Example 19:
Let be a relation on the set of ordered pairs of positive integers defined by if and only if . Show that is an equivalence relation.
Reflexive: Let . Since , . Thus is reflexive.
Symmetric: Let . Then , so . Thus is symmetric.
Transitive: Let and . Then and . Multiplying,
So . Thus is transitive.
Hence is an equivalence relation.
Want the reason behind every step, the basics and the checks?
Because and are positive, we may divide by them: . So two pairs are related exactly when they give the same fraction: , and are all "one half". "Has the same value as" is obviously reflexive, symmetric and transitive. The proof below writes this out using only multiplication of integers, and the one place where care is needed is the transitive step, where we cancel a non-zero factor. That is where "positive integers" is used.
- Given
- The set of all ordered pairs with and positive integers, and the rule .
- To show
- is reflexive, symmetric and transitive on .
- Plan
- (1) Get a feel for the rule with a few pairs. (2) Reflexive: compare with itself. (3) Symmetric: rewrite the other way round. (4) Transitive: combine two equations and cancel a positive factor. (5) Conclude and describe the classes.
Basics you need
The set consists of ordered pairs such as . The relation relates one pair to another pair. So a typical element of is a "pair of pairs", . Basics: sets and ordered pairs
Reflexive: every element is related to itself. Symmetric: . Transitive: and . Here are pairs. Basics: reflexive, symmetric, transitive
It is commutative () and associative, and we may multiply both sides of an equation by the same number. If and , then (cancellation). Basics: integers
For non-zero : (cross-multiplication).
Each property is a "for all" statement, so it must be proved for arbitrary pairs written with letters, never just for examples. Basics: proving and disproving
Solution
Try the rule on a few pairs
and Related? and yes and yes and yes and no The related pairs are those with the same ratio: , , but .
The condition is the cross-multiplied form of (Basic 4). Seeing this first tells us what to expect, an equivalence relation, and what the classes will look like. The proof itself will use only the integer equation .Cross-multiplication: , valid when and .Reflexive:
Take any . To compare it with itself, put and in the rule. The condition becomes
which is true because multiplication of integers is commutative. So for every , and is reflexive.
The rule compares two pairs, and . Reflexivity compares a pair with itself, so the second pair is the same as the first: , . The resulting equation holds for all integers, so no particular pair is needed.Commutativity: for all integers (Basic 3).Symmetric: swap the two pairs
Suppose , so . We must show . By the rule (with the roles of the pairs swapped), that means . Now
reading the equation from right to left, then using commutativity on each side. So , and is symmetric.
The main care here is getting the target right. "" puts first, so in the rule "first-entry of the first pair times second-entry of the second pair equals second-entry of the first pair times first-entry of the second pair" we get . Once that is written down, it is the given equation rearranged.An equation can be read in either direction (), and , .Transitive: combine and cancel
Suppose and . Then
We want , that is . Multiply the two equations side by side (left times left, right times right):
Rearrange each side, using commutativity and associativity, so that the factor stands out:
Since and are positive integers, , so and we may cancel it:
Hence , and is transitive.
We need a relation between and only, so the middle pair's entries must be removed. Multiplying the two equations makes both and appear on each side, as the common factor ; cancelling it removes them. Cancelling is only allowed for a non-zero factor, and this is exactly where the question's word positive is used. The textbook does the same thing in another way: from it gets , multiplies by , replaces on the left by , and simplifies to . Both routes divide by and .Cancellation: if and , then (Basic 3). Multiplying equal quantities by equal quantities gives equal results.Conclude, and describe the classes
is reflexive, symmetric and transitive, so is an equivalence relation on .
By Step 1, means . So each equivalence class is the set of all pairs that represent one positive fraction. For example
Each dot is a pair with . Pairs in the same class have the same ratio, so they lie on one straight line through the origin: for example on the blue line . Different lines through the origin never share a dot, so the classes do not overlap. This is the real meaning of the relation: it treats and as "the same" because they describe the same fraction . In fact, this is one standard way of constructing the positive rational numbers from pairs of integers. The picture also shows the partition: every dot lies on exactly one line through the origin.An equivalence relation splits the set into disjoint classes. Basics: equivalence classes
is reflexive (), symmetric () and transitive (multiply by and cancel the positive factor to get ). Therefore is an equivalence relation on , as the textbook shows. Its classes are the sets of pairs with a common ratio .
From the table: and . Transitivity predicts : and . ✓ In the cancellation step, , , , , , : and , , and . ✓ The verification script also tests all three properties by brute force on every pair with entries from 1 to 12 (144 pairs, and for transitivity every chain , among them).
- Cancelling without checking the factor is non-zero. If 0 were allowed as an entry, transitivity would fail: since , and since , but is not related to since . Positivity is essential.
- Getting the target of the symmetric step wrong. means , not .
- Proving transitivity with fractions without saying why division is allowed. Writing is fine, but only because ; say so.
- Treating as a set of numbers. Its elements are pairs, and relates pair to pair.
- is "equal ratios" in disguise. Recognising the hidden meaning tells you what to expect before you prove it.
- To eliminate the middle element in a transitivity proof, combine the two equations so that its entries appear as a common factor, then cancel, after checking the factor is non-zero.
- Conditions such as "positive" in a question are there for a reason. Find the step that uses them.
Every fact used, and where it comes from
| Fact | Used in | Source |
|---|---|---|
| Reflexive, symmetric, transitive; equivalence relation | Steps 2–5 | Definitions 3 and 4, Section 1.2 (Lesson: types of relations) |
| Commutativity and associativity of multiplication | Steps 2–4 | Properties of integers |
| Cancellation of a non-zero factor | Step 4 | Properties of integers ( since ) |
| for | Steps 1, 5 | Equality of fractions |
| Equivalence classes partition the set | Step 5 | Section 1.2, after Example 5 (Lesson: equivalence relations) |
| Table values, the zero counterexample, brute-force checks | Step 1, Check, Mistakes | Recomputed in the site's verification script |
Tell us if something on this page is wrong or unclear, or if you would like a fuller explanation, another method or a new feature. Send feedback on this problem →