5

0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


Localizations of innite subsets of
Andrzej Roslanowski

Institute of Mathematics
The Hebrew University of Jerusalem
Jerusalem, Israel
and
Mathematical Institute of Wroc law University
50384 Wroc law, Poland
Saharon Shelah

Institute of Mathematics
The Hebrew University of Jerusalem
Jerusalem, Israel
and
Department of Mathematics
Rutgers University
New Brunswick, NJ 08854, USA
done: September 1992
printed: October 6, 2003

The research was partially supported by Polish Committee of Scientic Research, Grant
KBN 654/2/91

The second author would like to thank Basic Research Foundation of The Israel Academy
of Sciences and Humanities for partial support. Publication number 501.
0
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 1
0 Introduction
0.1 Preliminaries
When we say reals we mean one of the nicely dened Polish spaces or their
(nite or countable) products like: the real line R, the Cantor space 2

, the
Baire space

or the space of innite sets of integers []

.
In the present paper we are interested in properties of forcing notions (or, gen-
erally, extensions of models of ZFC) which measure in a sense the distance
between the ground model reals and the reals in the extension. In particular we
look at the ways the new reals can be localized (or: aproximated) by old
reals. There are two extreme cases here: there are no new reals and the old reals
are countable. However, between these two extremes we have a wide spectrum
of properties among which the localizations by slaloms seem to be the most
popular. A systematic study of slaloms and related localization properties and
cardinal invariants was presented in [Bar1].
A slalom is a function S : []
<
such that (n)([S(n)[ = n + 1). We
say that a slalom S localizes a function f

whenever (n)(f(n) S(n)).

In this situation we can think that the slalom S is an approximation of the
function f. It does not determine the function but it provides some bounds
on possible values of f. Bartoszy nski, Cicho n, Kamburelis et al. studied the
localization by slaloms and those investigations gave the following surprising
result.
Theorem 0.1 (Bartoszy nski, [Bar2]) Suppose that V V

are models of
ZFC. Then the following conditions are equivalent:
1. Any function from

can be localized by a slalom from V.

2. Any Borel (Lebesgue) null set coded in V

can be covered by a Borel null

set coded in V.
On localizations by slaloms see Chapter VI of [Sh:b] too; other localizations of
slalom-like type appeared in [GoSh:448].
A stronger localization property was considered in [NeRo]. Fix a natural number
k 2. By a k-tree on we mean a tree T
<
such that each nite node
in T has at most k immediate successors in T. We say that a k-tree T localizes
a function f

whenever f is a branch through T (i.e. (n)(fn T)).

Clearly, if any function from

can be localized by a k-tree from V, k 2,

V V

then each function from

can be localized by a slalom from

V. Moreover the localization by a k-tree implies the localization by a k +1-tree
(but not conversely).
In the denition below we formulate general localization properties for Polish
spaces X, Y . In practice, however, these spaces will be various examples of reals
only.
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 2
Denition 0.2 Assume that X, Y are Polish spaces and R XY is a Borel
relation. Suppose that V V

are models of ZFC and that all parameters

we need are in V. We say that the pair (V, V

) has the property of the R-

localization if
(xXV

)(y Y V)((x, y) R)
If x X V

, y Y V and (x, y) R they we say that y R-localizes x.

In the examples we gave earlier X was

and Y was the space of slaloms or

the space of all k-trees, respectively. The respective relations should be obvious.
Those localizations were to approximate functions in an extension by objects
from the ground model. They are not useful if we consider innite subsets of .
Though each member of []

can be identied with its increasing enumeration,

the localization (either by slaloms or by k-trees) of the enumeration does not
provide satisfactory information on successive points of the set. The localization
gives us candidates for the n-th point of the set but the same candidates can
appear several times for distinct n. That led to a suggestion that we should
consider disjoint subsets of as sets of candidates for successive points of the
localized set (the approach was suggested by B. W eglorz). Now we have two
possibilities. Either we can demand that each set from the localization contains
a limited number of members of the localized set or we can postulate that each
intersection of that kind is large. Localizations of this kind are studied in section
1. In the second section we investigate localizations of innite subsets of by
sets of integers from the ground model. These localizations might be thought
as localizations by partitions of into successive intervals. A starting point for
our considerations was the following observation.
Proposition 0.3 Suppose that V V

are (transitive) models of ZFC. Then:

1. V

is unbounded in V

if and only if
for every set X []

there exists a set Y in[]

V such that
innitely often between two successive points of Y there are at least 2
points of X.
2. V

is dominating in V

if and only if
for every set X []

there exists a set Y []

V such that for

all but nitely many pairs of two successive points of Y there are at
least 2 points of X between them.
Now we try to replace the quantier for innitely many above by stronger quan-
tiers (but still weaker than for all but nitely many), like for innitely many
n, for both n and n +1. Finally, in section 3 we formulate several corollaries to
the results of previous sections for cardinal invariants related to the notions we
study.
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 3
0.2 Notation
Our notation is rather standard and essentially compatible with that of [Jec]
and [Kun]. In forcing considerations, however, we will use the convention that
a stronger condition is the greater one.
Basic Denitions 0.4 1. A tree on is a set T
<
closed under initial
segments. For the tree T the body [T] of T is the set
x

: (l )(xl T).
If t T then succ
T
(t) = s T : t s & lh(t) + 1 = lh(s).
2. By a model of ZFC we will mean a transitive model of (enough of ) ZFC.
Models of ZFC will be denoted by V, V

etc.
3. We will be interested in extensions of models, i.e. in pairs (V, V

) of
models such that V V

. If a property of an extension is dened then

we extend this denition to notions of forcing. We say that a notion of
forcing P has the property whenever for any generic lter G P over V
the extension V V[G] has the considered property.
4. We will use the quantiers (

n) and (

n) as abbreviations for
(m )(n > m) and (m )(n > m),
respectively.
5. The Baire space

of all functions from to is endowed with the

partial order

:
f

g (

n)(f(n) g(n)).
A family F

is unbounded in (

) if
(g

)(f F)(f

g)
and it is dominating in (

) if
(g

)(f F)(g

f).
6. The unbounded number b is the minimal size of an unbounded family in
(

); the dominating number d is the minimal size of a dominating

family in that order.
7. The size of the continuum is denoted by c, []

stands for the family of

innite co-innite subsets of .
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 4
0.3 Acknowledgement
We would like to thank Professor Uri Abraham for his helpful comments on the
paper.
1 R

k
, R

k
- localizations
In this section we show that a localization of innite subsets of suggested by
B. W eglorz implies that the considered extension adds no new real.
Denition 1.1 1. A partition of into nite sets is a sequence K
n
: n )
of disjoint nite sets such that

n
K
n
= .
2. T
k
is the set of all partitions K
n
: n ) of into nite sets such that
(n)([K
n
[ > k).
[Note that T
k
is a
0
2
-subset of ([]
<
)

so it is a Polish space.]
3. We dene relations R

k
[]

T
k
and R

k
[]

T
k+1
by
(X, K
n
: n)) R

k
(

n)([X K
n
[ k)
(X, K
n
: n)) R

k
(

n)([X K
n
[ k).
Their complements (in []

T
k
, []

T
k+1
) are denoted by cR

k
, cR

k
,
respectively.
If we want to approximate an innite co-innite subset of by an object in
a given model we can look for a separation of distinct members of the set by
a sequence of sets from the model. Thus we could ask if it is possible to nd
a partition of (in V) such that the localized set is a partial selector of the
partition. More generally we may ask for R

k
localization; recall denition 0.2.
Thus the R

k
-localization property means that for every innite set of integers X
from the extension there exists a partition K
n
: n ) T
k
from the ground
model such that for almost all n the intersection X K
n
is of size at most
k. The following result shows that the R

k
-localization fails if we add new reals.
Theorem 1.2 Suppose that V V

are models of ZFC such that V 2

,=
V

. Then there is a set X []

such that for no k there is a

partition K
n
: n ) V of such that
(n)([X K
n
[ k & [K
n
[ > k).
Consequently the extension V V

does not have the R

k
-localization property
(for any k).
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 5
Proof Let x 2

be a new real (i.e. x V

V) and let
X =
<
xi : i .
As we can identify with
<
we may think that X []

