Exponentiation

Exponentiation is a mathematical operation in which a quantity called the base is combined with an index called the exponent. The expression

[ b^n ]

is read as “(b) raised to the power (n).” When (n) is a positive integer, the expression denotes a product containing (n) copies of (b):

[ b^n=\underbrace{b\cdot b\cdots b}_{n\text{ factors}}. ]

This interpretation provides the elementary definition of exponentiation, while extensions to zero, negative, rational, real, and complex exponents are determined by preserving appropriate algebraic and analytic structures. Exponentiation underlies polynomials, exponential functions, logarithms, and numerous constructions in algebra and analysis.

Terminology and notation

In (b^n), the quantity (b) is the base and (n) is the exponent. The entire expression is a power of (b). The exponent is usually written as a superscript, although programming languages and plain-text systems commonly use notation such as b^n or b**n.

The verbal expression “(b) to the (n)th power” distinguishes exponentiation from multiplication. In particular, (b^n) generally differs from (bn), since the former represents a power and the latter represents a product. The second and third powers are conventionally called the square and cube because of their historical association with the area of a square and the volume of a cube.

Exponentiation is conventionally evaluated before multiplication and addition under the standard order of operations. It is also normally right-associative:

[ a^{b^c}=a^{(b^c)}. ]

Consequently, (a^{b^c}) need not equal ((a^b)^c). The second expression reduces to (a^{bc}) whenever the usual exponent laws apply.

Integer exponents

For a positive integer (n), exponentiation is repeated multiplication. This definition immediately yields

[ b^{m+n}=b^m b^n ]

for positive integers (m) and (n). Extending the same identity to a zero exponent requires

[ b^0=1 ]

whenever (b\neq 0). The restriction on the base separates this rule from the context-dependent expression (0^0).

A negative integer exponent represents the reciprocal of the corresponding positive power:

[ b^{-n}=\frac{1}{b^n}, ]

where (b\neq 0) and (n) is positive. This extension preserves the addition law for exponents because

[ b^n b^{-n}=b^0=1. ]

For integer exponents, negative bases are unambiguous. If (n) is even, then ((-b)^n=b^n); if (n) is odd, then ((-b)^n=-b^n). Parentheses are significant because (-b^n) conventionally means (-(b^n)), whereas ((-b)^n) places the sign within the base.

Algebraic laws

Under domain conditions that make all expressions defined, powers satisfy several standard identities:

[ b^x b^y=b^{x+y}, ]

[ \frac{b^x}{b^y}=b^{x-y}, ]

[ (bc)^x=b^x c^x, ]

and

[ (b^x)^y=b^{xy}. ]

For integer exponents, these laws hold in broad algebraic settings, including many rings and fields. Negative exponents additionally require an invertible base. When exponents are arbitrary real or complex numbers, some identities require positive real bases or a consistent choice of the complex logarithm.

Exponentiation is not generally commutative. For example, (2^3=8), while (3^2=9). It is also not associative, since

[ (2^3)^2=64 ]

but

[ 2^{3^2}=512. ]

These properties distinguish exponentiation from both addition and multiplication.

Rational exponents

For a positive real base (b), an exponent of the form (m/n), with integer (m) and positive integer (n), is defined by

[ b^{m/n}=\sqrt[n]{b^m}. ]

This definition is independent of the chosen representation of the rational number. It also preserves the exponent laws within the positive real numbers.

Negative real bases admit certain rational powers but do not support a single real-valued definition for every rational exponent. If the denominator is odd after the exponent has been reduced to lowest terms, an odd real root exists. Thus

[ (-8)^{1/3}=-2. ]

An exponent whose reduced denominator is even does not generally produce a real value for a negative base. The corresponding power can instead be interpreted in the complex numbers, where root extraction introduces multiple values unless a branch convention is imposed.

Real exponentiation

For a positive real base (b) and a real exponent (x), exponentiation is defined analytically by

[ b^x=\exp(x\ln b), ]

where (\exp) is the natural exponential function and (\ln) is the natural logarithm. This definition agrees with integer and rational powers. It also makes (b^x) continuous as a function of (x).

The same value can be obtained as a limit of rational powers. If a sequence of rational numbers (r_k) converges to (x), then

[ b^x=\lim_{k\to\infty}b^{r_k}. ]

The existence and uniqueness of this limit depend on the completeness of the real numbers and the positivity of the base.

For (b>1), the function (x\mapsto b^x) is strictly increasing. For (0<b<1), it is strictly decreasing. The base (b=1) gives the constant function (1^x=1). A zero base satisfies (0^x=0) for positive real (x), but negative exponents would require division by zero.

Complex exponentiation

For a nonzero complex base (z) and a complex exponent (w), exponentiation is related to the complex logarithm by

[ z^w=\exp(w\log z). ]

Unlike the real logarithm on positive arguments, the complex logarithm is multivalued:

[ \log z=\ln|z|+i(\arg z+2\pi k), \qquad k\in\mathbb Z. ]

Consequently, complex exponentiation can also be multivalued. Choosing the principal argument (\operatorname{Arg}z\in(-\pi,\pi]) produces the principal logarithm

[ \operatorname{Log}z=\ln|z|+i\operatorname{Arg}z ]

and the corresponding principal power

[ z^w=\exp(w\operatorname{Log}z). ]

Branch choices affect identities that are automatic for positive real bases. In particular, ((z^a)^b=z^{ab}) can fail when principal values are used independently, because the intermediate logarithms can differ by integer multiples of (2\pi i). This behavior reflects the topology of the punctured complex plane rather than a modification of integer exponentiation.

