Computable ordinal
In mathematics, specifically computability and set theory, an ordinal is said to be computable or recursive if there is a computable well-ordering of a computable subset of the natural numbers having the order type . This means that given any it is decidable whether , and given any it is decidable whether . Equivalently, is computable if it either is finite or is the order type of a computable well-ordering of all natural numbers.[1]
Examples
[edit]All computable ordinals are by definition countable. Conversely, for many countable ordinals, the "natural" witnesses for countability are also witnesses for computability. For example, the natural ordering of all natural numbers has order type . Since there exists a Turing machine that decides , this means that is a computable ordinal.
As another example, the following is the "canonical" construction of a well-ordering of all natural numbers with order type : An algorithm that decides can be as follows: Return true if is even and is odd, false if is odd and is even, and otherwise return . Therefore is also computable. In fact, with similar constructions, it can be shown that the successor of a computable ordinal and the sum, product, and power of a pair of computable ordinals are all computable.
The set of all computable ordinals is closed downwards, i.e., if is computable and , then is computable too.[1] This is because any well-ordering with order type has an initial segment with order type , where (for some fixed ) is a computable subset of .
Church–Kleene ordinal
[edit]The supremum of all computable ordinals is called the Church–Kleene ordinal, the first nonrecursive ordinal, and denoted by .[2] The Church–Kleene ordinal is a limit ordinal. An ordinal is computable if and only if it is smaller than .[3] Since there are only countably many computable binary relations, there are also only countably many computable ordinals. Thus, is countable.
The computable ordinals are exactly the ordinals that have an ordinal notation in Kleene's .[4]
See also
[edit]Notes
[edit]- ^ a b Sacks (1990), p. 9.
- ^ Sacks (1990), p. 10.
- ^ This follows immediately from downward closure and the definition of .
- ^ Sacks (1990), Theorem 4.4.
References
[edit]- Rogers, Hartley Jr. (1967), The Theory of Recursive Functions and Effective Computability, MIT Press, ISBN 0-07-053522-1
- Sacks, Gerald (1990), Higher Recursion Theory, Perspectives in mathematical logic, Springer-Verlag, ISBN 0-387-19305-7