Throughout history, we have continually refined our abilities, to invent navigation techniques using stars, apply medicinal plants to heal wounds, devise lunar calendars to track and predict seasons—and eventually develop farming techniques and early mathematical systems. Each method gave us a survival advantage. Eventually we developed the microscope and telescope that enabled vastly surpassing the limits of our evolved mind—triggering the discovery of entirely new stars, moons and medicines. The evolution of this capacity for method-making and tool-making marks one of the most fundamental changes in human history— and it enables science today. But what made this possible? How did we shift from our unaided mind and instinct to developing increasingly cumulative methods and tools? While evolution gradually honed and refined our inherent methodological abilities to observe, think and imagine throughout human history, we eventually began honing, refining and amplifying them into methods and tools. The bigger the innovation in method, the bigger the leap in progress. So can we best explain our species' success by our unique ability for cumulative methodmaking? Our success is indeed a remarkable story of realising the power of leveraging this capacity to conceive better tools that enable us to think, plan, test and predict better. We explain our uniqueness as cumulative method-makers and how evolution favoured those better at using this exceptional capacity— giving rise to a new kind of species called here homo methodologicus . Viewing humans as complex method-makers who extend our own minds better explains how we have been directly able to meet our needs, solve problems, make vast knowledge and eventually extraordinary discoveries. Here we dig deeper into our scientific origins that are traced from our evolutionary roots to contemporary scientists who aim to tackle our mind's evolved bottlenecks—including with AI tools. The 228 T HE ENGINE OF SCIENTIFIC DISCOVERY
What makes humans extraordinary is cumulative method-making that extends our minds. Our remarkable capacity to make methods and tools is the product of long and complex evolutionary feedbacks between nature and our biology, mind and the culture we built. In turn, these very methods and tools became our best way to understand those forces that shaped us—nature, our biology, mind, culture and even evolution itself. This evolutionary path we humans have taken has enabled our mind to gain increasingly complex knowledge—through early and simple methods we created that we gradually developed into ever more complex methods, until we were doing things no other species had done. Only our species has taken this evolutionary path to creating sophisticated tools. Indeed, many other smart animals have abilities for a sense of quantity, testing hypotheses, causal reasoning, communication and using rudimentary tools—from crows, to chimpanzees, to dolphins. But only our species has evolved to develop statistics, simulate complex ecosystems and devise controlled experiments by applying these extraordinary abilities in more complex ways. That is the leap: we no longer just adapted to nature—we learned to analyse, redesign and shape it with methods. Our method innovations let us see, measure and think in ways evolution never prepared us for. The remarkable turning point came when we began moving beyond intuition, instinct and trial and error. We began building more deliberate methods and external tools to solve our challenges—to better navigate, more efficiently hunt, better treat injuries, build stronger shelters. The more precisely our early ancestors could observe, test and predict, the better their chances of surviving and thriving—and passing down those new successful methods over time. Our evolved, expanding toolbox deeply links the first hunter-gatherers tracking animal footprints and inferring animal migrations to today's scientists running computer simulations of global migration and climate patterns. This extraordinary method-making capacity did not just become useful—but foundational to our species. Without it, we would not have invented early fire-hardened spears, then the first farming techniques, and eventually more systematic randomised experiments (Chapter 7). Before this evolutionary leap in thinking, science and discovery were out of our reach. So what gave us a survival edge in nature in our early history is at the same time what pushes the boundaries of modern science and enables us to understand and predict nature today. (Impatient readers less curious about the evolutionary origins of science can jump to Chapter 9.) Other animals have their own evolutionary strategies—and many are much faster, stronger and have a sharper range of senses than we do. Some birds detect and use earth's magnetic fields to navigate; bees can perceive ultraviolet light to detect nectar; some snakes can perceive heat signatures (infrared) in complete darkness. We did not evolve any of these and many other extraordinary abilities—nor for very cold tundras or very hot deserts. But only we can survive in all of them—not by adapting our genes (besides a few traits like skin colour), but by adapting our methods and tools. This is what sets our species apart. We gradually learned to become cumulative method-makers and toolmakers by more H systematically observing, experimenting and reasoning causally over time. This gave us the key edge over nature, biology and other species. Our ability to develop a set of methods for solving problems and amplifying our mind distinguishes us, more than other factors, from other species: our species is uniquely a complex method-making species, indeed a hyper-methodmaking species—homo methodologicus . This deep capacity to design and create better methods and tools that extend our mind, to test them and to refine them over time is not just what drives science—it is what has always made us human . It is being cumulative