This is the first book to systematically tackle the puzzle of what the engine of scientific discovery is. We analyse science's over 750 biggest discoveries—spanning across fields and history—and the broad range of traits of the discoverers, and provide a new lens on science's powerful methods and tools. This is an until now unexplored perspective to discoveries. The book takes a big picture view: to better understand science, its methods, its limits and its future, we have to study them systematically, combining methods and evidence from across fields. So what exactly do we cover in the book? Chapter 1: How do we spark new discoveries?To tackle the puzzle, we study science's biggest discoveries and link them to the methods and tools they used and to the discoverers themselves. We find a surprising pattern: across fields, science's major discoveries are sparked by a new method or tool that enables exploring the world in entirely new ways—from revealing unexpected bacteria to unknown particles. Often within just a few years. We uncover that powerful tools—from new x-ray devices and electron microscopes to statistical methods and spectrometers—have each unlocked dozens of major discoveries that were impossible without them. Inventing the laser for example triggered breakthroughs across fields that impact our lives from laser surgery to fibre optics—the backbone of high-speed internet and telecommunications we rely on. In Chapter 1, we lay out the core analysis that the other chapters build on— giving us a powerful new way to understand and accelerate discoveries. This leads us to explore how common expected and unexpected discoveries are in the next chapter. Chapter 2: What role do serendipity and chance play in discovery? For many, this is the most puzzling and mysterious feature of discovery—because it seems unpredictable. By analysing science's major discoveries, we uncover a striking pattern behind the surprise: breakthroughs often seen as serendipitous are actually triggered by using a powerful new tool, for the first time, that makes the unexpected observation possible. From pulsars (neutron stars) using a new massive radio telescope, to superconductivity using a new liquefier. New tools systematically enable other researchers to arrive at the same findings applying these same discovery tools. Serendipitous discoveries generally reduce the role of theory as they are sparked by new tools before anyone knew what to look for or how to explain it. In short: new tools are what commonly unlock surprise. It is far less about stumbling upon serendipitous insights and far more about engineering discoveries by constructing better ways to see, sense and think—and accelerate unexpected findings. So what if much of the randomness and serendipity associated with discovery so far has not come from the scientific system, scientists and their minds, or even unpredictable timing—but from something else entirely: not yet focusing deliberately on our tools' blind spots? Understanding the role of surprise in discovery brings us to the next chapter, where we explore just how cumulative science actually is—and how our discoveries fit into the overall picture of scientific progress. Chapter 3: Does science mainly go through revolutionary paradigm shifts—or is scientific progress mainly cumulative? Here we test this influential hypothesis of abrupt paradigm shifts—like the Copernican revolution. We uncover a remarkable pattern. Three key measures of scientific progress—science's major discoveries, methods and fields—each point to the same finding: the deeply cumulative evolution of science, not sudden ruptures that fundamentally challenge and overturn what we believe. In fact, we find that we have not entirely replaced any major scientific methods or tools we use across fields—from spectrometers to lasers. Likewise, we have not entirely abandoned any major scientific fields—from nuclear physics to biomedicine. They are not subject to radical paradigm shifts; rather, we expand them over time. Scientific discoveries are also highly cumulative: only 1% of over 750 major discoveries have ultimately been abandoned. We also explore the classic scientific method of testing hypotheses by observing and experimenting: is it compatible with the cumulative nature of science—and did science's major discoveries follow it? After examining the inner workings of science, we explore the broader external factors in the next chapter. Chapter 4: How does the broader environment—demographics, institutions and resources—help support the scientific process, and our powerful new tools? And more intriguingly, who are the scientists behind our major discoveries? By studying science's biggest discoverers and their broader traits, what emerges is a unique portrait of the greatest scientists in history. This also enables spotting what factors only apply to certain breakthroughs—and what connects across them. Strikingly, we find that interdisciplinary scientists, those with broader method training, are behind about half of science's major discoveries. The German Konrad Bloch for example completed degrees in chemical engineering and biochemistry, enabling him to apply new isotopic labelling methods to discover how cholesterol is regulated in our body. We also find that younger, more motivated scientists are the ones behind most of our greatest advances. And surprisingly, most discoverers work at less-known institutions at the time of their discovery. This flips the common assumption about where discovery happens—not just at top-tier institutions. We then broaden the lens from individual discoveries to also study scientific fields in the next chapter. Chapter 5: How do new scientific fields emerge?Fields embody our greatest scientific advances and accumulated bodies of knowledge. Here we trace the origins of science's major fields including hundreds of fields spanning across science, offering a unique opportunity to uncover the common force driving them. We find that the speed of opening