Scientific research and technological advancement have gone hand-in-hand since the invention of the wheel. Without research, we lack the knowledge base to advance the state of technology and, without technological advancement we lack the functional base for further scientific exploration. In their new book, The Genesis of Technoscientific Revolutions, Harvard University Professor of Technology and Public Policy, Venkatesh Narayanamurti, and Sandia National Laboratories Senior Scientist, Jeffrey Y. Tsao, explore the symbiotic relationship between these two concepts and how their interaction might be modulated to better serve the rapidly accelerating pace of 21st century technoscientific discovery.
Excerpted from THE GENESIS OF TECHNOSCIENTIFIC REVOLUTIONS: RETHINKING THE NATURE AND NURTURE OF RESEARCH by VENKATESH NARAYANAMURTI AND JEFFREY Y. TSAO, published by Harvard University Press. Copyright © 2021 by the President and Fellows of Harvard College. Used by permission. All rights reserved.
The Network Is Hierarchical: The Nesting of Questions and Answers
The way in which scientific and technological knowledge are hierarchical stems from the nesting discussed in the last chapter, both of scientific facts and explanations and of technological functions and the forms that fulfill them.
In science, at the top of the hierarchy are facts — raw patterns in observed phenomena. These patterns can be thought of as questions: Why does a particular pattern occur? Why when one releases a ball does the ball fall and fall faster the farther it has fallen? Explanations of those raw patterns come a level below in the hierarchy, and can be thought of as answers to those questions: Galileo’s sixteenth-century explanation of the observed distance-versus-time pattern was that the velocities of falling balls increase linearly with time. But this answer, or explanation, becomes itself another question: Why do the velocities of falling balls increase linearly with time? This question begs a deeper explanation, a deeper answer: Newton’s explanation was that gravity is a force, that uniform forces cause uniform acceleration, and that uniform acceleration causes linear increases in velocity. Scientific understanding is always incomplete, of course, so there is always a point at which we have no deeper explanation. This in no way detracts from the power of the explanations that do exist: science seeks proximate whys but does not insist on ultimate whys. The general theory of relativity explains Newton’s laws of gravity, even if its own origin is yet to be explained.
In technology, at the top of the hierarchy are human-desired functions. These functions present problems that are solved by forms below them in the hierarchy. Forms fulfill functions, but those forms present new problems that must be solved at successively deeper levels. Shifting from the problem-solution nomenclature to the equivalent question-answer nomenclature, we can say that the iPhone represented a technological question: How do we create an Internet-capable cellular phone with a software-programmable interactive display? A partial answer came in the form of multi touch capacitive surfaces, opening up a significant design space for user interaction when multiple fingers are used simultaneously. But the opaqueness of existing multitouch surfaces itself became a question: How do we make multi touch surfaces transparent so that the display is visible? The multi touch transparent surface display provided an answer.
In other words, science and technology are both organized into hierarchies of question-and-answer pairs, with any question or answer having two “faces.” One face, pointing downward in the hierarchy, represents a question to an answer just below it in the hierarchy. The other face, pointing upward in the hierarchy, represents an answer to a question just above it in the hierarchy. We emphasize that our depiction of questions as “above” answers and answers as “below” questions is arbitrary — it does not signify relative importance or value but is simply intended to be consistent with common usage. In science, an explanation is deeper and more “foundational” than the fact it explains, especially if it generalizes to explanations of many other facts. Special relativity is, in that sense, deeper than the constancy of c because it answers the question of why c is constant; it also answers the question of how much energy is released during nuclear fission and fusion. In technology, forms are deeper and more “foundational” than the functions they fulfill, especially if they have been adapted to fulfill many other functions. The multi touch transparent surface display is more foundational than the iPhone because it not only helps answer the question of how to create the iPhone, but also helps answer the question of how to create human-interactive displays in general. Rubber is more foundational than a bicycle tire because it not only helps answer the question of how to create a bicycle tire, but also helps answer the question of how to create a myriad of other kinds of tires.
The Network Is Modular: Facilitating Exploitation and Exploration
Closely related scientific questions and answers are organized into what we might call scientific domains, which we will refer to as scientific knowledge modules. Closely interacting technological problems and solutions are organized into engineered components, which we will refer to as technological knowledge modules.
Closely related scientific questions are often answerable within a scientific knowledge domain, or scientific knowledge module, drawing on multiple subdomains nested within the larger domain. A question related to some electron transport phenomenon in a particular semiconductor structure lies in the broad domain of semiconductor science but the answer might require an integrated understanding of both the subdomain of electron transport physics as well as the subdomain of the materials science of the synthesized structure. The subquestion associated with electron transport physics might require an integrated understanding of the subdomain of electrons in various kinds of structures (bulk materials, heterojunctions, nanostructures, coupled nanostructures) and of the sub-subdomains of interactions of electrons with phonons in those structures. The subquestion associated with the materials science of the synthesized structure might require an understanding of the sub-subdomains of substrates and epitaxy, thin films, or post materials synthesis fabrication. In other words, we can think of scientific knowledge domains as a modular hierarchy, and think of its subdomains as submodules and sub-submodules.
Closely related technological problems, likewise, are often solved by key technological components, or technological knowledge modules, perhaps integrating multiple subcomponents nested within the larger components. An iPhone is a component itself composed of many subcomponents, and each subcomponent is similarly subdivided. We can think of the “problem” of the iPhone as a component that is “solved” by its subcomponents — an enclosure, a display, a printed circuit board, a camera, and input / output ports. We can think of the “problem” of a printed circuit board as a subcomponent that is “solved” by sub-components that include low-power integrated circuit chips. Conversely, an iPhone is also a component that is itself nested in a hierarchy of use functions. An iPhone might be used as a solution to the problem of “running” a text-messaging app; a text-messaging app might be used as a solution to the problem of sending a mass text message to a friend group; the mass text message might be used as a solution to the problem of organizing the friend group into a protest in Times Square; and the protest in Times Square might be part of a solution to the problem of organizing a wider social movement for some human-desired social cause.
One might ask: Why is scientific and technological knowledge modular? They are modular because they are complex adaptive systems — systems sustained by and adapted to their environment by complex internal changes — and virtually all complex adaptive systems are modular (Simon, 1962). Complex adaptive systems both exploit their environments and explore their environments to improve that exploitation. Modularity enables efficiency, both in the exploitation of existing knowledge about the environment and exploration of that environment to create new knowledge.