and stronger reasoning skills. A second view is the ecological intelligence hypothesis. Our mind mainly evolved adapting to environmental challenges— just as certain birds have evolved extraordinary eyesight for high altitudes and use magnetic fields to perceive their altitude. Darwin and Wallace supported this view: that selection favoured individuals best able to adapt to nature. A third view is the cultural intelligence hypothesis. Our intelligence evolved through navigating the complex social world—through challenges we faced in complex social groups. These pushed our minds towards greater cooperation, cultural traditions and more complex language. Each of these competing hypotheses helps explain part of the puzzle—the broader context. But they do not explain or talk about how and why we directly became able to create methods and tools that go far beyond what nature endowed us with—and to eventually develop science. What is missing is an integrated approach that combines these views and places our evolved mind's methodological capacities at the centre. These abilities truly set our species apart. So how did they evolve to enable our ever more complex knowledge? Here we lay out a new view— the toolbox hypothesis—that highlights how our capacity to observe, solve problems, experiment and develop methods evolved in increasingly complex ways. It explains how this complex method-making capacity of our cognition evolved adapting to our natural and cultural environment over time. This toolbox hypothesis explains how evolution favoured those members and groups of our species who could more effectively apply our mind's method-making capacity. Throughout history, those better able to develop more effective methods for tracing animals, healing injuries, navigating using the stars, building stronger shelters and eventually growing food were more likely to survive and pass on their method-making capacities—and their genes. Over time, these early methods evolved into more complex ones. Those better at using this expanding toolbox had an edge: developing better plant-based medicines, lunar calendars to predict ocean tides and better technologies. Eventually, the best method-makers invented agricultural techniques and numerical systems to more effectively plan and produce food (Chapter 7). Each successful method added a survival advantage. Over generations, those advantages multiplied—through continual feedbacks. As our species migrated across the globe, we were forced to readapt and redesign our methods and tools over and over—facing unfamiliar climates, landscapes and threats. The result is our adaptive, method-making mind. Yet for psychologists like Steven Pinker, one of the great mysteries of our mind and human evolution has been precisely the question: how have we—given our evolutionary history—evolved to do science? How have we created calculus and surrealist art that are not directly tied to our survival? After all, evolution favours traits that help us survive and reproduce. Yet our early ancestors evolved abilities—from pattern recognition to numeracy—that helped us reason logically needed to plan hunting routes and food cycles. Similarly, we also evolved abilities for abstraction needed to carve stone tools and design shelters (Chapter 7). Today, to develop logic, calculus, abstract art or scientific discoveries, we rely on these same evolved abilities. But for purposes that often no longer immediately influence our survival. So the toolbox 236 T HE ENGINE OF SCIENTIFIC DISCOVERY hypothesis offers an answer to this mystery: our ability to develop science and complex society grew gradually from this evolving, flexible toolbox that once directly helped us survive. For most of our evolutionary history, the environment and biology shaped our minds—predators, droughts, diseases, as they shape all species. But over time, we learned to develop an extraordinary set of tools to manipulate the environment and our biology. This is how the edge shifted in our favour: we began shaping our environment just as much as it shaped us. Over time, our method-making capacity and sharing innovative tools across generations became more important than the environment, animals, plants or disease in shaping our mind's evolution. Why? Because the more effective the toolbox in our hands, the better we could survive and adapt to these changing external factors—through techniques to control fire and develop farming, clothing and shelters for nearly any environment. These were buffers we built against nature. What shaped the evolution of our mind shifted from external environmental challenges over much of our past to internal feedbacks: cognition, method-making and culture. Ever more sophisticated method-making and tool-making emerged, symbiotically with greater cooperation, language and imagination. Our tools turned into new selective pressures for our brain's evolution. Evolution selected better toolmakers—through evolutionary feedback loops. In a sense, our tools began shaping and evolving us. Ultimately, we develop and use increasingly complex tools because they improve our ability to survive and adapt. This kind of method-driven selection did not just apply to individuals, but also to groups. Tribes and communities that learned themselves or from others how to more efficiently exploit their environment were more likely to outgrow and outlive huntergatherers. Think of the extraordinary advantage for those who mastered experimental methods for agriculture and storing food or observational methods for predicting seasons. Some could then increasingly invent the remarkable techniques to domesticate animals, breed plants, irrigate fields and fertilise