new research domains is mainly determined by how quickly we create new tools: particle detectors launched high-energy physics, microscopy techniques triggered neuroscience and randomised controlled trials kick-started experimental economics. The pace at which science expands is not random. For tools enable a completely new perspective to the world—and without them, the fields would not be possible. Our extraordinary development of new statistical techniques, x-ray devices, microscopes and spectroscopes has fuelled a wave of new disciplines: each enabling over ten new fields to emerge. In fact, about a quarter of fields are themselves the new method or tool—from laser physics and computer science to x-ray crystallography and econometrics. After uncovering how powerful tool innovations spur scientific advances, the key question we need to answer to close the logical circle is how we make tool innovations themselves—the topic we tackle in the next chapter. Chapter 6: How do we create new methods and tools—that spark new discoveries and fields? To answer this, we begin by tracing the remarkable stories behind science's greatest toolmakers: from Ernest Lawrence's particle accelerator to Theodor Svedberg's centrifuge. In this foundational chapter, we bring together some of the key pieces of evidence. We show that—without powerful tools to detect or measure— more funding and collaborations are not enough, our scientific theories cannot be meaningfully developed or tested, and unexpected or serendipitous observations are not possible. So what if we no longer wait for new discovery tools to emerge by chance? How many big breakthroughs are we missing because we have not yet begun prioritising tool innovation? Here we map out the crucial pathways to create new tools—the discovery engine. We lay out a map of scientific methods—explored and unexplored method combinations that can help uncover and predict untapped opportunities. We introduce the idea of setting up methods labs and hubs—as incubators of innovation—that catalyse tool creation. What emerges is a new general theory of discovery—one that can provide a foundation for a new field targeted to developing methods: the Methodology of Science. After tackling the question of the foundations of science in Chapters 1–6, we then take a bigger leap—and explore the deeper origins of how we started science in the next chapter. Chapter 7: What are the early origins of science and our methods? Is the engine of science today the same as it was in our past? We find that the key turning point in scientific history came with the unexpected invention of tools that enhance human perception: the first microscope and telescope. They opened unimaginable new worlds and sparked most major discoveries of the 17th century. They put science on a no-return path of rapid discovery, triggered wider curiosity and kick-started modern observational and experimental science. Before their invention, science was limited by ancient tools and human senses—but their arrival broke those barriers, giving us direct access to microscopic and cosmic realms we never imagined. These extraordinary instruments did not emerge from scientific institutions, a large scientific community or formal scientific method. Instead, they inspired them— after observing the extraordinary discoveries these tools made possible. But to understand how we got there, we go further back: how did we develop our ability to do science in the first place? We cannot study what triggered our advances throughout human history as systematically as we can study contemporary discoveries using statistics. But we can trace what has driven dozens of transformational early breakthroughs— from the agricultural revolution to the scientific revolution. So while we, throughout the book, examine scientific methods and tools systematically applied in research— such as computational methods and x-ray devices—in this and the next chapter, we broaden the scope to study more basic, general methods and tools that go far beyond the narrow scientific context. And we turn to how we developed these method-making abilities in the next chapter. Chapter 8: Can we best explain our success, throughout history, by our unique human capacity for method-making—enabling us to better meet our needs, solve problems and develop vast knowledge? Throughout history, we have continually refined our abilities, to invent navigation techniques using stars, apply plant-based medicines to heal wounds, devise lunar calendars to 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 to developing increasingly cumulative methods? 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. Our success is a remarkable story of realising the power of leveraging this capacity to conceive better tools that enable us to think, test and predict better. We are complex method-makers who extend our own minds. Bringing together the evidence throughout the book leads us to the question of how we can best develop a general theory of discovery—a question we explore in the next chapter. Chapter 9: Can we develop a general theory of discovery, backed by different types of evidence and methods across fields? We cannot design an experiment to measure what our tools cannot detect. In science, the tools we use largely determine how and what we can study—and the questions we can ask. Powerful new tools—from superresolution microscopes to the gene-editing method CRISPR—in turn unlock entirely new types of questions and discoveries that were not imagined before. Here we take a step back and look at the bigger picture, merging and expanding the insights from earlier chapters—with new perspectives. While the search for grand explanations of how science progresses has largely been abandoned since Kuhn's concept of paradigm shifts, we expand the general theory of scientific progress here: the new methodsdriven discovery theory. This theory is backed by diverse kinds of evidence from five different domains: statistical analysis of science's major