1
BOREL AND BAIRE REDUCIBILITY
by
Harvey M. Friedman
Department of Mathematics
Ohio State University
Columbus, Ohio 43210
[email protected]
www.math.ohiostate.edu/~friedman/
May 25, 1999
INTRODUCTION
The Borel reducibility theory of Polish equivalence
relations, at least in its present form, was initiated in
[FS89]. There is now an extensive literature on this topic,
including fundamental work on the GlimmEffros dichotomy in
[HKL90], on countable Borel equivalence relations in [DJK94],
and on Polish group actions in [BK96]. Current principal
contributors include H. Becker, R. Dougherty, L. Harrington,
G. Hjorth, S. Jackson, A.S. Kechris, and A. Louveau, R. Sami,
and S. Solecki.
A Polish space is a topological space that is separable and
completely metrizable. The Borel subsets of a Polish space
form the least
σ
algebra containing the open subsets. A Borel
function from one Polish space to another is a function such
that the inverse image of every open set is Borel. Two Polish
spaces are Borel isomorphic if and only if there is a oneone
onto Borel function from the first onto the second. This is
an equivalence relation. Any two uncountable Polish spaces
are Borel isomorphic. See [Ke94].
We also consider Baire measurable subsets of a Polish space.
A nowhere dense set in a Polish space is a set whose closure
contains no nonempty open set. A meager subset of a Polish
space is a set which is the countable union of nowhere dense
sets. A Baire (measurable) subset of a Polish space is a set
whose symmetric difference with some open set is meager. A
comeager subset of a Polish space is a subset whose
complement is meager. We have the fundamental Baire category
theorem: A Polish space is not meager.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document2
A function from one Polish space to another is said to be
Baire if and only if the inverse image of every open set is
Baire.
A Polish equivalence relation is a pair (X,E), where X is a
Polish space and E is an equivalence relation on X. A Borel
(analytic) equivalence relation is a Polish equivalence
relation (X,E), where E is as a Borel measurable (analytic)
subset of X
2
.
Let (X,E
1
) and (Y,E
2
) be Polish equivalence relations. We say
that h:X
→
Y is a reduction from (X,E
1
) into (Y,E
2
) if and
only if for all x,y
∈
X, E
1
(x,y)
↔
E
2
(h(x),h(y)). We say that
(X,E1) is Borel reducible to (Y,E
2
) if and only if there is a
Borel reduction from (X,E
1
) to (Y,E
2
). We also say that (X,E
1
)
is Baire reducible to (Y,E
2
) if and only if there is a Baire
reduction from (X,E
1
) to (Y,E
2
).
Let STR = (STR,
≈
) be the analytic equivalence relation of all
structures of countable similarity type (in the sense of
model theory) whose domain is
ω
, under isomorphism. Also let
BR = (P(
ω
x
ω
),
≈
), be the analytic equivalence relation of
binary relations on
ω
under isomorphism. It is well known
that BR is not a Borel equivalence relation and BR is Borel
isomorphic to STR. It is generally simpler to work with BR.
The condition that a Polish equivalence relation E is Borel
This is the end of the preview.
Sign up
to
access the rest of the document.
 Fall '08
 JOSHUA
 Math, Equivalence relation, Descriptive set theory, comeager Borel

Click to edit the document details