A relation is the formal, mathematical name for what you already know informally as a table — a set of rows that all share the same set of columns.
Why This Matters
- Every rule you'll learn from here on is really a rule about relations, even when the word "table" is used instead.
- Knowing that "table" and "relation" point at the same thing means textbook definitions and everyday SQL talk are describing the same structure, just in different registers.
Core Concept
- A Relation is a set of tuples (rows) that all share the same set of attributes (columns).
- Formally, it comes from set theory: a relation is a subset of the Cartesian product of one or more domains.
- In practice, "relation" and "table" are used interchangeably — relation is the formal/academic term, table is the everyday/product term.
Deep Dive
- Because a relation is technically a set, strict theory says it can't contain duplicate rows and has no inherent row order — real SQL tables relax both of these (they allow duplicates and preserve insertion-adjacent order unless you sort), which is one of the small but real gaps between the pure relational model and how SQL actually works.
- Every relation has two fixed properties that don't depend on its current data: its degree (number of columns) and its schema (the column names and types) — you'll meet both formally in upcoming chapters.
Common Mistakes
Assuming "relation" means a relationship between two tables (like a foreign key link).
Why: This is the single most common mix-up with this term — "relation" in relational-database theory means a table itself, not a connection between tables.
Fix: Mentally substitute "table" every time you see "relation" in a formal definition — they mean the same thing.
Interview Questions
What is a relation in the relational model?
A set of rows (tuples) that all share the same set of columns (attributes) — it's the formal term for what's informally called a table.
Why might a strict definition of "relation" say it can't contain duplicate rows, when real SQL tables clearly can?
Because a relation is formally a set, and sets can't contain duplicate elements by definition. Real SQL relaxes this rule for practical reasons, which is one of several places SQL doesn't perfectly match the pure relational model it's based on.
Best Practices
Do
✓Use "table" and "relation" interchangeably once you know they mean the same thing.
✓Remember a relation's degree (columns) and schema are fixed by design, while its rows change constantly.
Don't
✗Don't confuse "relation" (a table) with "relationship" (a connection between tables) — they sound alike but mean different things.
Chapter Summary
A relation is the formal term for a table — a set of rows sharing the same columns. The word comes from set theory, which is why strict definitions of it can feel oddly mathematical compared to how SQL tables actually behave.