<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>Decidability frontier for fragments of logic with transitivity?</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Introduction</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Preliminaries</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>University of Opole</institution>
          ,
          <country country="PL">Poland</country>
        </aff>
      </contrib-group>
      <fpage>2</fpage>
      <lpage>13</lpage>
      <abstract>
        <p>Several decidable fragments of rst-order logic have been identi ed in the past as a generalisation of the standard translation of modal logic. These include: the uted fragment, the two-variable fragment, the guarded fragment and the unary negation fragment; some of them have been recently generalised or combined to yield even more expressive decidable logics (guarded negation fragment or uniform onedimensional fragment). None of the fragments allows one to express transitivity of a binary relation or related properties like being an equivalence, a linear or a partial order, that naturally appear in speci cations or in veri cation. The question therefore arises what is the impact of adding transitivity to these fragments and, where the cost is too high, how can these languages be tamed. In this talk we survey results concerning the decidability frontier of the above-mentioned fragments extended with transitivity. We discuss both, general and nite satis ability, as presence of transitivity axioms often allows one to express axioms of in nity. Simultaneously, we locate in the picture known description logics, discuss relevant technics, admire a few exotic results and state some open questions.</p>
      </abstract>
      <kwd-group>
        <kwd>First-Order logic Transitivity Complexity</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>-</title>
      <p>quanti ers are relativized by atomic formulas, (iii) negation is applied only to
subformulas with a single free variable. Moreover, by properly reusing variables:
(iv) the number of variables can be restricted to two.</p>
      <p>The four syntactic restrictions considered separately give rise to four
fragments of rst order logic: (i) the uted fragment F L, (ii) the guarded fragment
GF , (iii) the unary negation fragment U N F , and (iv) the two-variable fragment
F O2. All of the four languages have the nite model property, and hence are
decidable. As already mentioned in the abstract, none of them can express
transitivity of a binary relation or related properties like being a partial order or an
equivalence relation. Hence, their extensions are studied when it is assumed that
some of the binary symbols from the signature are interpreted in a special way
(as transitive relations, orderings, equivalences etc.).</p>
      <p>In the talk we review main properties of the above mentioned fragments
and describe the decidability frontier for their extensions with various transitive
relations. The presentation includes joint work with Georg Gottlob, Emanuel
Kieronski, Jakub Michaliszyn, Andreas Pieris, Ian Pratt-Hartmann, and Wieslaw
Szwast. In the next section we recall formal de nitions of the four fragments
mentioned above; more surveying material with several references can be found
e.g. in [7, 4]. In Section 2 we present supplementary material that is featured in
the talk but has not yet been published elsewhere.
1</p>
      <p>De nitions
We assume relational signatures not containing constants or function symbols.
De nition 1. The uted fragment [5]: Let x! = x1; x2; : : : be a xed sequence
of variables. We de ne the sets of formulas F L[k] (for k 0) by structural
induction as follows: (i) any atom (x`; : : : ; xk), where x`; : : : ; xk is a contiguous
subsequence of x!, is in F L[k]; (ii) F L[k] is closed under boolean combinations;
(iii) if ' is in F L[k+1], then 9xk+1' and 8xk+1' are in F L[k]. The set of uted
formulas is de ned as F L = Sk 0 F L[k]. A uted sentence is a uted formula
over an empty set of variables, i.e. an element of F L[0]. Thus, when forming
Boolean combinations in the uted fragment, all the combined formulas must
have as their free variables some su x of some pre x x1; : : : ; xk of x!; and
when quantifying, only the last variable in this sequence may be bound.
De nition 2. The guarded fragment [1], GF , is de ned as the least set of
formulas such that: (i) every atomic formula belongs to GF ; (ii) GF is closed under
logical connectives :; _; ^; !; and (iii) quanti ers are appropriately relativised
by atoms. More speci cally, in GF , condition (iii) is understood as follows: if
' is a formula of GF , is an atomic formula featuring all the free variables
of ', and x is any sequence of variables in , then the formulas 8x( ! ')
and 9x( ^ ') belong to GF . In this context, the atom is called a guard. The
equality symbol when present in the signature is also allowed in guards.
De nition 3. The unary negation fragment [6], U N F , consists of formulas in
which the use of negation is restricted only to subformulas with at most one free
variable. More precisely, U N F is de ned as the least set of formulas such that:
(i) every atomic formula of the form R(x) or x = y belongs to U N F ; (ii) U N F
is closed under logical connectives _, ^ and under existential quanti cation; (iii)
if '(x) is a formula of U N F featuring no free variables besides (possibly) x, then
:'(x) belongs to U N F .</p>
      <p>De nition 4. The two variable fragment: By the k-variable fragment of a logic
L, denoted Lk, we mean the set of formulas of L featuring at most k distinct
variables. In particular F Ok denotes the set of all rst-order formulas with at
most k variables.</p>
      <p>All of the above languages are incomparable in terms of expressive power,
have the nite model property and the satis ability problem (= nite satis
ability problem) is decidable.
