Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

ZFC's axiom of infinity states:

$$\exists x (\varnothing \in x \wedge \forall y \in x (y\cup \left \{y \right \} \in x)) $$

Isn't this set $ x $ really $\mathbb{N}$? It wouldn't be $\mathbb{N}$ if x would contain some set $z $ that is different from $\varnothing $, $ S (\varnothing)$, $ S (S (\varnothing)) $, etc. But we can't really prove that such a set $ z $ is an element of $ x $, or can we? I also know that this inductive set $ x $ is not unique, but that, in the light of my question, doesn't make any sense to me either. I feel like I'm missing a very peculiar technical point, and I don't know what it is. Any help is appreciated.

share|cite|improve this question
@drhab With your $x$, $y:=\mathbb N$ is $\in x$ but $y\cup \{y\}$ is not. You do need a limit ordinal (or at least add something that starts at some $S\notin \mathbb N$) – Hagen von Eitzen Feb 15 '14 at 19:06
@HagenvonEitzen thank you, I will delete my comment. – drhab Feb 15 '14 at 19:18
Another example from NFU (plus Infinity): the universal set $V$ contains $\emptyset$ and forall $x$ it contains $x\cup\{x\}$. But the universe is hardly just the natural numbers. – Malice Vidrine Feb 15 '14 at 19:25
I'm no expert on ZF, but the axiom does not rule out the possibility of other "junk" being in $x$ in additional to elements of $N$. To obtain $N$, you must obtain the intersection of all sets like $x$. – Dan Christensen Mar 19 '14 at 21:32

Write $$\phi(x):\Leftrightarrow (\emptyset \in x\land \forall y\in x\,(y\cup\{y\}\in x)). $$ Then the Axiom of Infinity states $$\tag1 \exists I\,\phi(I).$$ From such a set $I$ you can derive (using the other axioms) the existence of $$\tag2 \omega := \bigcap\{\,x\in\mathcal P(I):\phi(x)\,\}$$ and then show that this $\omega$ is what you want (and does not depend on the choice of $I$). However, trying to pack the minimality construction of $(2)$ into the axiom makes it look much more complicated than $(1)$ - and unnecessarily so, as we can prove just what we need.

share|cite|improve this answer

Another set which satisfies the axiom is $N''=\mathbb N \cup N'$ where $N'$ is the set containing $x_0 = \mathbb N$ (a set which is not in $\mathbb N$) and all the sets $x_{n+1} := x_n \cup \{ x_n\}$. If you define $n+1 := n \cup \{n\}$, $0 := \{\}$ and $\omega := \mathbb N$ you have $$ N'' = \{0, 1, 2, \dots, \omega, \omega+1, \omega+2,\dots\}. $$

share|cite|improve this answer

Axiom INF tells you that some set $x$ exists with property $\phi\left(x\right)$ as defined in the answer of Hagen von Eitzen.

Applying axiom SEP you find that $\omega:=\left\{ y\in x\mid\forall I\left[\phi\left(I\right)\Rightarrow y\in I\right]\right\} $ is a set.

Here $\omega=\left\{ 0,1,2,\dots\right\} $

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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