Induction is a kind of reasoning. A person reasons inductively when they infer that something they (or others) have yet to observe is or will be a certain way from some thing (or things) that they have observed. For example, someone might conclude that all swans (including the ones they haven’t seen) are white, or that the next swan they’ll see will be because all the swans they’ve seen before are white.
This kind of reasoning isn’t full-proof, but we rely it much of the time. The reason I can sit on chairs without feeling fearful is that I know that every time I’ve sat on a chair thus far in life, nothing terrible has happened. Nonetheless, it’s possible that the next time I sit on a chair it will fall apart, or explode by some freak accident. Inductive reasoning can still be useful, even if it doesn’t give us conclusions that are absolutely certain. However, even knowing this much – that inductive inferences give us conclusions that are likely to be true, even if not certain – is problematic. Think about how we might know that the conclusions of inductive arguments are likely to be true. Plausibly, we believe that inductive inferences are useful because every time we’ve reasoned inductively in the past, the conclusions we’ve reached have been true more often than not. But this justification itself relies on inductive reasoning! We’re making a claim about future pieces of inductive reasoning based on what we’ve observed about past pieces of inductive reasoning. So, justifying inductive reasoning seems to lead us in an annoying little circle! | |||