Morse–Kelley set theory

From Wikipedia, the free encyclopedia

In the foundation of mathematics, Kelley–Morse (KM) or Morse–Kelley (MK) set theory is a first order axiomatic set theory that is closely related to Von Neumann–Bernays–Gödel set theory (NBG). MK allows the bound variables in the schematic formula appearing in the axiom schema of Class Comprehension to range over proper classes as well as sets. NBG restricts these bound variables to sets alone. MK is a proper extension of the canonical set theory ZFC and cannot be finitely axiomatized. NBG is a conservative extension of ZFC, and can be finitely axiomatized.

Contents

[edit] MK axioms and ontology

NBG and MK share a common ontology. The universe of discourse consists of classes. Classes which are members of other classes are called sets. A class which is not a set is a proper class. The primitive atomic sentences involve membership or equality.

With the exception of Class Comprehension, the following axioms are the same as those for NBG, inessential details aside. The symbolic versions of the axioms employ the following notational devices:

  • The upper case letters other than M, appearing in Extensionality, Class Comprehension, and Foundation, denote variables ranging over classes. A lower case letter denotes a variable that cannot be a proper class, because it appears to the left of an ∈. As MK is a one-sorted theory, this notational convention is only mnemonic;
  • The monadic predicate \ Mx, whose intended reading is "'the class x is a set," abbreviates \exist W(x \in W);
  • The empty set \varnothing is defined by \forall x (x \not \in \varnothing);
  • The class V, the universal class having all possible sets as members, is defined by \forall x (x \in V). V is also the Von Neumann universe.


Extensionality: Classes having the same members are the same class.

\forall X \, \forall Y \, ( \forall z \, (z \in X \leftrightarrow z \in Y) \rightarrow X = Y).
Note that a set and a class having the same extension are identical. Hence MK is not a two-sorted theory, appearances to the contrary notwithstanding.

Foundation: Each nonempty class A is disjoint from at least one of its members.

\forall A [A \not = \varnothing \rightarrow \exist b (b \in A \and \forall c (c \in b \rightarrow c \not\in A))].

Class Comprehension: Let φ(x) be any formula in the language of MK in which x is a free variable and Y is not free. φ(x) may contain parameters which are either sets or proper classes. More consequentially, the quantified variables in φ(x) may range over all classes and not just over all sets; this is the only way MK differs from NBG. Then there exists a class Y=\{x \mid \phi(x)\} whose members are exactly those sets x such that φ(x) comes out true. Formally, if Y is not free in φ:

\forall W_1 ... W_n \exist Y \forall x [x \in Y \leftrightarrow (\phi(x, W_1, ... W_n) \and Mx)].

Pairing: For any sets x and y, there exists a set z = {x,y} whose members are exactly x and y.

\forall x \, \forall y \, [ (Mx \and My) \rightarrow \exist z \, (Mz \and \forall s \, [ s \in z \leftrightarrow (s = x \, \or \, s = y)])].
Pairing licenses the unordered pair in terms of which the ordered pair, \langle x,y \rangle, may be defined in the usual way, as \ \{\{x\},\{x,y\}\}. With ordered pairs in hand, Class Comprehension enables defining relations and functions on sets as sets of ordered pairs, making possible the next axiom:

Limitation of Size: C is a proper class if and only if V can be mapped one-to-one into C.

\forall C [\lnot MC \leftrightarrow \exist F ( \forall x [Mx \rightarrow \exist s (s \in C \and \langle x, s \rangle \in F)] \and
\forall x \forall y \forall s [(\langle x, s \rangle \in F \and \langle y, s \rangle \in F) \rightarrow x = y])].
The formal version of this axiom resembles the axiom schema of replacement, and embodies the class function F. The next section explains how Limitation of Size is stronger than the usual forms of the axiom of choice.

Power set: Let p be a class whose members are all possible subsets of the set a. Then p is a set.

\forall a \, \forall p \, [(Ma \and \forall x \, [x \in p \leftrightarrow \forall y \, (y \in x \rightarrow y \in a)]) \rightarrow Mp].

Union: Let s=\bigcup a be the sum class of the set a, namely the union of all members of a. Then s is a set.

\forall a \, \forall s \, [(Ma \and \forall x \, [x \in s \leftrightarrow \exist y \, (x \in y \and y \in a)]) \rightarrow Ms].

Infinity: There exists an inductive set y, meaning that (i) the empty set is a member of y; (ii) if x is a member of y, then so is x \cup \{x\}..

\exist y[My \and \varnothing \in y \and \forall z(z \in y \rightarrow \exist x [x \in y \and \forall w (w \in x \leftrightarrow [w = z \or w \in z])] )].


