Vectors

⚡ Key Takeaways

A vector is a quantity with both magnitude and direction, represented geometrically as an arrow or algebraically as an ordered list of components. Key result: vectors add tip-to-tail and obey the parallelogram law; they can be scaled, decomposed into components, and combined via the dot and cross products. Vectors apply everywhere direction matters — physics (force, velocity), engineering, computer graphics, and navigation. Formalized in the 19th century by Hamilton, Grassmann, Gibbs, and Heaviside.

💡 The Core Insight

A vector separates how much from which way. By splitting a directed quantity into perpendicular components, any vector problem in two or three dimensions reduces to independent one-dimensional arithmetic. Addition, scaling, and projection then become simple operations on numbers — but the geometric meaning (a length and a direction that exist regardless of coordinates) is always preserved underneath.

DefinitionVector — geometric and algebraic

A vector $\vec{v}$ is a quantity possessing both magnitude $|\vec{v}|$ and direction. Two equivalent representations:

Geometric: a directed line segment (arrow) from an initial point to a terminal point. Two arrows are the same vector if they have equal length and equal direction (parallel, same sense), regardless of starting location.

Algebraic: an ordered tuple of real numbers, the components, relative to a chosen basis:

1

\quad \vec{v} = (v_x,\, v_y,\, v_z) = v_x\hat{\imath} + v_y\hat{\jmath} + v_z\hat{k}

Derivation hint: project the arrow onto each coordinate axis; $\hat{\imath},\hat{\jmath},\hat{k}$ are unit vectors along $x,y,z$ , and each $v_x,v_y,v_z$ is the signed length of the projection. The magnitude (length, norm) is given by the Pythagorean theorem:

2

\quad |\vec{v}| = \sqrt{v_x^2 + v_y^2 + v_z^2}

Derivation hint: apply the Pythagorean theorem twice — once in the $xy$ -plane to get $\sqrt{v_x^2+v_y^2}$ , then again combining that with $v_z$ . The direction in 2D is captured by the angle $\theta$ from the positive $x$ -axis:

3

\quad \theta = \tan^{-1}\!\left(\frac{v_y}{v_x}\right)\quad(\text{adjusted for quadrant})

Derivation hint: in the right triangle formed by the components, the tangent of the angle is opposite over adjacent, $v_y/v_x$ ; check the signs of $v_x,v_y$ to place the quadrant correctly.

DefinitionVector operations

For vectors $\vec{a}=(a_x,a_y,a_z)$ , $\vec{b}=(b_x,b_y,b_z)$ and scalar $k$ :

Addition (component-wise):

4

\quad \vec{a}+\vec{b} = (a_x+b_x,\; a_y+b_y,\; a_z+b_z)

Derivation hint: place $\vec{b}$ 's tail at $\vec{a}$ 's tip; the resultant runs from $\vec{a}$ 's tail to $\vec{b}$ 's tip — each coordinate displacement simply adds.

Scalar multiplication:

5

\quad k\vec{a} = (ka_x,\; ka_y,\; ka_z)

Derivation hint: stretching the arrow by factor $k$ stretches each projected component by $k$ ; negative $k$ also reverses direction.

Dot (scalar) product:

6

\quad \vec{a}\cdot\vec{b} = a_xb_x + a_yb_y + a_zb_z = |\vec{a}|\,|\vec{b}|\cos\phi

Derivation hint: expand $(\vec a)\cdot(\vec b)$ in the orthonormal basis using $\hat\imath\cdot\hat\imath=1$ , $\hat\imath\cdot\hat\jmath=0$ ; equate with the law-of-cosines form to recover $\cos\phi$ .

Cross (vector) product (3D):

7

\quad \vec{a}\times\vec{b} = \big(a_yb_z - a_zb_y,\; a_zb_x - a_xb_z,\; a_xb_y - a_yb_x\big)

Derivation hint: expand the symbolic determinant with rows $(\hat\imath,\hat\jmath,\hat k)$ , $(a_x,a_y,a_z)$ , $(b_x,b_y,b_z)$ by cofactors along the first row. with magnitude $|\vec{a}\times\vec{b}| = |\vec{a}|\,|\vec{b}|\sin\phi$ , a vector perpendicular to both, direction given by the right-hand rule.

TheoremParallelogram Law and Triangle Inequality

Vector addition is commutative ( $\vec{a}+\vec{b}=\vec{b}+\vec{a}$ ) and associative, so the sum of two vectors equals the diagonal of the parallelogram they span. Furthermore the magnitudes satisfy the triangle inequality:

8

\quad |\vec{a}+\vec{b}| \le |\vec{a}| + |\vec{b}|

Derivation hint: square both sides; $|\vec a+\vec b|^2 = |\vec a|^2 + 2\vec a\cdot\vec b + |\vec b|^2$ and since $\vec a\cdot\vec b \le |\vec a||\vec b|$ , the cross term is bounded above. Equality holds only when $\vec{a}$ and $\vec{b}$ point in the same direction.

