# Difference between revisions of "Uniqueness quantification"

en>David Eppstein (→See also: unstub) |
(→Proving uniqueness: added missing verb) |
||

Line 1: | Line 1: | ||

{{ | {{redirect|Unique}} | ||

{{ | {{inline citations|date=January 2013}} | ||

In [[mathematics]] and [[logic]], the phrase "there is '''one and only one'''" is used to indicate that exactly one object with a certain property exists. In [[mathematical logic]], this sort of [[quantification]] is known as '''uniqueness quantification''' or '''unique existential quantification'''. | In [[mathematics]] and [[logic]], the phrase "there is '''one and only one'''" is used to indicate that exactly one object with a certain property exists. In [[mathematical logic]], this sort of [[quantification]] is known as '''uniqueness quantification''' or '''unique existential quantification'''. | ||

Line 8: | Line 9: | ||

== Proving uniqueness == | == Proving uniqueness == | ||

The most common technique to proving uniqueness is to first prove existence of entity with the desired condition; then, to assume there exist two entities (say, a and b) that should both satisfy the condition, and logically deduce their equality, i.e. ''a'' = ''b''. | |||

As a simple high school example, to show ''x'' + 2 = 5 has only one solution, we assume there are two solutions first, namely, ''a'' and ''b'', satisfying ''x'' + 2 = 5. Thus | As a simple high school example, to show ''x'' + 2 = 5 has only one solution, we assume there are two solutions first, namely, ''a'' and ''b'', satisfying ''x'' + 2 = 5. Thus | ||

Line 24: | Line 25: | ||

== Reduction to ordinary existential and universal quantification == | == Reduction to ordinary existential and universal quantification == | ||

Uniqueness quantification can be expressed in terms of the [[existential quantifier|existential]] and [[universal quantifier|universal]] quantifiers of [[predicate logic]] by defining the formula | Uniqueness quantification can be expressed in terms of the [[existential quantifier|existential]] and [[universal quantifier|universal]] quantifiers of [[predicate logic]] by defining the formula | ||

∃!''x'' ''P(x)'' to mean | ∃!''x'' ''P(x)'' to mean literally, | ||

:<math>\exists x P(x) \wedge \neg \exists | :<math>\exists x\,( P(x) \, \wedge \neg \exists y\,(P(y) \wedge y \ne x))</math> | ||

which is the same as, | |||

:<math>\exists x P(x) \wedge \forall y\,(P(y) \to x | :<math>\exists x \, ( P(x) \wedge \forall y\,(P(y) \to y = x)).</math> | ||

An equivalent definition that has the virtue of separating the notions of existence and uniqueness into two clauses, at the expense of brevity, is | An equivalent definition that has the virtue of separating the notions of existence and uniqueness into two clauses, at the expense of brevity, is | ||

:<math>\exists x P(x) \wedge \forall y\, \forall z\,((P(y) \wedge P(z)) \to y = z).</math> | :<math>\exists x\, P(x) \wedge \forall y\, \forall z\,((P(y) \wedge P(z)) \to y = z).</math> | ||

Another equivalent definition with the advantage of brevity is | Another equivalent definition with the advantage of brevity is | ||

:<math>\exists x\,\forall y\,(P(y) \leftrightarrow x | :<math>\exists x\,\forall y\,(P(y) \leftrightarrow y = x).</math> | ||

== Generalizations == | == Generalizations == | ||

Line 38: | Line 39: | ||

== See also == | == See also == | ||

* [[One-hot]] | * [[One-hot]] | ||

== References == | |||

*{{cite book|last=Kleene|first=Stephen|title=Introduction to Metamathematics|year=1952|publisher=Ishi Press International|pages=199}} | |||

*{{cite book|last=Andrews|first=Peter B.|title=An introduction to mathematical logic and type theory to truth through proof|year=2002|publisher=Kluwer Acad. Publ.|location=Dordrecht|isbn=1-4020-0763-9|pages=233|edition=2. ed.}} | |||

[[Category:Quantification]] | [[Category:Quantification]] | ||

[[Category:One]] | [[Category:One]] | ||

[[Category:Mathematical terminology]] | [[Category:Mathematical terminology]] | ||

## Revision as of 20:53, 30 January 2014

{{#invoke:Hatnote|hatnote}}Template:Main other Template:Inline citations

In mathematics and logic, the phrase "there is **one and only one**" is used to indicate that exactly one object with a certain property exists. In mathematical logic, this sort of quantification is known as **uniqueness quantification** or **unique existential quantification**.

Uniqueness quantification is often denoted with the symbols "∃!" or ∃_{=1}". For example, the formal statement

may be read aloud as "there is exactly one natural number *n* such that *n* - 2 = 4".

## Proving uniqueness

The most common technique to proving uniqueness is to first prove existence of entity with the desired condition; then, to assume there exist two entities (say, a and b) that should both satisfy the condition, and logically deduce their equality, i.e. *a* = *b*.

As a simple high school example, to show *x* + 2 = 5 has only one solution, we assume there are two solutions first, namely, *a* and *b*, satisfying *x* + 2 = 5. Thus

By transitivity of equality,

By cancellation,

This simple example shows how a proof of uniqueness is done, the end result being the equality of the two quantities that satisfy the condition. We must say, however, that existence/expressibility must be proven before uniqueness, or else we cannot even assume the existence of those two quantities to begin with.

## Reduction to ordinary existential and universal quantification

Uniqueness quantification can be expressed in terms of the existential and universal quantifiers of predicate logic by defining the formula
∃!*x* *P(x)* to mean literally,

which is the same as,

An equivalent definition that has the virtue of separating the notions of existence and uniqueness into two clauses, at the expense of brevity, is

Another equivalent definition with the advantage of brevity is

## Generalizations

One generalization of uniqueness quantification is counting quantification. This includes both quantification of the form "exactly *k* objects exist such that …" as well as "infinitely many objects exist such that …" and "only finitely many objects exist such that…". The first of these forms is expressible using ordinary quantifiers, but the latter two cannot be expressed in ordinary first-order logic.

## See also

## References

- {{#invoke:citation/CS1|citation

|CitationClass=book }}

- {{#invoke:citation/CS1|citation

|CitationClass=book }}