. Now the following

claim works.
Claim 1.2.1 Suppose that x 2

, x / V. Let K
n
: n ) V be a
sequence of disjoint subsets of
<
such that (n)([K
n
[ > k), k . Then
for some n the set K
n
xi : i has at least k + 1 points.
Proof of Claim: Suppose not. Thus for each n we have
[K
n
xi : i [ k and thus K
n
xi : i ,= .
First note that each K
n
is nite. If not then the tree
s 2
<
: (t K
n
)(s t)
has exactly one innite branch - the branch is x. As the tree is in V we would
get x V.
Let
u(n) = maxlh(s) : s K
n
and d(n) = minlh(s) : s K
n

(remember that each K

n
is nite). Choose an increasing sequence n

: <)
(in V) such that
( )(u(n

) < d(n
+1
))
(possible as the K
n
s are disjoint). Let
F

= sd(n

) : s K
n

.
Note that xd(n

) F

as an initial segment of x belongs to K

n

. Moreover
[F

[ k +1 as only one element of F

may be an initial segment of x and above

each member of F

there is an element of K
n

(remember [K
n

xi : i [
k). Clearly F

: ) V. Consider the set

A = y 2

: ( )(yd(n

) F

).
It is a nite set from V. But x A a contradiction.
Thus the R

k
-localization is the trivial one. The complementary cR

k
-localization
is not of special interest either. Every extension V V

has the cR

0
-localization.
The description of the cR

k
-localization for k > 0 is given by the following ob-
servation.
Proposition 1.3 Let V V

be an extension of models of ZFC. Then the

following conditions are equivalent:
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 6
1. For each k > 0 the pair (V, V

) has the cR

k
-localization.
2. For some k > 0 the pair (V, V

) has the cR

k
-localization.
3. V

is unbounded in V

.
Proof 2 3 Let an increasing function f

be given. Take an
increasing function f

such that
(n )(f(f

(n)) + 1 < f

(n + 1)).
Consider the set rng(f

) []

. Let K
n
: n ) V be the partition
of given for this set by the cR

k
-localization. Let k
n
= min K
n
and put
g(k
n
) = 1 + max K
n
. Extend g to putting g(m) = 0 if m / k
n
: n .
Clearly g V.
Note that [K
n
rng(f

)[ > k > 0 implies that f(k

n
) < g(k
n
). Consequently
(

m)(f(m) < g(m)).

3 1 Given k > 0. Let X []

. Dene
f(n) = minm > n : [X [n, m)[ > 2k for n .
Since V

is unbounded in V

we nd an increasing function g V

such that (

n)(f(n) < g(n)). Let k

n
be dened by:
k
0
= 0, k
n+1
= k + 1 +k
n
+g(k
n
).
Put K
n
= [k
n
, k
n+1
). Clearly K
n
: n ) T
k
V. Now suppose that
m K
n
is such that f(m) < g(m). As g is increasing we have g(m) < g(k
n+1
) <
k
n+2
. Consequently [[m, k
n+2
) X[ > 2k and hence either [K
n
X[ > k or
[K
n+1
X[ > k. Hence (

n)([K
n
X[ > k).
In a similar way one can prove the analogous result for the cR

k
-localization.
Proposition 1.4 Let V V

be models of ZFC. Then the following condi-

tions are equivalent:
1. For each k the pair (V, V

) has the cR

k
-localization.
2. For some k the pair (V, V

) has the cR

k
-localization.
3. V

is a dominating family in V

.
For the R

k
-localization we did not nd a full description. First note that the
requirement that members of the partition have to have at least k +2 elements
(i.e. R

k
[]

T
k+1
) is to avoid a trivial localization. If we divide into
k + 1-element intervals then for each set X []

innitely many intervals

contain at most k members of X.
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 7
Denition 1.5 Let V V

be models of ZFC and let k < l < . A set X

[]

is called (l, k)-large (over V) if for every sequence K

n
: n ) V
of disjoint l-element subsets of we have: (

n )([ K
n
X [ > k).
[Note that we do not require that K
n
: n ) is a partition of .]
Theorem 1.6 All R

k
-localizations (for k ) are equivalent, i.e. if an exten-
sion V V

has the R

k
-localization property for some k then it has the
R

k
-localization for each k .
Proof Let V V

be models of ZFC. Let k < l < .

Claim 1.6.1 If X []

is (l, k)-large, k + 1 < l < and m 1 then X

is (lm, lm(l k))-large.
Proof of Claim: Let K
n
: n ) V be a sequence of disjoint subsets of
of the size lm. Let K
n
= a
n,i
: i < lm be the increasing enumeration
and let K
A
n
= a
n,i
: i A for A lm. Fix a set A [lm]
l
and consider
the sequence K
A
n
: n ) []
l
. This sequence is in V and its members
are disjoint. Thus we nd N(A) such that (n N(A))([ K
A
n
X [ > k).
Let N = maxN(A) : A [lm]
l
. Then for each n N and A [lm]
l
we
have [ K
A
n
X [ > k and hence [ K
A
n
X [ < l k. Hence we conclude
that [ K
n
X [ < l k for each n N (just take a suitable A). Thus
[ K
n
X [ > lm(l k) for each n N and the claim is proved.
Claim 1.6.2 If X []

is (l, k)-large, k +1 < l < and m l k then

X is (m, m(l k))-large.
Proof of Claim: Let K
n
: n ) V be a sequence of disjoint m-element
sets. Put K

n
= K
ln
K
ln+1
. . . K
ln+(l1)
. Clearly K

n
: n ) V,
[K

n
[ = lm and the sets K

n
s are disjoint. It follows from 1.6.1 that the set X
is (lm, lm(l k))-large and hence there is N such that
(n N)([ K

n
X [ > lm(l k)).
So [ K

n
X [ < l k for n N and hence [ K
n
X [ < l k for n lN. This
implies (n lN)([ K
n
X [ > m(l k)) and the claim is proved.
Claim 1.6.3 Assume that the extension V V

has the R

0
-localization prop-
erty. Then it has the R

k
-localization for each k .
Proof of Claim: Let k > 0 and assume that the R

k
-localization fails. Then we
have a set X []

witnessing it, i.e. such that

(K
n
: n ) T
k+1
V)(

n )([ K
n
X [ > k).
Then, in particular, the set X is (l, k)-large for each l > k + 1. By claim
1.6.2 it is (l, 0)-large for each l 2. By the R

0
-localization we nd a partition
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 8
K
n
: n ) T
1
V such that (

n )(K
n
X = ). Each set K
n
we
partition into 2- and 3-element sets:
K
n
=

K
2
n,i
: i w
2
n

K
3
n,i
: i w
3
n

(everything should be done in V, of course). Next look at

K
2
n,i
: n , i w
2
n
), K
3
n,i
: n , i w
3
n
) V.
These are sequences of disjoint 2- (or 3-, respectively) element subsets of . At
least one of them is innite; for simplicity we assume that both are innite. As
X is both (2, 0)- and (3, 0)-large we nd N such that for each n N, j 2, 3
and i w
j
n
we have K
j
n,i
X ,= . But this implies that for each n N the
intersection K
n
X is not empty - a contradiction to the choice of K
n
: n ).
Claim 1.6.4 Assume that k , V V

has the R

k
-localization property.
Then the extension V V

has the R

0
-localization.
Proof of Claim: Suppose that the R

0
-localization fails and this is witnessed by
a set X []

. As earlier we conclude from this that the set X is (l, 0)-large

and hence, by 1.6.2, it is (l + k, k)-large for each l 2. By the R

k
-localization
we nd K
n
: n ) T
k+1
V such that (

n )([ K
n
X [ k). For
i let A
i
= n : [K
n
[ = i (some of these sets can be nite or even
empty). For n /

i2k+2
A
i
partition K
n
into 2- and 3-element sets to have
more than k pieces:
K
n
=

