Definition
A scalar invariant associated to a square matrix or endomorphism that encodes volume scaling factor, orientation sign, and invertibility: a matrix is invertible iff its determinant is nonzero.

Principle

Principle
Determinant is multiplicative under composition: det(AB)=det(A)det(B). It equals the product of eigenvalues (with algebraic multiplicity) and for n×n matrices is an alternating n-linear function of the columns (or rows).

Demonstration

Demonstration
Example: For A = [[2,1],[3,4]] over R, det(A)=2·4−1·3=8−3=5; A scales oriented area by factor 5 and is invertible since det≠0.

Misapplication

Misapplication
Using determinant as a numerically stable test of invertibility for large matrices—small determinants may be due to scaling and rounding—and attempting to define determinant for non-square matrices without passing to induced square operators (e.g., via Gram matrices).

Consequence

Consequence
Correct computation of determinant yields decisive information: invertibility, sign of orientation change, and global volume change; determinants appear in change-of-variable formulas and characteristic polynomial constant term.

Reversal

Reversal
Instead of summarizing a matrix by its determinant (single scalar), consider the full spectrum of eigenvalues or the singular value decomposition, which give finer directional scaling information and condition number.

Boundary

Boundary
Defined naturally for n×n matrices over commutative rings/fields; generalizations to operators on infinite-dimensional spaces require trace-class conditions or Fredholm determinant constructions; over noncommutative rings determinant-like invariants differ.

Semantic Tension

Semantic Tension
Tension between determinant and permanent: both are sums over permutations but permanent lacks sign factors and does not reflect orientation; also tension between determinant as algebraic invariant and numerical stability of its computation.

Synthesis

Synthesis
The determinant is a single scalar invariant of a square linear operator capturing invertibility, oriented volume scaling, and the product of eigenvalues; multiplicativity and alternating multilinearity are its organizing properties.