Induction versus Deduction

Some overlapping, hence confusion, between induction and deduction is possible because it is left to the author of the argu­ment whether to stress the probability, even when the probability is very high, or simply to equate very high probability with cer­tainty. Once an elderly colleague of mine took a few days off in a hurry to go for a thorough physical checkup. His son was a med­ical intern who possibly advised him. I do not know if there was any preceding anxiety, but his sudden disappearance caused some anxiety among his colleagues. Pleased to see him after a few days, I asked, “So, what did they say?” His answer: “They said, I am going to die.” Being slightly upset, I asked, “What do you mean?” He answered with a little chuckle, “They said, I am surely going to die. Only they don’t know when or how.” Now, catching on to his dry humor, I laughed with him. And then, he told me that everything was found to be “normal,” as is the case with most physicals. Case in point: Death is certain for all humans. He is a human, so he will die sooner or later. This is deduction. But how do we know that death is certain? The only evidence we have is that we do not know of anybody surviving, say, after age 200; the oldest recorded human was 122 at death. But we have not exhausted the testing of all the members of the class. In fact, we cannot, because there is no tangible proof that, in the distant future, humans will not evolve such that, maybe, one among them will survive forever. Surely none of us believes this, but the issue is not belief; it is proof. In this way, the conclu­sion is based on a “probable fact,” not just a fact; hence, the con­clusion itself may turn out to be only a probability, meaning, my colleague may, probably, live forever (God bless him)! This is induction.

1. Experiment: a situation that is deliberately set up by an individual or a group of investigators, either for the purpose of demonstrating or finding an object or phenomenon as expected, for verifying a hypothesis, or for validating a theory.
2. Empirical: that part of the method of science in which the reference to actuality or factual finding allows a hypothesis to be formed, which, on fur­ther verification, may become a general principle.
3. Verification: the demonstration of the correctness of a hypothesis by means of experiment or observation.
4. Hypothesis: a provisional assumption made as to how events—things or phenomena—are interre­lated, which is used as a guiding norm in devising experiments and making observations until it is verified or disproved by experimental evidence.

Broadly we may think of two levels of hypotheses:

1. 1. Preliminary hypothesis, wherein only the facts so far known are used for building the logical structure, a theory, of interrelations
   2. Final hypothesis, or “crucial test,” which is deducible from the theory, leading to new facts that can be experimentally verified. Observation of such facts directly or of the consequences implied by such facts serves as confirmation of the theory itself.

5. Theory: a coherent network of facts and principles. Whereas a hypothesis is a proposition, yet to be tested with evidence for its truth-value, a theory is based on inferences drawn from principles that have been established with independent evidence.

A clear demarcation between hypotheses and theory is not practical in that a hypothesis that has passed some tests of experimental evidence is often claimed to have acquired the status of a theory, though it may still have to undergo one or more crucial tests. A theory of high authority and accu­racy is often referred to as a law.

6. Thesis: an essay or dissertation upon a specific or definite theme, as an essay presented by a candi­date for a diploma or degree. The oral examina­tion preceding the confirmation of the diploma or degree is often referred to as a “defense.” In this context, a thesis is understood to be a proposition that the candidate advances and offers to defend by argument against objections.

Source: Srinagesh K (2005), The Principles of Experimental Research, Butterworth-Heinemann; 1st edition.