K
2
n,i
: i w
2
n

K
3
n,i
: i w
3
n
, [w
2
n
[ +[w
3
n
[ > k
(everything is done in V, of course). Consider the sequences
K
2
n,i
: i w
2
n
, n /

j2k+2
A
j
),
K
3
n,i
: i w
3
n
, n /

j2k+2
A
j
),
K
n
: n A
k+2
),. . . ,K
n
: n A
2k+2
)
(note that A
i
= for i < k + 2). These are sequences of disjoint sets of the
sizes 2, 3, k + 2, . . . , 2k + 2, respectively, and all sequences are in V. Since X is
(l +k, k)-large for each l 2 and it is (2,0)- and (3,0)-large we nd N such that
(a) if n > N, n A
i
, k + 2 i 2k + 2 then [K
n
X[ > k
(b) if n > N, n /

j2k+2
A
j
, i w
x
n
, x 2, 3 then K
x
n,i
X ,= .
The condition (b) implies that if n > N, n /

j2k+2
A
j
then [ K
n
X [ > k
(recall that we have more than k sets K
x
n,i
). Consequently [ K
n
X [ > k for
all n > N which contradicts the choice of K
n
: n ).
The above proof suggests to consider (m, 0)-large sets (over V) and ask if
the existence of such sets depends on m 2. The answer is given by the next
result.
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 9
Proposition 1.7 Suppose V V

are models of ZFC, m 2. Then there

exists an (m, 0)-large set over V if and only if there exists an (m + 1, 0)-large
set over V.
Proof Clearly each (m, 0)-large set is (m+1, 0)-large. So suppose now that
X []

is (m + 1, 0)-large over V. If it is (m, 0)-large then we are done.

So assume that X is not (m, 0)-large and this is witnessed by K
n
: n ) V
(so [K
n
[ = m, K
n
s are disjoint and (

n )(K
n
X = )). Let
Y = n : K
n
X ,= .
Clearly Y is innite co-innite. We are going to show that Y is (2, 0)-large (and
hence (m, 0)-large) over V.
Suppose that L
n
: n ) V is a sequence of disjoint 2-element sets.
Let K

n
=

lLn
K
l
. Thus [K

n
[ = 2m and K

n
s are disjoint. Obviously the
sequence K

n
: n ) is in V. Since the set X is (m + 1, 0)-large (and hence,
by 1.6.2, (2m, m1)-large) we have
(

n )([ K

n
X [ > m1).
But K

n
X ,= , L
n
= l
0
, l
1
imply that either K
l
0 X ,= or K
l
1 X ,=
and hence L
n
X ,= . Consequently Y is (2,0)-large.
Corollary 1.8 Let V V

be models of ZFC, m 2, k . Then the

following conditions are equivalent:
1. there is no (m, 0)-large set in []

over V
2. there is no (2, 0)-large set in []

over V
3. V V

has the R

0
-localization property
4. V V

has the R

k
-localization property.
Remark: One can consider a modication of the notion of (m, 0)-largeness
giving (probably) more freedom. For an increasing function f

V we say
that a set X []

is f-large over V if for every sequence K

n
: n ) V of
disjoint nite subsets of we have
either (n )([K
n
[ < f(n) + 2) or (

n )([K
n
X[ > f(n)).
Proposition 1.9 Let V V

be models of ZFC.
a) If V2

is not meager iv V

then the pair (V, V

) has the R

0
-localization
property.
b) If the pair (V, V

) has the R

0
-localization property then V

is unbounded
in V

.
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 10
Proof a) If V 2

is not meager in V

then
() (f V

)(Y V

[]

)(g V

)(

nY )(f(n) = g(n))
and V

is unbounded in V

(see [Bar1]). By proposition 1.3, the

pair (V, V

) has the cR

1
-localization property. Suppose that X []

.
By the cR

1
-localization we nd a partition K
n
: n ) V T
1
such that
(

n )([ K
n
X [ 2). In V

we dene
f(n) = K
n
X []
<
(for n ),
Y = n : [f(n)[ 2.
By () we nd g V, g : []
<
such that g(n) [K
n
]
2
and
(

n Y )(f(n) = g(n)).
Since f(n) = g(n) implies g(n) = K
n
X we get (

n )(g(n) X = ).
Hence we easily get that X can be R

0
-localized by a partition from V.
b) Since [K[ > 1 & K ( X) = implies [K X[ 2 we get that
R

0
-localization implies the cR

1
-localization. Now proposition 1.3 works.
The next result gives some bounds on possible improvements of the previous
one.
Proposition 1.10 1. The Cohen forcing notion has the R

0
-localization prop-
erty. Consequently, the R

0
-localization does not imply that the old reals
are a dominating family.
2. The Random real forcing does not have the R

0
-localization property. Con-
sequently, the localization is not implied by the fact that there is no un-
bounded real in the extension.
Proof 1. As in the extensions via the Cohen forcing the ground model
reals are not meager, we may apply 1.9.
2. The Random algebra B is the quotient algebra of Borel subsets of 2

modulo the ideal of Lebesgue null sets. We dene a B-name for an element of
[]

that cannot be localized:

Let l
0
= 0, l
k+1
= l
k
+ 2
k
2
(for k ).
For each k x disjoint Borel sets A
m
2

for l
k
m < l
k+1
such that (A
m
) = 2
k
2
, where is the Lebesgue measure on 2

X is a B-name for a subset of such that if m [l

k
, l
k+1
), k
then [[m /

X]]
B
= [A
m
]

.
It should be clear that
B
(k )([ [l
k
, l
k+1
)

X [ = 1).
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 11
Suppose now that K
n
: n ) T
1
V. Let k
0
n
= min K
n
and k
1
n
= max K
n
.
Suppose that l
m
k
0
n
< l
m+1
. Note that
k
1
n
< l
m+1
[[k
0
n
/

X & k
1
n
/

X]]
B
= 0
and
k
1
n
l
m+1
([[k
1
n
/

X]]
B
) 2
(m+1)
2
.
Hence
([[

X K
n
= ]]
B
) ([[k
0
n
/

X & k
1
n
/

X]]
B
) 2
(m+1)
2
.
Consequently, for each m 0:
([[(n )(l
m
k
0
n
& K
n

X = )]]
B
)

n
([[l
m
k
0
n
& K
n

X = ]]
B
) =
=

rm

k
0
n
[lr,lr+1)
([[

X K
n
= ]]
B
)

rm
2
r
2
2
(r+1)
2
=
1
3
2
12m
.
Hence we can conclude that ([[(

n )(K
n


X = )]]
B
) = 0 which means

B
K
n
: n ) does not R

0
-localize

X.
Though the random real forcing is an example of a forcing notion adding a
(2,0)-large set over V (without adding an unbounded real!) it does not seem
to be the minimal one. A canonical example of a forcing notion without the
R

0
-localization property is given below. (Recall that a forcing notion Q is -
centered if it can be presented as a countable union of sets which all nite
subsets have upper bounds in Q.)
Example 1.11 There is a -centered (Borel) forcing notion Q adding no dom-
inating real and without the R

0
-localization property.
Proof The forcing notion Q consists of pairs (u, /) such that u []
<
and / is a nite set of families of disjoint 2-element subsets of (so F /
F []
2
). The order of Q is given by
(u
0
, /
0
) (u
1
, /
1
) if and only if
u
1
(1 + max u
0
) = u
0
, /
0
/
1
and if K F /
0
and K u
1
then K u
0
.
For u []
<
, m let
Q
u
= (u, /) : (u, /) Q and Q
m
u
= (u, /) Q
u
: [/[ = m.
Since each Q
u
is obviously centered we get that Q is -centered. Let w be a
Q-name such that w
G
=

u : (/)((u, /) G) for each generic lter G Q

5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 12
over V. Note that
Q
w []

