3.128 min read

The Collapse Theorem

The Collapse Theorem for square matrices is really a consequence of counting: for $A \in \mathbb{R}^{n \times n}$ , we have $n$ inputs and $n$ outputs. If the map doesn't collapse (injective), it must cover everything (surjective) — because a faithful map on a finite-dimensional space from itself to itself can't "miss" anything.

More precisely: $\text{Rank}(A) + \text{Nullity}(A) = n$ . If $\text{Nullity}(A) = 0$ , then $\text{Rank}(A) = n = m$ , so the map is also surjective. And vice versa.

This is analogous to: a function from a finite set to itself is injective iff it is surjective (pigeonhole principle). Linearity and finite-dimensionality make this work.

Formal View

Corollary 3.12

For

A \in \mathbb{R}^{n \times n}

\text{Rank}(A) = n \Longleftrightarrow \text{Nullity}(A) = 0 \Longleftrightarrow A

is bijective.

Why This Matters

The Collapse Theorem dramatically simplifies work with square matrices — checking one property gives all.

To verify a square system $A\mathbf{x} = \mathbf{b}$ has a unique solution for all $\mathbf{b}$ : just check the null space
Matrix invertibility (Chapter 4) is exactly bijectivity of the associated map

Learning Resources

The Invertible Matrix Theorem

3Blue1Brown

All the equivalent conditions for invertibility unified.

10 min

Quiz

Question 1

For a square matrix $A$ , $\text{Nullity}(A) = 0$ implies:

Common Mistakes

Applying "injective implies surjective" to rectangular matrices — only valid for square matrices.