Note that p and s in Power Set and Union are universally, not existentially, quantified, as Class Comprehension suffices to establish the existence of p and s. Power Set and Union only serve to establish that p and s cannot be proper classes.

The above axioms are shared with other set theories as follows:

  • ZFC and NBG: Pairing, Power Set, Union, Infinity;
  • NBG (and ZFC, if quantified variables were restricted to sets): Extensionality, Foundation;
  • NBG: Limitation of Size;
  • ML: Extensionality, Class Comprehension (in NF, comprehension is restricted to stratified formulas).

[edit] Discussion

Monk (1980) and Rubin (1967) are set theory texts built around MK; Rubin's ontology includes urelements. These authors and Mendelson (1997: 287) submit that MK does what we expect of a set theory while being less cumbersome than ZFC and NBG.

MK is strictly stronger than ZFC and its conservative extension NBG, the other well-known set theory with proper classes. In fact, NGB—and hence ZFC—can be proved consistent if MK is. MK's strength stems from its axiom schema of Class Comprehension being impredicative, meaning that φ(x) may contain quantified variables ranging over classes. The quantified variables in NBG's axiom schema of Class Comprehension are restricted to sets; hence Class Comprehension in NBG must be predicative. (Separation with respect to sets is still impredicative in NBG, because the quantifiers in φ(x) may range over all sets.) The NBG axiom schema of Class Comprehension can be replaced with finitely many of its instances; this is not possible in MK. MK is consistent relative to ZFC augmented by an axiom asserting the existence of strongly inaccessible ordinals.

The only advantage of the axiom of limitation of size is that it implies the axiom of global choice. Limitation of Size does not appear in Rubin (1967), Monk (1980), or Mendelson (1997). Instead, these authors invoke a usual form of the local axiom of choice, and an "axiom of replacement,"[1] asserting that if the domain of class function is a set, its range is also a set. Replacement can do prove everything that Limitation of Size proves, except prove some form of the axiom of choice.

Limitation of Size plus I being a set (hence the universe is nonempty) renders provable the sethood of the empty set; hence no need for an axiom of empty set. Such an axiom could be added, of course, and minor perturbations of the above axioms would necessitate this addition. The set I is not identified with the limit ordinal ω, as I could be a set larger than ω. In this case, the existence of ω would follow from either form of Limitation of Size.

The class of von Neumann ordinals can be well-ordered. It cannot be a set (under pain of paradox); hence that class is a proper class, and all proper classes have the same size as V. Hence V too can be well-ordered.

MK can be confused with second-order ZFC, ZFC with second-order logic (representing second-order objects in set rather than predicate language) as its background logic. The language of second-order ZFC is similar to that of MK (although a set and a class having the same extension can no longer be identified), and their syntactical resources for practical proof are almost identical (and are identical if MK includes the strong form of Limitation of Size). But the semantics of second-order ZFC are quite different from those of MK; whatever is true in all models of second-order ZFC in which the classes are all the collections of the sets, is also provable (because the logic is second-order). Hence second-order ZFC "decides" many classical conundrums, such as the continuum hypothesis, although in a sense that is arguably not very practical. Second-order ZFC also uniquely determines the rank, if any, of the cumulative hierarchy up to the first inaccessible cardinal.

[edit] Model theory

ZFC, NBG, and MK each have models describable in terms of V, the standard model of ZFC and the von Neumann universe. Let the inaccessible cardinal κ be a member of V. Also let Def(X) denote the Δ0 definable subsets of X (see constructible universe). Then:

[edit] History

MK was first set out in an appendix to J. L. Kelley's (1955) General Topology, using the axioms given in the next section. The system of Anthony Morse's (1965) A Theory of Sets is equivalent to Kelley's, but formulated in an idiosyncratic formal language rather than, as is done here, in standard first order logic. The first set theory to include impredicative class comprehension was Quine's ML, that built on New Foundations rather than on ZFC.[2] Impredicative class comprehension was also proposed in Mostowski (1951) and Lewis (1991).

[edit] The axioms in Kelley's General topology

The axioms and definitions in this section are, but for a few inessential details, taken from the appendix to Kelley (1955). The explanatory remarks below are not his. The appendix states 181 theorems and definitions, and warrants careful reading as an abbreviated exposition of axiomatic set theory by a working mathematician of the first rank. Kelley introduced his axioms gradually, as needed to develop the topics listed after each instance of Develop below.