ProofThe component magnitude formula

We derive equation [2], $|\vec{v}| = \sqrt{v_x^2+v_y^2+v_z^2}$ .

Step 1. Place the vector's tail at the origin $O$ and tip at point $P=(v_x, v_y, v_z)$ .

Step 2. Drop $P$ to the point $Q=(v_x, v_y, 0)$ in the $xy$ -plane. Triangle $OQ$ lies flat in that plane with legs $v_x$ and $v_y$ , so by the Pythagorean theorem:

9

\quad |OQ| = \sqrt{v_x^2 + v_y^2}

Derivation hint: the projection onto the $xy$ -plane forms a right triangle with legs along the $x$ and $y$ axes.

Step 3. The segment $QP$ is vertical with length $|v_z|$ , and it is perpendicular to $OQ$ (one lies in the $xy$ -plane, the other along $z$ ). Triangle $OQP$ is therefore right-angled at $Q$ .

Step 4. Apply the Pythagorean theorem a second time:

10

\quad |\vec{v}|^2 = |OP|^2 = |OQ|^2 + |QP|^2 = (v_x^2 + v_y^2) + v_z^2

Derivation hint: substitute the result of Step 2 and $|QP|=|v_z|$ , noting $|v_z|^2=v_z^2$ .

Step 5. Taking the (positive) square root gives equation [2]. $\blacksquare$

∎

JW

Josiah Willard Gibbs

1839–1903

The American mathematical physicist Josiah Willard Gibbs, working at Yale, took the cumbersome quaternion algebra of Hamilton and distilled from it the practical, modern vector analysis used today — the dot product, cross product, and gradient/divergence/curl notation. His privately printed lecture notes (1881–84) and the 1901 textbook Vector Analysis (written up by his student E. B. Wilson) replaced four-component quaternions with clean three-component vectors. Working in parallel, the self-taught English engineer Oliver Heaviside developed an almost identical system to streamline Maxwell's equations of electromagnetism. Their pragmatic notation won out over the more abstract quaternion school, becoming the universal language of physics and engineering.

🧠 Intuition

Think of a vector as a set of walking instructions: "go 3 km east and 4 km north." The instructions are the components $(3,4)$ ; the actual straight-line distance you end up from the start is the magnitude $\sqrt{3^2+4^2}=5$ km; the compass heading is the direction. Crucially, where you start walking doesn't change the instructions — that is why a vector can slide around freely without changing identity. Adding two vectors is just following one set of instructions, then the next, from wherever you ended up (tip-to-tail). The dot product measures how much two vectors "agree" in direction: it is large and positive when they point similar ways, zero when perpendicular, negative when opposed. The cross product measures how much they "disagree," producing a new vector that sticks out perpendicular to both — its length is the area of the parallelogram they span.

Figure

Vector addition by the tip-to-tail (triangle) method, with the resultant as the parallelogram diagonal

ExampleResultant force on a hook

Given: Two forces act on a hook: $\vec{F}_1 = (6,\,0)\,\text{N}$ (pulling east) and $\vec{F}_2 = (0,\,8)\,\text{N}$ (pulling north). Find the resultant force's magnitude and direction.

Step 1 — Add components (equation [4]):

11

\quad \vec{R} = \vec{F}_1 + \vec{F}_2 = (6+0,\; 0+8) = (6,\,8)\,\text{N}

Derivation hint: add the $x$ -components together and the $y$ -components together.

Step 2 — Magnitude (equation [2]):

12

\quad |\vec{R}| = \sqrt{6^2 + 8^2} = \sqrt{36+64} = \sqrt{100} = 10\,\text{N}

Derivation hint: a 6–8–10 right triangle (a scaled 3–4–5).

Step 3 — Direction (equation [3]):

13

\quad \theta = \tan^{-1}\!\left(\frac{8}{6}\right) = \tan^{-1}(1.333) = 53.1^\circ

Derivation hint: both components positive ⇒ first quadrant, no adjustment needed.

Answer: $\boxed{|\vec{R}| = 10\ \text{N at } 53.1^\circ \text{ north of east}}$

ExampleAngle between two vectors via the dot product

Given: $\vec{a} = (2,\,3,\,1)$ and $\vec{b} = (1,\,-1,\,2)$ . Find the angle $\phi$ between them.

Step 1 — Dot product (equation [6]):

14

\quad \vec{a}\cdot\vec{b} = (2)(1)+(3)(-1)+(1)(2) = 2 - 3 + 2 = 1

Derivation hint: multiply matching components and sum.

Step 2 — Magnitudes (equation [2]):

15

\quad |\vec{a}| = \sqrt{4+9+1}=\sqrt{14}\approx 3.742,\qquad |\vec{b}| = \sqrt{1+1+4}=\sqrt{6}\approx 2.449

Derivation hint: square each component, sum, take the root for each vector separately.

Step 3 — Solve for the angle:

16

\quad \cos\phi = \frac{\vec{a}\cdot\vec{b}}{|\vec{a}|\,|\vec{b}|} = \frac{1}{(3.742)(2.449)} = \frac{1}{9.165} = 0.1091

Derivation hint: rearrange the geometric form of equation [6] to isolate $\cos\phi$ .