method-makers and tool-makers that makes us the species we are—giving us the power to vastly outsee and outthink nature, like no other animal can. In evolutionary terms, this matters. While Homo habilis was an early human species whose name means 'able or handy man' , many other animals are also able to use a stone tool. We have evolved far beyond that. While Homo sapiens means 'wise man' , this general description also falls short because it does not capture what we are: systematic method-builders who reshape how we think and can control nature and our biology through methods and tools themselves. We go beyond those classifications here. Instead, we are a cumulative method-making and tool-making species: homo methodologicus—using our mind's evolved methodological abilities of observing, solving problems, experimenting and developing and refining complex methods. To amplify our mind, early humans used these same abilities to create plant-based medicines and early agricultural techniques—and today's scientists apply and extend them to develop more complex vaccination methods and advanced agricultural trials. We invent new methods as we face new challenges—and to reduce our mind's constraints. What had driven our survival advantages over predators and environments is at the same time what has driven our great discoveries in modern science: the capacity to tackle our human limits using continually better methods and tools that expand our mind. Understanding these evolutionary origins of our toolbox—from our early basic tools to cutting-edge tools—and how it continues to shape our mind and methods today, we are in a better position to overcome our evolved limits. And today's extraordinary instruments enable detecting phenomena invisible to our senses throughout nearly our entire evolutionary history: from tracing sub-atomic particles to gravitational waves and exoplanets. But by designing new instruments, we have detected them since the 20th century (Chapter 1). We are all—whatever our genetic make-up—born with a universal toolbox . It enables us to perceive, apply trial and error and recognise patterns and is activated early in life as we begin interacting with the world around us. But as we grow, we also all inherit something even more powerful: an adaptive toolbox. This is a constantly evolving set of complex, invented methods that allow us to collectively make deeper knowledge and groundbreaking discoveries—many that improve our survival and others that now expand our understanding without an immediate benefit. So our methods have an evolutionary and an adaptive component. Together, we can think of them as our core and extended toolbox—and is the foundation of science itself (Figure 8.1). 230 T HE ENGINE OF SCIENTIFIC DISCOVERY Science Knowledge Discoveries, evidence, theories Adaptive toolbox Controlled experimenting, statistics, microscopes (methods and tools) Universal toolbox Combining abilities, systematic techniques, logical, rule-based reasoning (complex cognitive abilities) Vision, other senses (biological abilities) Nature and cultural context (we evolved in) Observing, problem solving, experimenting (basic cognitive abilities) Language, learning, teaching (social abilities) science is the accumulation of our methods and bodies of knowledge the context in which our universal toolbox evolved we develop methods using our universal toolbox (the goal is to tackle our biological, cognitive and social constraints) we develop knowledge using our (universal or adaptive) toolbox Figure 8.1 Methods drive science pyramid: how we use our toolbox to develop knowledge and science
With our remarkable method-making mind, a natural question arises: where does our cognition stop and the methods and tools we invent start? We use our mind's internal abilities (observation, pattern recognition etc.) to develop external methods that stretch far beyond what our biology can do alone. We build these methods by interacting with nature, the tools we have already built, and other people (Figure8.1). Methods, once invented, do not just exist in our heads: they become part of the outside world—external material artefacts—and can be shared and leveraged by others. Think of Babylonian mathematical tablets or today's computational and statistical programmes. Our sophisticated methods now drive every stage of the scientific process. They help researchers generate hypotheses using algorithms, collect and clean data using database systems, analyse and simulate experiments with statistical programmes and even discover potential cancer treatments with machine learning. Methods massively stretch what our mind can do. Statistical modelling lets us run experiments on climate futures. AI tools can rapidly spot patterns across millions of genetic sequences. They are like mind extensions—amplifying our brainpower with incredible speed, precision, memory and scale in ways that were entirely unimaginable just a few decades earlier. Yet the line can be blurred between internal cognition and external tools. Our cognition at times incorporates the extended tools it leverages, and ourmethods integrate the internal processes they rely on. To develop new methods, we depend on our mind's methodological abilities and—in contemporary science—also existing methods and tools. Without us, methods do not have meaning, context or questions to solve. H So the important question is not just 'at what point does our cognition end and methods start?' but rather: 'how can we better understand our mind-method synergies and limits and how can we continue refining our best methods to tackle them and better understand reality?' We return to this essential question later. It is clear that we are not biologically endowed with complex methods—in our brains. No single mind could alone come up with statistics. So how did we even develop such a powerful method in the first place? Did an individual not have to begin the process? Like most