- Practice
- About 19 min
- 5 practice questions
Properties of binary operations: commutative, associative, identity and inverse
Test a binary operation for the commutative and associative properties, find its identity element and inverses when they exist, and prove that an identity and an inverse are unique.
If something looks wrong, please tell us.
By the end of this lesson you can
- State the commutative, associative, identity and inverse properties.
- Verify or disprove each property for a given operation.
- Find the identity element and the inverse of an element by solving equations.
- Prove that the identity element and inverses are unique.
Ordinary addition of integers has several pleasant features: order does not matter, grouping does not matter, 0 changes nothing, and every integer can be undone by its negative. Abstract algebra asks which operations share these features. Checking them carefully is the main skill of this part of the chapter.
The four properties
- Commutative: a ∗ b = b ∗ a for all a, b ∈ S.
- Associative: (a ∗ b) ∗ c = a ∗ (b ∗ c) for all a, b, c ∈ S.
- Identity: there is e ∈ S with a ∗ e = e ∗ a = a for all a ∈ S.
- Inverse: given an identity e, for each a ∈ S there is b ∈ S with a ∗ b = b ∗ a = e. Then b is written a⁻¹, read "the inverse of a", not 1/a.
For ℚ under ×, the identity is 1 (n·1 = 1·n = n) and the inverse of 2 is 1/2; for ℤ under +, the identity is 0 and the inverse of m is −m. Remember: to prove a property you must show it for every element; to disprove it, one example is enough.
Checking an operation step by step
- Closure Is a ∗ b always in S?
- Commutative Compare a ∗ b with b ∗ a.
- Associative Expand (a ∗ b) ∗ c and a ∗ (b ∗ c).
- Identity Solve a ∗ e = a for e, for every a.
- Inverse Solve a ∗ b = e for b; is b in S?
Uniqueness of identity and inverse
Theorem (identity is unique). Suppose e₁ and e₂ are both identities in (S, ∗). Since e₁ is an identity, e₁ ∗ e₂ = e₂. Since e₂ is an identity, e₁ ∗ e₂ = e₁. Hence e₁ = e₂.
Theorem (inverse is unique, for an associative operation). Suppose a has inverses a₁ and a₂, so a ∗ a₁ = a₁ ∗ a = e and a ∗ a₂ = a₂ ∗ a = e. Then a₁ = a₁ ∗ e = a₁ ∗ (a ∗ a₂) = (a₁ ∗ a) ∗ a₂ = e ∗ a₂ = a₂. The middle step regroups, which uses associativity.
An identity, if it exists, is unique; so is each inverse (the proof regroups, using associativity). To disprove commutativity or associativity, one counterexample is enough.
Worked example 1: a ∗ b = a + b − 3 on ℤ
- Closure: a + b − 3 is an integer.
- Commutative: a ∗ b = a + b − 3 = b + a − 3 = b ∗ a.
- Associative: (a ∗ b) ∗ c = (a + b − 3) + c − 3 = a + b + c − 6, and a ∗ (b ∗ c) = a + (b + c − 3) − 3 = a + b + c − 6. Equal.
- Identity: a ∗ e = a means a + e − 3 = a, so e = 3 ∈ ℤ. Check 3 ∗ a = 3 + a − 3 = a.
- Inverse: a ∗ a′ = 3 means a + a′ − 3 = 3, so a′ = 6 − a ∈ ℤ. For example the inverse of 5 is 1, since 5 ∗ 1 = 5 + 1 − 3 = 3.
All five properties hold, just as for + on ℤ.
Worked example 2: subtraction on ℤ
Closure holds. Commutative fails: 4 − 5 = −1 but 5 − 4 = 1. Associative fails: (4 − 5) − 7 = −8 but 4 − (5 − 7) = 6. An identity would need a − e = a and e − a = a for every a; the first gives e = 0, but then 0 − a = −a ≠ a when a ≠ 0. So there is no identity, and without an identity there can be no inverses.
