# Is a non-empty set a partition of itself?

I have a homework assignment where, given some definitions, I need to prove that every set has a partition. There were some very elaborate ideas on how to prove this, but I realized that the definitions we were given allow a set to be a partition of itself. And if that is the case, then there clearly exists a partition of every set.

I'm not sure if I'm taking advantage of the simplicity of the definitions we are given. I tried to search if a set is a partition of itself, and not much if anything came up. The claim "every set is a partition of itself" seems to violate our english definition of partition, but be consistent with the mathematical definitions I have found.

For the sake of this discussion, I would think it be best to only consider non-empty sets because there seems to be some variety in how partitions are defined that prohibit the empty set from having partitions.

-
Yes, you´re right , but for the empty set itself you will have just the empty set. – user40276 Apr 7 '14 at 19:56

No, a set is not a partition of itself. But rather the singleton $\{X\}$ is a partition of $X$ whenever $X$ is non-empty.
In the other end of the spectrum, $\{\{x\}\mid x\in X\}$ is a partition of $X$, again when $X$ is non-empty. And if $X$ has at least two elements, then the two suggestions are indeed distinct.
Actually I'd say that the empty set is a partition of itself: $\varnothing$ is a collection of disjoint nonempty sets whose union is $\varnothing$. – Henning Makholm Apr 7 '14 at 20:30
@Reck: As Henning points out, one can use the empty set as a partition of itself. But I find it philosophically unsatisfying. If you definitions permit this, then you're about done (and you can note that the singletons partition gives $\varnothing$ when $X=\varnothing$). – Asaf Karagila Apr 7 '14 at 20:44