Newtonpedia ES
Home › Mathematics › Hyperreal number
Hyperreal number
Image: Wikimedia Commons
Mathematics · 1948

Hyperreal number

An extension of the real numbers that adds infinite and infinitesimal quantities while keeping the basic rules of algebra.

Year
1948
People involved
Edwin Hewitt
Section
Mathematics

History

Hyperreal numbers extend the real numbers with quantities that are infinitely large and quantities that are infinitely small, called infinitesimals. Their appeal is that they give a rigorous meaning to ideas that mathematicians and physicists had used for centuries without fully justifying them, such as the derivatives and integrals written by Leibniz. An infinitesimal is a quantity smaller than any positive real number, and its reciprocal is an infinite number.

Before modern formalization, the idea of infinitesimal magnitudes appeared intuitively in ancient Greek mathematics. In the definition usually attributed to that tradition, a length a is infinitesimal relative to b if no whole-number multiple of a ever exceeds b. This clashes with the Archimedean property of the reals, which says that adding enough copies of any quantity eventually surpasses any other. Archimedes avoided infinitesimals in his proofs and preferred the method of exhaustion. Later, Gottfried Leibniz proposed a theory of infinitesimals and infinite numbers. It lacked solid logical foundations, yet it let physicists carry out the calculations they needed, especially for differential equations. In the nineteenth century, Cauchy, Weierstrass, Cantor and Bolzano gave precise definitions of limit and continuity, and calculus rested on them. That rigor came at a cost: the resulting formalism was heavy and not very intuitive.

Edwin Hewitt introduced the term "hyper-real" in 1948. The modern formalization owes most to Abraham Robinson, who in the 1960s showed that the hyperreals are logically consistent if the reals are. This removed the worry that proofs involving infinitesimals might be unsound, provided they followed the logical rules he set out. Robinson called the resulting field nonstandard analysis. Wilhelmus Luxemburg also made important contributions in the same decade.

The central tool is the transfer principle: any statement of the form "for every number x..." that holds for the reals also holds for the hyperreals. So addition is commutative in both systems. Differences appear in statements that quantify over sets, for example, the hyperreals contain an element larger than every natural number, something impossible among the reals. Within this framework, a derivative is defined as the standard part of a difference quotient with an infinitesimal increment, and an integral as the standard part of a suitable infinite sum. The standard part of a finite hyperreal is the nearest real number to it.

Nonstandard analysis has not replaced classical analysis, which remains the common language of mathematics. It offers an alternative route to differentiation, integration and reasoning about limits that some find more intuitive. Its main present-day role is in the logical foundations of calculus and in model theory, where it remains an active research tool.

Timeline

  • 1948Edwin Hewitt introduces the term "hyper-real".
  • 1955Łoś proves the theorem underlying the transfer principle for ultrapowers.
  • 1960Wilhelmus Luxemburg carries out early work related to nonstandard analysis.
  • 1966Abraham Robinson develops nonstandard analysis and shows the hyperreals are logically consistent.

Trivia and anecdotes

A disputed name

The name is somewhat misleading, since there is no single ordered field that is referred to as the hyperreals.

Other milestones in Mathematics