Systems of Linear Equations

A system of linear equations is a collection of two or more linear equations involving the same variables. Solving it means finding values for the variables that satisfy all equations simultaneously. A system can have exactly one solution, no solution, or infinitely many solutions — geometrically corresponding to lines/planes that intersect at a point, never meet, or coincide. Used everywhere from circuit analysis to economics; formalized through the 18th–19th centuries (Gauss, Cramer).

A linear equation is one in which each variable appears only to the first power, with no products of variables, no roots, and no variables in denominators. A system of such equations asks: which values of the unknowns make every equation true at once? This matters because an enormous range of real problems — balancing chemical reactions, pricing goods, fitting models to data — reduces to finding the simultaneous solution of linear constraints. The surprising fact is that no matter how many equations or variables are involved, a linear system can only ever have zero, one, or infinitely many solutions — never exactly two or three. This trichotomy is a deep structural feature of linearity itself.

The Core Insight

Each linear equation in two variables represents a line; the solution to a system is the set of points lying on all the lines at once. Solving algebraically (substitution or elimination) is just a bookkeeping method for finding that geometric intersection. The same idea scales: in three variables each equation is a plane, and solutions are where planes meet. Linearity guarantees that the solution set is itself "flat" — a point, a line, a plane, or empty — which is exactly why only three outcomes are possible.

Formal Statement

Linear Equation

A linear equation in the variables $x_{1}, x_{2}, \dots, x_{n}$ is an equation that can be written in the form

a_{1} x_{1} + a_{2} x_{2} + \dots + a_{n} x_{n} = b

⟳A linear equation contains only first-degree terms — no

x_{i}^{2}

, no

x_{i} x_{j}

, no

x_{i}

, no

1/ x_{i}

Here the $a_{i}$ are the coefficients (known constants), $b$ is the constant term, and the $x_{i}$ are the unknowns. If all $a_{i} = 0$ and $b \neq = 0$ , the equation has no solution; if all $a_{i} = 0$ and $b = 0$ , every assignment satisfies it.

System of Linear Equations

A system of $m$ linear equations in $n$ unknowns is a list

⎩ ⎨ ⎧ a_{11} x_{1} + a_{12} x_{2} + \dots + a_{1 n} x_{n} = b_{1} a_{21} x_{1} + a_{22} x_{2} + \dots + a_{2 n} x_{n} = b_{2} ⋮ a_{m 1} x_{1} + a_{m 2} x_{2} + \dots + a_{mn} x_{n} = b_{m}

⟳The first subscript

i

a_{ij}

labels the equation (row); the second subscript

j

labels the variable (column).

A solution is an ordered tuple $(x_{1}, \dots, x_{n})$ that satisfies every equation. The solution set is the collection of all such tuples. Two systems are equivalent if they have the same solution set.

For the common case of two equations in two unknowns:

{a_{1} x + b_{1} y = c_{1} a_{2} x + b_{2} y = c_{2}

⟳Each equation, with

a_{i}, b_{i}

not both zero, plots as a straight line in the

x y

-plane.

Trichotomy of Solutions

Every system of linear equations has exactly one of the following:

No solution (the system is inconsistent),
Exactly one solution (consistent and independent),
Infinitely many solutions (consistent and dependent).

No linear system can have, for example, exactly two solutions.

Cramer's Rule for 2×2 Systems

For the system in equation [3], define the determinant $D = a_{1} b_{2} - a_{2} b_{1}$ . If $D \neq = 0$ , the unique solution is

x = \frac{c _{1} b _{2} - c _{2} b _{1}}{a _{1} b _{2} - a _{2} b _{1}}, y = \frac{a _{1} c _{2} - a _{2} c _{1}}{a _{1} b _{2} - a _{2} b _{1}}

⟳Each numerator is the determinant of the coefficient matrix with one column replaced by the constants

(c_{1}, c_{2})

If $D = 0$ , the system has either no solution or infinitely many. See Determinants and Cramer's Rule.

Why It Holds / Derivation

Solving a 2×2 System by Elimination

We solve the general system [3] to derive Cramer's rule and confirm the trichotomy.

Step 1. Multiply the first equation by

b_{2}

and the second by

b_{1}

a_{1} b_{2} x + b_{1} b_{2} y = c_{1} b_{2}, a_{2} b_{1} x + b_{1} b_{2} y = c_{2} b_{1}

⟳We chose multipliers that make the

y

-coefficients equal so they cancel on subtraction.