Notations appearing below and now well-known are not defined. Peculiarities of Kelley's notation include:

  • He did not distinguish variables ranging over classes from those ranging over sets;
  • domain f and range f denote the domain and range of the function f; this peculiarity has been carefully respected below;
  • His primitive logical language includes class abstracts of the form \ \{x : A(x)\}, "the class of all sets x satisfying A(x)."


Definition: x is a set [and hence not a proper class] iff for some y, x \in y.


I. Extent: For each x and each y, x=y iff for each z, z \in x when and only when z \in y.

I is a variant of the axiom of extensionality, except that its scope includes proper classes as well as sets.

II. Classification (schema): An axiom results if in

For each β, \beta \in \{\alpha:A\} iff β is a set and B,

'α' and 'β' are replaced by variables, ' A ' by a formula Æ, and ' B ' by the formula obtained from Æ by replacing each occurrence of the variable which replaced α by the variable which replaced β.

Develop: Algebra of sets. Existence of the null class and the universal class V.


III. Subsets: If x is a set, there exists a set y such that for each z, if z \subseteq x, then z \in y.

Develop: V is not a set. Existence of singletons. Separation provable. Proof sketch of Power Set: for any class z which is a subclass of the set x, the class z is a member of the set y whose existence III asserts. Hence z is a set.


IV. Union: If x and y are both sets, then x \cup y is a set.

The import of IV is that of Pairing above, whose proof sketch goes as follows. The singleton {x} of a set x is a set because it is a subclass of the power set of x (by two applications of III). Then IV implies that {x,y} is a set if x and y are sets.

Develop: Unordered and ordered pairs, relations, functions, domain, range, function composition.


V. Substitution: If f is a [class] function and domain f is a set, then range f is a set.

The import of V is that of the "axiom of replacement" appearing in textbook treatments of NBG and MK.

VI. Amalgamation: If x is a set, then \bigcup x is a set.

The import of VI is that of Union above. V and VI may be combined into one axiom.[3]

Develop: Cartesian product, injection, surjection, bijection, order theory.

VII. Regularity: If x \neq \varnothing there is a member y of x such that x \cap y = \varnothing.

The import of VII is that of Foundation above.

Develop: Ordinal numbers, transfinite induction.

VIII. Infinity: There exists a set y, such that \varnothing \in y and x \cup \{x\} \in y whenever x \in y.

VIII asserts the unconditional existence of two sets, the infinite inductive set y, and the null set \varnothing. \varnothing is a set simply because it is a member of y. Up to this point, evertyhing that has been proved to exist is a class, and Kelley's discussion of sets was entirely hypothetical.

Develop: Natural numbers, N is a set, Peano axioms, integers, rational numbers, real numbers.


Definition: c is a choice function iff c is a function and c(x) \in x for each member x of domain c.

IX. Choice: There exists a choice function c whose domain is V - \{\varnothing\}..

IX is very similar to the axiom of global choice derivable from Limitation of Size.

Develop: Equivalents of the axiom of choice. As is the case with ZFC, the cardinal numbers require some form of this axiom. Theorems of Cantor and Schroeder-Bernstein.

[edit] Notes

  1. ^ See, e.g., Mendelson (1997), p. 239, axiom R.
  2. ^ The locus citandum for ML is the 1951 ed. of Quine's Mathematical Logic. However, the summary of ML given in Mendelson (1997), p. 296, is easier to follow. Mendelson's axiom schema ML2 is identical to the above axiom schema of Class Comprehension.
  3. ^ Kelley (1955), p. 261, fn †.

[edit] References

  • John L. Kelley 1975 (1955) General Topology. Springer. Earlier ed., Van Nostrand. Appendix, "Elementary Set Theory."
  • Lemmon, E. J. (1986) Introduction to Axiomatic Set Theory. Routledge & Kegan Paul.
  • David K. Lewis (1991) Parts of Classes. Oxford: Basil Blackwell.
  • Mendelson, Elliott (1997). Introduction to Mathematical Logic. Chapman & Hall. ISBN 0-534-06624-0.  The definitive treatment of the closely related set theory NBG, followed by a page on MK. Harder than Monk or Rubin.
  • Monk, J. Donald (1980) Introduction to Set Theory. Krieger. Easier and less thorough than Rubin.
  • Morse, A. P., (1965) A Theory of Sets. Academic Press.
  • Mostowski, Andrejz (1951) "Some impredicative definitions in the axiomatic set theory," Fundamenta Matheticae 37: 111-24.
  • Rubin, Jean E. (1967) Set Theory for the Mathematician. San Francisco: Holden Day. More thorough than Monk; the ontology includes urelements.

[edit] External links

From Foundations of Mathematics (FOM) discussion group:

Languages