soil. By changing how we thought and interacted, our expanding toolbox sparked survival rates and population growth to rise at a pace not yet experienced in history. Early cities and societies emerged and grew at the pace we expanded our toolbox—and knowledge accumulated faster than ever before. These were tipping points in human history. And at their heart, better method-making was driving them. By developing agriculture and then early civilisations came a method revolution . With more stable food supplies, more people could dedicate their time to testing and crafting better tools. This kicked off a method-making surge. Throughout history, human populations that rapidly increased their ability to survive and innovate generally did so by extending their methodological toolbox. These method innovations, by stretching our mind, have driven our species' advances and success. And we can observe the methods at play in each advance. (Even the satisfying thrill and eurekalike feeling when creating a new tool that works or making a new discovery may be evolution's way of rewarding us for solving complex problems, signalling they are important to us.) While social and emotional intelligence helps us bond and form communities, those with greater methodological intelligence —who excel at testing and refining tools—were more likely to optimise strategies for gathering food, engineer better shelters and unlock new knowledge, giving them an inevitable survival advantage. Method-making is not just an outcome of our intelligence—it is central to what made our intelligence possible and helps define it. Our minds created better methods, and those methods in turn refined our minds—from one generation to the next. Our method-making capacity is both an inseparable product and driving force of our evolution. It is an essential part of explaining what made our unique, evolved species and mind possible. It shaped who we became. How our mind evolved is, more than any other species, an outcome of our own making—the result of being method-makers, designing new and better ways to solve problems. We depend, with each new method, on being taught how to use them: from arithmetic, to techniques for herbal remedies, to types of experiments. Ultimately, this toolbox view explains why we are the only species to build science. It explains how we became able to develop complex tools and knowledge—and eventually learn to create experimental controls, solve differential equations, develop AI and make discoveries that no other species can. If our methods and tools shaped our brains in the past, then developing more powerful tools today—better statistical, computational or experimental methods—is not just scientific progress, it might also have evolutionary spinoffs. Box 8.1 Our toolbox functions like an immune system—universal and adaptive Think of our scientific toolbox like an immune system: both have universal parts we are born with and adaptive parts we later develop. We constantly scan our environment for problems to solve, much like our immune system constantly scans for threats to tackle. We are born into the world with a universal immune system—broad but unspecialised. We are initially more often sick because our basic defences have to learn. But over time, we adapt, develop antibodies and build up immunities and a stronger defence system to better tackle bacteria and viruses and better survive: our adaptive immune system. Our toolbox works the same way. We are also born into the world with a universal toolbox—also broad but unspecialised. Our mind and senses are constrained to our niche of the world we can observe, and we hit the limits of our bare mind. But over time, we adapt, create and master more complex methods and tools that enable us to tackle more complex challenges and better survive and understand the world: ouradaptive toolbox. Our mind develops new methods, just as our body develops new antibodies—by learning from experience. Our immune system learns by facing immunological challenges (new viruses and bacteria). Our toolbox learns by facing real-world challenges (new problems in nature and society). Our toolbox naturally divides into sub-toolkits—just like our immune system into subsystems. We use our evolved universal toolbox to observe, experiment 238 T HE ENGINE OF SCIENTIFIC DISCOVERY an d reason, and our adaptive toolbox to produce mathematics, microscopes and controlled experiments. Our immune cells recombine genes to create seemingly endless antibody variations to tackle immunological problems we face. We invent and recombine methods and tools to create seemingly endless method innovations to tackle scientific and technological problems we face (Table 6.2). Our toolbox is our best means to confront the challenges we encounter through our diverse methods (experimental, mathematical etc.)—just as with our immune system (cells, antibodies and proteins). Our toolbox is our cognitive defence and survival system. It protects us against disease by enabling us to design clinical trials to test how effective our vaccines are. It helps us prevent food shortages and deal with natural disasters by allowing us to devise methods to better plan and reduce risks. (Our powerful immune system is very effective but can at times attack the body, causing autoimmune disease or failing to stop cancerous cells. Our powerful toolbox is also very effective but can at times have unintended effects, attacking science and society through environmental degradation and nuclear weapons.) Ultimately, our adaptive toolbox lets us confront vast challenges and understand the world in ways no other species can. Without it, science and technology would not be possible.
