Mathematical Analysis

Mathematical analysis is the branch of mathematics that studies change, accumulation, and approximation with precision.

11 frameworks, from 370 BCE to today · Analysis · Mathematics

still a working view ended · each leads to its entry below

Overview

Mathematical analysis is the branch of mathematics that studies change, accumulation, and approximation with precision1. Its central objects are limits, continuity, derivatives, integrals, and infinite series, and its central question is deceptively simple: what does it mean for a quantity to approach a value, and when can we trust that it does12? That question has driven the field from ancient geometry through the invention of calculus, the nineteenth-century campaign to put calculus on a logical footing, and the twentieth-century creation of abstract spaces and generalized functions1. The history of analysis is largely a history of making the infinite safe to use.

Ancient Roots: Exhaustion and the Geometry of the Infinite

Long before anyone wrote a limit symbol, Greek geometers confronted the infinite through geometry1. The method of exhaustion, associated with Eudoxus of Cnidus (fourth century BCE) and used brilliantly by Archimedes (third century BCE), computed areas and volumes by inscribing and circumscribing figures with polygons of ever more sides. Archimedes used it to show that the area of a parabolic segment is four-thirds the area of a certain inscribed triangle, and to bound the value of π between roughly 3.1408 and 3.1429. The method worked by a double contradiction: assume the area is larger than the claimed value, show a sufficiently fine polygon contradicts this; assume it is smaller, and get a contradiction the other way3. The infinite process was never completed—it was only used to rule out alternatives. This avoidance of actual infinity was characteristic of Greek mathematics, and it left the field without a general technique for handling infinite processes.

The Calculus of Newton and Leibniz

The decisive break came in the late seventeenth century, when Isaac Newton (working in the 1660s and 1670s, publishing his Principia in 1687) and Gottfried Wilhelm Leibniz (publishing his calculus papers in 1684 and 1686) independently created what became infinitesimal calculus. Their insight was algorithmic: derivatives and integrals could be computed by systematic rules, and the two operations were inverse to each other14. Leibniz's notation—$dy/dx$ and the integral sign—proved extraordinarily durable. The power of the method was immediate: it solved problems of tangents, areas, maxima and minima, and motion that had resisted classical geometry.

But the logical foundation was shaky5. Infinitesimals were quantities said to be smaller than any positive number yet not zero, and critics objected that such things were incoherent5. Bishop George Berkeley's 1734 polemic The Analyst mocked them as "the ghosts of departed quantities." The calculus worked, but no one could say precisely why. For over a century, analysis advanced on the strength of its results while its foundations remained contested.

The Epsilon-Delta Rigorization

The repair came in the nineteenth century, chiefly through Augustin-Louis Cauchy and Karl Weierstrass16. Cauchy's Cours d'analyse (1821) began recasting limits as the fundamental concept, defining continuity and convergence in terms of quantities becoming arbitrarily small. Weierstrass, lecturing in Berlin from the 1860s onward, completed the transformation with the epsilon-delta formulation: a sequence converges to $L$ if, for every $\epsilon > 0$, there is an $N$ such that all terms beyond $N$ lie within $\epsilon$ of $L$. A function is continuous at a point if, for every $\epsilon > 0$, there is a $\delta > 0$ such that inputs within $\delta$ of the point give outputs within $\epsilon$ of the value1.

This was more than notation. It eliminated infinitesimals from the official language of analysis and replaced them with finite, checkable inequalities. It also exposed pathologies that intuition had hidden1. Weierstrass exhibited a function continuous everywhere but differentiable nowhere (presented in 1872), shattering the assumption that continuity implied smoothness. The epsilon-delta framework made such examples rigorous and forced mathematicians to distinguish carefully between concepts they had conflated.

Arithmetizing the Continuum

A parallel campaign sought to ground the real numbers themselves1. If limits were to be defined rigorously, the number line could not remain an intuitive picture1. Richard Dedekind, in Stetigkeit und irrationale Zahlen (1872), defined a real number as a "cut" in the rationals—a partition of the rationals into two sets with every member of one less than every member of the other. Georg Cantor, in the same period, defined reals via Cauchy sequences of rationals. Both constructions built the continuum out of the rationals, which in turn rested on the integers. This "arithmetization of analysis" meant that the entire edifice of calculus could, in principle, be derived from the natural numbers. Giuseppe Peano and others pushed further, axiomatizing the natural numbers themselves (Peano's axioms, 1889).

