Scientific and technological knowledge
Science, craft and technology are not synonyms. Telling them apart changes what can be demanded of each one, and explains why an engineer has to decide even without having all the information.
01Science, craft, technology and engineering
The words are used as synonyms and they are not. Telling them apart is not an academic whim: it changes what you can demand of each one and which questions it answers.
| Activity | What it seeks | How it is validated | Example in electronics |
|---|---|---|---|
| Science | To explain: true knowledge about how the world is | By testing predictions against experience | Band theory explains why doped silicon conducts |
| Craft | To make: practical know-how, often passed on from person to person | Because it works, even if nobody quite knows why | Knowing how to solder well, “nudging” an adjustment until the stage stops oscillating |
| Technology | To make with a foundation: craft supported by scientific knowledge | By its effectiveness, measured and reproducible | The manufacturing process of an integrated circuit |
| Engineering | To solve a specific problem, with finite resources and deadlines | Because the system meets the specification, within the accepted cost and risk | Designing a power supply that delivers 5 V ±2% at 3 A for ten years |
Science can afford to say “we don’t know yet.” Engineering cannot: a decision has to be made anyway, with the information available, and the engineer has to answer for the result. That asymmetry explains almost everything else: safety margins, standards, type tests and professional responsibility.
02How scientific knowledge is validated
The classical scheme —observe, formulate a hypothesis, deduce consequences, test them— has an asymmetry that Karl Popper placed at the center: no number of confirmations proves a theory, but a single contrary experiment refutes it. That is why the criterion for calling a claim scientific is that it be falsifiable: that you can say in advance which result would leave it in trouble.
“This cable improves the sound, but only trained ears can perceive it” is built so that it can never fail, and so it says nothing verifiable. In electronics the honest equivalent is: “this cable lowers the contact resistance by 3 mΩ, measured this way.” That can be checked, and it can be disproved.
Thomas Kuhn added the historical dimension: science does not advance only by accumulating data. It works within a paradigm —a set of accepted assumptions, methods and problems— until anomalies pile up and a scientific revolution comes along and replaces it. The move from classical physics to quantum physics, without which the transistor would not exist, is the most frequently cited case.
03Technological knowledge has rules of its own
It is not simply “applied science.” It has its own form, criteria and problems:
- It seeks effectiveness, not truth. A technological rule says “to obtain A under conditions B, do C.” It is judged by whether it works and at what cost, not by whether it is true.
- It deliberately uses false models. The ideal diode, the source with no internal resistance, the amplifier with infinite gain: everyone knows they do not exist, and they are used anyway because they simplify things and give good enough results. Knowing when a model stops being useful is part of the trade.
- It incorporates tacit knowledge. Some know-how is not written down anywhere and is learned by doing: why one circuit board oscillates and the one next to it doesn’t, how you “smell” an overheated component.
- It is codified in standards and datasheets. An IRAM or IEC standard (IRAM is the Argentine Standards and Certification Institute; IEC, the International Electrotechnical Commission) is crystallized technological knowledge: decades of tests and accidents condensed into a procedure.
A circuit can work on the bench by sheer chance —a parasitic capacitance that happens to filter out the noise— and fail in production. What separates the skilled technician from the engineer is being able to explain why it works, which is the only thing that makes it possible to predict when it will stop working.
04Science and technology push each other in both directions
The “linear model” —science discovers, technology applies, industry produces— is convenient and often false. The history of electronics shows both directions:
| Case | What came first |
|---|---|
| Steam engine and thermodynamics | The engine came first (Watt, 1769) and the theory that explains it came later (Carnot, 1824). Technology led the way |
| Transistor | First quantum mechanics and solid-state physics, then the device (Bardeen, Brattain and Shockley, 1947). Science led the way |
| Optical fiber | Optics was already known; it took a breakthrough in materials, ultrapure glass, to make the idea viable |
| Integrated circuit | No new science and no new theory: an idea about process (Kilby and Noyce, 1958-59) that changed the whole industry |
Today people speak of R&D&I: research, development and innovation, as activities that feed back into each other. And to describe what stage a technology has reached, technology readiness levels (TRL, from 1 to 9) are used, running from the basic principle observed to the system proven in its real environment. They help avoid mistaking a laboratory result for a product.
Read the case and decide. There is no trick: the idea is to discuss the boundaries, which are often blurry.
05What this means at the workbench
- Reading a datasheet is reading codified science. Every curve came from a physical model and from a measurement made with a declared method. The test conditions —temperature, voltage, time— are part of the data, not fine print.
- Measuring is testing a hypothesis. Before measuring, it pays to write down what you expect to find. If the result matches, the model is useful; if not, there is something to learn, and that is more valuable.
- Documenting is making your own work falsifiable. A report that does not say which instrument, under what conditions and with what uncertainty something was measured lets no one —not even you, six months later— check anything.
- Standards are not bureaucracy. They are the accumulated knowledge of the profession, and they are often written after an accident.
06Activities
Choose an everyday component —the white LED, the lithium battery, the power MOSFET— and reconstruct its history: what scientific knowledge made it possible, what technical problem it solved, what industry pushed it and in which decade it became mass-market. Identify whether science led or followed.
Collect five advertising claims for electronic products and rewrite each one in falsifiable form, indicating which measurement would put it to the test and with which instrument.
Take three models used in class (ideal diode, ideal operational amplifier, resistor with no inductance) and determine experimentally under what conditions they stop describing real behavior. You can start with the instrument simulator and continue at the bench.
07Common misconceptions
- Believing that technology is only “gadgets.” A quality-management method or a communication protocol is technology too.
- Assuming that science always comes first. Often practice leads and theory arrives later to explain it.
- Mistaking a model for reality and being surprised when the real circuit behaves differently.
- Confusing correlation with cause when interpreting a measurement: two things that change together are not always related.
- Discarding the tacit knowledge of the shop floor because it isn’t published.
08Self-assessment
What does it mean for a claim to be falsifiable?
That you can specify in advance which observation would contradict it. It is Popper’s criterion for telling a scientific claim from one that is not.
Difference between craft and technology.
Craft is know-how, often without an explanation of why; technology is that same doing grounded in scientific knowledge, which makes it possible to predict, optimize and transfer.
Why use models that are known to be incorrect?
Because they simplify the calculation and give results that are close enough within a known range. What matters is knowing what that range is.
What is a paradigm according to Kuhn?
The set of assumptions, methods and problems accepted by a scientific community during a period. When anomalies overwhelm it, a scientific revolution takes place.
What are TRL levels for?
To locate the stage of maturity a technology has reached, from the idea to the system proven under real conditions, and to avoid mistaking a laboratory prototype for a product.
09Further reading
- Mario Bunge. La ciencia, su método y su filosofía. Siglo Veinte / Sudamericana (in Spanish). Short and sharp: the distinction between science, craft and pseudoscience, written by an Argentine physicist and philosopher.
- Thomas S. Kuhn. The Structure of Scientific Revolutions. University of Chicago Press, 1962 (Spanish edition: “La estructura de las revoluciones científicas,” Fondo de Cultura Económica). The book that introduced the idea of the paradigm and changed the way the history of science is understood.
- Karl R. Popper. The Logic of Scientific Discovery. Hutchinson, 1959 (Spanish edition: “La lógica de la investigación científica,” Tecnos). The original formulation of the falsifiability criterion.
- Carl Mitcham. ¿Qué es la filosofía de la tecnología? Anthropos (in Spanish). An orderly introduction to the questions specific to technological knowledge, as distinct from scientific knowledge.