2</p>
      <p>Fluted Fragment with Transitive Relations
We show two undecidability results for the uted fragment with two variables,
F L2, extended with transitive relations. We employ the apparatus of tiling
systems (cf. e.g. [3]).</p>
      <p>A tiling system is a tuple C = (C; CH ; CV ), where C is a nite set of tiles,
and CH , CV C C are the horizontal and vertical constraints. Let S be any of
the spaces N N, Z Z or Zt Zt. A tiling system C tiles S, if there exists a
function : S ! C such that for all (p; q) 2 S: ( (p; q); (p + 1; q)) 2 CH and
( (p; q); (p; q + 1)) 2 CV . It is known that the problem whether a given tiling
system tiles any of the spaces N N, Z Z, Zt Zt is undecidable. To reduce one
of the problems to the ( nite) satis ability problem for some logic L we proceed
as follows.</p>
      <p>Let C = (C; CH ; CV ) be a tiling system. We write an L-formula C such that
the following reduction properties hold
(i)
(ii)</p>
      <p>C is satis able i C tiles N N or Z
C is nitely satis able i C tiles Zt</p>
      <p>Z, and
Zt, for some t
1.</p>
      <p>Properties (i) and (ii) above yield undecidability of the satis ability, respectively,
nite satis ability, problem for L.</p>
      <p>In order to encode tilings using two-variable logics it is useful to de ne
structures that are grid-like. A structure G = (G; h; v) with two binary relation h
and v is grid-like, if one of the standard grids N N, Z Z or Zt Zt can
be homomorphically embedded into G. To show that a structure G is grid-like
it su ces to require that G j= 8x(9yh(x; y) ^ 9yv(x; y)) (note that this is an
F L2-formula) and the following con uence property holds</p>
      <p>G j= 8x; x0; y; y0((h(x; y) ^ v(x; x0) ^ v(y; y0) ! h(x0; y0)):
(*)
The con uence property above uses four variables and is not uted. We will
enforce it (or a variation of it) in F L2 using transitive relations.</p>
      <p>
        Undecidability of F L2 with two transitive relations and equality
Suppose the signature contains two transitive relations b (blue) and r (red), and
additional unary predictes ci;j (0 i 3, 0 j 3) called colours. We write a
formula 'grid capturing several properties of the intended expansion of the Z Z
grid as shown 0i3n Fig1u3re 1. 2T3 here,33each e03lemen1t3 with2c3oordi3n3ates (k; l) satis es
ci;j , where i = k mod 4 and j = l mod 4 and the transitive relations connect
only some elements that are close in the grid. The formula 'g32rid is a conjunction
of the statements (
        <xref ref-type="bibr" rid="ref1">1</xref>
        )-(
        <xref ref-type="bibr" rid="ref7">7</xref>
        ) described below02; in the formulas in this subsection we
02 12 22 32 12 22
assume that addition in subscripts of the cij s is always understood modulo 4.
01 11 21 31 01 11 21 31
10
13
12
11
10
20
23
22
21
20
30
33
32
31
30
00
03
02
01
00
10
13
12
11
10
20
23
22
21
20
30
33
32
31
30
00
03
02
01
00
^
0 i;j 3
(
        <xref ref-type="bibr" rid="ref1">1</xref>
        ) there is an initial element: 9xc00(x).
(
        <xref ref-type="bibr" rid="ref2">2</xref>
        ) the cij s enforce a partition of the universe: 8x W_0 i 3 W_0 j 3ci;j (x).