discoveries, evolutionary biology, cognitive science, scientific practice, and philosophy of science . Integrating the five different explanations, this unifying theory offers a new understanding and roadmap for discovery. It is commonly about overcoming the limits of our methods, mind and senses with powerful new tools that outthink, outsee and outimagine our own minds. Our tools do not just assist science, they largely define it. This leads us to explore our scientific bounds in the next chapter. Chapter 10: What are the current limits of science—and what exactly shapes those limits? Can we keep pushing forward into the edges of the universe, atoms, genes and human societies? These are key unanswered questions, but the answers are often less mysterious than they seem. So here we unpack our present boundaries—after we tackle the foundations and origins of science in earlier chapters; and the three puzzles are deeply interconnected. Yet when studying our boundaries, researchers commonly explore what we are not yet able to explain—from dark matter in physics, to how life began in biology. But here we explain how our current scientific frontiers are shaped by five factors: our scientific methods and tools set the present limits of what we can observe, test and discover in most of science today, but our evolved mind shapes how we develop those methods—while the time and place we live in, our human perspective and our social context help shape what we study within these limits. Science's boundaries are not fixed, they are the shifting edges of a map we redraw with each major new tool. Still, there are many mysteries we cannot unlock with our current tools today: from the size of the universe and superstrings to how consciousness emerges. But there is a paradox: understanding what drives the current limits of science is the best way we have to expand and redefine these very limits—the topic we turn to in the next chapter. Chapter 11: How can we push our present limits of science—and make new advances that expand the frontier faster? Inventing tools with cutting-edge innovations— sharper resolving power, greater computational strength, more statistical power— continually break the boundaries at the frontier. From developing the gene-editing method CRISPR to rewrite genes, to creating quantum computers to solve problems we otherwise could never do, the power of our tools is enormous. So how do we actually push our limits and accelerate discoveries? To break new ground in any field, we generally first have to spot where our tools run into bottlenecks— and then tackle them. It is about amplifying our ability to see, measure and theorise about the world with enhanced lenses. Whether better fighting climate change with cutting-edge carbon-capture methods or better mapping the brain's complexity with more advanced neuroimaging and AI analysis, our progress depends on constantly pushing our tools' boundaries. Here we map out the steps we need to take to expand the top ten most influential discovery tools—from x-ray devices to statistical methods. Researchers best equipped at the edge of science are those who can best see the gaps in our tools and what we know—and test new ones to fill these gaps. Here we also explore deeper questions: are there fundamental limits to what we can discover in some domains? And what can the future of science look like—and how can we power it with new tools, including AI tools? As we go through the chapters of the book, we take a journey across the foundations, boundaries and future of science—by tracing science's major discoveries and turning points. Along the way, we uncover a powerful framework—built on new evidence and a new theory—for understanding discovery and methods. And this lays a foundation for two fields: the science of science and the methodology of science . Defining discovery, methods and science—and how we study them To do science and make breakthroughs, scientists develop and use methods and tools, gather and analyse data, provide explanations or a theory based on the results and, in extraordinary cases, uncover a major discovery—the hallmark of scientific progress. Here we instead go backwards: we begin with science's major discoveries and the discoverers behind them, gather and analyse data on how those discoveries are made and the methods and tools they used, to ultimately develop a theory of discovery based on the common patterns across them. How do we study and uncover something as complex as scientific progress? Just as there is no one method in biology or physics to tackle major questions, there is also no one method to study science itself. How science and discovery are currently explained can be grouped into five main approaches: explanations by historians exploring rich case studies of individual discoveries, scientometricians tracking highimpact articles and how often they are cited, psychologists examining discoveries indirectly with experiments on how scientists think, computer scientists modelling the discovery process and philosophers exploring the concept of what discovery means. Scientometricians, computer scientists and psychologists analyse quantitative data, while historians and philosophers offer in-depth, narrative insight. To unpack how discovery works, we combine the different types of methods and evidence across different fields into a coherent framework. As the backbone of the analysis, we statistically study science's over 750 biggest discoveries—the method we use most in the book; while we also explore the personal motivations behind the discoverers themselves (Chapters 1–6). We draw on qualitative methods and insights from cognitive science and evolutionary biology, and from the history and philosophy of science (Chapters 7–9). We also analyse science's top methods and tools themselves—from lasers to spectrometers (Chapters 10–11). In short, we turn the methods of science onto science itself: applying scientific tools and techniques to study science and its best methods. That