Step 4: $\phi = \cos^{-1}(0.1091) = 83.7^\circ$ .

Answer: $\boxed{\phi \approx 83.7^\circ}$ (nearly perpendicular).

⚠️ Common mistake

Vectors are not just "numbers with arrows on top." Several frequent errors:

Adding magnitudes instead of vectors. $|\vec{a}+\vec{b}| \ne |\vec{a}|+|\vec{b}|$ in general (see the triangle inequality, [8]). Two 5 N forces can produce a resultant anywhere from 0 N to 10 N depending on their angle.
Confusing the dot and cross products. The dot product yields a scalar (a number); the cross product yields a vector. Writing $\vec{a}\cdot\vec{b}$ as a vector or $\vec{a}\times\vec{b}$ as a number is a category error.
Forgetting quadrant adjustment in $\tan^{-1}$ . A calculator returns angles only in $(-90^\circ, 90^\circ)$ ; for vectors pointing left or down you must add $180^\circ$ or otherwise correct.
Position vs. free vectors. A free vector can slide anywhere; a position vector is pinned to the origin. Treating them interchangeably causes errors in geometry problems.

Counterexample

A student claims "the magnitude of a sum equals the sum of magnitudes." Take $\vec{a}=(3,0)$ and $\vec{b}=(-3,0)$ . Then $|\vec{a}|=3$ and $|\vec{b}|=3$ , so the claim predicts $|\vec{a}+\vec{b}|=6$ . But $\vec{a}+\vec{b}=(0,0)$ , the zero vector, with magnitude $0\ne 6$ . The two displacements cancel exactly. This shows magnitude is fundamentally non-additive for vectors unless they are parallel and same-sense.

connection

Vectors sit at the crossroads of geometry, algebra, and physics:

They generalize the Number Line and Coordinate Geometry from one signed magnitude to many directed quantities.
The dot product is the gateway to Trigonometry (it encodes $\cos\phi$ ) and to the abstract notion of inner products in Linear Algebra.
Collections of vectors form vector spaces, the central object of Linear Algebra, where matrices act as linear transformations on vectors.
In physics, vectors underlie Newton's Laws of Motion (forces and accelerations), Kinematics (displacement, velocity), and Electromagnetism (electric and magnetic fields).
Differentiating a vector function leads to Vector Calculus (gradient, divergence, curl).

🔬 Where this appears

⏳ History

- c. 1580 (Simon Stevin): Demonstrates the parallelogram of forces in studies of statics and inclined planes — an early physical use of vector addition.

- 1687 (Isaac Newton): Principia implicitly combines forces and velocities by the parallelogram rule, though without modern notation.

- 1799 (Caspar Wessel) / 1806 (Jean-Robert Argand): Geometric representation of complex numbers as directed segments in the plane — a 2D proto-vector idea.

- 1843 (William Rowan Hamilton): Invents quaternions, a four-dimensional algebra; coins the word "vector" for the imaginary (directed) part.

- 1844 (Hermann Grassmann): Publishes Ausdehnungslehre ("Theory of Extension"), an abstract,

n

-dimensional theory of directed magnitudes far ahead of its time but little read.

- 1873 (James Clerk Maxwell): Uses quaternion methods in his Treatise on Electricity and Magnetism, spotlighting the need for a cleaner vector calculus.

- 1881–1884 (J. W. Gibbs): Distributes lecture notes establishing modern vector analysis — separate dot and cross products in three dimensions.

- 1893 (Oliver Heaviside): Independently formulates and popularizes vector operations to recast Maxwell's equations into their compact modern form.

- 1901 (E. B. Wilson): Publishes Vector Analysis, based on Gibbs's lectures — the textbook that standardized today's notation.

open-problem

The "arrow with magnitude and direction" picture, while perfect for introductory work, conceals deeper structure that mathematicians and physicists must confront:

What is a vector really? The modern definition is anything that transforms correctly under change of coordinates. A list of three numbers is only a genuine vector if its components rotate the way an arrow's do — distinguishing true vectors from arbitrary triples. This leads to tensors and the geometry of how quantities behave under symmetry.
Axial vs. polar vectors. The cross product yields a "pseudovector" (axial vector) that flips sign under a mirror reflection differently from ordinary (polar) vectors — subtle but physically important in magnetism and chirality.
Curved spaces. On a sphere or in curved spacetime, you cannot freely slide a vector from one point to another without ambiguity; comparing vectors at different points requires parallel transport and the machinery of differential geometry, where the naive "free vector" idea breaks down.
Beyond three dimensions. The neat cross product exists only in 3 (and trivially 7) dimensions; its generalization requires the exterior/wedge product of Grassmann's algebra.

Formal Statement

Why It Holds / Derivation

Building Intuition

Worked Examples

Common Misconceptions

Connections

Real-World Applications

Historical Development

Subtleties & Limits

See Also