(
        <xref ref-type="bibr" rid="ref3">3</xref>
        ) transitive paths do not connect distinct elements of the same colour:
8x(cij (x) ! 8y((b(x; y) _ r(x; y)) ^ cij (y) ! x = y))
(
        <xref ref-type="bibr" rid="ref4">4</xref>
        ) each element belongs to a 4-element blue clique; we write the following
conjuncts for each i; j 2 f0; 2g:
      </p>
      <p>8x(cij (x) ! 9y(b(x; y) ^ ci+1;j (y)))
8x(ci+1;j (x) ! 9y(b(x; y) ^ ci+1;j+1(y)))
8x(ci+1;j+1(x) ! 9y(b(x; y) ^ ci;j+1(y)))</p>
      <p>
        8x(ci;j+1(x) ! 9y(b(x; y) ^ cij (y)))
(
        <xref ref-type="bibr" rid="ref5">5</xref>
        ) each element belongs to a 4-element red clique; we write the following
conjuncts for each i; j 2 f1; 3g:
      </p>
      <p>8x(cij(x) ! 9y(r(x; y) ^ ci+1;j(y)))
8x(ci+1;j(x) ! 9y(r(x; y) ^ ci+1;j+1(y)))
8x(ci+1;j+1(x) ! 9y(r(x; y) ^ ci;j+1(y)))</p>
      <p>
        8x(ci;j+1(x) ! 9y(r(x; y) ^ cij(y)))
(
        <xref ref-type="bibr" rid="ref6">6</xref>
        ) A group of formulas saying that some pairs of elements connected by r are
also connected by b:
^ 8x(cii(x) !8y(r(x; y) ^ (ci;i 1(y) _ ci 1;i(y)) ! b(x; y)))
i=0;2
^ 8x(cii(x) !8y(r(x; y) ^ (ci;i+1(y) _ ci+1;i(y)) ! b(x; y)))
i=1;3
^ 8x(ci;i+1(x) !8y(r(x; y) ^ (ci;i+2(y) _ ci 1;i+1(y)) ! b(x; y))) (6c)
i=0;2
^ 8x(ci;i 1(x) !8y(r(x; y) ^ (cii(y) _ ci;i 2(y)) ! b(x; y)))
i=1;3
(
        <xref ref-type="bibr" rid="ref7">7</xref>
        ) A group of formulas saying that some pairs of elements connected by b are
also connected by r:
^ 8x(cii(x) !8y(b(x; y) ^ (ci;i 1(y) _ ci 1;i(y)) ! r(x; y)))
i=0;2
^ 8x(cii(x) !8y(b(x; y) ^ (ci;i+1(y) _ ci+1;i(y)) ! r(x; y)))
i=1;3
^ 8x(ci;i+1(x) !8y(b(x; y) ^ (ci;i+2(y) _ ci 1;i+1(y)) ! r(x; y))) (7c)
i=0;2
^ 8x(ci;i 1(x) !8y(b(x; y) ^ (cii(y) _ ci;i 2(y)) ! r(x; y)))
i=1;3
It should be clear that the structure shown in Figure 1 is a model of 'grid. One
can note that 'grid has also nite models expanding a toroidal grid structure
Z4m Z4m (m &gt; 0) obtained by identifying elements from columns 0 and 4m
and from rows 0 and 4m.
      </p>
      <p>Let us introduce the following de nitions:
h(x; y) :=b(x; y) ^
_ (cij(x) ^ ci+1;j(y)) _ r(x; y) ^
i=0;2
j=0;1;2;3
_ (cij(x) ^ ci+1;j(y))
i=1;3
j=0;1;2;3
v(x; y) :=b(x; y) ^ _ (cij(x) ^ ci;j+1(y)) _ r(x; y) ^ _ (cij(x) ^ ci;j+1(y))
i=0;1;2;3 i=0;1;2;3
j=0;2 j=1;3
(6a)
(6b)
(6d)
(7a)
(7b)
(7d)
Let A j= 'grid. We show that every model (A; h; v) of 'grid is grid-like. We
rst show that A satis es 8x(9yh(x; y) ^ 9yv(x; y)). One considers several cases
depending on the values of the unary predicates.</p>
      <p>
        Let a 2 A and assume A j= c00(a). By (
        <xref ref-type="bibr" rid="ref4">4</xref>
        ) there are a1; a2; a3; a4 2 A such
that A j= b(a; a1) ^ c10(a1) ^ b(a1; a2) ^ c11(a2) ^ b(a2; a3) ^ c01(a3) ^ b(a3; a4) ^
c00(a4). By (
        <xref ref-type="bibr" rid="ref3">3</xref>
        ) a = a4 and by transitivity of b the elements a; a1; a2; a3 form a
blue clique in A. Hence, A j= h(a; a1) ^ v(a; a3).
      </p>
      <p>
        The same argument works if a has one of the colours c02; c20 or c22 and,
similarly, applying (
        <xref ref-type="bibr" rid="ref5">5</xref>
        ) instead of (
        <xref ref-type="bibr" rid="ref4">4</xref>
        ) when a has the colours c11; c31; c13 or c33.