means combining large-scale data analysis to uncover the common patterns across discoveries, rich case studies to explain the biggest discoveries in detail and conceptual analysis to understand scientific concepts deeper. Integrating these mixed methods, we gain a deeper and broader understanding. So to uncover what drives breakthroughs, we need an integrated approach that overcomes the shortcomings of relying on a single method. Because the current explanations of discovery are built around the particular method that researchers use. The method applied does not just guide research, it shapes—and constrains—the very questions researchers are able to ask and the answers they are likely to find. If we use the common method of scientometricians, or philosophers of science, or historians of science and so on, then we can only look at problems that large-scale data on citations, or conceptual analysis, or individual case studies can address. Take scientometricians who show for example that the most impactful scientific work appears randomly throughout a scientist's career—a finding based on big data analysis of article citations. But we are left asking how scientists triggered that high-impact research. Take philosophers like Karl Popper who argue that there is no logic behind how discoveries arise—an idea based on conceptual reasoning. But we are left without an analysis of discoveries to test that claim. Take historians like Thomas Kuhn who popularised the notion of paradigm shifts shaping science—an insight drawn from few case studies of famous theories. But we are left with a focus on select theories that does not allow us to spot broader patterns—or generalise about science, but that is precisely what Kuhn still did. This sparked an enormous debate on what actually drives scientific progress that persists today. What is clear is that discovery is far too complex to study or understand using just one method and exploring just one or a few factors—and it is the reason why it is still a puzzle. Here we do not start with a method and then explore what questions it can tackle. Instead, we begin with the big question: what actually sparks science's major discoveries? And we let that question guide which methods we need. So what is the best way to tackle this question? To answer it, we take the most direct route possible: we open up science's over 750 biggest discoveries across fields and history and—one by one—collect and analyse detailed data for each discovery. This means studying how each breakthrough was made, what methods and tools they used, what the traits of the discoverers are and what kind of broader environment they worked in. This enables us to spot the common thread driving them. The aim has been to provide the most systematic, comprehensive and in-depth analysis of discoveries to date. So instead of just focusing on a discovery in detail, or tracking high-impact papers from high altitude, we merge the different approaches. Throughout the book, we zoom in and zoom out—like turning a microscope on the discovery process itself. What is remarkable is that the broad statistical patterns from two independent sources (all nobel-prize and major non-nobel discoveries)—and the more detailed findings from different methods across fields—are deeply consistent with each other. This kind of cross-disciplinary convergence—that the evidence consistently points in the same direction—is rare in studies on discovery. And it reveals something fundamental and consistent about the hidden patterns driving discovery. It is this kind of broad, evidence-rich approach that enables a bigger picture of scientific progress. How do we define major discoveries?The structure of DNA, the human genome, the periodic table of elements, gravitational waves, relativity theory and CRISPR gene editing reflect groundbreaking discoveries. These are common examples of major discoveries. A major discovery is defined as a new experimental, methodological or theoretical breakthrough that marks an entirely new way to understand the world, opens new paths of inquiry and later has a proven, lasting impact on science. Here we study science's major discoveries—spanning all nobel-prize and major non-nobel discoveries—that form the foundation of the major fields across the natural and social sciences (as we explore in depth in the next chapter). For both scientists and the general public, these discoveries are often the most exciting aspect of science. Science in general is often defined as the study of the 'world through observation, experimentation, and the testing of theories' (Oxford English Dictionary). But as we go through the evidence in the coming chapters, we will develop a broader understanding that gets at the core of what science is. What are scientific methods and tools? Scientific methods are systematic techniques—such as controlled experimental, statistical and computational methods; scientific tools are systematic instruments—such as telescopes, x-ray devices and electron microscopes; both are used to study the world and extend our ability to observe, measure and analyse, and are general-purpose. This means they can be applied to different questions and often domains. In practice, researchers often use the words method and tool interchangeably—and we do too for simplicity. What matters is what they do: both give us better ways to see and think. Together, methods and tools make up what we call our scientific toolbox. (To be clear, we are not referring here to cognitive abilities like observation and hypothesising, or theoretical frameworks.) The origins: the book's methods-driven and interdisciplinary approach As the book offers a new way to tackle this fundamental puzzle, it is worth briefly explaining how this approach to studying science and discovery came about, before we begin. After finishing my PhD, I worked on applied research projects with governments and the World Bank for five years. It was an exciting mix of research and policy, working on