I've been seeing arguments of this form for awhile:

  1. (agreed true premise) Some predicate P was applicable to behavior/event/thing A in the past. Variations: P(A) may have been true at some particular past time t, or over all past time T.
  2. (fallacious consequence) Therefore P(A) is true now. Variations: P(A) must be true now and all futures.

But I haven't known what to call them (e.g., when "calling out" their utterers). The problem seems to me to be due to ignoring (empirical) context and quantity (but has nothing to do with logical quantification). Some examples:

  1. Mass immigration was "good for America" in its (past) history. Therefore mass immigration must be "good for America" now. (P=good for America, A=mass immigration)
  2. (even more logically problematic) Humans have been burning fossil fuels for centuries, and "there was no crisis then." Therefore, to claim that fossil fuel combustion is "a crisis now, one must believe that we've been in crisis for centuries." (P=not crisis-producing, A=fossil fuel combustion)

Does this fallacy (or cognitive bias, but this failure "feels logical" to me--ICBW) have a widely-accepted name? Given the extensive and systematic attention to logic and rhetoric over millenia, I suspect this has been previously taxonomized.

1 Answer 1


It is a generalization from induction. The same way you can presume the sun will rise tomorrow because it has risen every day, you can presume that burning coal tomorrow will not cause massive, simultaneous coastal floods because you have done it every day last century and noticed only limited floods that seem to have other causes.

But generalization from induction is never fully warranted. There will be a day that the sun does not rise.

And is partially warranted only when there is a sustained mechanism that explains the consistency. The day the sun does not rise, that will be because it has stopped shining, or because we have left orbit.

The fact of its reliability depends upon the low probability of those two changes. This depends upon the low probability of other aspects of astronomy changing, and the interdependence of the whole system lets us combine those slim chances into a negligible chance that tomorrow is the day the sun will not rise.

  • Obviously correct--dunno how I missed the inductive aspect of these claims, esp as I am "a big fan" of Hume.
    – TomRoche
    Commented Aug 20, 2015 at 19:15
  • 1
    It is kind of scary to people that generalization from induction is never fully warranted. It is a hard argument to make in practice, and it makes a lot of people hate science. So passing over bad induction is a normal thing to do.
    – user9166
    Commented Aug 20, 2015 at 19:33

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .