For a relation on a single set A, the most revealing picture is the directed graph. Draw one node for each element of A. Draw one arrow from a to b for each pair (a, b) \in R. That is the entire construction. What makes this picture powerful is that every major property of the relation — reflexive, symmetric, transitive, antisymmetric — becomes a visual pattern you can recognise at a glance.

The picture

Take A = \{1, 2, 3, 4\}. A relation on A is any subset of A \times A = \{ all ordered pairs of elements from A \} — sixteen possible pairs in this case. The directed graph shows:

Directed-graph slider for relation propertiesA horizontal slider runs from zero to four, selecting one of five sample relations on a four-element set. Above the slider, four nodes are arranged at the corners of a square and various arrows and self-loops appear between them to depict the current relation. A readout at the top names the current case and the caption below each case describes which properties hold. 0 2 4 drag to change the relation 1 2 3 4
A directed graph of the "same parity" relation on $A = \{1, 2, 3, 4\}$. Nodes $1$ and $3$ (both odd) have self-loops and two-way arrows between them. Nodes $2$ and $4$ (both even) have self-loops and two-way arrows between them. Odd and even nodes are not connected, because $1$ and $2$ do not have the same parity. The picture shows at a glance that the relation is reflexive (every self-loop present), symmetric (every arrow has a reverse twin), and transitive (every two-hop chain has its shortcut). It is an equivalence relation, partitioning $A$ into two blocks.

How each property reads off the graph

Reflexive: every node has a self-loop.

Irreflexive: no node has a self-loop.

Symmetric: for every straight arrow a \to b (with a \neq b), the reverse arrow b \to a is also present. Self-loops are trivially symmetric.

Antisymmetric: no two distinct nodes have two-way arrows between them. An arrow from a to b excludes the reverse unless a = b.

Transitive: for every two-arrow path a \to b \to c, the shortcut arrow a \to c is also drawn. Self-loops participate in chains but never break them.

Equivalence relation: reflexive + symmetric + transitive. Visually: every self-loop drawn; every arrow has its reverse; every chain closes. The graph breaks into disjoint "cliques" — groups of nodes with complete back-and-forth arrows — one clique per equivalence class.

Partial order: reflexive + antisymmetric + transitive. Visually: self-loops everywhere, no two-way arrows between distinct nodes, and every chain closes. The graph looks like a hierarchy with a clear direction.

Five sample cases the slider walks through

Case 0 — empty relation. No arrows at all. Irreflexive, symmetric (vacuous), antisymmetric (vacuous), transitive (vacuous). Not reflexive (no self-loops), hence not an equivalence relation.

Case 1 — identity relation \{(1,1), (2,2), (3,3), (4,4)\}. Only self-loops. Reflexive, symmetric, antisymmetric, transitive. The smallest equivalence relation; every element is its own class.

Case 2 — same parity (the picture shown). Two cliques (\{1, 3\} and \{2, 4\}), each with self-loops and two-way arrows. Equivalence relation. Partitions A into odds and evens.

Case 3 — strict less-than, \{(1,2), (1,3), (1,4), (2,3), (2,4), (3,4)\}. Arrows only in one direction. Irreflexive (no self-loops), antisymmetric, transitive. Not a partial order because reflexivity fails — this is a strict partial order.

Case 4 — full relation A \times A. Every possible arrow. Reflexive, symmetric, transitive. An equivalence relation (everyone is in one big class). Also not antisymmetric (since two-way arrows exist between distinct nodes).

What the graph emphasises that the list hides

When a relation is given as a list of pairs, properties like symmetry and transitivity can be tedious to check — you scan pair by pair. The directed graph compresses the information spatially, so patterns pop:

Why: a directed graph lays out every pair in two dimensions rather than stringing them in a list. Patterns that are global (symmetry, transitivity) become local (twin arrows, closed triangles) when drawn spatially. The graph is the relation, rendered in a form that foregrounds its structure.

Conventions for drawing

For small sets (|A| \leq 6 or so), arrange the nodes in a circle or at the corners of a polygon. This keeps arrows from tangling and makes chains easier to trace.

For larger sets, or for relations with heavy structure, the reduction diagram (Hasse diagram for partial orders) drops redundant arrows — specifically, every transitive shortcut is erased and every self-loop is implicit. This is a cleaner picture for posets, but you lose the ability to read reflexivity and transitivity visually. Use the full directed graph when studying properties; use the Hasse diagram when the relation is a known partial order and you want to see its hierarchy.

Arrows are typically drawn as curves when there are two between the same pair of nodes (one for each direction), with arrowheads near the target. Self-loops are small circles or teardrop shapes attached to a single node.

Spotting properties — a walk-through

Take the relation R = \{(1,1), (2,2), (1,2), (2,1), (1,3)\} on A = \{1, 2, 3\}. Draw the graph:

None of the three big properties hold. This relation fits no named category — it is just a generic relation.

Contrast with R = \{(1,1), (2,2), (3,3)\}: all three self-loops, no other arrows. Reflexive. Symmetric (vacuously, no non-loop arrows to check reverses for). Transitive (vacuously). Equivalence relation. The identity relation.

Why the graph picture generalises

Everything you learn from drawing directed graphs of small relations transfers directly to:

The habit of reading properties off pictures, trained on small relations, generalises to every domain where "which things relate to which" is the central question. It is a core visual literacy for modern mathematics and computing.

The reflex

Once drawn, every major property is visible in under thirty seconds for small sets. For JEE-level questions, small sets are the norm, and the graph picture is routinely the fastest path to the answer.

Related: Relations · Reflexivity Tester: Missing Self-Loops Glow Red · Symmetry Check: Missing Reverse Arrows Glow Red · Transitivity Chain Finder