complex, real-world challenges and tackling them by thinking across disciplines. Whether we worked on public health, energy or agriculture, we constantly had to integrate different methods—and pull together perspectives on the environment, human behaviour and social systems to find solutions that work. That applied, hands-on research led me to two key insights that lay the foundation of this book. First, I realised that the specific method we use shapes what we see. If we apply big datasets and statistical models, our answers are different from if we run controlled experiments, if we employ mathematical methods or if we apply rich, in-depth qualitative methods. The method we use is not a neutral lens—and gives different kinds of insights. Second, I realised how adopting one disciplinary perspective also does not take us very far in understanding and tackling a complex challenge—whether malnutrition, climate adaptation or fostering innovation. In the real world, complex challenges always cut across disciplinary boundaries—and the only way to tackle them is by crossing those boundaries and combining multiple methods. There is no way around it—and this is clear doing applied research. But in academia, research generally stays siloed, shaped by researchers' training in a specific method and specialised in a field. Time, resources and institutional incentives also reinforce that narrow focus. So when I returned to academic research, I brought this big-picture mindset with me—but this time to study science itself. I realised there is a pattern running through breakthroughs I came across in public health, economics and medicine: each breakthrough seemed to rely on using new method innovations. Developing randomised controlled trials, new statistical methods or novel 'natural experiment' techniques revealed completely new kinds of findings. I began to wonder, could there be a deeper principle at play—that new methods drive new breakthroughs across science? I set out to test it. To my surprise, no scientific study has yet explored the role of new methods and tools in discovery across science, and no study has yet explored the wide range of factors—at the same time—that may play a role in the discovery process. That big-picture approach was lacking. Over fifteen months, I did extensive full-time research to compile the comprehensive data on science's major discoveries across fields. I read each of the over 750 discovery-making publications to meticulously track down and extract the data. For each of these breakthroughs, I compiled detailed data ranging from the discoverers' age, their education level, their gender and their university at the time of the discovery, to the methods and tools they used, if their discovery was serendipitous, and much more. In total, the study covers 32 different variables for each discovery— and over 20,000 individual data points were collected and verified, one by one. This time-intensive research lays the backbone of this book. It involved building a dataset from the bottom up—detailed to zoom in on any one major discovery, and broad to zoom out across science's major discoveries, enabling rich analysis at different levels. It allows asking and answering new kinds of questions about the role of each of these factors—and crucially, at the same time. At the core: what common driver links across science's major discoveries—and what influences only certain breakthroughs? What patterns show up when we look across centuries, not just decades; or across scientific fields, not just some; or across methods and tools, not just citations or theories, and so on? It offers a new, broader understanding of how discoveries emerge—that earlier research studying one or a few factors has overlooked. The book is the result of this work that developed over many years—the biggest and most demanding project I have ever taken on. The motivation and input for this book is shaped by these experiences. The different methods I learned working in these different domains enables tackling the complex puzzle from multiple angles. Much of this research has been published in peer-reviewed scientific journals (under a creative commons licence), and this book brings the findings together—and goes far beyond them—by connecting the dots into one big picture. Another book of mine, Science of Science , qualitatively studies the foundations of science across 14 different fields—from psychology and philosophy of science, to computer science and biology, sociology and economics of science. Before we get started, let us take a final step back. In a nutshell, we will explore how the story of discovery is not unpredictable and mysterious—and not what most people think: science's major discoveries are not commonly driven by greater funding, larger teams, bigger theories or serendipity. When we examine science's over 750 major discoveries systematically, a different and clear pattern emerges: new tools and methods are the most overlooked factor and commonly the central driver of discovery. What is surprising is that, so far, our methods and tools have been largely created in an unplanned and unstructured way. But if we want to speed up breakthroughs, we have to rethink how science is structured and organised—and shift how universities, journals, scientific awards and funding bodies incentivise science. In Part I of the book, we take on some of the biggest questions about science: how do we drive new discoveries, scientific fields and science in general? To answer these questions, we have to trace science's major discoveries themselves, map the methods and tools they used, and explore the broad range of traits of the discoverers behind them—spotting the common patterns and insights along the way. Understanding what drives scientific advances will then allow us to tackle other fundamental puzzles about science: what are the deeper origins of science—and what are its current limits and how can we push beyond them? These are the challenges we turn to in Parts II and III .