Dev

Building Without Science: Lessons from Ancient Engineering Wisdom for the Modern World

Aeon's latest video explores beautiful structures built solely with rules of thumb, not science or math, and considers implications for modern technological development. The value of empirical rules learned from historical construction methods.

4 min read

Building Without Science: Lessons from Ancient Engineering Wisdom for the Modern World
Photo by Jason Leung on Unsplash

Building Without Science: Lessons from Ancient Engineering Wisdom for the Modern World

Aeon’s video “The engineering method,” published on April 20, 2026, vividly depicts how humanity constructed complex and beautiful structures before the establishment of systematic scientific and mathematical knowledge. This video goes beyond a mere historical retrospective, prompting a reevaluation of the importance of “rules of thumb” in modern technological development. How did craftsmen build pyramids and aqueducts before the light of science? How can that wisdom be applied to today’s AI and software development? This article delves into the core of the video, exploring everything from the background to its future implications.

The Mystery of Ancient Engineering: Miracles Born from Rules of Thumb

Looking back at history, many astonishing structures were built without a full understanding of scientific principles. For example, the Great Pyramid of Giza in Egypt, completed around 2500 BC, stands 146 meters tall and is composed of about 2.3 million stone blocks, its precision astonishing even modern engineers. At the time, mathematics was still in its infancy, and the laws of physics had not been established. The construction relied solely on simple tools (saws, levers, inclines), the craftsmen’s empirical rules, and organized labor. The video points out that the pyramid’s construction was made possible by the accumulation of “rules of thumb”—empirical rules of “this works.” Techniques such as using wet sand to reduce friction when transporting stone blocks and the prototype of a level for ensuring perfect horizontality were employed. These were not scientifically proven but wisdom born from trial and error.

Similarly, the aqueducts of the Roman Empire (construction began around 312 BC) established the technology to transport water over long distances with a slight incline (about 0.5%) without mathematical calculations. The arch structure of the aqueducts was achieved simply by stacking stones based on rules of thumb, and its durability has lasted for 2000 years. The pointed arches and flying buttresses seen in medieval Gothic cathedrals (e.g., Notre-Dame de Paris) were also supported by an empirical understanding of mechanics, creating a sense of religious uplift. These examples demonstrate that even without science, human creativity and rules of thumb can solve complex problems.

Application to Modern Technology Development: Reevaluating Rules of Thumb

In modern times, where science and technology are highly advanced, data-driven approaches are mainstream. This is especially true in AI and machine learning, which are built on vast datasets and mathematical models (e.g., linear algebra, calculus). However, as Aeon’s video suggests, the value of rules of thumb remains important and should perhaps be reevaluated. This is because many real-world problems cannot be fully represented by data.

Let’s look at specific examples. In software development, design patterns (e.g., Singleton, Factory Pattern) function as rules of thumb. These are “reusable solutions” derived from experience; though not mathematically proven, they improve code quality and maintainability. In DevOps practices, continuous integration (CI) steps and infrastructure configuration best practices are widely accepted as industry rules of thumb.

In the field of AI, early expert systems (1970s-80s) adopted a rule-of-thumb approach and were effective in medical diagnosis and financial risk assessment. In modern machine learning, boosted decision trees (e.g., XGBoost) combine multiple rules of thumb (decision trees) to enhance prediction accuracy. Furthermore, in the development of self-driving cars, alongside data-driven deep learning, rules of thumb serving as traffic rules and safety foundations are indispensable. For example, the simple rule of “stop at a red traffic light” complements complex neural networks.

The video advises modern engineers not to rely too heavily on science and to utilize rules of thumb as “heuristics.” This also impacts the innovation process. In startup development environments, “lean development”—which involves rapidly releasing a minimum viable product (MVP) and building rules of thumb from user feedback—is mainstream. This is similar to the ancient trial-and-error approach, supplementing scientific hypothesis testing with empirical rules.

Impact on Industry and Future Outlook: A Bridge to a Sustainable Future

The wisdom of ancient engineering offers new perspectives on modern challenges, particularly sustainability and resilience. In the construction industry, low

Source: Aeon

Comments

← Back to Home