Step 2. Subtract the second from the first; the

y

-terms vanish:

(a_{1} b_{2} - a_{2} b_{1}) x = c_{1} b_{2} - c_{2} b_{1}

⟳

b_{1} b_{2} y - b_{1} b_{2} y = 0

, isolating

x

Step 3. If

D = a_{1} b_{2} - a_{2} b_{1} \neq = 0

, divide to obtain

x = \frac{c _{1} b _{2} - c _{2} b _{1}}{D}

, matching [4]. By a symmetric elimination of

x

we get

y

. Step 4. If

D = 0

, equation [6] reads

0 \cdot x = c_{1} b_{2} - c_{2} b_{1}

. If the right side is nonzero, this is impossible — no solution. If the right side is also zero, the two original equations are proportional and describe the same line — infinitely many solutions.

This exhausts all cases, proving the trichotomy for $n = 2$ . $■$

Carl Friedrich Gauss | 1777–1855

Though systematic elimination was known in ancient China, Gauss formalized the elimination procedure now bearing his name while computing the orbit of the asteroid Ceres around 1801. His method — reducing a system to row-echelon form through systematic operations — remains the foundation of computational linear algebra. Gauss treated linear systems as routine tools in his astronomical and geodetic work, and the method's reliability made large-scale hand computation feasible long before computers.

Building Intuition

Picture two straight roads on a flat map. Each road is the graph of one equation. A solution to the system is a place that lies on both roads at once — an intersection. Three things can happen: the roads cross once (one solution), run parallel and never meet (no solution), or are the same road drawn twice (infinitely many shared points). This is why a system can never have exactly two solutions: two distinct straight lines cannot cross twice. The algebra of elimination is simply a way of finding that crossing without drawing the map.

Three possible outcomes for a system of two lines

<svg viewBox="0 0 600 220" xmlns="http://www.w3.org/2000/svg">
<rect width="600" height="220" fill="#0a0e17"/>
<!-- Panel 1: one solution -->
<line x1="20" y1="180" x2="180" y2="40" stroke="#94a3b8" stroke-width="2"/>
<line x1="20" y1="60" x2="180" y2="170" stroke="#d4a853" stroke-width="2"/>
<circle cx="103" cy="108" r="4" fill="#e2e8f0"/>
<text x="100" y="205" fill="#e2e8f0" font-size="13" text-anchor="middle">One solution</text>
<text x="100" y="30" fill="#e2e8f0" font-size="11" text-anchor="middle">lines cross</text>
<!-- Panel 2: no solution -->
<line x1="220" y1="170" x2="380" y2="50" stroke="#94a3b8" stroke-width="2"/>
<line x1="220" y1="190" x2="380" y2="70" stroke="#d4a853" stroke-width="2"/>
<text x="300" y="205" fill="#e2e8f0" font-size="13" text-anchor="middle">No solution</text>
<text x="300" y="30" fill="#e2e8f0" font-size="11" text-anchor="middle">parallel lines</text>
<!-- Panel 3: infinite solutions -->
<line x1="420" y1="180" x2="580" y2="50" stroke="#94a3b8" stroke-width="5" opacity="0.5"/>
<line x1="420" y1="180" x2="580" y2="50" stroke="#d4a853" stroke-width="2"/>
<text x="500" y="205" fill="#e2e8f0" font-size="13" text-anchor="middle">Infinitely many</text>
<text x="500" y="30" fill="#e2e8f0" font-size="11" text-anchor="middle">same line</text>
</svg>

Worked Examples

A Unique Solution by Elimination

Given: Solve

{2 x + 3 y = 12 4 x - y = 5

Step 1. Make the

y

-coefficients opposites. Multiply the second equation by

3

12 x - 3 y = 15

Step 2. Add to the first equation

(2 x + 3 y = 12)

14 x = 27 ⟹ x = \frac{27}{14}

Step 3. Substitute into

4 x - y = 5

4 \cdot \frac{27}{14} - y = 5 ⟹ \frac{108}{14} - y = 5 ⟹ y = \frac{108}{14} - \frac{70}{14} = \frac{38}{14} = \frac{19}{7}

Check (Cramer's rule):

D = (2) (- 1) - (4) (3) = - 2 - 12 = - 14 \neq = 0

, confirming a unique solution.

x = \frac{27}{14}, y = \frac{19}{7}

(Pure numbers — no units in this abstract example.)

An Application with Units

Given: A vendor sells apples at \

2 e a c han d banana s a t

1 each. A customer buys 7 pieces of fruit total and pays \$11. How many of each?

Let $a$ = number of apples, $b$ = number of bananas.