. If K F /, (u, /) Q and max u < min K

then (u, /)
Q
K w ,= . Hence we conclude that

Q
w is (2, 0)-large over V.
So Q does not have the R

0
-localization property. Suppose now that is a
Q-name for a member of

.
Claim 1.11.1 Let u []
<
, m, n . The there is f(u, m, n) < such that:
for every p Q
m
u
there is q = (u
q
, /
q
) p such that q decides the
value of (n) and max u
q
< f(u, m, n).
Proof of Claim: The space A of all families F []
2
of two-element disjoint
subsets of can be equipped with a natural topology. For F A, N the
N-th basic open neighbourhood of F is
F

A : K N : K F = K N : K F

.
This topology is compact. It introduces a (product) topology on Q
m
u
such that if
q Q
u
, q p, p Q
m
u
then for some open neighbourhood V of p (in Q
m
u
) each
member of V has an extension in Q
u
. Applying this fact and the compactness
of Q
m
u
we get the claim.
Now we dene a function g

putting
g(k) = 1 + maxl : (u k)(m, n k)(v)
(u v f(u, m, n) and (q Q
v
)(q (n) = l)).
Given p Q
m
u
, l . Take k > maxl, m, maxu. By the denition of
f(u, m, k) we nd v such that u v f(u, m, k) and some q Q
v
, q p
decides the value of (k). By the denition of the function g, the condition q
forces (k) < g(k).
Remark: An example of a forcing notion P with the R

0
-localization property
and such that

P
V 2

is meager
is an application of a general framework of [RoSh:470] and will be presented
there.
2 Between dominating and unbounded reals
In this section we are interested in some localizations which are between the cR

k
-
localization and cR

k
-localization (so between not adding a dominating real and
not adding an unbounded real). The localizations are similar to that considered
in the previous section. The dierence is that we will consider partitions of
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 13
into intervals and we will introduce quantiers stronger than

n but weaker
than

n.
For an innite subset X of let
X
: X be the increasing enumeration
of X. A set X []

can be identied with the partition

[
X
(n),
X
(n + 1)) : n )
of
X
(0), so essentially []

T
0
. Now, for k > 0 and an increasing function

, we dene relations S
k
, S
+
, S
+
, S

+
[]

[]

:
(X, Y ) S
k
(

n)(i < k)([ [

Y
(n +i),
Y
(n +i + 1)) X [ 2)
(X, Y ) S
+
(m)(n)(i < m)([ [
Y
(n+i),
Y
(n+i +1)) X [ 2)
(X, Y ) S
+
(

n)(i < 2
n
)([ [
Y
(2
n
+i),
Y
(2
n
+i + 1)) X [ 2)
(X, Y ) S

+
(

n)(i < (n))([ [

Y
(n +i),
Y
(n +i + 1)) X [ 2).
[The relation S
+
appears here for historical reasons only: it determines a car-
dinal invariant which was of serious use in [RoSh:475].]
Note that S
1
S
2
S
3
. . . S
+
S
+
S

+
(remember that is increasing).
If the function is increasing fast enough (e.g. (n) > 2
2n
) then S
+
S

+
.
It should be clear that if we consider S

-localizations we could put any integer

greater than 2 in place of 2 in the denitions above. However we do not know
if replacing 2 by 1 provides the same notions of localizations.
Proposition 2.1 Let V V

be models of ZFC.
(a) The pair (V, V

) has the S
1
-localization property if and only if V

is
unbounded in V

.
(b) If the pair (V, V

) has the S
k+1
-localization property then it has the S
k
-
localization. The S
+
-localization property implies the S
k
-localization for
each k > 0 and is implied by both the S

+
and the S
+
-localization properties
( - an increasing function).
(c) If V

is dominating in V

then the extension V V

has the
S

+
-localization property for every increasing function V

.
Proposition 2.2 If V V

are models of ZFC such that V2

is not meager
in V

, V

is increasing then the pair (V, V

) has the S

+
-localization
property.
Proof The proof is almost the same as that of 1.9(a). Suppose that X
V

[]

. Let X
0
[X]

be such that for each n

[ [
X0
(n),
X0
(n + 1)) X [ > 3(
X0
(n) + 4) + 6.
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 14
As

V is unbounded in

we nd a set Y
0
V

such that the

set
X
1
=
Y0
(n) : n & 2 [ [
Y0
(n),
Y0
(n + 1)) X
0
[
is innite. Let f : []
<
be such that for every
Y0
(n) X
1
we have
f(
Y0
(n)) = [
Y0
(n),
Y0
(n + 1)) X
(so, for
Y0
(n) X
1
, [ f(
Y0
(n)) [ > 3(
X0
(k) + 4) + 6 where k is the rst
such that
Y0
(n)
X0
(k) <
Y0
(n + 1)). By () from the proof of 1.9 we nd
a function g V, g : []
<
such that
(

n )(
Y0
(n) X
1
& g(
Y0
(n)) = f(
Y0
(n))).
Next, using g and Y
0
(both are in V) we dene sets Y
1
, Y V []

:
Y
1
=

n
g(
Y0
(n)) [
Y0
(n),
Y0
(n + 1))
and
Y
(n) =
Y1
(3n) for each n. Note that if n is such that
Y0
(n) X
1
and
g(
Y0
(n)) = f(
Y0
(n)) then
Y
1
[
Y0
(n),
Y0
(n + 1)) = X [
Y0
(n),
Y0
(n + 1))
is of the size > 3(
Y0
(n) + 4) + 6 and consequently Y S

+
-localizes X.
Proposition 2.3 1. The Cohen forcing notion has the S

+
-localization prop-
erty for each increasing function

.
2. If V V

are models of ZFC, the pair (V, V

) has the S

+
-
localization property and V

is dominating in V

then the
extension V V

has the S

+
-localization property.
3. The iteration of the Cohen forcing notion and the random real forcing has
the S

+
-localization.
Consequently, the S

+
-localization property implies neither that the ground model
reals are dominating in the extension nor that the old reals are not meager.
Theorem 2.4 For each k > 0, the S
k
-localization property does not imply the
S
k+1
-localization property.
Proof To prove the theorem we will dene a forcing notion Q
k
which will
possess the S
k
-localization property but not S
k+1
. The forcing notion is similar
to that used in [Sh:207], [BsSh:242] and is a special case of the forcing notions
of [RoSh:470].
We start with a series of denitions.
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 15
1. A function n is a nice norm on A if:
n : T(A) and it is monotonic
(i.e. B C A n(B) n(C)),
if B C A, n(C) > 0
then either n(B) n(C) 1 or n(C B) n(C) 1,
n(A) > 0, if a A then n(a) 1.
2. A creature is a tuple T = T, nor, L, R) such that
) T
<
is a nite nonempty tree,
) for each t T, either succ
T
(t) = or [succ
T
(t)[ = k or [succ
T
(t)[ > k
[so we have three kinds of nodes in the tree T]
) nor is a function with the domain
dom(nor) = t T : [succ
T
(t)[ > k
and such that for t dom(nor), nor(t) is a nice norm on succ
T
(t)
[nor stands for norm],
) if s succ
T
(t) then either [succ
T
(t)[ ,= k or [succ
T
(s)[ ,= k
[i.e. we do not have two successive k-ramications in T],
) L, R : T are functions such that for each t T:
L(t) R(t)
if s succ
T
(t) then [L(s), R(s)] [L(t), R(t)]
if s
1
, s
2
succ
T
(t) are distinct
then [L(s
1
), R(s
1
)] [L(s
2
), R(s
2
)] =
if succ
T
(t) = then L(t) = R(t)
[L stands for left and R is for right].
3. We will use the convention that if a, b are indexes and T
a
b
is a creature
then its components are denoted by T
a
b
, nor
a
b
, L
a
b
and R
a
b
, respectively.
4. Let T = T, nor, L, R) be a creature. We dene its weight |T| and its
contribution cont(T):
|T| = minnor(t)(succ
T
(t)) : t dom(nor) & [succ
T
(t)[ > k
[if dom(nor) = then we put |T| = 0]
cont(T) = L(t) : t T & succ
T
(t) =
[recall that if t is a leaf in T then L(t) = R(t)].
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 16
5. Let T
i
(i = 0, 1) be creatures. We say that the creature T
1
renes T
0
(we
write: T
0
T
1
) if:
a) T
1
T
0
, L
1
= L
0
T
1
, R
1
= R
0
T
1
,
b) if t T
1
then
[succ
T1
(t)[ > k i [succ
T0
(t)[ > k,
[succ
T1
(t)[ = k i [succ
T0
(t)[ = k, and
[succ
T1
(t)[ = i [succ
T0
(t)[ =
(in other words we keep the kind of nodes),
c) nor
1
(t) = nor
0
(t)T(succ
T1
(t)) for all t dom(nor
1
).
6. Let T
0
, T
1
, . . . , T
n
be creatures. We say that the creature T is built of the
creatures T
0
, . . . , T
n
(we will write it as T (T
0
, . . . , T
n
)) if there is a
maximal antichain F of T such that for each t F, for some i n:
s T : t s = tr : r T
i
,
if s = tr T, r T
i
then L(s) = L
i
(r), R(s) = R
i
(r) and nor(s) =
nor
i
(r) (if dened).
7. If n k, H : T(n + 1) is a nice norm on n + 1 (i.e. it satises the
conditions listed in (1)) then S
H
(T
0
, . . . , T
n
) is the creature T such that
T = i)t : i n, t T
i
, nor()) = H, nor(i)t) = nor
i
(t)
and similarly for L and R.
Clearly S
H
(T
0
, . . . , T
n
) (T
0
, . . . , T
n
).
8. We dene gluing k creatures similarly to the operation S
H
above. Thus
S(T
0
, . . . , T
k1
) = T is a creature such that
T = i)t : i < k, t T
i

