I noticed in my discrete math textbook that the Rosen makes reference to the universal set, denoted $U$, and establishes various results, like the complement of a set $A$ with respect to $U$, or the identity law $A \cap U = A$. But in what sense are we calling $U$ the universal set so that such a set is valid in elementary set theory? In other words, why is it legitimate in the author's case to speak of $U$ where it seems to be implied somehow that $U$ is not the set of all sets? Do I have to state that $U$ is specific universal set as to avoid confusion as to whether or not it is the set of all sets (but still calling the set $U$ rather than a concrete universe like $\mathbb{Z}$)?

With that question proposed, if we let $U$ be specific universal set (not the set of all sets), and define $\emptyset_1$ and $\emptyset_2$ to be two empty sets, then we can conclude that $\emptyset_1 = \emptyset_2$ by noting

$$U = {\emptyset_1}^c$$ $$U = {\emptyset_2}^c$$

implies that

$$U^c = ({\emptyset_1}^c)^c = \emptyset_1$$ $$U^c = ({\emptyset_2}^c)^c = \emptyset_2$$

Because $U^c = U^c$, then $\emptyset_1 = \emptyset_2$.

Of course, another way to do this proof would be to show that the two empty sets are subsets of each other.


In those examples, we are working in the so-called "naive" set theory; in this case, the "universal" set is simply a collection of objects assumed as the domain where the variables range.

It can be the set $\mathbb N$ of natural numbers, or the set $Humanity$ of all human beings.

It is not intended to be the set of "everything".

With this approach, we can treat the elementary properties of sets : the so-called algebra of sets.

For example, if the domain (or universe) $U$ is the set $\mathbb N$ of naturals and $A$ is the set of odd numbers, then $A^c$ is the complement of $A$ "in $U$", i.e. the set of even numbers.

Regarding the "unicity" of $\emptyset$, this is straightforward.

We know that :

for all $x$, $x \notin \emptyset$.

But we have also the principle of extensionality for sets :

if for all $x, x \in A \leftrightarrow x \in B$, then $A=B$.

Applying it to $\emptyset_1$ and $\emptyset_2$, we clearly have :

for all $x, x \in \emptyset_1 \leftrightarrow x \in \emptyset_2$

because both are empty.

Thus, we conclude with :

$\emptyset_1 = \emptyset_2$.

  • $\begingroup$ Ah, so proving that the two empty sets are subsets of each other demonstrates their equality. Definitely easier. Just for hiccups and giggles, is the proof that I wrote valid? $\endgroup$ – A. Lovelace Sep 26 '15 at 19:29
  • $\begingroup$ @A.Lovelace - yes, it is. $\endgroup$ – Mauro ALLEGRANZA Sep 26 '15 at 19:51
  • $\begingroup$ thank you for your assistance, sir $\endgroup$ – A. Lovelace Sep 26 '15 at 19:57

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.