disciplinary areas What are the most transformative tools across different disciplinary areas? New research domains cluster around key methods that trigger them. The methods and tools that acted as catalysts in opening most fields within physics-related disciplines are new mathematical techniques, lasers, spectroscopes and cathode-ray oscillographs. Within chemistry-related fields, these groundbreaking methods and tools are new spectroscopic methods, x-ray methods and mathematical techniques. Within medicineand biology-related fields, these are new optical microscopes and other vision-enhancing tools like x-ray devices, spectroscopes and electron microscopes. Within economicsand social science-related fields, these are new statistical methods and economic modelling methods (as mapped out in Figure 5.3). In biology for example, without the microscope—our classic instrument—several fields could not have emerged, from microanatomy that studies the structure of tissues and organs, to bacteriology that investigates bacteria and their links to disease. The historical pattern is striking: before we change the way we explore and view the world, we first change the tools we design to study the world. Because we can observe these changes in the rise of new disciplinary branches before and after the invention of the key tools, new disciplines do not emerge randomly (Appendix Figure 5.2). Microelectronics Mass Spectrometry Fiber Optics Laser Spectroscopy DNA Sequencing Nanotechnology Bioenergetics Metrology Molecular Theory Cryo-Electron Microscopy Modern Surface Chemistry Molecular Spectroscopy Protein NMRSpectroscopy Proteomics Spectroscopy Materials Science ModernStereochemistry Molecular Biology X-Ray Spectroscopy Protein Engineering Quantum Electronics Quantum Optics Neurophysiology Paleogenomics Bacteriology ClimateChange Economics Industrial Organization Empirical Finance Economic Forecasting FinancialEconomics International Economic Policy International Macroeconomics Modern Economic Geography Modern Economics cathode ray tube microscope DNA sequencing x-ray method statistics spectroscope microscope, electron Cellular Pathology economic modelling statisticsEconometrics DevelopmentEconomics Corporate Finance Behavioral Finance Cytogenetics Maleriology Microanatomy Microbiology Microscopy Modern Botany, Modern Zoology Neuroscience Modern Cell Biology MRI Evolutionary Genetics RadiationGenetics Ultrafast OpticalScience(Quantum) CondensedMatter Physics Biophysics Complex Systems Condensed Matter Physics Dynamical Systems Theoretical Astrophysics Thermodynamics Exoplanetary Science Neutron Spectroscopy X-Ray Crystallography cathode ray tubex-ray method spectroscope mathematics laser/maser Physics-related fields Chemistry-related fields Medicine-and biology-related fields Economics-and social science-related fields cathode ray tube x-ray method spectroscope mathematics laser/maser electrophoresis DNA sequencing X-ray method laser/maser mathematics spectroscope electrophoresisDNA sequencing X-ray method laser/maser mathematics spectroscope microscope, electron cathode ray tube microscope DNA sequencing x-ray method economic modelling statistics microscope, electronQuantum Computing Figure 5.3 Mapping the central methods and tools that open new fields—a network analysis The data reflect the central methods and tools we used in kick-starting established fields since 1600—reflecting 18, 16, 15 and 13 fields ( from the top, left to right). Each method or tool was applied in developing at least three fields within any disciplinary areas. Take the fascinating field of exoplanetary science. It launched in 1995 with the groundbreaking study published in Nature— A Jupiter-mass companion to a solartype star . The field was made possible by a new echelle spectrograph, invented in 1993. New tools expose entirely new realms of exploration, they are indispensable for shedding new light on the known and revealing the unknown. It is those who T conceive and refine these tools who amplify our research scope and fields by tackling the bottlenecks of our methods and mind. But new tools do not just give birth to fields; for over a quarter of them, they are the defining feature of the discipline itself—and not what we study using it. Fields like electron microscopy (1933), x-ray crystallography (1913), mass spectrometry (1919) and neutron spectroscopy (1955) emerged with the tool's invention. Such methodological fields are often inherently interdisciplinary as we leverage these tools across the broad domains of chemistry, biology, medicine and physics—with each tool earning a Nobel prize. Reinvention and fusion are also important: dozens of fields have emerged when we merge cutting-edge tools from different domains together, like computational chemistry, quantum interferometry and statistical mechanics. Most major method discoveries—supported at times by interdisciplinary work—establish new fields While new fields are consistently driven by method innovations, not every method innovation fuels a new