and nor, L, R are dened naturally. Once again, S(T

0
, . . . , T
k1
)
(T
0
, . . . , T
k1
).
9. For a creature T we dene its upper half T
uh
= T
uh
, nor
uh
, L
uh
, R
uh
) by:
T
uh
= T, L
uh
= L, R
uh
= R but
nor
uh
(t)(A) = max0, nor(t)(A) [
|T|
2
]
whenever t T, [succ
T
(t)[ > k and A succ
T
(t). Above, [x] stands for
the integer part of x.
[It is routine to check that T
uh
is really a creature and that |T
uh
| =
|T| [
1
2
|T|], cont(T
uh
) = cont(T).]
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 17
10. For creatures T
0
, . . . , T
n
, the closure of T
0
, . . . , T
n
under the operations
of shrinking (rening), taking the upper half and building creatures is
denoted by

(T
0
, . . . , T
n
). Thus T
0
, . . . , T
n

(T
0
, . . . , T
n
), if T

(T
0
, . . . , T
n
) then
T
uh

(T
0
, . . . , T
n
), T T

(T
0
, . . . , T
n
),
and if T

0
, . . . , T

(T
0
, . . . , T
n
) then
(T

0
, . . . , T

m
)

(T
0
, . . . , T
n
).
[Note that

(T
0
, . . . , T
n
) is nite (up to isomorphism).]
Now we may dene our forcing notion Q
k
:
Conditions are sequences w, T
0
, T
1
, T
2
, . . .) such that w []
<
, T
i
are
creatures, |T
i
| and
max(w) < L
0
()) R
0
()) < L
1
()) R
1
()) < . . .
(recall that T
i
= T
i
, nor
i
, L
i
, R
i
)).
The order is given by
w, T
0
, T
1
. . .) w

, T

0
, T

1
. . .) if and only if
for some increasing sequence n
0
< n
1
< n
2
< . . . <
w w

i<n0
cont(T
i
) and (i )(T

(T
ni
, . . . , T
ni+11
)).
We say that a condition w

, T

0
, T

1
. . .) Q
k
is a pure extension of a condition
w, T
0
, T
1
. . .) Q
k
if
w, T
0
, T
1
. . .) w

, T

0
, T

1
. . .) and w = w

.
One can easily check that (Q
k
, ) is a partial order and that the relation
of pure extension is transitive. The proof that the forcing notion Q
k
has the
required properties is broken into several claims. The rst claim is of a technical
character, but it implies in particular that Q
k
is proper. (The proof of this claim
is straightforward and we will omit it.)
Claim 2.4.1 If p Q
k
,
n
(for n < ) are Q
k
-names for ordinals then there
is a pure extension q = w, T
0
, T
1
, . . .) of p such that
1. |T
i
| : i ) is increasing,
2. for each n, v R
n
()) and i n,
if some pure extension of w v, T
n+1
, T
n+2
. . .) decides the value of
i
then w v, T
n+1
, T
n+2
. . .) decides it.
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 18
Claim 2.4.2 Suppose that T = T, nor, L, R) is a creature, |T| 15 and
B []

. Then there is a creature T

T such that |T

| |T| 14 and
(n )(i k)([ cont(T

) [
B
(n +i),
B
(n +i + 1)) [ < 2).
Proof of Claim: We prove this essentially by the induction on [T[ (or the height
of T). We show how to eliminate ) and apply the inductive hypothesis to T
above t (for t succ
T
())).
Case 1: k = [succ
T
())[
If, for each t succ
T
()), succ
T
(t) = then [ cont(T) [ = k and there is no
problem. So there are t succ
T
()) such that [succ
T
(t)[ > k (here we use the
requirement () of the denition of creatures); above each such t we can apply
the inductive hypothesis and shrink suitably the tree T. However the problems
coming from distinct t could accumulate. Therefore for each such t we rst
choose a set A = A
t
succ
T
(t) such that nor(t)(A) |T| 7 and one of the
following occurs:
for some m, (s A)(
B
(m) L(s) R(s)
B
(m + 1))
there are 1 m
0
< m
1
such that
(s A)(
B
(m
0
) L(s) R(s)
B
(m
1
)),
and if t
0
succ
T
()), t
0
,= t then
either R(t
0
) <
B
(m
0
1) or
B
(m
1
+ 1) < L(t
0
).
To nd such a set A we use the second property of nice norms. First we look
at the set
A
0
= s succ
T
(t) : (m )(
B
(m) L(s) R(s)
B
(m + 1)).
If nor(t)(A
0
) nor(t)(succ
T
(t)) 1 |T| 1 then we easily nish: either for
some m
nor(t)(s succ
T
(t) :
B
(m) L(s) R(s)
B
(m + 1)) |T| 7
or we have m

0
< m

0
< m
0
< m
1
< m

1
< m

1
such that
nor(t)(s succ
T
(t) :
B
(m
0
) L(s) R(s)
B
(m
1
)) |T| 7 and
(s
0
, s
1
succ
T
(t))(
B
(m

0
) L(s
0
)
B
(m

0
) &
B
(m

1
) R(s
1
)
B
(m

1
)).
So suppose that nor(t)(A
0
) < |T| 1. In this case nor(t)(succ
T
(t) A
0
)
|T| 1. For each s succ
T
(t) A
0
there is m such that
L(s) <
B
(m) < R(s).
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 19
Removing 4 extreme points from succ
T
(t) A
0
(the rst two and the last
two, counting according to the values of L) we get the required set A (with
nor(t)(A) |T| 5).
Now we would like to apply the induction hypothesis above each t succ
T
())
restricting ourselves to successors of t from the suitable set A
t
(if applicable).
A small diculty is that we have decreased the norm of the creature above
those t (possibly by 7, as the result of restricting to A
t
). But now we apply the
procedure described in the case 2 below and we pass to B
t
A
t
such that
nor(t)(b
t
) nor(t)(A
t
) 7 |T| 14.
Next we apply the inductive hypothesis above each s B
t
, t succ
T
())
dom(nor). In this way we construct T

as required (the point is that restricting

to the sets B
t
A
t
causes that what happens above distinct t succ
T
()) is
isolated in a sense).
Case 2: k < [succ
T
())[
Then one of the following possibilities occurs (i = 0, 1):
()
i
nor())(t succ
T
()) : (m)(
B
(2m+i) L(t) R(t) <
<
B
(2m+i + 1))) |T| 3
() nor())(t succ
T
()) : (L(t), R(t)) B ,= ) |T| 3
If one of cases ()
0
, ()
1
holds then the creature rening T and determined by
the respective set of successors of ) can serve as T

