How do we “know” something in the world to be true?
Understanding comes from personal observation and experience, of course, but that source can be rife with error. Previous experience (or desire) can skew how we interpret new observations (justified or not); unnoticed or underrated complications or errors in the observation can cause our interpretation to be incorrect; or we may make an erroneous broad inference from a single observation or from too-few, outlier observations.
One time-tested way to improve the reliability of what we think is true is through methodical procedures, working with others: science.
"Science" is not simply a body of knowledge. Science is as a dynamic and interactive process of learning about the world, a way of gathering and organizing information to build understanding.
The classic scientific process has specific steps: 1) make an observation or ask a question; 2) develop a "hypothesis," a prediction or question that can be observed or tested; 3) make further, focused observation to test the hypothesis; 4) apply that information to build or alter theories; 5) repeat.
In everyday language “theory” usually means a guess or an opinion. In science, however, a theory is an underlying construct built over time by testing many related hypotheses. As theories are fine-tuned, improved - or even thrown out - with new observations, they produce new questions and predictions while they better explain the world.
Other guidelines apply to the process of science. Observations must be empirical, based on the senses. We often use tools, such as telescopes and microscopes, to make such observations more precise and more reliable. Observations (including setting up a specific observation, an "experiment") must be reproducible - that is, other people must be able to make the same observation, or run the same experiment, with the same results. Our observations need substantiation because “eyewitness accounts” are quite fallible. (Other people must have seen the UFO, or observed indisputable evidence of the UFO, for your sighting to be credible.)
In science, hypotheses and theories must be "falsifiable" - that is, the questions or statements, even the broadest concepts, must have the potential of being proven false.
Experiments are controlled observations that usually involve some set-up or manipulation. A typical experiment would compare two groups or subjects, the test group with the manipulation and the control group with no manipulation.
Non-experimental observations can be science, too: using thoughtful, meticulous observations to understand patterns - then using those patterns to make predictions that are “tested” by later observation.
The critical importance of questioning makes the scientific process distinctively skeptical. The requirement of reproducibility and the weaving of various data to build theories make science a community effort, with many people adding their input to eventually reach consensus or majority opinion.
This process is not a straight, plodding line - there are side-trails and backslides, and there are flashes of inspiration and creativity that leap to fresh insights. And part of the beauty of this system is that even “wrong” answers add to the body of knowledge.
Our image of a meticulous, well-educated scientist with lots of technical equipment has some validity: Having solid knowledge in a particular field makes it easier to ask good questions; technology greatly expands the observations that generate and answer questions; keeping records fuels more questions while documenting the process for others to scrutinize or follow.
Anyone can be a "scientist" by making rigorous observations, then asking straight-forward, answerable questions to develop a fundamental understanding of what's observed.
The key result of the scientific method is that it does change our understanding. Science - and knowledge - progresses as new information tailors or supplants old. Explanations evolve as the science community expands their observations and hones our grasp of the processes behind those observations. "Good" theories fit well with the empirical information, continuously building a harmonious framework for understanding our world.
As our knowledge base grows and our technology improves, our understanding of what is true evolves and improves.