by five different perspectives Can we reduce how we make discoveries and fields to a universal theory that we can test with evidence? Can we describe the discovery process in a way that holds up to evidence across fields? Equations like Einstein's E = mc 2 and Newton's F = ma formalise knowledge in fields like physics—but can the method-driven process of how we spark breakthroughs also be better formalised? A good scientific explanation generally needs to combine solid evidence with a theory explaining 'how' (the causal mechanism). By examining science's major discoveries and fields, a remarkable pattern emerges: new methods and tools area necessary condition and commonly the key driver for how science's major discoveries and fields emerge. Other factors like funding, teamwork, institutions, greater education or serendipity support breakthroughs but are not always necessary—they vary by discovery and by context. And theories cannot be meaningfully developed or tested without building on methods (mathematical, experimental). What is at the core of progress—its common thread: the development of new tools to see and measure the world through new lenses (Chapters 1–6). Here, we combine that evidence across science with this theoretical explanation. This new methods-driven discovery theory is straightforward: we unlock science's major discoveries and fields by using current scientific methods and tools and developing a new method or tool that enables us to observe, solve problems, experiment and theorise and collect new evidence of theworld—in ways that are impossible without the new method or tool (Figure 9.4).¹ The discovery process can be reduced to these methodological features. The key is inventing a new method or tool that generally makes it possible to see, test and think in ways we could not before (Chapters1–6). While disciplines vary widely in what they study—from chemistry to neuroscience and astronomy—they are consistently powered by the same engine of how they study and advance: new methods and tools. Take the nobel-prize-winning discovery of cell structure. This remarkable breakthrough was triggered using a new electron microscope to uncover cellular structures no one had seen before—while experimenting with mice and using other methods such as statistics and a centrifuge. This optical innovation transformed how we understand life at the cellular level and opened modern cell biology. Many think scientific progress is mainly driven by new findings and theories. But digging deeper, we find that creating and applying new methods and tools are commonly what drive those new breakthrough findings and theories—by reshaping how we perceive, experiment, imagine and theorise in ways not possible before (Chapters 1–2). This is not just a theory about tools of discovery—it is a theory about science's central driver of perception, experimentation, imagination and theorising. This theory ¹ We can express the theory as: (M and T + nM or nT) + (O, P, E, Th) + W(1, 2, 3…) = D T explains how we make science's major discoveries and fields—and we can express it simply as: ++ = Evidence of the world (observations or experiment findings previously inaccessible without the new tool) New discoveries and fields We develop tools – or adopt them from other fields – to observe, solve problems experiment and theorise Current methods and tools New method or tool Broad factors, such as existing research, basic funding and the scientific community, support developing tools and expanding science Figure 9.4 The new methods-driven discovery theory This general theory links scientific progress to the current gaps in our tools. We can use the theory as a guide to spur future breakthroughs by tackling current method bottlenecks and extending our discovery tools. Our big unsolved challenges—curing neurodegenerative disease, understanding our brain's complexity, halting climate change, etc.—stall when our current tools cannot see or test deeply enough. In short: bigger new methods are more likely to produce bigger, more transformative discoveries, while relying on smaller conventional methods—ones we are familiar with—are more likely to produce smaller conventional knowledge.
Imagine if a method and tool revolution transformed the future of science as dramatically as six game-changing tools did for the 17th-century scientific revolution (Chapter 7). What if scientific labs began competing to invent tools of discovery, not just to apply them? This can spark a tool revolution across scientific fields—but also across science policy, funding priorities and even the way we think about science and discovery itself. Imagine a world where research labs invest as much time in making method innovations as they do in running experiments and collecting and analysing data. Imagine global open-source platforms where scientists prototype new research tools across disciplines—from more advanced real-time brain imaging in neuroscience, to ultra-efficient carbon capture methods in climate science, to designing materials in engineering with more advanced AI tools. Or imagine even AI tools that are built as much for designing new experimental methods as for running existing methods.Such a transformation would fundamentally change our focus of the eureka moment: from just single discoveries to equally celebrating the moment we invent powerful new tools that often unlock many more discoveries . 256 T HE ENGINE OF SCIENTIFIC DISCOVERY An awareness within the scientific community of this new tools-driven discovery principle would have the power to drive a method revolution in science—a fundamental change in mindset among scientists. It means researchers would recognise that developing new methods and tools is how we spark new advances. If we shift our focus to extending what our methods and tools can do, we introduce a powerful new way to address science's biggest problems—from better tackling cancer to better protecting our ecosystem. A method revolution would fundamentally place innovation in tools at the centre of the scientific and discovery process. Until now, we have made our tools in an ad-hoc, scattered way—without recognising that the driving force of science is creating and upgrading the very tools we use. A method revolution would be fuelled by the scientific community—including scientists, universities, journals, funders—deliberately prioritising the design of new methods in a strategic, targeted way to accelerate the pace of discoveries. We would start seeing the foundation of science not as our body of knowledge but as the remarkable toolbox it is that we invent and continually refine to develop new breakthroughs. If we want to better understand the world and science, we first have to understand—and improve—the ways we study the world and do science. When we think of science, what is the first thing that comes to mind? Most think of our bodies of knowledge: laws of physics, biological mechanisms, chemical elements. But is it not just as important to understand the way we generate that knowledge— to understand our tools and how we develop them? The tools we have invented up to now largely set the scope of what we can explore and what we can know. There is an inseparable interplay between our available tools and the discoveries we can make. Before we can make a breakthrough, we first have to come up with a way to expand how we study the world. Science's enormous success is largely explained by upgrading our toolbox over and over—from massive particle accelerators discovering fundamental particles to CRISPR rewriting genes. The success of our best methods in tackling our large challenges in science, technology and society in the past highlights the power of deliberately extending the tools we have today. A methodological shift is also needed in how we conceive science itself. We should no longer divide scientific research into two categories, experiment and theory—or experimental scientists and theoretical scientists. Methods and tools are what make both developing experiments and theories possible in the first place. They determine our experiments (how we design, run and assess them) and govern our theories (how we formulate, test, refine and validate them) (Chapter 6). That is why we introduce here a third essential dimension to science: method. Without methods and tools, science is missing a central pillar. Recognising this method-experiment-theory distinction gives us a much clearer, more realistic picture of the actual nature of science and scientific progress.