field. This raises a crucial question: why do some major method discoveries establish new fields while others do not? To answer this, we first examine the extent to which major method discoveries (all nobel-prize-winning and major non-nobel method discoveries) trigger new fields. An extraordinary pattern emerges: these major new methods and tools reflect about one in four major discoveries in science, but among them, a remarkable 82% opened a new field. These include field-triggering tools like laser cooling launched in 1985 by Steven Chu at Bell Labs; DNA amplification pioneered in 1985 by Kary Mullis at Cetus Corporation; and neutron spectroscopy created in 1955 by Bertram Brockhouse at the Atomic Energy of Canada. This striking finding—that major method discoveries are much more likely to lead to new fields than not—holds true across time and disciplinary areas (Appendix Figure 5.4). But can broader demographics, institutions and geographic location support new fields arising? To test this, we compare a control group of major method discoveries that did not establish new fields with those that did. This enables us to examine the differences, between the two groups, in factors that can support disciplines arising. Among method discoverers, a striking 65% of those who have triggered new fields have worked interdisciplinarily, while this figure drops to just 39% for those who have not sparked a field (Figure 5.4a). But it is not just about combining two scientific communities through new collaboration networks; rather, new fields are more likely to emerge when we fuse methods across disciplines—either integrating methodological approaches from different domains or applying methods in completely new domains. Other factors, like discoverers' level of education, gender and age, show little to no difference between establishing new fields and not establishing them. These factors seem less important behind fields emerging. Method discoverers who launched new fields were more likely to work at a top 50 university and be based in North America, but these factors are not statistically significant when controlling for the range of demographic factors. 0.55 0.65 0.66 0.24 0.39 0.55 0.34 0.08 0.77 0.48 0.39 0.70 0.26 0.35 0.26 0.00 0.83 0 .2 .4 .6 .8 Percentage Established field No field Major method discoveries since 1500 leading to a field or not Discoverer has interdisciplinary degrees Education: Discoverer is professor Education: Discoverer has PhD only Discoverer worked interdisciplinary Discoverer at top 50 university Discoverer lived in North America Discoverer aged 18–34 Discoverer(s) female (at least one) Discovery earned Nobel prize 0.65 0.60 0.71 0.25 0.44 0.67 0.29 0.06 1.00 0.47 0.37 0.79 0.21 0.32 0.42 0.26 0.00 1.00 0 .2 .4 .6 .8 1 Percentage Established field No field Contemporary science: nobel-prize method discoveries leading to a field or not Discoverer has interdisciplinary degrees Education: Discoverer is professor Education: Discoverer has PhD only Discoverer worked interdisciplinary Discoverer at top 50 university Discoverer lived in North America Discoverer aged 18–34 Discoverer(s) female (at least one) Discovery earned Nobel prize 0.26 Figure 5.4 Major method discoveries leading to new established fields compared to those that did not, by features at the time the discovery/field emerged The data reflect a total of 85 major method discoveries since 1500, with 62 leading to established fields and 23 not leading to a new field (Figure a). And the data represent a total of 67 nobel-prize-winning method discoveries, with 48 and 19 discoveries in the two groups (Figure b). So we next explore what predicts whether major method discoveries establish new fields or do not. To do this, we use logistic regression to analyse the demographics, institutions and geographic location of these discoverers (as independent variables). A method discoverer working interdisciplinarily is the only significant predictor of T a new field emerging, while controlling for these other factors—and considering the small sample of less than a hundred major method discoveries (Appendix Figure 5.3). In other words, when we step beyond the boundaries of our own field and blend methods, we increase the odds of breaking new ground and domains. Box 5.2 How Archer Martin and Richard Synge developed simple chromatography methods that transformed chemistry and kick-started new fields within chemistry and biology In chemistry, the first step is to isolate a substance from natural materials like plants or animals. The next step is to identify the substance and determine its composition. To achieve this, the Russian Mikhail Tsvet pioneered the method of chromatography in 1906, yet it was not until the Austrian Richard Kuhn refined the method in 1931 that it began spreading through the scientific community. Kuhn, who received his PhD in chemistry at just 22, used chromatography to discover a new type of carotene (a vital component of vitamin A) and provide new insights into vitamins B2 and B6—research that