Observational science is thousands of years old, but until we developed the scientific method, there was no reliable mechanism for separating fact from fancy.
Science, as practised by ancient and classical scholars, was impressive.
The longest running set of recorded observations dates back to the 1st millennium BCe, from the Babylonian Astronomical Diaries. Muslim scholars in Persia were documenting medical observations 1,000 years ago, and Aristotle himself was a keen observer of natural phenomena.
However, until the age of enlightenment, we had no way to know for sure which explanations of natural phenomena were correct. For example, although Aristotle correctly asserted that whales and seals were not fish, he also stated that men had more teeth than women did.
Observing and recording natural phenomena is called "descriptive science."
The mathematics required to analyze recorded data developed in ancient China, Babylon, Greece, Egypt, India and, of course, the Islamic empire in North Africa and the Middle East.
The development of mathematics and science were global endeavours and included contributions from many cultures.
I am always surprised by "post-modernists" who accuse science of being Eurocentric or having a "Western" bias.
Descriptive science allowed us to make generalizations about the world using "deductive logic."
Galileo deduced somewhat controversial facts about our solar system, which opened the door to modern astronomy. This process, however, could take him only so far.
During the 16th and 17th centuries, great minds like Francis Bacon, Descartes and Newton built on Galileo's observations by using what is called "inductive logic."
They created explanations, or hypotheses, that explained how nature worked and then they did something that changed the world. They generated predictions arising from those hypotheses and tested them.
I cannot overstate how profound an advance this was. If your hypothesis is the correct one, then your predictions will be borne out with evidence. If they are not, then your explanation must be incorrect and you must reject it.
Hypotheses must be "falsifiable" (i.e. they must be testable in a way that can show us whether or not they are wrong). If a hypothesis is not falsifiable, then it's not testable, and it's not science.
Here is where the line between "science" and "non-science" is drawn.
For example, the hypothesis that the Earth is spherical can be tested by watching a ship going out to sea on a calm day. As the ship goes further from shore, it will disappear from the bottom up as the lower part of the ship is obscured by the curvature of the Earth, the reason ancient sailors did not think the Earth was flat.
Great scientists are very creative; they try to imagine every possible explanation for an observation and then try to shoot them down (falsify them) by testing them. The last explanation standing is tentatively, cautiously accepted as "the best current explanation." Good scientists will never use the terms "proof" or "proven to be true."
After a hypothesis has been tested extensively from all sides, has accumulated mountains of evidence and has no remaining challengers, we accept it as "fact" and move on.
For example the theory of gravity is now the accepted explanation for the observation that unsupported objects will fall towards the ground and the theory of evolution is now the accepted explanation for many phenomena, such as changes in the fossil record that correspond with geological changes and the close physical resemblance between species that share very similar DNA.
Sadly, the word "theory" in common language does not have the same meaning as it does in scientific parlance.
The scientific revolution caught on rather quickly (in historical terms) and set the world on fire during a time we now call "the Enlightenment." During the following century, inductive science brought us a doubling of life expectancy, international transportation and the industrial revolution.
Since then, we have landed men on the moon, sent probes to other planets and improved living conditions for many people in profound ways.
Einstein worked out relativity, and now we have GPS in our phones, global communications and the Internet. Bacteria and viruses were discovered, and we now have vaccines for many of them. In all the generations before my parents, nobody knew enough about viruses to treat smallpox. Now, it's eradicated.
Over the last two centuries, we have refined and expanded the rules and methods for scientific testing. If followed correctly, these rules allow researchers to collect and analyze data in a way that eliminates human bias and reveals the truth about natural processes. These methods and rules are collectively called "the scientific method."
Different kinds of science use different parts of it, but at its core there are some simple requirements. Hypotheses must be falsifiable and testable. Experiments must be designed in a way that eliminates human bias. Experiments must have appropriately sized samples and must have proper control groups.
The results must also be repeatable by other scientists. If you are getting different results than everyone else, then you're probably doing it wrong.
There are many subsets of rules, depending on the discipline, that comprise the "gold standard" of experimentation, and we are adding best practices every day. When done correctly, science is the most powerful tool we have. It's a set of rules that is most definitely not a belief system.
Despite protestations from certain sociologists, science does not run on belief or faith, but on evidence and fact. How do we know it works? It delivers the goods.
Blythe Nilson is an associate professor of biology at UBC Okanagan and advisory fellow of the Centre For Inquiry Okanagan.

READ MORE

Kitab fi al-jadari wa-al-hasbah (A Treatise on Smallpox and Measles, translated 1848) by the Islamic physician Al-Razi, 9th century. Britannica Concise Encyclopedia.

Newton, Isaac (1642-1727). The Principia: a new Translation, Guide by I. Bernard Cohen ISBN 0-520-08817-4 University of California (1999)

Great book on Newton: The Clockwork Universe, Isaac Newton, the Royal Society, and the Birth of the Modern World By Edward Dolnick. Harper/HarperCollins Publishers. $27.99.

Great science podcasts for non-scientists: Science Talk, RadioLab, Science Fridays, Quirks and Quarks, The Science Show, Star Talk and AstronomyCast among many.