of our methods, statistics was not invented in one leap. In early civilisations, we first created systems of mathematics. By the 1600s, European states were collecting basic population data. And mathematicians like Pascal and Fermat were laying the groundwork for probability theory. In the 1800s, Gauss and Laplace formalised statistical reasoning, and in the early 1900s, the pioneering statistician Ronald Fisher introduced game-changing techniques like randomised trials and analysis of variance. In the 2000s, scientists took these methods further by combining statistical analysis with digital computing power and machine learning—tools and methods that keep evolving today (Box 5.1). What developed over many generations is a highly composite method called statistics—a multi-layered invention that drastically extends scientific cognition and much of what we can understand in the world. Our methods are typically the result of continually reworking and refining existing methods—and statistics, like the microscope, is no exception. Since our mind and senses evolved adapting to our environmental and cultural niche, we face constraints in how we understand the world especially beyond our niche. Science pushes vastly beyond our evolutionary environment and that is where our raw abilities hit a wall—when studying phenomena from viruses to distant galaxies. That is why it is so crucial to focus more on our tools. When we design a better method or tool, we are actually asking how we can tackle a problem by reducing a human constraint we face in understanding the world. We are asking how we can answer a question by improving our current cognitive or methodological capacities—how we can hack, or work around, the limits of our mind and existing methods. Think of how we stretch our memory, perception or reasoning with algebra, microscopes, computers or regression analysis. Think of what it actually takes to study thousands or even millions of observations using regression analysis—for example the effects of a changing climate or new drugs for medical patients. Such analysis was not possible or even thinkable until we developed contemporary statistical methods and computers in the mid-20th century. Merging these two general-purpose methods—statistics with computational tools—transformed nearly every major scientific field: from experimental physics to medicine, psychology and economics. It marked the birth of large-scale, high-speed statistical analysis that characterises science today.
and gaining knowledge—and advancing science Our species has become ever more methodologically adaptive. Unlike any other species, we have diversified and upgraded our methods as a strategy to meet our needs, tackle challenges and make sense of the world. This flexible ability—shifting between different methods in our toolbox—has enabled us to better control and 232 T HE ENGINE OF SCIENTIFIC DISCOVERY discover layers of nature previously out of reach. After all, our biology, mind and senses we are born with have not changed much in the past few thousand years. The reason we understand the world much better today is not better brains we are born with—it is better tools and methods we inherit and create. Our knowledge has surged at the pace we have developed and shared more powerful methods and tools to build that very knowledge (Chapter 7). Our early ancestors could not explain well the mysteries of shooting stars and eclipses. Today, we can—because of inventions like spectrometers and telescopes. We have gone from guessing why lightning strikes to measuring its electric charge using high-speed sensors. We no longer blame disease on the supernatural, but understand how bacteria and viruses cause disease by using microscopes and controlled experiments. But our current tools only enable us to partially understand many other phenomena—from cancer and our complex brain to the origin of life and the mechanics of the global economy. And there is so much we only poorly understand: from the size and nature of the universe to how organisms like us evolved to become self-aware and practice morality. Why do we not fully understand many phenomena? Often because we have not yet invented the tools we need to shed light on them. Remarkably, we are both born and made into method-makers. Theevolved methodmakers in each of us has been an inevitable part of our species' cognition. Every child is born able to observe, test and count. The trained method-makers in each of us has become an inevitable part of our cognition in more recent history. Every child is also trained in a set of inherited, increasingly complex methods. Over the past few centuries—especially since the rise of public education in the 1800s—we have passed down increasingly sophisticated methods, from geometry and algebra to logic and basic scientific reasoning. Each of us has gone through a massive process of method socialisation—for example to be able to read books on science like this one. This process enables us to turn a basic ability to quantify into an ability to calculate complex probabilities using statistical methods. In science and society we heavily depend on statistics and experiments, but we do not intuitively know how to do statistical analysis or controlled experiments (Chapter 7). Our toolbox is adaptive, and grows with us, in three powerful ways. We adapt our methods over time to meet new challenges. Our methods enable us to better adapt to nature itself. And perhaps most fascinating, our methods adapt how our mind works. When we learn statistics, many begin thinking in probabilities—what are the odds of catching a virus, being affected by extreme weather or even living past 100. Our methods become thinking tools, helping us reason and imagine in new ways—in and out of science. The more we expand our toolbox, the more we stretch and reshape the boundaries of how our mind can think and imagine—and even think about thinking and imagination (Chapter 6).