Normal form (abstract rewriting)

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search

In abstract rewriting,[1] an object is in normal form if it cannot be rewritten any further, i.e. it is irreducible. Depending on the rewriting system, an object may rewrite to several normal forms or none at all. Many properties of rewriting systems relate to normal forms.

Definitions [edit]

Stated formally, if (A,→) is an abstract rewriting system, xA is in normal form if no yA exists such that xy, i.e. x is an irreducible term.

An object a is weakly normalizing if there exists at least one particular sequence of rewrites starting from a that eventually yields a normal form. A rewriting system has the weak normalization property or is (weakly) normalizing (WN) if every object is weakly normalizing. An object a is strongly normalizing if every sequence of rewrites starting from a eventually terminates with a normal form. An abstract rewriting system is strongly normalizing, terminating, noetherian, or has the (strong) normalization property (SN), if each of its objects is strongly normalizing.[2]

A rewriting system has the normal form property (NF) if for all objects a and normal forms b, b can be reached from a by a series of rewrites and inverse rewrites only if a reduces to b. A rewriting system has the unique normal form property (UN) if for all normal forms a, bS, a can be reached from b by a series of rewrites and inverse rewrites only if a is equal to b. A rewriting system has the unique normal form property with respect to reduction (UN) if for every term reducing to normal forms a and b, a is equal to b.[3]


This section presents some well known results. First, SN implies WN.[4]

Confluence (abbreviated CR) implies NF implies UN implies UN.[3] The reverse implications do not generally hold. {a→b,a→c,c→c,d→c,d→e} is UN but not UN as b=e and b,e are normal forms. {a→b,a→c,b→b} is UN but not NF as b=c, c is a normal form, and b does not reduce to c. {a→b,a→c,b→b,c→c} is NF as there are no normal forms, but not CR as a reduces to b and c, and b,c have no common reduct.

WN and UN imply confluence. Hence CR, NF, UN, and UN coincide if WN holds.[5]


For example, using the term rewriting system with a single rule g(x,y)→x, the term g(g(4,2),g(3,1)) can be rewritten as follows, applying the rule to the outermost occurrence [note 1] of g:

g(g(4,2),g(3,1)) → g(4,2) → 4.

Since no rule applies to the last term, 4, it cannot be rewritten any further, and hence is a normal form of the term g(g(4,2),g(3,1)) with respect to this term rewriting system. The rule system is strongly normalizing, since each rule application properly decreases term size and hence there cannot be an infinite rewrite sequence starting from any term. In contrast, the two-rule system { g(x,y) → x, g(x,x) → g(3,x) } is weakly, [note 2] but not strongly [note 3] normalizing, although each term not containing g(3,3) is strongly normalizing. [note 4] The term g(4,4) has two normal forms in this system, viz. g(4,4) → 4 and g(4,4) → g(3,4) → 3, hence the system is not confluent.

Another example: The single-rule system { r(x,y) → r(y,x) } has no normalizing properties (not weakly or strongly), since from any term, e.g. r(4,2) a single rewrite sequence starts, viz. r(4,2) → r(2,4) → r(4,2) → r(2,4) → ..., which is infinitely long.

Untyped lambda calculus[edit]

The pure untyped lambda calculus does not satisfy the strong normalization property, and not even the weak normalization property. Consider the term (application is left associative). It has the following rewrite rule: For any term ,

But consider what happens when we apply to itself:

Therefore the term is not strongly normalizing. And this is the only reduction sequence, hence it is not weakly normalizing either.

Typed lambda calculus[edit]

Various systems of typed lambda calculus including the simply typed lambda calculus, Jean-Yves Girard's System F, and Thierry Coquand's calculus of constructions are strongly normalizing.

A lambda calculus system with the normalization property can be viewed as a programming language with the property that every program terminates. Although this is a very useful property, it has a drawback: a programming language with the normalization property cannot be Turing complete, otherwise one could solve the halting problem by seeing if the program type-checks. That means that there are computable functions that cannot be defined in the simply typed lambda calculus (and similarly there are computable functions that cannot be computed in the calculus of constructions or System F), for example a self-interpreter.[6]

See also[edit]


  1. ^ Each occurrence of g where the rule is applied is highlighted in boldface.
  2. ^ Since every term containing g can be rewritten by a finite number of applications of the first rule to a term without any g, which is in normal form.
  3. ^ Since to the term g(3,3), the second rule can be applied over and over again, without reaching any normal form.
  4. ^ For a given term, let m and n denote the total number of g and of g applied to identical arguments, respectively. Application of any rule properly decreases the value of m + n, which is possible only finitely many times.


  1. ^ Franz Baader; Tobias Nipkow (1998). Term Rewriting and All That. Cambridge University Press. ISBN 9780521779203.
  2. ^ Ohlebusch, Enno (1998). "Church-Rosser theorems for abstract reduction modulo an equivalence relation". Rewriting Techniques and Applications. Lecture Notes in Computer Science. 1379: 18. doi:10.1007/BFb0052358. ISBN 978-3-540-64301-2.
  3. ^ a b Klop, J.W.; de Vrijer, R.C. (February 1989). "Unique normal forms for lambda calculus with surjective pairing". Information and Computation. 80 (2): 97–113. doi:10.1016/0890-5401(89)90014-X.
  4. ^ "logic - What is the difference between strong normalization and weak normalization in the context of rewrite systems?". Computer Science Stack Exchange. Retrieved 12 September 2021.
  5. ^ Ohlebusch, Enno (17 April 2013). Advanced Topics in Term Rewriting. Springer Science & Business Media. pp. 13–14. ISBN 978-1-4757-3661-8.
  6. ^ Riolo, Rick; Worzel, William P.; Kotanchek, Mark (4 June 2015). Genetic Programming Theory and Practice XII. Springer. p. 59. ISBN 978-3-319-16030-6. Retrieved 8 September 2021.