Chapter 35 — Further Reading

This is the chapter where abstract textbooks are at their best, because the entire point is to work with linear maps before — and often without — coordinates. Axler's Linear Algebra Done Right is the definitive companion here: it is built from the ground up around linear maps rather than matrices, which is exactly this chapter's thesis.

Core textbooks

  • Sheldon Axler, Linear Algebra Done Right (4th ed.), Chapter 3 ("Linear Maps") — the primary companion to this chapter. Axler develops linear maps, the matrix of a linear map relative to chosen bases, null space (kernel) and range (image), and the Fundamental Theorem of Linear Maps — his name for what we called Rank–Nullity — entirely coordinate-free, deferring matrices until the structure is clear. His treatment is the natural next read after §35.6–§35.8, and his deliberate avoidance of determinants until the very end makes the operator-versus-matrix distinction unusually crisp. If you read one source on this chapter, read this; it is the book whose philosophy ours shares most closely. Chapter 3 also covers products and quotients of vector spaces and the algebra of operators, extending §35.3.

  • Gilbert Strang, Introduction to Linear Algebra (5th/6th ed.). Strang stays close to matrices and $\mathbb{R}^n$, but his sections on the four fundamental subspaces and on linear transformations are the concrete grounding this chapter generalizes — especially valuable for seeing that kernel/image are the null/column spaces you already know. His chapter on linear transformations explicitly builds the matrix of a transformation from the images of basis vectors, exactly our §35.4. Best read as the $\mathbb{R}^n$ anchor before climbing to the abstraction.

  • Stephen Boyd & Lieven Vandenberghe, Introduction to Applied Linear Algebra (VMLS). The most application-forward of the three and freely available online. Its treatment of linear functions, the matrix-of-a-map idea, and especially its many engineering examples make the operator viewpoint concrete; the material on linear dynamical systems is the applied face of Case Study 1, and its coverage of features/embeddings connects to how encoders and maps appear in data science.

On the abstract / proof-based treatment

  • Friedberg, Insel & Spence, Linear Algebra (5th ed.), Chapter 2 ("Linear Transformations and Matrices"). The standard rigorous undergraduate development: linearity, the matrix representation $[T]_\beta^\gamma$ (their notation for our $[T]_{C\leftarrow B}$), composition and matrix multiplication, invertibility and isomorphisms, and the change-of-coordinates matrix. Its careful, theorem-proof style is ideal if you want every claim in §35.4–§35.8 stated and proved in full generality, including the isomorphism theorems.

  • Paul Halmos, Finite-Dimensional Vector Spaces. A classic, elegant, and demanding account of exactly the coordinate-free viewpoint of this chapter — linear transformations, their matrices, kernels and ranges, and the structure of operators — written by a master expositor. The book that taught generations of mathematicians to see the operator before the matrix. Read for the prose and the perspective; it pairs naturally with Axler.

  • Kenneth Hoffman & Ray Kunze, Linear Algebra (2nd ed.). A rigorous standard reference; its chapters on linear transformations and on the algebra of polynomials in an operator are the right preparation for the operator-as-polynomial idea (Case Study 1) and for the Jordan-form theory of Chapter 36.

On the differentiation-operator and differential-equations connection (Case Study 1)

  • Gilbert Strang, Differential Equations and Linear Algebra. Strang's dedicated text develops precisely the bridge of Case Study 1: linear differential operators, the solution space as a kernel, particular-plus-homogeneous as preimage-plus-kernel, and the matrix exponential. The ideal next step toward Chapter 37, and it makes the operator view of $\frac{d}{dx}$ explicit throughout. See also the derivative as an operator for the calculus foundation.

  • Any standard ODE text (e.g., Boyce & DiPrima, Elementary Differential Equations). For the classical mechanics of finding particular and homogeneous solutions; read it knowing that "the homogeneous solution space" is the kernel of a linear operator and "two arbitrary constants" is its dimension — the chapter's reframing makes the standard methods feel inevitable.

On linear codes (Case Study 2)

  • Richard Hamming, Coding and Information Theory, and the original 1950 paper "Error Detecting and Error Correcting Codes." The source of the Hamming codes that Case Study 2 models. Hamming's construction of generator and parity-check matrices is linear algebra over the binary field, and the syndrome decoding we described is reading a kernel — a beautiful, historically pivotal application of this chapter's ideas.

  • MacWilliams & Sloane, The Theory of Error-Correcting Codes. The comprehensive reference, for readers who want the full algebraic theory of linear codes as subspaces (images of generator maps / kernels of parity-check maps). Graduate level; the first chapters suffice to see the Chapter 35 structure in full.

Free online resources

  • MIT OpenCourseWare 18.06 (Strang), lectures on linear transformations and the four subspaces. Video development of the matrix-of-a-transformation idea and rank–nullity, with Strang's characteristic geometric emphasis.
  • 3Blue1Brown, Essence of Linear Algebra, especially "Linear transformations and matrices" and "Change of basis." The single best visual intuition for "a matrix is a transformation in coordinates" and for how the matrix changes with the basis — the animated version of §35.4 and §35.7.
  • Axler's Linear Algebra Abridged (free PDF from the author) covers the linear-maps material of this chapter if you want to sample the full text before buying.

Where to go next in this book

Chapter 36 (Jordan Normal Form) studies the operators whose matrices cannot be diagonalized in any basis — the nilpotent $D$ of this chapter is the prototype — and repairs them with generalized eigenvectors. Chapter 37 (the matrix exponential) is differentiation's grandest application, solving $\mathbf{x}'=A\mathbf{x}$; and Chapter 40 surveys functional analysis, where the linear maps of this chapter become operators on the Hilbert spaces of Chapter 34, and a differential equation is once more a question about kernels and images.