Science is the objective pursuit of reliable knowledge. Although one might "know" something through authority, faith, or intuition, scientific method is distinct in that it must be possible for other investigators to ascertain the truth of scientific theories. Its founded on objective observation, the formulation of hypotheses that fit the data and predict other posibilities, repeatable experiments that can fail as well as succeed, and analysis and review by the scientific community.
What is Science?
Science enquires into the nature of the Universe and the world around us. We see interesting things and wonder what they are and how they work. What are stars made of? Why do they shine? What is lightning? How do rainbows form? Why do children tend to look like their parents? There's an infinity of such questions!
The word science has its origins in the Latin verb scire, meaning "to know." Although one might "know" something through authority, faith, or intuition, scientific method is distinct in that it must be possible for other investigators to ascertain the truth of scientific theories.
For our present purpose we'll briefly define science as the pursuit of reliable knowledge of the natural world.
'Reliable' means that when we apply it, e.g. to deduce something else, such as a prediction of some future occurences - then it doesn't let us down. For example, I've observed that the sun rises every morning (at least, those mornings on which I've been up early enough, and looked). On the basis of that knowledge, I have predicted that the sun will continue to rise every morning. The next time I checked, it did indeed rise, and now I have more confidence that my knowledge that 'the sun rises every morning' is reliable. I'm not saying that it's 'correct', or 'true'; only that it works well enough for me, for the time being. I'll trust it until such a time as I see a need to modify it (e.g. astronomers report a massive black hole racing towards the solar system...).
Fallibility and Skepticism
How can we obtain reliable knowledge?
Appeal to Authority
As children, we learn to ask our parents or teachers. "Mom/Dad, why doesn't the Moon fall down?" "Son, it's because it's too high for the Earth's gravity to reach". Well, we soon learn too that although some people might be right annoyingly often, nobody is infallible.. How about looking up the answer in books? Presumably, facts that are stated in print, especially in reputable publications, by reputable authors, are likely to be 'reliable'. This is a good method most of the time, but most people know that "you can't believe everything you read", and there have been plenty of instances of supposedly trustworthy publications getting things plain wrong.
Authority is not always a reliable source of knowledge, whether its people, books, or anything else. For practical purposes we usually accept what they say as probably correct because we can't check everything. However, trying to answer questions of nature correctly requires extreme care - history has shown repeatedly that people can draw the wrong conclusions from the evidence of their senses.
Misinterpretation of data
Drop a lead ball and a feather at the same time (in air) and the lead ball reaches the ground before the feather. Therefore heavy objects fall faster than light objects? No! In a vacuum the feather, unimpeded by air, reaches the ground together with the lead ball.
Another example is that the sun can be seen to rise in the east, and travel in an arc over to set in the west. Therefore the sun goes around the Earth? No! But this is what people truly believed for most of mankind's history.
Even given reliable data, we may easily draw the wrong conclusions, perhaps because we don't understand the principles of logic, or the limitations of statistics. Suppose that a survey shows that 20% of all road accidents are caused by drunk drivers. So 80% of all road accidents are caused by sober drivers. Therefore, it's safer to drive drunk... Why is this a false argument? There are many more like this that appear in the media, and everyday life. Particularly prevalent is the fitting of theories to random data - because there'll always be patterns you can see after the fact. Shuffle a deck of cards, then drop them face up on the floor. Oh, look! There are three kings together - it must mean something! Even in a truly random sequence, 'unlikely things' must happen sometimes.
Not only do we too easily draw the wrong conclusions from the given evidence, but we can all too easily 'see' things different than how they really are, or are not even there at all! For example, optical illusions.
Evolution and Revolution
Modern science has arisen from strenuous efforts over the past few centuries (especially since Galileo, Bacon, etc) to ensure that our evidence (data), and the conclusions we draw from it, are watertight. This has evolved into what we now call 'scientific method', which is based on the use of repeatable experiments and objective reasoning. A theory is not held to be true because some important people say it is, or because its proponents talk louder and longer than anyone else, or because it promises to keep my skin soft and smooth forever...
In fact, a theory is never held to be 'true' in any absolute sense; science does not prove anything (only mathematics does that, of mathematical statements) - it only offers tentative working models of the physical world, as agreed upon by several independent researchers after they have impersonally tested the evidence for themselves, and considered if there might be better explanations.