. In case () divide the set

A = t succ
T
()) : (L(t), R(t)) B ,=
into four disjoint subsets, each containing every fourth member of A. One of
these subsets (call it A

) has the norm |T| 7. For each t A

apply
the inductive hypothesis to the creature given by s T : t s. Note that
either it is of the weight |T| or succ
T
(t) = or [succ
T
(t)[ = k and for
s succ
T
(t) we have succ
T
(s) = . The last two cases are trivial and actually
should be considered separately (compare Case 1). In this way we get the
required creature T

.
Claim 2.4.3 Q
k
does not have the S
k+1
-localization property.
Proof of Claim: Let w be a Q
k
name such that if G Q
k
is a generic then
w
G
=

w : (T
0
, T
1
, . . .))(w, T
0
, T
1
, . . .) G). We claim that the S
k+1
-
localization always fails for w
G
, i.e. that if B V []

then

Q
k
(

n )(i k)([ w [
B
(n +i),
B
(n +i + 1)) [ < 2).
Let p = w, T
0
, T
1
, . . .) Q
k
be given. We may assume that
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 20
1. (l )([ [R
l
()), L
l+1
())) B [ > 2)
2. (l )(|T
l
| > 15).
For each T
i
take the creature T

i
T
i
given by claim 2.4.2 for T
i
and B. Look
at the condition q = w, T

0
, T

1
, . . .) Q
k
. Clearly q p and if
B
(n) > max(w)
then
q
Q
k
(i k)([ w [
B
(n +i),
B
(n +i + 1))[ < 2).
This proves the claim.
The next claim explains why we introduced the operation of taking the upper
half of a creature as a part of the denition of (the order of) Q
k
Claim 2.4.4 Let p = w, T
0
, T
1
. . .) Q
k
, m . Suppose that is a Q
k
-
name for an ordinal. Then there are n
0
and a nice norm H on n
0
such that
H(n
0
) m and
if T

S
H
(T
uh
0
, . . . , T
uh
n01
), |T

| > 0, w w

L
0
())
then there exist v cont(T

) and T

0
, T

1
, . . .) such that
, T

0
, T

1
, . . .) , T
n0
, T
n0+1
, . . .)
and w

v, T

0
, T

1
, . . .) decides the value of .
Proof of Claim: This is essentially 2.14 of [Sh:207].
Dene the function H : []
<
by:
H(u) 0 always
H(u) 1 if [u[ > 1 and for each T

i
T
uh
i
, |T

i
| > 0 (for i u), for every w

,
w w

L
0
()) there is v

iu
cont(T

i
) such that some pure extension
of w

v, T
l
, T
l+1
, . . .) decides the value of (where l = max u + 1).
H(u) n + 1 if for every u

u either H(u

) n or H(u u

) n (for n > 0).

As H is monotonic it is enough to nd u such that H(u) m. The existence
of the u can be proved by induction on m, for all sequences w, T
0
, T
1
, . . .).
Let us start with the case m = 1. Suppose that T

i
T
uh
i
, |T

i
| > 0 (for i ).
For each i choose a creature T

i
T
i
such that
cont(T

i
) = cont(T

i
) and |T

i
|
1
2
|T
i
|
(possible by the denition of the upper half of a creature). Then for each
w

L
0
()):
w

, T

0
, T

1
, T

2
, . . .) Q
k
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 21
and thus we nd n(w

) , v(w

) []
<
such that
v(w

n<n(w

)
cont(T

n
) =

n<n(w

)
cont(T

n
) and
some pure extension of
w

v(w

), T

n(w

)
, T

n(w

)+1
, . . .)
(and so of w

v(w

), T
n(w

)
, T
n(w

)+1
, . . .)) decides the value of .
Let M(T

0
, T

1
, T

2
, . . .) be the rst M such that for every w

L
0
()) there is
v

n<M
cont(T

n
) such that
some pure extension of w

v, T
M
, T
M+1
, . . .) decides the value of .
The space
T

0
, T

1
, T

2
, . . .) : (i )(T

i
T
uh
i
)
equipped with the natural (product) topology is compact and the function
M : T

0
, T

1
, T

2
, . . .) M(T

0
, T

1
, T

2
, . . .)
is continuous. Hence the function M is bounded, say by n
0
. Clearly H(n
0
) 1.
Now suppose that we always can nd a set of the norm m 1. Thus we nd an
increasing sequence l
i
: i ) such that H([l
i
, l
i+1
)) m for each i. Consider
the space of all increasing

such that (i )((i) [l

i
, l
i+1
)) - it is a
compact space. For each from the space we may consider w, T
(0)
, T
(1)
, . . .)
and the respective function H

. By the induction hypothesis we nd n = n

such that H

(n) m. But H

(n) H((i) : i < n). By the compactness

we nd one n such that m H((i) : i < n) for each . Hence we conclude
that H(l
n+1
) m+ 1.
Main Claim 2.4.1 1. The forcing notion Q
k
has the S
k
-localization prop-
erty.
2. Moreover the following stronger condition is satised by Q
k
:
(S

k
)
Q
k
Suppose that N is a countable elementary submodel of H(
+
7
),
, <

), p = w, T
0
, T
1
, . . .) Q
k
N. Assume that Y []

is such
that
(X []

N)((X, Y ) S
k
).
Then there is a condition q p which is (N, Q
k
)-generic and such
that
q
Q
k
(X N[

G
Q
k
] []

)((X, Y ) S
k
).
Proof of Main Claim: 1) Since in the next section we will need the property
(S

k
)
Q
k
, we will present the proof of it fully below. Here we sketch the proof of
the S
k
-localization property for readers not interested in the stronger property
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 22
(needed for iterations). So suppose that

X is a Q
k
-name for an element of []

and p = w, T
0
, T
1
, . . .) Q
k
. We may assume that |T
n
| n. We inductively
dene integers
0 = b
0
< b
1
< b
2
< . . . and 0 = n
0
< n
1
< n
2
< . . .
and nice norms H
m
on [n
m
, n
m+1
). Suppose we have dened b
m
, n
m
.
Let
m
be a Q
k
-name for an integer such that

Q
k
[[b
m
,
m
)

X[ > 2.
Modifying the tail (above n
m
) of p we may assume that if n
m
n, w w

R
n
()) and some pure extension of w

, T
n+1
, T
n+2
, . . .) decides the value of
m
then w

, T
n+1
, T
n+2
, . . .) does it already (see 2.4.1). Applying 2.4.4 we nd
n
m+1
> n
m
and a nice norm H
m
on [n
m
, n
m+1
) such that
H
m
([n
m
, n
m+1
)) m + 1 and if T

S
Hm
(T
uh
nm
, . . . , T
uh
nm+11
),
w w

L
nm
()) then there is v cont(T

) such that some pure

extension of
w

v, T
nm+1
, T
nm+1+1
, . . .)
decides the value of
m
and thus w

v, T
nm+1
, T
nm+1+1
, . . .) does
it.
Let b
m+1
be an integer larger than all possible values forced to
m
in the con-
dition above.
Now for each l we put
T

l
= S

S
H
lk
(T
uh
n
lk
, . . . , T
uh
n
lk+1
1
), . . . , S
H
(l+1)k1
(T
uh
n
(l+1)k1
, . . . , T
uh
n
(l+1)k
1
)

,
and then
q = w, T

0
, T

1
, . . .), B = b
0
, b
1
, b
2
, . . ..
Check that q p and
q
Q
k
the set B S
k
localizes

X
(or see the end of the proof of 2) below).
2) The construction of the condition q required in (S

k
)
Q
k
is similar to that in
1). Here, however, we have to take care of all names for elements of []

from
the model N (as well as names for ordinals to ensure the genericity).
Let
n
: n ) enumerate all Q
k
-names from N for ordinals and let

