Induction Problem & Correspondence Theory of Truth
The Problem of Induction
The results of observation and experimentation provide evidence for a scientific theory, but cannot prove the theory is correct. Even a modest empirical generalization, such as that all water boils at the same temperature, goes beyond what can be strictly deduced from evidence. If scientific theories expressed no more than the evidence supporting them, they would have little use. They could not predict the course of nature and would lack explanatory power.
The non-demonstrative, or inductive, link between evidence and theory poses a fundamental problem in the theory of knowledge: the problem of induction. David Hume, the 18th-century Scottish philosopher, provided its classic formulation. Hume considered simple predictions based on past observations, such as predicting the sun will rise tomorrow, given that it always has in the past. Life would be impossible without anticipating the future, but Hume argued that these inferences are rationally indefensible. This conclusion may seem unbelievable, but Hume’s argument remains unanswered. He admitted that inductive inferences are reasonably reliable, or we would not be alive to discuss the problem. However, he stated we only have reason to rely on induction if we have reason to believe it will remain reliable in the future, and demonstrated that this reasoning is impossible. The crux of the problem is that assuming induction will be reliable in the future is itself a prediction, justifiable only by induction, leading to circular reasoning. Maintaining that induction will work in the future because it has worked in the past assumes the very thing it attempts to justify. If this skeptical argument is valid, inductive knowledge seems impossible, and there’s no rational argument against someone believing, for example, that it’s safer to exit a room through the window than the door.
The problem of induction relates directly to science. Without an answer to Hume’s argument, there’s no reason to believe any aspect of a scientific theory that goes beyond what has been observed. The issue isn’t that scientific theories are never completely certain—this is an obvious truth. Rather, the issue is that we have no reason to suppose, for example, that water we haven’t tested will boil at the same temperature as water we have tested. Philosophers have continuously tried to resist this skeptical conclusion. Some argue that scientific methods of weighing evidence and making inferences are rational by definition. Others argue that past successes of inductive systems justify future use without circularity. A third approach argues that while we cannot prove induction will work in the future, we can demonstrate it will if any prediction method does, making its use reasonable. Using more recent theories, some philosophers argue that the current reliability of induction, something Hume acknowledged, is sufficient for inductive knowledge.
Karl Popper offered a radical response, forming the basis of his influential philosophy of science. He agreed with Hume that inductive inferences are rationally unjustifiable. However, he argued this doesn’t threaten the rationality of science, whose inferences are deductive. Popper’s central idea is that while evidence never proves a theory true, it can prove it false. Numerous black crows don’t prove all crows are black, but a single white crow proves the generalization false. Scientists can thus know a theory is false without induction. Furthermore, when choosing between two theories, there’s a rational preference for the theory not yet refuted; it could be true, while the refuted theory is known to be false. Induction never enters the picture, neutralizing Hume’s argument.
This solution faces objections. If true, scientists would have no reason to believe any theories or hypotheses are even approximately correct, or that any predictions are true, as these require induction. Popper’s position seemingly doesn’t allow scientists to know a theory is false, as evidence contradicting a theory might itself be flawed. The inductive inferences scientists make are unavoidable.
Theory of Truth as Correspondence
Everyone is interested in truth. Attaining it feels good; it connects us to reality. However, having true beliefs doesn’t explain what truth is.
We consider truth a property of thoughts, statements, or assertions. The correspondence theory posits a match between a statement and reality—between belief and reality. If I think “I see a cat” while looking at a cat, my thought is true because it corresponds to the fact it describes. There is a thought with content, and a fact (reality) that makes the thought true.
We should distinguish between truth and fact. A truth is a thought, a mental representation, or an accurate description of the world. A fact is anything describable or representable by truths. A truth is true because it relies on a corroborating fact. Without facts, there are no truths. However, there may be facts we haven’t observed, which aren’t truths for us until we encounter or describe them.
Aristotle formulated this theory in his Metaphysics: “To say of what is that it is not, or of what is not that it is, is false, while to say of what is that it is, and of what is not that it is not, is true…”. If a sentence or thought describes an event as we interpret the world, and this corresponds to the facts, the statement is true.
There are two ways to conceive of correspondence. First, as a strict coincidence between utterance and reality, an absolute copy reflecting reality like a mirror. The structure of a sentence determines its correspondence to reality. Bertrand Russell championed this view. Second, as a reciprocal relationship, where a thought means the same as or is adjusted to reality, in a broader sense. Aristotle held this view.
Could there be truths not strictly corresponding to facts? If someone says I am not a cat, they are correct. But is there a fact referring to the non-reality, the evidence that I am not a cat? Where do we “find” this fact? How can we understand a negative fact?
The correspondence theory faces other problems. Martin Heidegger argued that the notion of truth derives from a more primitive notion connected to the being of things, prior to any judgment. Truth lies in being, before any relationship. Another difficulty is the concept of correspondence itself. Our culture associates linguistic signs with objects. These signs represent the object in the human mind. But how does this mental representation occur?
The correspondence theory, as a descriptive approach to truth, has been contrasted with other theories, including the pragmatic theory, coherence theory, semantic theory, and redundancy theory.