Understanding 'Linear Algebra Done Wrong' by Sergei Treil: A Guide to Rigorous Linear Algebra

What is "Linear Algebra Done Wrong"?

When students first encounter the title Linear Algebra Done Wrong by Sergei Treil (Brown University), they often assume the book contains errors. In reality, the title is a provocative critique of how linear algebra is typically taught. Most introductory courses use a "cookbook" style—focusing on rote memorization of algorithms, such as how to perform row reduction or calculate a determinant, without explaining why these methods work or what they mean in a broader mathematical context.

Treil’s approach is "wrong" only in the sense that it deviates from the standard, calculation-heavy curriculum. Instead, it introduces students to rigorous mathematics. This means moving away from just finding the answer and moving toward providing formal definitions, constructing logical proofs, and understanding the abstract structures of vector spaces. For a student, this transition is the jump from being a calculator to being a mathematician.

The Mechanism: Moving from Calculation to Abstraction

To understand the framework of this approach, one must understand the three pillars of rigorous linear algebra: Vector Spaces, Linear Transformations, and Inner Product Spaces.

1. Vector Spaces and Bases

In a basic course, a vector is often just a list of numbers (a column). In Treil's rigorous approach, a vector space is defined by a set of axioms. A vector can be a function, a polynomial, or a matrix. The focus shifts to linear combinations and bases. A basis is not just a set of vectors, but the smallest set of vectors that can span the entire space while remaining linearly independent. Understanding this ensures that you can handle dimensions in any context, not just $\mathbb{R}^n$.

2. Linear Transformations as Mappings

Rather than seeing a matrix as a grid of numbers, this approach treats a matrix as a representation of a linear transformation. A transformation is a rule that moves a vector from one space to another while preserving the operations of addition and scalar multiplication. The "cookbook" method teaches you to multiply $Ax = b$; the rigorous method teaches you that $A$ is a map, and $x$ is being transformed into $b$.

3. Inner Product Spaces and Orthogonality

Rigorous linear algebra deepens the concept of "distance" and "angle" through inner product spaces. This allows for the development of the Gram-Schmidt process, which transforms a set of messy, linearly independent vectors into an orthonormal basis—a set of vectors that are all perpendicular to each other and have a length of one. This is the foundation for advanced topics like Least Squares solutions and Singular Value Decomposition (SVD).

Practical Example: Solving a Homogeneous System

Consider the problem of solving $Ax = 0$.

The Cookbook Way: You perform Gaussian elimination to get a Reduced Row Echelon Form (RREF), identify the free variables, and write down the general solution.

The Rigorous Way: You identify the Null Space (or Kernel) of the transformation. You recognize that the solution set is actually a subspace of the original vector space. By analyzing the pivots of the matrix, you are essentially determining the dimension of the kernel and the rank of the transformation. You aren't just finding $x$; you are describing the geometry of the space that the matrix "collapses" to zero.

Common Misconceptions and Pitfalls

• "Proofs are just for theorists": Many students believe that if they can get the right numerical answer, they understand the concept. This is a mistake. In higher-level math and physics, the structure of the solution is more important than the number.
• Confusing Matrices with Transformations: A common error is thinking that the matrix is the transformation. The matrix is merely a representation of the transformation relative to a specific basis. If you change the basis, the matrix changes, but the transformation remains the same.
• Over-reliance on $\mathbb{R}^n$: Students often struggle when asked to apply linear algebra to polynomials or matrices because they only visualize vectors as arrows in 3D space.

Actionable Steps for Mastering Rigorous Linear Algebra

If you are transitioning from a calculation-based course to a proof-based one, follow these steps:

1. Master the Definitions: Never start a proof without writing down the formal definition of the terms involved (e.g., "What exactly defines a linearly independent set?").
2. Visualize the Geometry: For every matrix operation, ask: "Is this stretching the space, rotating it, or collapsing it into a lower dimension?"
3. Verify via Small Examples: When attempting a formal proof, test your logic with a simple $2 \times 2$ matrix to see if the property holds.
4. Connect to Subspaces: Whenever you find a solution set, identify which fundamental subspace it belongs to (Column Space, Null Space, etc.).
5. Practice the Gram-Schmidt Process: Manually orthogonalize a set of three vectors to understand how projection works in higher dimensions.

Practice Questions for Self-Check

1. Conceptual: Why is a set of vectors that spans a space not necessarily a basis?
2. Application: If a linear transformation $T: V \to W$ is injective (one-to-one), what does that tell you about the null space of its matrix representation?
3. Computational: Given vectors $x_1 = (1, 1, 1)^T$ and $x_2 = (0, 1, 2)^T$, use the Gram-Schmidt process to find an orthogonal basis for the plane they span.
4. Theoretical: Explain the relationship between the rank of a matrix and the dimension of its image (column space).

For those who want to engage directly with the text or need guided help working through these complex proofs, Vidyora provides a platform to open the referenced book and chat with it, allowing you to ask for deeper examples or clarifications on specific theorems in real-time.