5
$\begingroup$

I'm reading up on "Set Theory and Logic" by Stoll and came upon notation for relations that I haven't seen before. I've seen $x\sim{y},$ and $xRy$ before but Stoll uses this one. $$(x,y)\in{\rho}$$ Now I admit that the prior two are specifically binary relations and I haven't seen ternary or n-ary relations like that. Is the advantage of Stoll's notation apparent through n-ary relations? An example of the ternary relational notation would be $$(3,5,8)\in{+}$$ where $+$ is addition. Or is this notation not really used? As I mentioned I don't recall seeing this written before but I was curious about its popularity

$\endgroup$
2
  • 1
    $\begingroup$ I'd say the advantage of that notation is that you are constantly reminded of what a relation IS (viz. a subset of S x S, for some set S). The elements are order pairs. $\endgroup$ – The Chaz 2.0 Sep 1 '13 at 5:26
  • $\begingroup$ I agree with you. I thought it was easier to see that as well but I'm surprised I've never seen it $\endgroup$ – Eleven-Eleven Sep 1 '13 at 5:29
6
$\begingroup$

Formally a relation from a set $X$ to a set $Y$ is a subset of $X\times Y$; if I call this subset $\rho$, the familiar notation $x\mathbin{\rho}y$ is an abbreviation for the more formal $\langle x,y\rangle\in\rho$. Since the most familiar binary relations are typically written with this infix notation (e.g., $x\le y$, $A\supseteq B$, etc.), the more informal $x\mathbin{\rho}y$ is often felt to be more intuitive, but it does obscure the set-theoretic nature of relations a bit.

In short, Stoll’s notation is entirely standard, and you’re likely to encounter it any time you have to deal with relations in the abstract rather than with specific relations.

Actually, $+$ on, say, the integers is a function from $\Bbb Z\times\Bbb Z$ to $\Bbb Z$. Functions are just a special kind of relation, so it’s a relation from $\Bbb Z\times\Bbb Z$ to $\Bbb Z$ and therefore a subset of $(\Bbb Z\times\Bbb Z)\times\Bbb Z$; formally one would write $\big\langle\langle 3,5\rangle,8\big\rangle\in+$, though there is a natural correspondence between $(\Bbb Z\times\Bbb Z)\times\Bbb Z$ and the set of ordered triples of integers.

$\endgroup$
2
  • $\begingroup$ I've only had the basic intro to set theory that every undergraduate gets in Discrete and Abstract/Analysis but this is my first attempt at Set Theory as a field of study. Do you recommend Stoll's book or is there a better one out there? $\endgroup$ – Eleven-Eleven Sep 1 '13 at 5:36
  • $\begingroup$ @Christopher: You could do a lot worse for an introduction, though as I recall elementary set theory is just part of Bob’s book. If you already have the book available, I’d use it as a starter and then move up to this, which is excellent and will take you much further into set theory proper. $\endgroup$ – Brian M. Scott Sep 1 '13 at 5:43
3
$\begingroup$

Relations are represented by sets of ordered pairs; so the advantage of this notation agrees with that representation. In practice it is not all that common - certainly outside of set theory you almost always see infix notation for binary relations. For ternary and higher order relations this gets less practical (I can't think of a single example of "infix-like" notation in a ternary relation right now) and you're more likely to see $R(x,y,z)$ or $(x,y,z) \in R$.

$\endgroup$
1
  • $\begingroup$ I agree with you as well. I tried to think of the infix notation for at least ternary and I drew a blank. I like the notation but suppose it's in it's home in set theory. $\endgroup$ – Eleven-Eleven Sep 1 '13 at 5:31

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.