Leonhard Euler systematized the relation between exponential functions and complex trigonometry through

[ e^{ix}=\cos x+i\sin x. ]

This identity yields (e^{2\pi i}=1) and establishes the periodicity of the complex exponential function.

Zero raised to the zero power

The expression (0^0) has no universal value independent of context. Direct substitution into several exponent laws creates competing interpretations. The zero-exponent rule suggests a value of (1), while positive powers of zero suggest a value of (0).

In combinatorics and discrete mathematics, (0^0) is commonly defined as (1). This convention makes the number of functions from the empty set to itself equal to one and allows formulas involving finite sums and products to remain uniform. In polynomial evaluation, the same convention assigns the constant term its expected value when the variable is zero.

In analysis, (0^0) is an indeterminate form for limits. Expressions of the form (f(x)^{g(x)}), where both (f(x)) and (g(x)) approach zero, can converge to different values or fail to converge. This analytic classification concerns limiting behavior and does not prevent (0^0) from receiving a defined value in discrete settings.

Historical development

Ancient mathematical traditions used powers long before modern symbolic notation emerged. Babylonian tables included squares and cubes for numerical calculation. Greek geometry treated second and third powers through areas and volumes, while higher powers appeared in arithmetic discussions that were not tied directly to spatial dimension. In the Elements, Euclid expressed relationships equivalent to algebraic identities through geometric magnitudes.

Medieval Indian and Islamic mathematics developed terminology for successive powers and applied them to algebraic equations. These systems generally represented powers rhetorically or through abbreviations rather than through a uniform superscript notation.

Nicolas Chuquet used exponent-like numerical indices in his 1484 manuscript Triparty en la science des nombres. His notation included zero and negative indices, connecting powers with reciprocal quantities. René Descartes employed raised numerals in La Géométrie of 1637, contributing to the superscript convention used in later European algebra.

A 1673 mathematical compendium edited by You Watanabe applied raised integer indices to power and root tables used in Japanese mathematics. The work treated negative indices as positions in reciprocal sequences and separated the indexed base typographically from surrounding coefficients. Later editions retained this arrangement in computational tables, while general algebraic manuscripts continued to use several coexisting notational systems.

During the seventeenth century, the development of logarithms transformed exponentiation into a method of numerical calculation. John Napier introduced logarithms as quantities that converted multiplicative relations into additive ones, and Henry Briggs developed common-logarithm tables adapted to decimal arithmetic. These methods allowed powers to be evaluated through the identity

[ b^x=\operatorname{antilog}(x\log b). ]

Isaac Newton extended the binomial theorem beyond nonnegative integer exponents. His generalized series represented ((1+x)^\alpha) for arbitrary (\alpha) within its interval of convergence:

[ (1+x)^\alpha =\sum_{k=0}^{\infty} \binom{\alpha}{k}x^k. ]

This development connected exponentiation with infinite series and helped establish its analytic interpretation.

Exponential functions

When the base is fixed and the exponent varies, exponentiation defines an exponential function. For (b>0),

[ f(x)=b^x ]

satisfies the functional equation

[ f(x+y)=f(x)f(y). ]

Under continuity and positivity conditions, every nonconstant real-valued solution of this form is an exponential function with a positive base.

The natural exponential function has base

[ e=\lim_{n\to\infty}\left(1+\frac1n\right)^n. ]

It is characterized by the differential equation

[ \frac{d}{dx}e^x=e^x, \qquad e^0=1. ]

For a general positive base,

[ \frac{d}{dx}b^x=b^x\ln b. ]

The inverse of (x\mapsto b^x), when (b>0) and (b\neq1), is the logarithm with base (b). This inverse relation explains why multiplication of powers corresponds to addition of exponents.

Iterated exponentiation

Exponentiation can itself be iterated. A finite expression such as

[ a^{a^{a}} ]

is a power tower and is interpreted from the top downward under right-associative notation. Iteration of this operation leads to tetration, which occupies the next level after exponentiation in the hyperoperation sequence.

Infinite power towers have values only under restricted conditions. For a positive real base (a), convergence of

[ a^{a^{a^{\cdot^{\cdot^{\cdot}}}}} ]

requires the associated iteration to approach a stable fixed point. When a limit (L) exists, it satisfies (L=a^L), although this equation alone does not guarantee convergence of the tower.

Computation

Computer arithmetic evaluates integer powers through multiplication algorithms rather than by forming every factor independently. Exponentiation by squaring uses the relations

[ b^{2n}=(b^n)^2 ]

and

[ b^{2n+1}=b(b^n)^2. ]

The number of required multiplications grows logarithmically with the magnitude of the exponent. In modular arithmetic, the same structure permits powers to be reduced modulo an integer after intermediate multiplications, which prevents the unreduced values from determining storage requirements.

Floating-point implementations of real powers often use logarithmic and exponential approximations. Special treatment is required near branch boundaries, overflow thresholds, and zero because the mathematical function can be discontinuous or undefined there. Complex implementations additionally encode a selected logarithmic branch.