Saturday, February 5, 2011

What is science?

From Wikipedia :
"Science (from the Latin scientia, meaning "knowledge") is an enterprise that builds and organizes knowledge in the form of testable explanations and predictions about the world."


From that definition one can pick up that science is (a) the empirical collection of data and (b) the attempt to explain this data through a testable theory that not only explains the data but also make predictions about what will happen in a given situation.

 Empirical data is data collected through observation or experiment(data must be able to be accurately recreated under the same circumstances)

 What is a theory?
A theory is an attempt to explain a fact, for instance Newton asked why a apple always fell towards the Earth and never up or sideways etc. and he explained this phenomena through his Theory of Gravity,

It is however impossible to collect every shred of data there is out there, there is just too much of it, so scientists have to make certain generalizations about the data to come up with a theory that explains a fact.

Scientists use two types of generalizations
  • Inductive reasoning - a generalization based on on a large number of of observations e.g look at many living organisms under a microscope, see that they are all made of cells, deduce all living organisms made out of cells
  • Deductive reasoning - When several generalizations are used to predict a result e.g 1. All living organisms made out of cells 2. Hums are living organisms. Deduce humans are made of cells.
Generalizations are necessary but also prone to error, and therefore any proposed theory must not only be checked for accuracy of the deductions(this is where Peer Review  important)   and the testability of a theory.

But what does it mean if a theory is testable?
First of all, a theory cannot be proven, never ever. It can have countless experiments supporting it, but none can ever prove it. Science rather relies on disproof of theories. A theory must be falsifiable meaning that an experiment can used to test its predictions, and these predictions and results must be unambiguous.
This is an important test for pseudoscience, proponents of pseudoscience  often call scientists to disprove their theory, but in its nature it is impossible to devise an experiment that can test its predictions unambiguously.


Scientific theories although oftentimes perfectly accurate, are never considered to be 100% certain. Science through its error-checking mechanisms only improves previous theories, making them more accurate, explain more phenomena etc.

Science never asks you to blindly believe that what it says is true, on the contrary, it asks to to question more and more.




"I am only a child playing on the beach, while vast oceans of truth lie undiscovered before me." Sir Isaac Newton


Suggested reading : The Demon Haunted World: Science as a Candle in the Dark by Carl Sagan


Thank you for reading my first blog entry, and comments would be greatly appreciated.

No comments:

Post a Comment