Properties — row operations, multiplicativity: Foundations

This page assumes you have seen nothing. We name every symbol the parent note leans on, draw the picture behind it, and say why the topic can't live without it. Read top to bottom — each idea is a rung for the next.

1. Number, scalar, and the letter $k$

The picture: a scalar is a stretch factor. Multiplying something by $k = 2$ makes it twice as long; $k = \frac{1}{2}$ shrinks it to half; $k = - 1$ flips its direction.

Why the topic needs it: the rule "scale a row by $k$ " and the corollary $det (k A) = k^{n} det A$ are entirely about what a scalar does to length and area. Before we can talk about stretching a row, we need the idea of a stretch number.

2. Vector and "a row"

The picture: in figure s01 the row $(2, 3)$ is drawn as an arrow reaching across and up to the point $(2, 3)$ . The numbers are how far right and how far up.

Figure — Properties — row operations, multiplicativity

Why the topic needs it: the whole derivation writes the determinant as a function of its rows, $D (R_{1}, \dots, R_{n})$ . To understand "swap two rows" or "add a multiple of one row to another", you must first see each row as an arrow you can move around.

3. Matrix — a stack of rows, and a machine

The picture: figure s02 shows both readings side by side. On the left, $A$ is two arrows (its rows). On the right, $A$ acts as a machine: feed in the little unit square, and it comes out stretched and tilted into a parallelogram.

Why the topic needs it: all the objects in the parent note are matrices, named by capital letters: $A$ and $B$ are the general machines, $I$ is the do-nothing machine (§7), an $E$ is a machine that performs one row operation (§8b), and a $U$ is a triangular machine (§10). Each is introduced properly in its own section below. "Row operations" edit the stack reading; products like $A B$ use the machine reading. You must hold both pictures at once.

4. The unit square, area, and orientation

The picture: figure s03. The machine turns the unit square (area $1$ ) into a parallelogram. Two things can happen:

the parallelogram has a bigger or smaller area — that's the scaling factor;
the corners can wind the opposite way around (clockwise instead of counter-clockwise) — that's a flip, shown by the little curved arrow reversing.

Why the topic needs it: this is the definition of the determinant the parent opens with — "signed scaling factor of the volume of the unit cube". The size of the parallelogram is $∣ det A ∣$ ; the sign records whether space was flipped. Row swaps flip; row scalings resize; shears slide corners sideways without changing area.

5. The determinant symbol $det$

The picture: the shaded parallelogram in figure s03; its signed area is $det A$ .

Why the topic needs it: it is the topic. Everything else measures how $det$ reacts to edits.

5b. From area (2D) to volume (3D and beyond)

The picture: figure s05. Three row-arrows in space frame a slanted box; $det A$ is how many unit cubes fit inside it, with a $-$ sign if the three arrows form a left-handed set instead of a right-handed one (the 3D version of "mirror-flipped").

Why the topic needs it: every rule you meet — swap flips the sign, scale a row by $k$ multiplies by $k$ , shear changes nothing, $det (k A) = k^{n} det A$ — is stated for general $n \times n$ matrices. The 2D pictures are just the easiest-to-draw case; picture the parallelepiped whenever the text says "volume", and trust that the same verbal rule holds in every dimension. That transfer is the whole reason we prove the rules from the axioms rather than from a single drawing.

6. Subscripts, arrows, and the operation notation

The picture: think of the rows as arrows from §2. A swap trades two arrows' positions; a scale stretches one arrow; a replacement slides one arrow's tip along the direction of another (a shear), shown in figure s04 — notice the base stays put and the area is unchanged.

Why the topic needs it: these three arrows are the three elementary row operations the whole chapter revolves around.

7. The identity matrix $I$

$I_{2} = (1001), det I = 1.$

The picture: the "do-nothing" machine leaves the unit square as the unit square — area $1$ , no flip. That's why $det I = 1$ , the normalization axiom.

Why the topic needs it: it is the reference point. $det I = 1$ anchors the whole system, and $det (A^{- 1}) = 1/ det A$ comes from $A A^{- 1} = I$ .

8. Multiplication $A B$ and the inverse $A^{- 1}$

The picture: feed the unit square through $B$ (area $\to det B$ ), then through $A$ (area multiplies again by $det A$ ). Total area factor $= det A \cdot det B$ . That is $det (A B) = det A det B$ seen.

Why the topic needs it: multiplicativity — the second headline of the parent — is literally "areas multiply when machines chain".