Nor is any theory ever immune from challenge; they may serve for a long time as foundation for much else in science, but if they fail to account perfectly for everything you'd expect, then they may be superceded by a 'better' theory - as happened with Newton's laws of gravitation and motion. Those laws aren't wrong - at everyday scales, but for near-light velocities, or strong gravitation fields, Einstein's theories work better.
In science, all claims are tentative, subject to revision on the basis of new evidence. Although science cannot provide one with hundred percent certainty, yet it is the most, if not the only, objective mode of pursuing knowledge. Science is above all else a critical and analytical activity and the scientist is pre-eminently a person who requires irrefutible evidence before he or she delivers an opinion.
Occasionally, experimental data just cannot be explained with any reasonable extension of existing theories, and someone will try a radical new approach (or 'paradigm' as Thomas Kuhn calls it). This new approach may be so strange and against prevailing 'common sense', that other (perhaps older) scientists are unable to accept it, but eventually the weight of the evidence and the predicitng power of the new theory win out, and the old theory is abandoned. The change in world view can be so radical as to constitute a 'revolution', rather than the more gradual evolution of 'normal science'.
Objective Experimentation and Observation.
Science depends heavily on the repeatability of experiments, and on their giving consistent (e.g. almost identical) results. This repeatability hinges on objective comparison of observations of different researchers studying the phenomenon. Objectivity indicates the desire to observe things as they are, without manipulating the observational results to accord with some preconceived world view. All observation is potentially contaminated, whether by our theories or our worldview or our past experiences.
Scientists, like anyone else, may be swayed by some preconceptions to look for certain experimental results rather than others. Scientists are people and suffer the flaws of humanity too, and it cannot truthfully be said that every scientific theory has arisen from a perfectly executed process of 'scientific method'. Scientists have desires, opinions, and biases that may sometimes influence them in the selection of their data and hypotheses - even, very ocasionally, to the point of fraudulence. We trust that this is very rare, and that science is a self-correcting process with checks and balances - such as empirical replication, and peer reviews of published work. The scientific community as a whole, however, judges the work of its members by the objectivity and rigor with which that work has been conducted.
An infamous example of an experiment that couldn't be repeated to give the same results again, involved "cold fusion" - a supposed process by which nuclear fusion could be achieved at room temperatures instead of the extremely high temperatures normally needed for fusion to occur. The researchers involved in the original experiment were found to have been less than objective in their methods.
The scientific method is founded upon direct observation of the world around us. A scientist examines critically and attempts to avoid all sources of bias. Wherever possible, observations are made to generate accurate measurements. Look at these optical illusions. Even seeing something with your own eyes may lead you to the wrong conclusions. The remedy is to devise reliable ways to measure things, e.g. with accurate rulers. Which of these lines is longer? In fact, they are both the same length (to within a small error). The arrowheads on the lines "trick" the human visual perception system, so an accurate ruler is required to avoid bias.
Science is a method that allows one to obtain, with the highest degree of certainty possible, reliable knowledge about nature. Scientific method establishes the principles that guide scientific research and experimentation, Scientific method is practiced within a context of scientific thinking, which is based on three things:
- using empirical evidence (empiricism),
- practicing logical reasonsing (rationalism),
- and possessing a skeptical attitude (skepticism).
Very generally, a scientist begins with some observations (data); through some creative process a hypothesis is generated about how these observations came to be (e.g. a conceptual or mathematical model). This hypothesis is used to predict something that was not part of the original data. An experiment devised and executed to see if the predictions of our hypothesis hold true. If not then we are faced with several possibilities, e.g. our hypothesis needs to be revised, the experiment was carried out incorrectly, or the analysis of the results from that experiment was in error.
- Observe and describe some phenomenon or group of phenomena.
- Formulation of a tentative description, called a hypothesis, that is consistent with what you have observed. The hypothesis often takes the form of a conceptual model, causal mechanism or a mathematical relation.
- Use of the hypothesis to predict the existence of other phenomena, or to predict quantitatively the results of new observations.
- Performance of experimental tests of the predictions by several independent experimenters and properly performed experiments.
- Comparison of the experimental results with theoretical predictions, and attempt to reconcile differences, e.g. by modifying the hypothesis or repeating the experiment under different conditions.