A
n
: n
) list all names (from N) for innite subsets of . Of course, both sequences
are not in N but all their initial (nite) segments are there.
Now we inductively dene sets B
n
= b
n
0
, b
n
1
, b
n
2
, . . . []

and conditions
q
n
= w, T
n
0
, T
n
1
, T
n
0
, . . .) Q
k
such that B
n
, q
n
N:
To start with we put B
0
= , q
0
= p = w, T
0
, T
1
, . . .).
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 23
Arriving at stage n > 0 we have dened B
n1
, q
n1
N. We dene B
n
, q
n
applying the following procedure inside the model N (so the result will be there;
compare this procedure with that in part (1)):
Let
0
be a Q
k
-name for an integer such that
Q
k
(i < n)([
0


A
i
[ > 2). We
modify q
n1
(passing to a pure extension of it) and we assume that
()
0,0,...,n1
if v R
n1
i
()), i and there exists a pure extension of
w v, T
n1
i+1
, T
n1
i+2
, . . .) deciding the value of one of
0
,
0
, . . . ,
n1
then
w v, T
n1
i+1
, T
n1
i+2
, . . .) decides it already
(see 2.4.1). Next, by Claim 2.4.4, we nd n
0
and a nice norm H
0
on n
0
such
that
H
0
(n
0
) 1 and if T

S
H0
((T
n1
0
)
uh
, . . . , (T
n1
n01
)
uh
), |T

| > 0,
w w

L
n1
0
()) then there exists v cont(T

) such that
w

v, T
n1
n0
, T
n1
n0+1
, T
n1
n0+2
e . . .) decides the value of
0
.
Let T
n
0
= S
H0
((T
n1
0
)
uh
, . . . , (T
n1
n01
)
uh
) and let b
n
0
be greater than all possible
values of
0
(i.e. the values forced in the condition on H
0
above). Let
1
be a
Q
k
-name for an integer such that
Q
k
(i < n)([ [b
n
0
,
1
)

A
i
[ > 2). We modify
the tail of q
n1
and we assume ()
1,0,...,n1
(for i n
0
, q
n1
). Next we
choose n
1
> n
0
and a nice norm H
1
on [n
0
, n
1
) such that
H
1
([n
0
, n
1
)) 2 and if T

S
H1
((T
n1
n0
)
uh
, . . . , (T
n1
n11
)
uh
), |T

| >
0, w w

L
n1
n0
()) then there exists v cont(T

) such that
w

v, T
n1
n1
, T
n1
n1+1
, . . .) decides the value of
1
.
Let T
n
1
= S
H1
((T
n1
n0
)
uh
, (. . . , T
n1
n11
)
uh
) and let b
n
1
be greater than all possibil-
ities for
1
in the above property.
We continue in this fashion and we determine integers n
0
< n
1
< n
2
< . . .,
b
n
0
< b
n
1
< b
n
2
< . . . and nice norms H
0
, H
1
, H
2
. . . and we dene creatures
T
n
i
= S
Hi
((T
n1
ni1
)
uh
, . . . , ((T
n1
ni1
)
uh
). Finally we let B
n
= b
n
0
, b
n
1
, b
n
2
, . . . and
q
n
= w, T
n
0
, T
n
1
, . . .) (actually we should have taken more care while getting
()
i,0,...,n1
in constructing T
n
i
: this is necessary for lim
i
|T
n
i
| = ).
Suppose now that Y []

is a set S
k
-localizing []

N. Then for each

n we have
(

i)(m < k)([ [

Y
(i +m),
Y
(i +m+ 1)) B
n
[ 2).
We inductively dene increasing sequences i
n
: n ), l
n
: n ) of integers
and a sequence T

n
: n ) of creatures:
Let i
0
be such that
(m < k)([ [
Y
(i
0
+m),
Y
(i
0
+m+ 1)) B
1
[ 2).
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 24
Let j
0
m
(for m < k) be such that
B1
(j
0
m
) is the rst element of
[
Y
(i
0
+m),
Y
(i
0
+m+ 1)) B
1
.
Let T

0
= S(T
1
j
0
0
+1
, T
1
j
0
1
+1
, . . . , T
1
j
0
k1
+1
). Note that the creatures T
1
j
were ob-
tained as results of the operation S
Hj
(for some norms H
j
) and hence their
roots (i.e. )) are > k-splitting points, thus no danger can appear in this proce-
dure. Finally we choose l
0
such that T

(T
0
, . . . , T
l01
).
Assuming that we have dened i
n
, l
n
, T

n
, take i
n+1
> i
n
+k such that
(m < k)([ [
Y
(i
n+1
+m),
Y
(i
n+1
+m+ 1)) B
n+2
[ 2)
and if
Bn+2
(j) [
Y
(i
n+1
),
Y
(i
n+1
+ 1)) then T
n+2
j

(T
ln
, . . . , T
m
) (for
some m > l
n
) and |T
n+2
j
| n + 1. Now, let j
n+1
m
(for m < k) be such that

Bn+2
(j
n+1
m
) is the rst element of [
Y
(i
n+1
+ m),
Y
(i
n+1
+ m + 1)) B
n+2
.
Finally, T

n+1
= S(T
n+2
j
n+1
0
+1
, . . . , T
n+2
j
n+1
k1
+1
) and l
n+1
is such that
T

n+1

(T
ln
, . . . , T
ln+11
).
The condition q is w, T

0
, T

1
, . . .). Clearly q p. To show that it is (N, Q
k
)-
generic suppose that N is a Q
k
-name for an ordinal, say =
n
, and let
q

= w

, T

0
, T

1
, . . .) q be a condition deciding . Look at the construction
of q
n+1
= w, T
n+1
0
, T
n+1
1
, . . .). Because of ()
0,0,...,n
, if v M
n+1
i
()) and
there is (in N) a pure extension of w v, T
n+1
i+1
, T
n+1
i+2
, . . .) deciding
n
then
w v, T
n+1
i+1
, T
n+1
i+2
, . . .) decides it already.
Let i be such that w

M
n+1
i
()). Then there exists a pure extension of
the condition w

, T
n+1
i+1
, T
n+1
i+2
, . . .) deciding
n
, e.g. that one which can be
obtained from q

(note that q

and w

, T
n+1
i+1
, T
n+1
i+2
, . . .) are compatible). By
the elementarity of N there is such an extension in N. This implies that
w

, T
n+1
i+1
, T
n+1
i+2
, . . .) decides
n
(and the decision belongs to N). The condi-
tions q

and w

, T
n+1
i+1
, T
n+1
i+2
, . . .) are compatible, so the values given by them to

n
are the same and we are done (with the genericity).
Now we want to show that
q
Q
k
(X N[

G
Q
k
] []

)((X, Y ) S
k
).
Let

A N be a Q
k
-name for an innite subset of , say

A =

A
m
. We are going
to prove that
q
Q
k
(

n )(i < k)([ [

Y
(n +i),
Y
(n +i + 1))

A
m
[ 2).
Let q

= w

, T

0
, T

1
, . . .) q. Fix l > m. Look at T

l
for some m

, m

,
l < m

< m

we have T

(T

m
, . . . , T

m
) and w

i<m
cont(T

i
). So we
nd n > l and t T

l
such that T

l
above t comes from T

n
by decreasing norms
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 25
(like in the operation of taking the upper half but possibly with other values
subtracted) and rening. Note that neccessarily [succ
T

l
(t)[ = k and T

l
above
each successor of t renes some T
n+1
j
n
i
+1
(for i < k) modulo decreasing norms by
some values, which does not inuence contributions. Next we nd v cont(T

l
)
(actually v = v
0
. . . v
k1
, v
i
is included in the contributions of the part
above that successor of t which renes T
n
j
n
i
+1
) such that w

v, T

l+1
, T

l+2
, . . .)
decides the values of the names
j
n
0
+1
, . . . ,
j
n
k1
+1
(dened in the construction
of T
n+1
j
s) and forces them to be bounded by b
j
n+1
0
+1
, . . . , b
j
n+1
k1
+1
, respectively.
By the choice of
j
s this means that
w

v, T

l+1
, T

l+2
, . . .)
Q
k
(i < k)(j < n)([ [b
n
j
n+1
i
, b
j
n+1
i
+1
)