8b. The elementary matrix $E$

For example, starting from $I_{2}$ and swapping its two rows gives the swap matrix $E = (0110), det E = - 1.$

The picture: $E$ is a "one-move machine" — it does exactly one swap, one scale, or one shear and nothing else. Its determinant is therefore the exact factor that move applies: $- 1$ for a swap, $k$ for a scale, $1$ for a shear.

Why the topic needs it: the parent's algebraic proof of $det (A B) = det A det B$ writes $A$ as a chain of these one-move machines $E_{1} E_{2} \dots E_{m}$ . You must know what the letter $E$ stands for before that proof makes sense. See Elementary Matrices.

9. Product notation $\prod$ , powers $A^{k}$ , and exponent $n$

Why the topic needs it: the REF formula $det U = \prod_{i} u_{ii}$ and the corollaries $det (A^{k}) = (det A)^{k}$ , $det (k A) = k^{n} det A$ all use this compact notation.

10. Triangular matrix and "pivot"

$U = 200310 13 - 3, det U = 2 \cdot 1 \cdot (- 3) = - 6.$

Intuition Why triangular

$det$ = product of the diagonal Geometric: the first row-arrow lies along the $x$ -axis with length $u_{11}$ (its later entries don't matter for the base). The second arrow rises to height $u_{22}$ above that axis — the zeros below the diagonal mean each new arrow adds a fresh perpendicular height and nothing sideways that could collapse the box. So the slanted box is really "base $\times$ height $\times$ height $\dots$ ", i.e. $u_{11} \cdot u_{22} \dots u_{nn}$ . Shearing (the non-zero entries above the diagonal) only slides corners sideways and never changes that product. Algebraic: expand the determinant along the first column. Because every entry below $u_{11}$ is $0$ , only the top-left term survives, leaving $u_{11}$ times the determinant of the smaller triangular block. Repeat all the way down and you are left with $u_{11} u_{22} \dots u_{nn}$ .

Why the topic needs it: the practical way to compute any $det$ is to shear-and-swap the matrix into triangular form, then read off the pivot product. This is the bridge to Gaussian Elimination & Row Echelon Form.

11. The three defining properties (the vocabulary of the derivation)

The picture: two equal rows = two identical arrows = a degenerate parallelogram with no area. That single fact (A) is what forces the sign-flip on a swap.

Why the topic needs it: the parent derives every rule from just (M), (A), (N). These three words are the axioms — see Multilinear Alternating Forms.

Prerequisite map

Equipment checklist

Cover the right side and test yourself — you are ready for the parent topic when every line comes instantly.

What does a scalar

k

do to a length, geometrically?

Stretches it by factor

k

(flips direction if

k < 0

What is a row, seen as a picture?

An arrow from the origin to the point with those coordinates.

Two readings of a matrix?

A stack of row-arrows, and a machine that maps the unit square to a parallelogram.

What number is

det A

geometrically?

The signed area of that output parallelogram (

+

keeps orientation,

-

flips it).

In 3D, what shape does the unit cube become and what is

det

A slanted box (parallelepiped);

det

is its signed volume.

det (a c b d) =

a d - b c

What do the arrows

\leftrightarrow

and

\to

mean on rows?

\leftrightarrow

swaps two rows;

\to

replaces a row (scale it, or add a multiple of another).

Which row operation is a shear, and what happens to area?

R_{i} \to R_{i} + k R_{j}

; area (and

det

) is unchanged.

What is

I

and what is

det I

The do-nothing matrix (1's on diagonal);

det I = 1

What is an elementary matrix

E

The identity with one row operation done to it;

E A

performs that operation on

A

What does

A B

mean and what does

det (A B)

equal?

B

then

A

;

det (A B) = det A det B

Why does

det (k A) = k^{n} det A

, not

k det A

All

n

rows are scaled, so one factor of

k

comes out per row.

det

of a triangular matrix, and why?

The product of its diagonal pivots; column expansion leaves only the top-left term at each step (or: base

\times

successive heights).

State the three defining properties of

det

Multilinear in each row, Alternating (equal rows

\Rightarrow 0

), Normalized (

det I = 1

Ready? Every symbol the parent Row Operations & Multiplicativity note uses now has a plain meaning and a picture. Next stops: Determinant — Definition (cofactor / Leibniz), Elementary Matrices, and Volume, Orientation & the Determinant.