tools—including AI systems At the heart of AI lies the bold ambition: to replicate and even enhance our core methodological capacities (observing, experimenting, solving problems) that evolution has endowed us with—our human intelligence. From complex causal reasoning and imagination to designing experiments and running statistical analyses, scientists are attempting to translate many of the cognitive capacities we have developed and honed over millions of years into computer code. It is about trying to build systems that can carry out the wide range of reasoning abilities we have evolved—systems that can think, learn, infer and solve ever more complex problems. Flexibly emulating these faculties demands an extremely deep understanding of our mind's architecture itself: how it evolved, how we learn from experience, how we solve problems we have never faced before and what it takes to try and recreate these capacities. It is about trying to reverse-engineer ourselves. This brings us to a central puzzle in understanding this shift—from the evolutionary roots of our cognitive abilities to designing complex artificial systems: how do we best try and reconstruct these remarkably evolved capacities? How do we best link and power our scientific tools—from statistical methods to microscopes—with current AI systems? How can we understand the logic behind such systems? Should we model our methodological abilities and tools as large language models or deep H neural networks? Or probabilistic reasoning systems or dynamic network systems? Or as something not yet conceived—systems that combine all these approaches into a single adaptive architecture? Each model offers a different lens into how to capture the complexities of human thought. But none fully captures its richness. So how do we best rebuild something that evolved over millions of years and was never designed to be understood in parts? And how do we best understand the design of tools that extend—and even reinvent—our evolved mind including our extraordinary method-making capacity? It involves testing whether replicating specific capacities is best achieved by isolating cognitive functions or by integrating them into a unified architecture. These questions are not just methodological and technical, they are also deeply philosophical and ethical. Ultimately, this reflects a challenge at the core of science: how to understand the complexities of our mind and tools and tackle their bottlenecks with ever more sophisticated tools. We as method-makers may enter a new phase with tools like deep machine learning, where we could build methods that further build themselves: metatools. The evolution of method-making did not stop with stone tools, compasses or telescopes—it continues to accelerate in the digital age vastly faster than we evolved.
To solve the puzzle of the origins of science, we need to step back and ask deeper questions: what gave our species the power to adapt to nearly every environment on the planet, intervene in nature and domesticate plants and animals? What has driven our ability to generate scientific and technological knowledge—and eventually develop modern science? How can we overcome the evolved limits of our own mind and senses? What trait best describes humans as the unique species we are? And how have we been able to increasingly influence our mind's remarkable evolution over time? The answer that ties these diverse questions together: our mind's evolved methodological abilities to observe, experiment and solve problems—our universal toolbox—that we have developed into increasingly powerful methods and tools—our adaptive toolbox. With this toolbox, we can better explain—as the unique, cumulative method-making species we are—the origins and foundations of science and technology. While evolution gradually sharpened these inherent abilities throughout human history, we eventually began sharpening and expanding them into ever more complex methods and physical tools. We began shaping the tools that vastly expanded our own evolved mind. We are not just a species that thinks in complex ways, we are the species that builds methods and tools to think better: homo methodologicus. Methodmaking has become its own evolutionary force—one that we can now steer. What first began as a survival advantage that only our species learned to master—as cumulative method-makers and toolmakers—became both what made us human and the very engine of discovery itself. 240 T HE ENGINE OF SCIENTIFIC DISCOVERY Fast forward to the remarkable invention of the microscope and telescope at the turn of the 1600s. For the first time, our tools drastically surpassed our evolved human abilities for perceiving and imagining like never before. This marked the crucial turning point in science: when the driving force of discovery shifted from our unaided mind to our ever more complex tools and methods themselves that enabled peering into the stars and deep into the cell. The power of these tools—as extraordinary extensions and upgrades to our mind—simply became undeniable and redefined what science today is: more tool-driven than mind-driven. That shift has only accelerated. No one imagined or predicted how merging statistics with computing power would transform science. But today, most of science is no longer imaginable without this powerful hybrid method combination (Table 6.2). These once-unimaginable tools are now behind almost everything from vaccine trials to climate models to the search for dark matter. Yet precisely because these extraordinary tools have become so common in science, we take them for granted. Every day we use them across science, but rarely take the time to ask the crucial question: what better tools could take us further that we have not yet built? Because we are comfortable with today's tools, we direct far less attention than we should to pushing these methods even further—to inventing the next generation of tools that can unlock tomorrow's breakthroughs. Getting too comfortable with our current toolbox means we stop pushing its boundaries. But each time we hit a wall using our tools in science— each time the predictions fall short, the electron microscope cannot go further—it is not failure. It is a critical sign: it means we are getting closer to the frontier, if we learn to upgrade those very methods and tools to break the limits of science. Ultimately, our species' biggest discovery is not a single tool or theory—it is realising that using our methodological abilities we can develop better methods and tools that enable us to better discover, understand and control the world. With our toolbox, we developed farming techniques and early medicines thousands of years ago. Then we eventually invented statistical surveys and controlled experiments to understand the mass population growth brought about by such agriculture and the effectiveness of our medical treatments for diseases. We invented technologies to extract many natural resources, and then eventually also built tools that produce satellite and temperature data to understand the causes of climate change that such natural resource depletion contributes to. The discovery of our toolbox and our ability to expand it has opened up the possibility for us to understand and tackle our big challenges—from feeding billions to fighting health pandemics to managing the planet's resources. But like Pandora's Box, we have also given rise to new and larger challenges. Better curing diseases leads to longer lives and bigger populations. Better controlling natural resources can speed up climate change. Again, we turn to science and its tools to tackle these very challenges.