Is every non empty subset of the integers well ordered and does this mean that every subset contains a least element?

Are the positive rationals well ordered? i believe not.

Is this because of the fact that this set has no minimal element? Does every subset strictly smaller than the set of positive rationals have a least element? i believe it does but not sure


  • 1
    $\begingroup$ The well ordering principal is a theorem about natural numbers... $\endgroup$ Nov 14 '18 at 19:35
  • 1
    $\begingroup$ @CarlosBacca In general, an well-order is a total order such that every nonempty subset has a least element. The well-ordering principle says that the natural numbers are well-ordered. The argument is correct- since the positive rational numbers do not have a least element, then it is a nonempty subset of itself without a least element. $\endgroup$
    – Kevin Long
    Nov 14 '18 at 19:41
  • $\begingroup$ Oh, thank you i see $\endgroup$ Nov 14 '18 at 19:42
  • $\begingroup$ im not quite sure the argument is correct actually he said something a bit different you could have the set of all positive rational numbers 1/q such that q is greater 5 This is a subset of the set of positive rationals without a least element $\endgroup$ Nov 14 '18 at 19:51

When a linear order is non-well-ordered, lots of its proper subsets are also non-well-ordered. For example, consider the set of negative even integers or the set of reciprocals of integer powers of $2$: these are proper subsets of the integers and the positive rationals respectively, and neither has a least element.

The definition of well-orderedness - "Every subset has a least element" - can feel weird at first. It's helpful to think instead in terms of descending sequences: a linear order $(L,<)$ is well-ordered iff it has no infinite descending sequence $a_1>a_2>a_3>...$. Thinking this way it should be clear e.g. that the positive rationals aren't well-ordered, because we can "count down" $${1\over 1}>{1\over 2}>{1\over 3}>{1\over 4}>...$$ An important thing to keep in mind is that a counterexample to well-orderedness - that is, an infinite descending sequence in the linear order in question - will not be unique: we can't speak of "the" descending sequence, but rather "a" descending sequence.

  • $\begingroup$ A basically-irrelevant comment, to be ignored until the answer is understood: when we move into more advanced set theory - specifically, when we start to consider the axiom of choice - the two notions of well-orderedness above (least elements vs. no descending sequences) turn out to not be quite the same in some contexts. But that's something to ignore right now - I mention it only because if you look around this site for further related questions, you might see this issue mentioned, and I want to get out ahead of any possible confusion. $\endgroup$ Nov 14 '18 at 19:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.