earned him the Nobel prize. At this time, Archer Martin was studying biochemistry at Cambridge. In 1938, Martin began working at Wool Industries Research Association and Richard Synge joined as a research student the following year. The two British biochemists soon began working together on wool felt and its amino acid composition. Synge's studentship received funds from the wool industry thanks to Hedley Marston, who advised him to study the amino acid makeup of wool and to start by improving the methods for analysing amino acids: 'If you work steadily at that for five years, you will revolutionize the whole of protein chemistry' . This challenge was a source of inspiration for developing a new method, along with a technique he came across in a key study. He described this technique as 'countercurrent fractional extraction [and] until then it had not occurred to me that an extraction column could be used to separate two substances of rather similar partition coefficient' . His realisation underscores a crucial lesson: scientific progress often hinges on being aware of available methods to tap their potential. Synge did not need five years—he achieved his goal in just three. In 1941, he completed his PhD and published the landmark study with Martin on the powerful new method of partition chromatography, paving the way for the explosive surge of chromatography that transformed chemical analysis. Creating partition (paper) chromatography—with Synge at just 27 and Martin at 31—earned them the Nobel. The method is straightforward: substances— like amino acids—are separated in a mixture to establish what they are made of. The process itself is surprisingly simple: a single drop of the chemical substance is placed on a strip of paper. The paper is then soaked in a solvent (like chloroform or alcohol) and the different components begin to spread out, forming distinct coloured marks that reveal the mixture's components. That is it. This simple 132 T HE ENGINE OF SCIENTIFIC DISCOVERY method requires just water, filter paper and a solvent—all very cheap lab supplies (Picture 5.2). As Martin himself remarked in his Nobel speech: ' All of the ideas are simple and had peoples' minds been directed that way the method would have flourished perhaps a century earlier' . Any chemist can use the method—no expensive equipment or research teams required. The powerful method enables us to separate proteins, amino acids, carbohydrates and sugars. It has led to many new medical treatments and biological advances that have benefited our lives, laying also the foundation for new fields like bioenergetics and signal transduction. Picture 5.2 Paper chromatography method. Reproduced from light the minds. Before closing, think of major fields that have stagnated and those that have recently grown rapidly. Now think of how tool development relates to these differences. It is applied fields across science—like experimental physics, economics and biology— that are largely thriving. In contrast, theoretical fields—like theoretical physics, economics and biology—have largely stagnated, at times locked in long-standing debates (Chapter 10). Applied research thrives because of new, frontier-opening tools, methods and data they produce. Think of the James Webb space telescope exploring the early universe, the LIGO interferometer detecting gravitational waves, and sequencing and high-throughput tools decoding the entire human genome. Can method innovation explain this vast divide? Indeed, the main driver of a field's growth or stagnation is commonly the power and novelty of its tools and methods. The fields that expand most rapidly commonly apply the most powerful new tools—think of the field of AI that is driven by new machine learning methods, to genomics powered by DNA sequencing methods, and genetic engineering driven by the CRISPR gene-editing method. So when do fields actually grow fastest? We find here four pathways: one, through such method inventions that allow asking questions not possible before. Two, through upgrading—or three, integrating—computational, statistical and experimental tools that enable researchers to analyse such previously intractable problems with massive, complex data. Four, through cross-disciplinary borrowing that can involve fields also combining tools—like neuroscience merging tools from biology, computing and cognitive science, and behavioural economics mixing methods from psychology and economics. And when do fields stagnate? Not because we run out of ideas or papers, but because we run out of ways to explore, test and generate new findings—using new methods. The key insight is simple: fields that begin treating tool-building and method-design as central—not auxiliary—will be the ones generally unlocking the new frontiers of research. T