The cost was abstraction: the real line became a completed infinite set with a specific structure, and questions about that structure became mathematical questions. Cantor's work on the cardinality of infinite sets—showing that the reals are uncountable while the rationals are countable—grew directly out of this program and opened a new field.

Measure Theory and the Lebesgue Integral

The Riemann integral, the standard integral of nineteenth-century calculus, integrates functions by slicing the domain into small intervals7. It fails for functions with too many discontinuities8. Henri Lebesgue's 1902 dissertation, Intégrale, longueur, aire, reversed the strategy: slice the range instead of the domain, and measure the size of the sets on which the function takes values in each slice. This required a theory of measure—a systematic way to assign sizes to subsets of the line—which Lebesgue built on earlier work by Émile Borel.

The Lebesgue integral integrates a far larger class of functions, behaves better under limits (the monotone convergence and dominated convergence theorems let one swap limits and integrals under checkable conditions), and made the function spaces of the twentieth century possible. Measure theory also gave probability a rigorous foundation when Andrey Kolmogorov axiomatized it in 1933. The trade-off was that measure theory itself required careful set-theoretic foundations, and the discovery of non-measurable sets (Vitali, 1905) showed that not every set can be assigned a size consistently.

Functional Analysis: Functions as Points

The early twentieth century saw analysis shift from studying individual functions to studying whole spaces of functions. The key move was to treat a function as a point in a space and ask geometric questions: how far apart are two functions, what does it mean for a sequence of functions to converge, what is the "shape" of the space? David Hilbert's work on integral equations (1904–1910) and his introduction of what became Hilbert space provided the model. Stefan Banach's Théorie des opérations linéaires (1932) systematized the theory of complete normed vector spaces—Banach spaces—and the linear operators between them.

Functional analysis supplied the language for quantum mechanics (states as vectors in Hilbert space, observables as operators), for partial differential equations, and for approximation theory. The Hahn-Banach theorem, the open mapping theorem, and the uniform boundedness principle became standard tools. The field also developed abstract spaces beyond Banach spaces—locally convex spaces, Fréchet spaces—to handle the function spaces that analysis actually needed.

Distributions and Generalized Functions

A persistent nuisance was that many operations analysts wanted—differentiating a discontinuous function, taking the Fourier transform of a constant—were undefined in classical terms. Sergei Sobolev (1930s) and Laurent Schwartz (1950–1951) resolved this by introducing distributions, also called generalized functions. A distribution is defined not by its values at points but by how it acts on a class of well-behaved "test functions" through integration9. The Dirac delta, long used informally in physics as a function that is zero everywhere except at one point and integrates to one, became a rigorous distribution9. Schwartz's Théorie des distributions (1950–1951) earned him a Fields Medal and gave analysis a flexible calculus for objects that classical theory could not handle. Distributions became indispensable in the theory of partial differential equations9.

Constructive and Intuitionistic Analysis

Not everyone accepted the classical foundations. L. E. J. Brouwer, from 1907 onward, developed intuitionism, a philosophy of mathematics that rejected the law of excluded middle (the principle that every statement is either true or false) for infinite collections and demanded constructive proofs—proofs that exhibit the object claimed to exist. In intuitionistic analysis, the real line itself behaves differently: not every real number is either rational or irrational in the constructive sense, and certain classical theorems fail. Brouwer's student Arend Heyting formalized intuitionistic logic in 1930.

A related but distinct tradition is Russian constructivism, associated with Andrey Markov Jr. and others from the 1950s, which demanded algorithmic constructibility. Errett Bishop's Foundations of Constructive Analysis (1967) showed that a large portion of classical analysis could be rebuilt constructively, which surprised mathematicians who assumed constructivism would be crippled. These schools remain minorities, but they have influenced computer science (via type theory and proof assistants) and forced classical mathematicians to notice which results depend on non-constructive principles.

