In 1928, Alexander Fleming returned from a two-week holiday to find that a mold had contaminated one of his petri dishes and killed the surrounding bacteria. Most researchers would have tossed the plate and started over. Fleming stopped, looked closer, and asked a different question: what is that mold doing to those bacteria? That moment of structured curiosity - treating an unexpected observation as information rather than inconvenience - is where empirical inquiry begins.
Empirical inquiry is the practice of building knowledge through direct observation and evidence rather than through authority, tradition, or intuition alone. The word "empirical" traces to the Greek word for experience. The central commitment is simple: claims about the world should be tested against the world. What makes this harder than it sounds is that human beings are extraordinarily skilled at seeing what they expect to see.
Why Observations Need Structure
An observation on its own is not science. Fleming noticed the mold. But noticing is just the beginning. To turn an observation into useful knowledge, you have to isolate it, record it precisely, and ask whether it could be explained by something other than what you think is happening.
Consider what Fleming had to rule out before he could claim the mold was producing something antibacterial. Maybe the bacteria around the mold were already weak. Maybe a contaminating chemical had accidentally entered from another experiment. Maybe the plate was exposed to unusual temperature or light. Each alternative explanation is a rival hypothesis - a competing story that fits the same facts. Empirical inquiry is the systematic process of eliminating rival hypotheses until only the most defensible explanation remains.
This is not a process that ends in certainty. It ends in probability - in claims that are better supported than their alternatives. That is not a limitation. It is the feature. Science that admits its uncertainty and shows you the evidence is more trustworthy than authority that demands your belief.
The Observation-Question Cycle
Every empirical investigation starts with an observation, then generates a question from that observation. The question has to be answerable. "Why does the universe exist?" is a profound question. It is not an empirical one in any practical sense, because no experiment you could design would test it. "Does penicillin inhibit bacterial growth in controlled conditions?" is empirical. You can measure bacterial colonies. You can compare plates with and without the substance. You can run the experiment again.
Good empirical questions share a structure: they specify what you are measuring, they imply a comparison (with or without, before and after, more or less), and they can be answered with data. When a question lacks one of those three elements, the investigation that follows will produce ambiguous results that do not resolve anything.
Key Point: A well-formed empirical question specifies what is being measured, implies a comparison, and can be answered with observable data. If your question cannot be answered by observing something in the world, it belongs to philosophy or theology - both valuable, but not empirical science.
Observation Bias and How It Creeps In
The trap built into human observation is confirmation bias - the tendency to notice evidence that supports what you already believe and overlook evidence that complicates it. Scientists are not immune. In the 1970s, researcher Robert Rosenthal demonstrated that experimenters who were told their lab rats were "maze-bright" reported significantly better performance from those rats than experimenters told their rats were "maze-dull." The rats were randomly assigned from the same population. The difference lived entirely in the observers.
This is why empirical science develops procedures to protect observations from the observer's expectations. Blind and double-blind protocols, pre-registered hypotheses, standardized measurement tools - these are not bureaucratic obstacles. They are defenses against the very human tendency to find what you are looking for. You will encounter these tools in detail in later lessons. For now, the core lesson from Fleming's petri dish is that the value of an observation depends on the discipline with which it was made.