The scientific method is a process for creating models of the natural world that can be verified experimentally. The scientific method requires making observations, recording data, and analyzing data in a form that can be duplicated by other scientists. In addition, the scientific method uses inductive reasoning and deductive reasoning to try to produce useful and reliable models of nature and natural phenomena. Inductive reasoning is the examination of specific instances to develop a general hypothesis or theory, whereas deductive reasoning is the use of a theory to explain specific results. In 1637 René Descartes published his Discours de la Méthode in which he described systematic rules for determining what is true, thereby establishing the principles of the scientific method.
The subject of a scientific experiment has to be observable and reproducible. Observations may be made with the unaided eye, a microscope, a telescope, a voltmeter, or any other apparatus suitable for detecting the desired phenomenon. The invention of the telescope in 1608 made it possible for Galileo to discover the moons of Jupiter two years later. Other scientists confirmed Galileo's observations and the course of astronomy was changed. However, some observations that were not able to withstand tests of objectivity were the canals of Mars reported by astronomer Percival Lowell. Lowell claimed to be able to see a network of canals in Mars that he attributed to intelligent life in that planet. Bigger telescopes and satellite missions to Mars failed to confirm the existence of canals. This was a case where the observations could not be independently verified or reproduced, and the hypothesis about intelligent life was unjustified by the observations. To Lowell's credit, he predicted the existence of the planet Pluto in 1905 based on perturbations in the orbits of Uranus and Neptune. This was a good example of deductive logic. The application of the theory of gravitation to the known planets predicted that they should be in a different position from where they were. If the law of gravitation was not wrong, then something else had to account for the variation. Pluto was discovered 25 years later.
Real science hops from failure to failure, from several falsifiable hypotheses in confused competition to the next set, until a consensus evolves around a surviving paradigm that often uses aspects of its predecessors, adding unexpected novel ideas that lead to productive questions and more definitive tests, as disparate data starts to fit an overall unifying view. — R. Murray
The apparatus for making a scientific observation has to be based on well-known scientific principles. The telescope, for instance, is based on magnification of an image using light refraction through lenses. It can be proved that the image perceived through the telescope corresponds to that of the object being observed. In other words, you can trust observations made through telescopes. This is in contrast to magic wands, divining rods, or other devices for which no basis in science can be found. A divining or dowsing rod is a "Y" shaped branch of a tree, which is supposed to be able to help to identify places where there is underground water. The operator holds the divining rod by the top of the "Y", and the single end is supposed to dip when the operator passes over a section of land where there is water. What is the force that makes the divining rod dip? How does the divining rod "sense" the water? A scientist would try to answer these questions by experiments. Place the divining rod on a scale, for example, and then put a bowl of water under the divining rod. Is there a change of weight that indicates force? In another experiment the scale with the divining rod may be placed over a place known to have underground water, and over another place known to be dry. If these experiments show no force being exerted on the divining rod, we have to conclude that divining rods cannot be used as instruments for detecting water. We also have to conclude that any movement of the rod is accomplished by the hands of the person holding it, no matter how much the person denies it.
The scientific method requires that theories be testable. If a theory cannot be tested, it cannot be a scientific theory. Testing of scientific ideas can include the classical experimental method, replication, attempted refutation, prediction, modeling, inference, deduction, induction and logical analysis. Step 2 involves inductive reasoning, as described above. This approach can be used to study gravitation, electricity, magnetism, optics, chemistry, etc. Sometimes more than one theory can be proposed to explain observable events. In such cases, different predictions made with each theory can be used to set up experiments that select one theory over another. In the 17th century there were competing theories about whether electromagnetic radiation, such as visible light, consisted of particles or waves. At the beginning of the 20th century Max Planck postulated that energy can only be emitted or absorbed in small, discrete packets called quanta. This seemed to favor the particle theory, particularly after Einstein demonstrated that light behaves like a stream of particles in photoelectric cells. However, diffraction experiments with electrons, which were considered particles because they had a measurable weight, showed all the characteristics of waves. In 1926, Erwin Schrödinger developed an equation that described the wave properties of matter, and this became the foundation for the branch of physics called quantum mechanics.