Worked example 3: a ∗ b = a + 2b on ℤ
1 ∗ 2 = 1 + 4 = 5 but 2 ∗ 1 = 2 + 2 = 4, so ∗ is not commutative. (a ∗ b) ∗ c = (a + 2b) + 2c = a + 2b + 2c, while a ∗ (b ∗ c) = a + 2(b + 2c) = a + 2b + 4c; with c = 1 these differ, so ∗ is not associative.
Worked example 4: m ∗ n = m + n + mn on ℤ
Closure and commutativity are clear. Associative: (m ∗ n) ∗ p = (m + n + mn) + p + (m + n + mn)p = m + n + p + mn + mp + np + mnp, and expanding m ∗ (n ∗ p) gives the same symmetric expression, so ∗ is associative. Identity: m + e + me = m gives e(1 + m) = 0 for all m, so e = 0. Inverse: m + m′ + mm′ = 0 gives m′ = −m/(1 + m). This is an integer only for m = 0 (m′ = 0) and m = −2 (m′ = −2), and m = −1 has no inverse at all. So the inverse property fails on ℤ.
Worked example 5: multiplication on the even integers
The product of two even integers is even, so the set E of even integers is closed under ×, and × is commutative and associative. But an identity would have to be 1, and 1 ∉ E. So E under × has no identity, and therefore no inverses.
| Operation | Closure | Commutative | Associative | Identity | Inverse |
|---|---|---|---|---|---|
| + on ℤ | Yes | Yes | Yes | 0 | −a |
| − on ℤ | Yes | No | No | None | None |
| a + b − 3 on ℤ | Yes | Yes | Yes | 3 | 6 − a |
| a + 2b on ℤ | Yes | No | No | None | None |
| a + b + ab on ℤ | Yes | Yes | Yes | 0 | Only for 0 and −2 |
| × on even integers | Yes | Yes | Yes | None | None |
Key takeaways: check closure first, then commutativity and associativity by expanding both sides, then solve a ∗ e = a for an identity that works for every a, then solve a ∗ a′ = e and check that a′ lies in the set. Identity and inverses are unique when they exist. A single counterexample disproves any "for all" property.
Check your understanding
-
1. On ℤ, a ∗ b = a + b − 3. What is the identity element?
- 0
- 3
- −3
- 1
Show answer
Answer: B. a ∗ e = a means a + e − 3 = a, so e = 3. Check: 3 ∗ a = 3 + a − 3 = a.
-
2. On ℤ, a ∗ b = a + b − 3. What is the inverse of 5?
- −5
- 3
- 1
- 6
Show answer
Answer: C. The identity is 3, so we need 5 + b − 3 = 3, giving b = 1. Check: 5 ∗ 1 = 5 + 1 − 3 = 3.
-
3. On ℤ, a ∗ b = a + 2b. What are 1 ∗ 2 and 2 ∗ 1?
- 5 and 4
- 4 and 5
- 5 and 5
- 3 and 3
Show answer
Answer: A. 1 ∗ 2 = 1 + 2(2) = 5 and 2 ∗ 1 = 2 + 2(1) = 4. Since 5 ≠ 4, the operation is not commutative.
-
4. In an algebraic structure (S, ∗), how many identity elements can there be?
- Exactly two
- At most one
- Any number
- Always at least one
Show answer
Answer: B. If e₁ and e₂ were both identities, then e₁ = e₁ ∗ e₂ = e₂. So an identity, if it exists, is unique; but it need not exist (subtraction on ℤ has none).
-
5. Which property fails for ordinary multiplication on the set of even integers?
- Closure
- Commutativity
- Associativity
- Existence of identity
Show answer
Answer: D. Products of even integers are even, and × is commutative and associative. The only possible identity is 1, which is not even, so there is no identity in the set.
Sign in to keep track of the lessons you finish.