A
j
[ 2).
Since m < l n and
Y
(i
n
+i)
Bn+1
(j
n
i
) <
Bn+1
(j
n
i
+ 1) <
Y
(i
n
+i +1)
(for all i < k) we get
w

v, T

l+1
, T

l+2
, . . .)
Q
k
(i < k)([ [
Y
(i
n
+i),
Y
(i
n
+i + 1))

A
m
[ 2).
As l > m was arbitrary we are done. The Main Claim and the theorem are
proved.
Remark: More examples of forcing notions distinguishing the localization prop-
erties we have introduced in this section will be presented in [RoSh:470]. They
are (like the forcing notion Q
k
) applications of the general schemata of that
paper.
3 Cardinal coecients related to the localiza-
tions
In this section we discuss cardinal coecients related to the localization prop-
erties introduced earlier.
Following Vojt as (cf [Voj]) with any relation R X Y we may associate
two cardinal numbers (the unbounded and the dominating number for R):
b(R) = min[B[ : (y Y )(x B)((x, y) / R)
d(R) = min[D[ : (x X)(y D)((x, y) R).
For purposes of applications these cardinals are introduced for relations R
X Y such that dom(R) = dom(cR) = X and rng(R) = rng(cR) = Y . Note
that for each such relation we have
b(R) = d(cR
1
), d(R) = b(cR
1
), b(R
1
) = d(cR), d(R
1
) = b(cR).
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 26
All results of the previous sections provide information on dominating numbers
d(R) for the considered relations. Let
b = b(

) = min[F[ : F

& (x

)(y F)(

n )(x(n) < y(n))

d = d(

) = min[F[ : F

& (x

)(y F)(

n )(x(n) y(n))
non(K) = min[X[ : X 2

& X is not meager .

These are three of ten cardinal invariants forming the Cicho n Diagram. For more
information on the cardinals related to measure and category see [BJSh:368] or
[CiPa].
Corollary 3.1
1. (see 1.2) (k )(d(R

k
) = c)
2. (see 1.3) (k > 0)(d(cR

k
) = b)
3. (see 1.4) (k )(d(cR

k
) = d)
4. (see 1.6) (k )(d(R

k
) = d(R

0
))
5. (see 1.9) b d(R

0
) non(K)
Corollary 3.2
1. Con(d(R

0
) < d)
2. Con(d < d(R

0
))
Proof For the rst model add
2
Cohen reals to a model of CH. Then in
the extension we will have d =
2
and d(R

0
) =
1
(see 1.10). The second model
can be obtained by adding
2
random reals to a model of CH, which results in
a model for d =
1
and d(R

0
) =
2
(see 1.10).
Corollary 3.3 Let

be an increasing function. Then:

1. (see 2.1) b = d(S
1
) d(S
2
) . . . d(S
k
) . . . d(S
+
),
d(S
+
) d(S
+
), d(S

+
) d, and
if is increasing fast enough (e.g. (n) > 2
2n
) then d(S
+
) d(S

+
),
2. (see 2.2) d(S

+
) non(K). .
Corollary 3.4 Let

be an increasing function. Then

Con(d(S

+
) =
1
+non(K) = d =
2
).
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 27
Proof Start with a model for CH and add to it rst
2
Cohen reals (what
causes d =
2
but keeps d(S

+
) =
1
) and next add
2
random reals (what
preserves d =
2
, d(S

+
) =
1
but causes non(K) =
2
), see 2.3.
Corollary 3.5 Let k > 0. Then
Con(d(S
k
) < d(S
k+1
)).
Proof Let Q
k
be the forcing notion from the proof of 2.4. It is proper (see
2.4.1). To get the respective model it is enough to take the countable support
iteration of the length
2
of the forcing notions Q
k
over a model of CH. As Q
k
does not have the S
k+1
-localization we easily get that in the resulting model we
will have d(S
k+1
) =
2
. The only problem is to show that the iteration has the
S
k
-localization property (to conclude that in the extension d(S
k
) =
1
). But
this is an application of 3 Chapter XVIII of [Sh:f]. We may think of S
k
as
a relation on

(after canonical mapping). Keeping the notation of [Sh:f] we

put:
S o
<1
(H(
1
)
V1
) and for a S, g
a

is such that for each f a

,
f S
k
g
a
, g = g
a
: a S);

= 1;

R = R
0
) = S
k
).
Note that (

R, S, g) strongly covers i it covers i S is stationary (for strongly
covers we are in Possibility B). The property (S

k
)
Q
k
of claim 2.4.1(2) guaran-
tees that the forcing notion Q
k
is (

R, S, g)-preserving. Hence Theorem 3.6 of
Chapter XVIII of [Sh:f] applies to this situation and the iteration is (

R, S, g)-
preserving. Consequently we are done.
Problem 3.6 Are the following consistent:
1. b < d(S
2
) < d(S
3
) < . . . d(S
k
) < d(S
k+1
) < . . . ?
2. There exists a sequence

: <
1
)

of increasing functions such

that
< <
1
d(S

+
) < d(S

+
)?
References
[Bar1] T.Bartoszy nski, Combinatorial aspects of measure and category,
Fundamenta Mathematicae 127 (1987):225239.
5
0
1


r
e
v
i
s
i
o
n
:
1
9
9
5
-
0
6
-
2
8







m
o
d
i
f
i
e
d
:
1
9
9
5
-
0
6
-
2
8


[RoSh 501] October 6, 2003 28
[Bar2] T.Bartoszy nski, Additivity of measure implies additivity of cat-
egory, Transactions of the American Mathematical Society 281
(1984):209213.
[BJSh:368] T.Bartoszy nski, H.Judah, S.Shelah, The Cicho n Diagram, Journal
of Symbolic Logic, 58 (1993):401423.
[BsSh:242] A.Blass, S.Shelah, There may be simple P
1
and P
2
points and
Rudin-Keisler ordering may be downward directed , Annals of Pure
and Applied Logic, 33 (1987):213-243.
[CiPa] J.Cicho n, J.Pawlikowski, On ideals of subsets of the plane and on
Cohen reals, Journal of Symbolic Logic, 51 (1986):560-569.
[GoSh:448] M.Goldstern, S.Shelah, Many simple cardinal invariants, Archive
for Mathematical Logic, 32 (1993):203-221.
[Kun] K.Kunen, Set Theory (An introduction to Independence Proofs),
NorthHolland, Amsterdam, 1983.
[Jec] T.Jech, Set Theory, Academic Press, New York 1978.
[NeRo] L.Newelski, A.Ros lanowski, The ideal determined by the unsymmet-
ric game, Proceedings of the American Mathematical Society, 117
(1993):823-831.
[Sh:207] S.Shelah, On cardinal invariants of the continuum, Proceedings of
the Conference in Set Theory, Boulder, June 1983; ed. J. Baum-
gartner, D. Martin and S. Shelah, Contemporary Mathematics
31 (1984):183-207.
[RoSh:470] A.Ros lanowski, S.Shelah, Norms on possibilities: a missing chapter
of Proper and Improper Forcing, in preparation.
[RoSh:475] A.Ros lanowski, S.Shelah, More forcing notions imply diamond,
Archive for Mathematical Logic, to appear.
[Sh:b] S.Shelah, Proper Forcing, Lecture Notes in Mathematics 940,
Springer 1982.
[Sh:f] S.Shelah, Proper and Improper Forcing, Springer, to appear.
[Voj] P.Vojt as, Topological cardinal invariants and the Galois-Tukey cat-
egory, Proceedings of the Conference on recent developments of
General Topology and its applications, in memory of F.Hausdor,
Academic Verlag.