How can waves behave like particles and particles behave like waves? Some scientific facts are very hard to comprehend. Yet, these are observable phenomena verified over and over again by many people all over the world. The behavior of the speed of light is another physical fact that is hard to understand. The speed of light in a vacuum is approximately 299,792 kilometers per second. The speed is reduced by about 3% in air and by 25% in water. A famous experiment conducted by Michelson and Morely at the end of the 19th century showed that the speed of light was the same perpendicular to the orbit of the earth and parallel to the orbit of the earth. The orbital speed of the earth of 29 kilometers per second could not be detected in the measurement of the speed of light. Einstein's theory of relativity is based on the constancy of measurement of the speed of light for all observers. A train has its headlight on. The speed of the light emanating from the train is the same whether the train is moving toward you or not! It is hard to accept, but many experiments for over one hundred years have come to the same conclusion.
Science has some well-known limitations. Science works by studying problems in isolation. This is very effective at getting good, approximate solutions. Problems outside these artificial boundaries are generally not addressed. The consistent, formal systems of symbols and mathematics used in science cannot prove all statements, and furthermore, they cannot prove all TRUE statements. Kurt Gödel showed this in 1931. The limitations of formal logical systems make it necessary for scientists to discard their old systems of thought and introduce new ones occasionally. Newton's gravitational model works fairly well for everyday physical descriptions, but it is not able to account for many important observations. For this reason, it has been replaced by Einstein's general theory of relativity for most celestial phenomena. Instead of talking about gravity, we now are supposed to talk about the curvature of the four-dimensional time-space continuum. Scientific observations are also subject to physical limits that may prevent us from finding the ultimate truth. The Heisenberg Uncertainty Principle states that it is impossible to determine simultaneously the position and momentum of an elementary particle. So, if we know the location of a particle we cannot determine its velocity, and if we know its velocity we cannot determine its location. Jacob Bronowski wrote that nature is not a gigantic formalizable system because to formalize it we would have to make some assumptions that cut some of its parts from consideration, and having done that, we cannot have a system that embraces the whole of nature.
The application of the scientific method is limited to independently observable, measurable events that can be reproduced. The scientific method is also applicable to random events that have statistical distributions. In atomic chemistry, for example, it is impossible to predict when one specific atom will decay and emit radiation, but it is possible to devise theories and formulas to predict when half of the atoms of a large sample will decay. Irreproducible results cannot be studied by the scientific method. There was one day when many car owners reported that the alarm systems of their cars were set off at about the same time without any apparent cause. Automotive engineers were not able to discover the reason because the problem could not be reproduced. They hypothesized that it could have been radio interference from a passing airplane, but they could not prove it one way or another. Mental conceptual experiences cannot be studied by the scientific method either. At this time there is no instrumentation that enables someone to monitor what anybody else conceives in their mind, although it is possible to determine which part of the brain is active during any given task. It is not possible to define experiments to determine objectively which works of art are "great", or whether Picasso was better than Matisse. So-called miracles are also beyond the scientific method. A person has tumors and faces certain death, and then, the tumors start shrinking and the person becomes healthy. What brought about the remission? A change in diet? A change in mental attitude? It is impossible to go back in time to monitor all variables that could have caused the cure, and it would be unethical to plant new tumors into the person to try to reproduce the results for a more careful study.
The scientific method relies on critical thinking, which is the process of questioning common beliefs and explanations to distinguish those beliefs that are reasonable and logical from those which lack adequate evidence or rational foundation.
Arguments consists of one or more premises and one conclusion. A premise is a statement that is offered in support of a claim being made. Premises and claims can be either true or false. In deductive arguments the premises provide complete support for the conclusion. If the premises provide the required degree of support for the conclusion then the argument is valid, and if all its premises are true, then the conclusion must be true. In inductive arguments the premises provide some degree of support for the conclusion. When the premises of inductive arguments are true, their conclusion is likely to be true. Arguments that have one or more false premises are unsound.