Scientists like Galileo, Newton, Mendeleev, Hooke and Mendel were pioneers in testing new methods and evidence. Yet they could not foresee whether and how their individual contributions would fit the construction of an immense system of knowledge. What they were contributing—eventually leading to the fields of physics, chemistry and biology—became clearer over the centuries. Today, we have a far clearer view of the evolving edifice of science and its ever-expanding structure and complexity. While we are not fully aware of the immensity of what is beyond our planet, we have amassed vast bodies of complex knowledge, from laser physics and genetics to climate science and AI, that were incomprehensible just a few generations ago. A key goal of the science of science is to understand how these bodies of knowledge grow—from the 17th century to the cutting-edge developments of today. Because each publication that opened a new field used a new tool—and the study, including often experiments, could only be conducted with that tool—we find that new fields consistently emerge through new tools. Because they enable novel insights and testing those insights that would not have been possible before. Through the new tools we design, we can trace what new fields fundamentally rely on: new microscopes leading to microbiology, computers launching computer science and radio telescopes giving rise to radio astronomy. For these necessary tools allow us to observe what is otherwise too small, too vast, too fast, too far beyond our mind's capacities to imagine. The history of science is ultimately a history of expanding our human senses—crafting new ways to detect, measure and understand our world that open new domains of knowledge. Here we identify the fundamental principle that fields share in common: we consistently kick-start new fields through newly invented methods and instruments that reflect a new way to perceive the world not previously feasible (Figure5.1). Yet nearly all scientists commonly just use conventional methods they are trained in to study a problem. We also uncovered that most major method discoveries do not just launch a new field but often multiple new fields. Science's most powerful tools are rarely confined to their discipline of origin; instead, most spill over into other disciplinary areas to unexpectedly trigger new fields that their inventors never anticipated. This underscores the causal link of new fields driven by new tools that would not have been possible without them. This new methods-to-fields principle holds across history and disciplinary areas and helps redefine the predictability of new fields emerging after and where we make new method advances. Unlike traditional explanations—including paradigm shifts in theories, evolving research programmes or splitting or merging scientific communities—this principle does not focus on just outputs but on what precedes and causes new fields in the first place. Beyond new tools, there are supporting factors that can help influence when a field emerges—like funding, collaborations, and developments in other fields—but they cannot directly start a field on their own. This tool-driven principle highlights the need for us to redirect much greater attention to upgrading our toolbox. By prioritising the development of new methods and technologies, we can create an environment for accelerating new breakthroughs and fields. Ultimately, the origin of a major breakthrough is the central event, and whether the breakthrough takes place in an established field or creates a new field is often less important. Expanding our tools is where the frontier of research lies, yet we do not give enough attention to this frontier research. A guiding principle for researchers seeking to break new ground is: when we hit upon an interesting problem facing a method or tool we are using or come across an idea of how to tackle that problem, we should drop everything and pursue it—because history shows that this is generally how new scientific advances and fields are born. The most promising thing to hear in science is often not just 'I have a new idea' but rather 'I have a new method that I can apply' . In the final chapter, we sketch out the constraints to our best tools—the bottlenecks that, if we overcome, would vastly advance science. Shifting our research focus to this powerful principle of tool innovation would speed up how we spur new advances and the pace at which we can enter unmapped terrain at the borders of science. Such a shift would mark a tool revolution in science itself. But this raises key questions: if we become better at developing new methods, could we better anticipate and predict tomorrow's new scientific advances and domains? If we learn to recognise the signs of methodological bottlenecks and methodological leaps before their impact unfolds, could we actively create and shape the future of science? And crucially, what concrete steps do we need to take to design our next big methods and tools? We tackle these key questions in the next and foundational chapter. But one thing is clear: tomorrow's great scientists—those best equipped to tackle society's pressing challenges—are generally those who are best at developing new tools or who take advantage of the new tools.