When a set has an empty set as an element, e.g.$ \{\emptyset, a, b \}$. What is the powerset?

Is it: $$ \{ \emptyset, \{ \emptyset \}, \{a\}, \{b\}, \{\emptyset, a\} \{\emptyset, b\}, \{a, b\}, \{\emptyset, a, b\}\}$$


$$ \{ \emptyset, \{a\}, \{b\}, \{\emptyset, a\} \{\emptyset, b\}, \{a, b\}, \{\emptyset, a, b\}\}$$

Or $$ \{ \{\emptyset\}, \{a\}, \{b\}, \{\emptyset, a\} \{\emptyset, b\}, \{a, b\}, \{\emptyset, a, b\}\}$$

The confusion arises for me because, the powerset of every non-empty set has an empty set. Well the original set already has the empty set. So we don't need a subset with an empty set.

Somehow, the first one seems correct. Yet, I can't seem to accept it.

  • 1
    $\begingroup$ The first one: $\;\emptyset\;$ is one of the elements of the given set, besides being a subset of it. $\endgroup$ – DonAntonio Oct 12 '16 at 12:39
  • 13
    $\begingroup$ Let $c$ denote $\varnothing$. What is the power set of $\{a,b,c\}$? Now write $\varnothing$ instead of $c$ again. $\endgroup$ – Asaf Karagila Oct 12 '16 at 13:16

The first one is correct.

This is because $\emptyset$ and $\{\emptyset\}$ are different. The first is an empty set whereas the second is a set whose only element is the empty set.

Both are subsets of the given set. This is because the $\emptyset$ is the subset of every set, and as it happens to be an element of the given set, the set containing it as its element is also its subset.


If a set $A$ is such that $\emptyset\in A$, its power set must necessarily contain these two sets:

  • $\emptyset$ (like all other power sets), corresponding to selecting nothing from $A$ (not even $\emptyset$, which is something)
  • $\{\emptyset\}$, corresponding to selecting $\emptyset$ only

Therefore only the first of your proposed answers is correct, as you think.


Your suggestions differ by having $\emptyset$ and/or $\{\emptyset\}$ included or not.

  • We have $\emptyset\in\mathcal P(X)$ because $\emptyset\subseteq X$ (which would hold for any other $X$ as well)
  • We have $\{\emptyset\}\in\mathcal P(X)$ because $\{\emptyset\}\subseteq X$ (which is the case because $\emptyset\in X$ in this specific problem)

Therefore, your first variant is correct (and the other two are incorrect because $\emptyset\ne\{\emptyset\}$).

  • 1
    $\begingroup$ Your second bullet is strangely phrased to me; we have $\{\emptyset\}\in\mathcal{P}(X)$ simply because $\emptyset\in X$ in this specific problem. The fact that $\emptyset\subseteq X$, which is true for every set $X$, has nothing to do with it? $\endgroup$ – Inactive - avoiding CoC Oct 12 '16 at 15:47

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.