Every Relation Is Just a Subset of A × B

The word "relation" sounds grand. It hints at something abstract — a hidden link, a bond, a rule connecting one set to another. That mystique is exactly what derails students. Once you pull the word down to its mathematical skin, every theorem, every property test, every JEE question becomes mechanical. A relation from A to B is a subset of A \times B — literally nothing more than a checklist of the ordered pairs you have decided to keep.

The one-line reframe

Write this on the inside of your notebook:

R \subseteq A \times B

That is the entire definition. A relation is a set. The elements of that set happen to be ordered pairs. You are allowed to pick any ordered pairs you like from A \times B, and whichever bundle you pick is the relation.

Why this framing helps: it replaces a vague idea ("some pairs are related") with a concrete object (a subset). Subsets you already know. You can list their elements, count them, take unions, check membership. Every question about R becomes a question about a set — and set questions have crisp answers.

A relation you already know, rewritten

Take A = \{1, 2, 3\} and the relation "a is less than b" on A. Most students picture an arrow. The subset view writes it flat:

R = \{(1, 2),\ (1, 3),\ (2, 3)\}

Three ordered pairs. That is the whole relation. Want to know if 2 is related to 1? Check if (2, 1) \in R. It is not. You have your answer in one lookup — no philosophy, no "less than" reasoning. Everything the relation is, is sitting in that list.

Visualising the subset

Three shaded cells. That is the whole relation. Every unshaded cell is an ordered pair that $R$ did not include — it is in $A \times B$, but not in $R$.

The grid makes the subset view unmistakable. A \times B gives you the full 3 \times 3 = 9 cells. Picking a relation means picking a subset of those cells. You can pick any subset, including the empty one (no pairs — the empty relation) or all nine (the universal relation).

Why this collapses property tests

Every property test for a relation becomes a question about the subset. No abstract reasoning required — only checking membership.

Reflexive? Is every pair (a, a) with a \in A in the subset? (For our example, check (1,1), (2,2), (3,3) against the list. None are in R, so not reflexive.)
Symmetric? For every (a, b) in the subset, is (b, a) also in it? (Pick (1, 2) \in R; is (2, 1) \in R? No. So not symmetric.)
Transitive? For every chain (a, b), (b, c) in the subset, is (a, c) also in it? (Pick (1, 2) and (2, 3); is (1, 3) \in R? Yes. You would need to check all such chains, but for this tiny example transitivity holds.)

Each test is a question about the list. No intuition, no philosophy — just reading the list and checking.

Counting relations becomes easy

If A has n elements and B has m elements, then A \times B has nm ordered pairs. A relation is a subset, and the number of subsets of a set with nm elements is 2^{nm}. Done.

For a relation on a single set A (where B = A), |A \times A| = n^2, so there are exactly 2^{n^2} relations. A set of size 3 has 2^9 = 512 possible relations. That number looked mysterious before the subset view; now it is the obvious count of subsets.

The "rule" and "list" are the same thing

When a JEE problem writes R = \{(a, b) : a \mid b,\ a, b \in \{1, 2, 3, 4\}\} (divides), your first move is to expand it into a list:

R = \{(1,1), (1,2), (1,3), (1,4), (2,2), (2,4), (3,3), (4,4)\}

The "rule" is a compact description of the list. Expanding it makes the list visible so you can test properties on it directly. For small domains, always expand. For larger domains, expand a sample and test the rule abstractly — but the subset view is still the foundation.

translate the rule into a subset

Problem: On A = \{1, 2, 3\}, define R = \{(a, b) : a + b \text{ is even}\}. Write R as a subset and test reflexivity.

Step 1 — expand to a list. Which pairs (a, b) \in A \times A have a + b even? Check all nine:

(1,1): 2 ✓
(1,2): 3 ✗
(1,3): 4 ✓
(2,1): 3 ✗
(2,2): 4 ✓
(2,3): 5 ✗
(3,1): 4 ✓
(3,2): 5 ✗
(3,3): 6 ✓

So R = \{(1,1), (1,3), (2,2), (3,1), (3,3)\}. Five pairs out of nine.

Step 2 — test reflexivity. Need (1,1), (2,2), (3,3) all in R. Look at the list: yes, all three are there. Reflexive ✓.

The rule disappeared the moment we wrote the subset. Every question is now a lookup.

The takeaway habit

Whenever you see "relation," whisper "subset of A \times B" to yourself. Then write the subset out (or at least mentally picture the grid with shaded cells). From that point on, you are doing set arithmetic, not abstract reasoning. Property checks become lookups. Counting relations becomes counting subsets. Union, intersection, and complement of relations are just the same operations on sets. The scariness of "relation" evaporates when you remember it is just a list of allowed ordered pairs.