Arguments are subject to a variety of fallacies. A fallacy is an error in reasoning in which the premises given for the conclusion do not provide the needed degree of support. A deductive fallacy is a deductive argument where the premises are all true but reach a false conclusion. An inductive fallacy consist of arguments where the premises do not provide enough support for the conclusion. In such cases, even if the premises are true, the conclusion is not likely to be true.
Common fallacies are categorized by their type, such as Ad Hominem (personal attack), and appeals to authority, belief, fear, ridicule, tradition, etc. An example of an Ad Hominem fallacy would be to say "You do not understand this because you are American (or Chinese, etc.)". The national origin of a person (the premise) has nothing to do with the conclusion that a person can understand something or not, therefore the argument is flawed. Appeals to ridicule are of the form: "You would be stupid to believe that the earth goes around the sun". Sometimes, a naive or false justification may be added in appeals to ridicule, such as "we can plainly see the sun go around the earth every day". Appeals to authority are of the form "The president of the United States said this, therefore it must be true". The fact that a famous person, great person, or authority figure said something is not a valid basis for something being true. Truth is independent of who said it.
Direct or Experimental evidence. The scientific methods relies on direct evidence, i.e., evidence that can be directly observed and tested. Scientific experiments are designed to be repeated by other scientists and to demonstrate unequivocably the point that they are trying to prove by controlling all the factors that could influence the results. A scientist conducts an experiment by varying a single factor and observing the results.
When appropriate, "double blind" experiments are conducted to avoid the possibility of bias. If it is necessary to determine the effectiveness of a drug, an independent scientist will prepare the drug and an inert substance (a placebo), identifying them as A and B. A second scientist selects two groups of patients with similar characteristics (age, sex, etc.), and not knowing which is the real drug, administers substance A to one group of patients and substance B to the second group of patients. By not knowing whether A or B is the real drug, the second scientist focuses on the results of the experiment and can make objective evaluations. At the end of the experiment, the second scientist should be able to tell whether the group receiving substance A showed improvements over those receiving substance B. If no effect can be shown, the drug being tested is ineffective. Neither the second scientist nor the patients can cheat by favoring one substance over another, because they do not know which is the real drug.
Anecdotal, Correlational, or Circumstantial Evidence. "Where there is smoke, there is fire" is a popular saying. When two things occur together frequently, it is possible to assume that there is a direct or causative relationship between them, but it is also possible that there are other factors. For example, if you get sick every time that you eat fish and drink milk, you could assume that you are allergic to fish. However, you may be allergic to milk, or only to the combination of fish with milk. Correlational evidence is good for developing hypotheses that can then be tested with the proper experiments, e.g., drink milk only, eat fish only, eat fish and milk together.
There is nothing wrong with using representative cases to illustrate an inductive conclusion drawn from a fair sample. The problem arises when a single case or a few selected cases are used to draw a conclusion which would not be supported by a properly conducted study.
Argumentative Evidence consists of evaluating facts that are known and formulating a hypothesis about what the facts imply. Argumentative evidence is notoriously unreliable because anybody can postulate a hypothesis about anything. This was illustrated above with the example about the "channels" of Mars implying intelligent life. The statement "I heard a noise in the attic, it must be a ghost" also falls in this category.
Testimonial Evidence. A famous football player appears on television and says that Drug-XYZ provides relief from pain and works better than anything else. You know that the football player gets paid for making the commercial. How much can you trust this evidence? Not very much. Testimonials are often biased in favor of a particular point of view. In court proceedings, something actually experienced by a witness (eyewitness information) has greater weight than what someone told a witness (hearsay information). Nevertheless, experiments have repeatedly demonstrated that eyewitness accounts are highly unreliable when compared with films of the events. The statement "I saw a ghost last night." is an example of testimonial evidence that probably cannot be verified and should not be trusted. On the other hand, the statement "I saw a car crash yesterday." can be objectively verified to determine whether it is true or false by checking for debris from the accident, hospital records, and other physical evidence.Next Subjective Perceptions