The world around us consists of astonishingly complex mysteries—from galaxies colliding across the cosmos to genes shaping life itself—that we can only uncover and T understand using the tools we have invented. Recognising that new methods and tools are the engine of discovery offers us a new way to understand and advance science: by redirecting our attention to extending our extraordinary toolbox. What emerges is a new methods-powered explanation of the foundations of science, and a new general theory that explains how we develop breakthroughs by innovating and re-innovating new tools. This is the core of the new methods-driven discovery theory: inventing better ways to look for and find answers. It explains how our ability to access the world is constrained by the available tools we have designed so far. Simply put: we cannot do science in ways other than by using our available methods and tools. New tools shape the very direction science takes. From quantum computing that opens up an entirely new way to think about information itself, to the gene-editing method CRISPR that makes biology programmable. Expanding the research frontier is about breaking through the bottlenecks of what our current tools and mind can handle—it is about innovative toolmaking. Science is, in many ways, about amplifying how we see, experiment, infer and imagine with cutting-edge spectrometers, telescopes and supercomputers. This methods-led principle of science is backed by multiple independent strands of evidence and methods—as we laid out here. New methods are what enable us to revise our biggest questions, answers and theories under new light—and develop entirely new ones. If we want better answers, we have to change the way we approach the questions—and that generally demands new ways to explore them. Take the ten most powerful tools of discovery that have all been continually upgraded (Table 6.1). The great method-makers who built these groundbreaking innovations have arguably had more impact on scientific breakthroughs than other scientists. For we have implemented each of these tools to catalyse multiple discoveries (Chapter 1). Yet most people have likely not ever heard of these method discoverers—the most impactful researchers in history. But all are likely familiar with discoveries made possible by harnessing these great tools: from the discovery of DNA's double-helix structure (using x-ray crystallography) to the discovery of cell structure (using electron microscopes). These tool innovators are the forgotten heroes of science. We need to rewrite science textbooks to make them the key protagonists of science. After all, our best methods and tools may well be the greatest and most efficient things humans have ever created to understand and shape the world. The future of science depends largely on the toolmakers and method-makers who give us new ways to see, explore and imagine. If we all became aware of the power of toolmaking on discovery-making, it could spark a method revolution in how we do science. If we taught young scientists to see toolmaking as an essential part of discovery—as important as finding answers. If institutions and funding agencies prioritised method innovation as much as chasing big results. And ultimately, if we take the needed steps and time to identify our own method constraints and contribute to expanding science's toolbox—to unlock new domains we cannot imagine today. But this requires us to reshift our attention from the shine of discovery to the engine that powers discoveries—so that discovery tools begin to share the spotlight.
P A RT I I I TH E PR E SEN T LIMIT S A N D FU T UR E OF S CI ENCE A N D DIS CO V ER Y In Part I of the book, we asked big questions: how do we drive new discoveries, scientific fields and ultimately science? The answer: by inventing powerful new methods and tools that make new ways to see, measure and experiment possible. In Part II, we then tackled the question about the deeper origins of science: how did we start science in the first place? The answer: by using our mind's evolved methodological abilities that we have developed into ever more sophisticated methods and tools. These enabled us to develop scientific and technological knowledge—and eventually give rise to the innovative techniques that sparked the agricultural, scientific, industrial and digital revolutions. After mapping out the foundations and origins of science, other fundamental questions at the frontier of science arise: what are the current limits of science? How can we break through these limits faster? Can expanding our toolbox play as powerful a role in explaining the future of science as it does the past and present? In Part III, we turn to these questions—some of the most important questions we tackle in the book, because the answers will shape what science can do next. Pulling the three parts of the book together, we gain a new picture of science—how our toolbox is at the foundations, origins and limits of science. Strikingly, we uncover how the foundations and boundaries of science are primarily the foundations and boundaries of the evolving methods and tools we create. We now combine these parts in the last two chapters.