I beg your pardon for poor editing, I'm just Latex beginner. This exercise comes from book on set theory.

Prove that for any finite set A with any partial order $r$, there exists $X\subset N \setminus \{0\}$ with divisibility relation | , such that sets $\langle A ,r\rangle$ and $\langle X ,|\rangle$ are isomorphic.

The answer book suggests defining mapping $a \mapsto p(a)$ and then a function :

$f\colon A \rightarrow X$

$f(a) = \prod_{a\leq x}\ p(x)$

It suffices to show that this function establishes an isomorphism in between set $A$ and that is clear.


I'd like to ask why can't we use more trivial mapping. My first idea was to use the following formula: $a \mapsto 2^a$ and then define a function based on the product in the same way they did in the book. As far I can see set of powers of $2$ does the same job as the set of primes. On the other hand I suppose similiar answer would show up in the answer key if it were correct. Could someone tell me what I'm doing wrong? Is it because we are not told that $A$ consists of integers?

  • $\begingroup$ If you click edit, you'll be able to see some of the changes I made to your typesetting. Hopefully you can see how it looks a bit more professional now. $\endgroup$ – Dan Rust Feb 4 '15 at 18:13
  • $\begingroup$ Thanks, it's much better now. Couple more days and I'll get better :) $\endgroup$ – Pasato Feb 4 '15 at 18:17
  • $\begingroup$ yeah it's easy to get to grips with after you've used it a few times. You tend to learn little tricks along the way, like how \colon has better spacing than if you just use : Such as $f\colon X\to Y$ is better spaces than $f:X\to Y$. $\endgroup$ – Dan Rust Feb 4 '15 at 18:19

Of course the $a's$ need not be integers, but even if you mapped $a$s to integers, say $a\mapsto g(a)$, with the mapping $a\mapsto 2^{g(a)}$, the corresponding divisibility order is a total order, whereas a general partial order need not be totally ordered.

| cite | improve this answer | |
  • $\begingroup$ Thanks. That would mean that my mapping makes every two elements of A comparable whereas it doesn't have to be a case? $\endgroup$ – Pasato Feb 4 '15 at 18:14
  • $\begingroup$ Or in other words : it would be correct if we were told there was a total order on A ? $\endgroup$ – Pasato Feb 4 '15 at 18:20
  • $\begingroup$ Yes, I think it could work if there were a total order on $A$. $\endgroup$ – paw88789 Feb 4 '15 at 18:24

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.