Nonstandard Analysis

In 1960, Abraham Robinson found a way to rehabilitate infinitesimals rigorously. Using model theory, he constructed hyperreal numbers—an extension of the reals containing infinitesimal and infinite quantities—in which the epsilon-delta definitions of calculus could be replaced by infinitesimal reasoning that looked like the informal calculus of Newton and Leibniz but was logically sound. Nonstandard analysis (Robinson's Non-standard Analysis, 1966) offered an alternative foundation and has been used in economics, probability, and the study of differential equations. It remains a specialized approach rather than a replacement for standard methods, but it settled a centuries-old dispute by showing that infinitesimals could be made coherent after all.

The Present

Analysis today is plural. The epsilon-delta framework and measure theory remain the standard language of the field, taught in every university and used across mathematics and its applications. Functional analysis underpins quantum theory, numerical analysis, and the study of partial differential equations. Distributions are routine in engineering and physics9. Constructive and nonstandard approaches persist as active research programs, each with its own journals and communities. Open questions remain: the foundations of measure theory still interact with set theory (the existence of certain large cardinals affects what can be measured), the classification of operator algebras is incomplete, and the constructive rebuilding of analysis continues. What unites the field is the same question the Greeks faced: how to reason rigorously about the infinite and the infinitesimal, and what it costs to do so.

References

  1. Mathematical analysis (introduction, Ancient, Origins, Modernization, Continuity), Wikipedia
  2. History of calculus, Wikipedia
  3. Method of exhaustion, Wikipedia
  4. Fundamental theorem of calculus (introduction, History), Wikipedia
  5. Nonstandard analysis, Wikipedia
  6. Limit (mathematics) (History), Wikipedia
  7. Riemann integral, Wikipedia
  8. Lebesgue integral, Wikipedia
  9. Distribution (mathematical analysis), Wikipedia

The frameworks of Mathematical Analysis

Each framework is a way of working the field has used: when it held, and the question it answers.

  1. 370 BCE-210 BCE

    Method of Exhaustion

    How can we compute areas and volumes exactly?

    Fly to its tower in Nousfarer

  2. 1665-1820

    Calculus

    How can we compute rates of change and accumulations?

    Fly to its tower in Nousfarer

  3. 1821-Present

    Epsilon-Delta Rigorization

    How can we make the notion of limit precise?

    Fly to its tower in Nousfarer

  4. 1872-1970

    Arithmetization of Analysis

    What are the real numbers and how do they form a continuum?

    Fly to its tower in Nousfarer

  5. 1902-Present

    Measure Theory and the Lebesgue Integral

    How can we assign sizes to complicated sets and integrate more functions?

    Fly to its tower in Nousfarer

  6. 1904-Present

    Functional Analysis

    How can we study spaces of functions as geometric objects?

    Fly to its tower in Nousfarer

  7. 1907-Present

    Intuitionistic and Constructive Analysis

    How can we do mathematics constructively without excluded middle?

    Fly to its tower in Nousfarer

  8. 1936-Present

    Distribution

    How can we differentiate and integrate generalized functions?

    Fly to its tower in Nousfarer

  9. 1950-Present

    Russian Constructive Analysis

    How can we make analysis algorithmic and computable?

    Fly to its tower in Nousfarer

  10. 1960-Present

    Nonstandard Analysis

    Can infinitesimals be made rigorous?

    Fly to its tower in Nousfarer

  11. 1967-Present

    Bishop Constructive Analysis

    How can we develop analysis with explicit constructions?

    Fly to its tower in Nousfarer

How the ideas connect

Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.

Method of ExhaustionCalculusEpsilon-Delta RigorizationArithmetization of AnalysisMeasure Theory and the Lebesgue Integ…Functional AnalysisIntuitionistic and Constructive Analy…DistributionRussian Constructive AnalysisNonstandard AnalysisBishop Constructive Analysis

Beyond the field

Where Mathematical Analysis's ideas came from in other fields, and what they led to there: the strongest 12 of 101 links.

A city in Nousfarer: a red-carpeted stair climbing terraces to a crowned tower on an acropolis, more towers round it, forest and snowy mountains beyond.
A city in Nousfarer, the game played in the browser.

Mathematical Analysis is a planet in Nousfarer

Its city has a tower for each of these 11 frameworks. Land, walk in, and climb one concept at a time.

Land on Mathematical Analysis

Related fields

Fields its ideas are linked to

More in Analysis

See where Mathematical Analysis sits on the Map of Human Knowledge