Cancellation in Vector Addition

Why Cancellation Needs a Proof

For real numbers, if \(x+a=b\), we solve by subtracting \(a\). The same idea works in a vector space, but its justification comes from the vector space axioms: addition is associative and commutative, there is a zero vector, and every vector has an additive inverse. These properties let us construct a solution and prove that no second solution is possible.

Throughout, let \(V\) be a vector space over \(\mathbb{R}\). Its elements are vectors, its zero vector is \(\mathbf{0}\), and the additive inverse of \(\mathbf{w}\in V\) is written \(-\mathbf{w}\). Thus \(\mathbf{w}+(-\mathbf{w})=\mathbf{0}\). The vector space axioms also give \((\mathbf{u}+\mathbf{v})+\mathbf{w}=\mathbf{u}+(\mathbf{v}+\mathbf{w})\) and \(\mathbf{u}+\mathbf{v}=\mathbf{v}+\mathbf{u}\). We will use these facts, along with the identities for scalar multiplication, to reason about vector equations.

Cancellation means that a common addend can be removed from both sides of an equality. The crucial condition is that the same vector is added to both sides. We will prove this directly from the axioms, rather than assuming that cancellation is an additional vector space property.

The Cancellation Law

Proof. Assume first that \(\mathbf{u}+\mathbf{w}=\mathbf{v}+\mathbf{w}\). Add \(-\mathbf{w}\) to both sides; equality is preserved when the same vector is added to equal vectors. By associativity and the additive-inverse axiom,

The additive-identity axiom gives \(\mathbf{u}=\mathbf{v}\). Now assume \(\mathbf{w}+\mathbf{u}=\mathbf{w}+\mathbf{v}\). Commutativity rewrites this as \(\mathbf{u}+\mathbf{w}=\mathbf{v}+\mathbf{w}\), so the first part gives \(\mathbf{u}=\mathbf{v}\). Thus cancellation holds whether the common addend is on the right or the left. \(\square\)

The conclusion concerns the vectors being added, not the common addend. For example, if \(\mathbf{x}+\mathbf{w}=\mathbf{y}+\mathbf{w}\), cancellation gives \(\mathbf{x}=\mathbf{y}\). The proof works because \(\mathbf{w}\) occurs on both sides in the same way. It does not license removing different vectors from opposite sides of an equation.

Solving an Unknown on Either Side

There are two common forms: \(\mathbf{x}+\mathbf{a}=\mathbf{b}\), where the unknown is on the left of the known addend, and \(\mathbf{a}+\mathbf{x}=\mathbf{b}\), where it is on the right. The additive inverse tells us what to add to undo \(\mathbf{a}\). Associativity then places \(\mathbf{a}\) next to its inverse, while the zero vector leaves the unknown unchanged.

Proof. For the first equation, set \(\mathbf{x}=\mathbf{b}+(-\mathbf{a})\). Substitution gives

So this vector is a solution. If \(\mathbf{x}_1\) and \(\mathbf{x}_2\) are both solutions, then \(\mathbf{x}_1+\mathbf{a}=\mathbf{b}=\mathbf{x}_2+\mathbf{a}\). Additive cancellation gives \(\mathbf{x}_1=\mathbf{x}_2\), proving uniqueness.

For the second equation, set \(\mathbf{x}=(-\mathbf{a})+\mathbf{b}\). Then

Thus this vector is a solution. If \(\mathbf{x}_1\) and \(\mathbf{x}_2\) are both solutions, commutativity gives \(\mathbf{x}_1+\mathbf{a}=\mathbf{b}=\mathbf{x}_2+\mathbf{a}\) from \(\mathbf{a}+\mathbf{x}_1=\mathbf{b}=\mathbf{a}+\mathbf{x}_2\). Cancellation gives \(\mathbf{x}_1=\mathbf{x}_2\). Finally, \(\mathbf{b}+(-\mathbf{a})=(-\mathbf{a})+\mathbf{b}\) by commutativity, so both forms have the same solution. \(\square\)

This proof establishes two separate claims for each equation. Substitution proves existence: the proposed vector really does solve the equation. Cancellation proves uniqueness: if any vector solves it, that vector must equal the one already found. A candidate alone is not a uniqueness argument; the cancellation step supplies that part.

Worked Solutions in an Abstract Vector Space

Common Pitfall: Cancel Only a Common Addend

A frequent error is to treat terms on opposite sides as though they were a common addend. For a concrete example, take a vector \(\mathbf{u}\ne\mathbf{0}\). Then

The equality is valid because both sides equal \(\mathbf{u}\). It does not permit cancelling \(\mathbf{u}\) from the left side against \(\mathbf{0}\) on the right: those are different addends. The cancellation theorem applies to \(\mathbf{x}+\mathbf{w}=\mathbf{y}+\mathbf{w}\), or to \(\mathbf{w}+\mathbf{x}=\mathbf{w}+\mathbf{y}\), with the same \(\mathbf{w}\) on both sides.

A related mistake is to use \(\mathbf{a}\) rather than \(-\mathbf{a}\) to undo addition by \(\mathbf{a}\). If \(\mathbf{u}\ne\mathbf{0}\), the equation \(\mathbf{x}+2\mathbf{u}=5\mathbf{u}\) does not have \(\mathbf{x}=7\mathbf{u}\): substituting gives \(7\mathbf{u}+2\mathbf{u}=9\mathbf{u}\), which is not \(5\mathbf{u}\) when \(\mathbf{u}\ne\mathbf{0}\). The correct inverse is \(-2\mathbf{u}\), so \(\mathbf{x}=5\mathbf{u}-2\mathbf{u}=3\mathbf{u}\); checking gives \(3\mathbf{u}+2\mathbf{u}=5\mathbf{u}\).

Why the Axioms Matter

These results are basic, but they are useful whenever a vector equation must be solved without choosing coordinates. The same reasoning works for every real vector space because it uses only the addition axioms: it does not depend on what the vectors are or how they are represented. The two equation forms also clarify the role of commutativity. The additive inverse handles the known addend, associativity brings it together with its inverse, and commutativity lets us pass between left and right forms. Cancellation then turns a verified solution into a unique solution.

When solving, keep the logic in order: identify the addend, use its additive inverse, simplify with the axioms, and substitute the result into the original equation. If uniqueness is required, compare two possible solutions and apply cancellation to the common addend. This distinguishes a correct candidate from a proof that it is the only candidate.