{a + b = 7 (count) 2 a + b = 11 (dollars)

Step 1. Subtract the first equation from the second:

(2 a + b) - (a + b) = 11 - 7 ⟹ a = 4

Step 2. Substitute back:

4 + b = 7 ⟹ b = 3

. Check:

2 (4) + 1 (3) = 8 + 3 = 11

✓ and

4 + 3 = 7

✓.

a = 4 apples, b = 3 bananas

Common Misconceptions

"More equations always means a smaller solution set or a unique answer." False. Adding equations can leave the solution unchanged (if they are redundant), make a system inconsistent, or have no effect. What matters is not the count of equations but how many are independent. A system with three equations in two unknowns is overdetermined and usually has no solution — but if all three lines pass through one point, the unique solution survives. "If I get $0 = 0$ while solving, I made a mistake." Not necessarily —

0 = 0

signals that the equations were dependent, giving infinitely many solutions. Conversely, arriving at

0 = 5

(or any false statement) correctly signals no solution. These are meaningful outcomes, not errors.

Consider the system

{x + y = 1 2 x + 2 y = 3

A naive solver might expect a unique intersection. But the lines have the same slope and different intercepts — they are parallel. Eliminating: multiply the first by 2 to get

2 x + 2 y = 2

, then subtract from the second:

0 = 1

, which is false. No solution exists. This shows that two equations in two unknowns do not guarantee a solution.

Connections

Systems of linear equations are the gateway to Matrices, where the coefficients are packaged into an array and the system written compactly as $A x = b$ . Solving then connects to Matrix Inverse (via $x = A^{- 1} b$ when $A$ is invertible), Determinants (which detect whether a unique solution exists), and Gaussian Elimination as the general algorithm. The geometric viewpoint links directly to Linear Functions and the graphing of Lines and Slopes. At a higher level, the trichotomy is a special case of the structure of solution sets in Vector Spaces and Linear Transformations.

Real-World Applications

Electrical circuits: Kirchhoff's voltage and current laws produce linear systems whose solutions give the currents in each branch (see Kirchhoff's Laws).
Chemistry — balancing equations: The numbers of atoms of each element yield linear equations in the stoichiometric coefficients.
Economics — supply and demand: Market equilibrium is the intersection of linear supply and demand curves, a 2×2 system.
Computer graphics: Transforming and intersecting objects (e.g., finding where a ray hits a plane) reduces to solving linear systems.
Nutrition / diet planning: Mixing foods to hit target amounts of protein, fat, and carbohydrates gives a linear system (often solved via Linear Programming).
GPS and navigation: Trilateration from satellite distances linearizes into a system solved for position.
Network flow and traffic modeling: Conservation at each node yields one linear equation per junction.

Historical Development

c. 200 BCE — China: The Nine Chapters on the Mathematical Art presents a method essentially identical to Gaussian elimination ("fangcheng") for solving systems with rod numerals.
c. 250 CE — Greece: Diophantus solves simultaneous equations, though without systematic notation.
1545 — Italy: Cardano gives a rule resembling a 2×2 determinant in Ars Magna.
1683 — Japan: Seki Takakazu develops determinant-like methods for eliminating variables.
1693 — Germany: Leibniz independently writes about determinants in correspondence, anticipating modern notation.
1750 — Switzerland: Gabriel Cramer publishes the rule now bearing his name in Introduction à l'analyse des lignes courbes algébriques.
c. 1801 — Germany: Gauss systematizes elimination during his orbit computations for Ceres.
1888 — England: Wilhelm Jordan refines the method into Gauss–Jordan elimination, reducing to reduced row-echelon form.
1858 — England: Arthur Cayley founds matrix algebra, giving the system the compact form $A x = b$ .

Subtleties & Limits

For small hand-computed systems the methods here are exact, but real applications raise deeper questions. When systems are overdetermined (more equations than unknowns) with no exact solution — as in fitting noisy experimental data — one seeks a best approximate solution via Least Squares rather than an exact one. When coefficients are only known approximately, a system may be ill-conditioned: tiny changes in the data produce huge changes in the solution, a central concern of numerical analysis. For systems with millions of variables (climate models, search engines), the practical challenge is not whether a solution exists but how to compute it efficiently and stably — an active area connecting linear algebra to computer science. The boundary between "solvable in principle" and "solvable in practice" remains a rich frontier.

Systems of Linear Equations

The Core Insight

Formal Statement

Linear Equation

System of Linear Equations

Trichotomy of Solutions

Cramer's Rule for 2×2 Systems

Why It Holds / Derivation

Solving a 2×2 System by Elimination

Carl Friedrich Gauss | 1777–1855

Building Intuition

Three possible outcomes for a system of two lines

Worked Examples

A Unique Solution by Elimination

An Application with Units

Common Misconceptions

Connections

Real-World Applications

Historical Development

Subtleties & Limits

See Also

Before this

Where this leads