In the context of a discussion on induction, it is a reasonable thing to ask what a piece of inductive reasoning looks like. What is it about it that specifically makes it appropriate to call it "inductive"? This argument: This A is B, This A is B, .... ----------------- All As are Bs or even This A is B, This A is B, .... --------------------- Probably As are Bs is, at least, some sort of recognizable pattern of an argument that can be called *inductive* to contrast it with a deductive argument like This A is a B This A is a B --------- Some As are Bs The idea here is that people think there are perfectly good arguments like the above that are not deductive and so let's call them inductive! But they are not *good* arguments at all, they never are, no matter how many cases are piled up in the premises. It is not just that they are not deductive, it is that they do not seem to have any *reasoning power*, there seems not even a *weak* force between the premises and the conclusion. Reasoning power? I refer to the power that avoids The Gambler's Fallacy. You see, no matter how many times a penny comes up tails, it does not follow in any way at all that it will come up tails on the next throw. It is not even probable! Nor is the likelihood of heads any better. There is no reasoning connection between the premise data and the conclusion. Some people say that there is a more sophisticated idea of induction that does not involve the above simplistic patterns. OK. I am listening. What are these more sophisticated ideas that identify something aptly to be called induction? It is no use merely pointing to the various things scientists do because they do too many things! The inductive bit gets lost in the haze! Some people have thought to say that scientists *induce* things by thinking up patterns that the data in the premises of so called inductive arguments suggest to their minds. But the trouble with this is that this does not make for any actual argument. Patterns are sometimes ten a penny. Any finite set of data points, any number of so called inductive premises likely fit an infinite number of possible patterns. It is often a remarkable achievement for humans to even think of one! But that act of thinking up a pattern, a possible theory, is not any kind of persuasive *argument* in itself. That may well be called part of a man's efforts to think through a scientific problem, it might even loosely called reasoning. But that bit in itself is not any persuasive forceful reasoning. That scientist X thinks up one pattern and scientist Y thinks up another contrary pattern can be described as both of them inducing different things from the data. But there is nothing in this kind of psychological induction to say the least thing about whether one is good *reasoning* and the other bad. It is just a psychological trick that trained and gifted scientists get up to! The testing of theories is the main game but that game is a game of deduction.