      </p>
      <p>
        Consider now the case A j= c10(a). By (
        <xref ref-type="bibr" rid="ref4">4</xref>
        ) there is a0 2 A such that A j=
b(a; a0) ^ c11(a0), hence A j= v(a; a0). Moreover, by (
        <xref ref-type="bibr" rid="ref5">5</xref>
        ) there are a1; a2; a3; a4 2 A
such that A j= r(a; a1)^c13(a1)^r(a1; a2)^c23(a2)^r(a2; a3)^c20(a3)^r(a3; a4)^
c10(a4). By (
        <xref ref-type="bibr" rid="ref3">3</xref>
        ) a = a4 and by transitivity of r the elements a; a1; a2; a3 form a
red clique in A. By (6d) A j= b(a; a1), hence A j= h(a; a1).
      </p>
      <p>Remaining cases are shown similarly.</p>
      <p>
        Now, we show the con uence property (*). Let a; a0; b; b0 2 A and A j=
h(a; b) ^ v(a; a0) ^ h(b; b0). One needs to consider several cases depending on the
colour of a, in each of them showing that A j= h(a0; b0). For instance:
{ A j= c00(a). Then A j= c01(a0)^b(a; a0)^b(a; b)^c10(b)^b(b; b0)^c11(b0). By (
        <xref ref-type="bibr" rid="ref4">4</xref>
        )
b0 is a member of a blue clique containing elements of colours c11; c01; c00; c10.
Since by (
        <xref ref-type="bibr" rid="ref3">3</xref>
        ) the relation b does not connect distinct elements of the same
colour, a0 belongs to the blue clique of b0 and A j= b(a0; a). Now, by
transitivity of b, A j= h(a0; b0).
{ A j= c30(a). Then A j= c31(a0) ^ b(a; a0) ^ r(a; b) ^ c00(b) ^ b(b; b0) ^ c01(b0).
      </p>
      <p>
        Similarly as above, b belongs to a red clique of a, hence A j= r(b; a). By (6a),
A j= b(b; a). Moreover, b0 is in a blue clique of b, and so A j= b(b0; b). By
transitivity of b, A j= b(b0; a0). Now, by (7c), A j= r(b0; a0). By (
        <xref ref-type="bibr" rid="ref5">5</xref>
        ), a0 is a
member of a red clique that, by (
        <xref ref-type="bibr" rid="ref3">3</xref>
        ), must contain b0. Hence h(a0; b0) holds.
Remaining cases can be shown in a similar way. Hence, every model of 'grid is
grid-like. Now we ensure that we also can assign tiles to elements of the grid-like
models using uted formulas. The task in F O2 is easy, it su ces to require that
(8) each node encodes precisely one tile: 8x( W_C2C Cx),
(9) adjacent tiles respect CH and CV ; the rst condition can be written as follows:
^ 8x(Cx ! 8y(h(x; y) !
C2C
      </p>
      <p>_
C0:(C;C0)2CH</p>
      <p>C0y)); ;
(9a)
and similarly for CV .</p>
      <p>Formula (8) is uted, formulas in (9) are not, but can be written as uted using
rst-order tautologies. E.g. each conjunct in (9a) can be written as follows:
^
i=0;2;j=0;1;2;3</p>
      <p>^
8x(Cx ^ cij (x) ! 8y(b(x; y) ^ ci+1;j (y) !
8x(Cx ^ cij (x) ! 8y(r(x; y) ^ ci+1;j (y) !
_
Let2 C be the conjunction of 'grid with the properties (8) and (9) written in
F L , as explained. Now it is routine to show that the reduction properties (i)
and (ii) hold. If C tiles any of the spaces Z Z or Zt Zt, for some t, we expand
the grids to our intended models. In the opposite direction, when G j= C and
G is in nite we obtain a tiling of Z Z; in case G is nite we obtain a tiling of
Zt Zt with t divisible by 4. As a result we simultaneously conclude that the
satis ability and the nite satis ability problems for F L2 with two transitive
relations are undecidable.</p>
      <p>One can also observe that the above formulas are guarded or can easily be
rewritten as guarded. Hence we also get the following
Corollary 1. The ( nite) satis ability problem for the intersection of the uted
fragment with equality with the two-variable guarded fragment is undecidable in
the presence of two transitive relations.</p>
      <p>Undecidability of F L2 without equality and with three
transitive relations
Suppose the signature contains transitive relations b (black), g (green) and r
(red), and additional unary predicates e, e0, f , l, ci;j (0 i 5, 0 j 2)
and di;j (0 i 2, 0 j 5); we refer to the ci;j 's and to the di;j 's as
colours. In this section addition in subscripts of the ci;j 's, it is always
understood modulo 6 in the rst position, and modulo 3 in the second position, i.e.
ci+a;j+b denotes c(i+a)mod 6;(j+b)mod 3; and, addition in subscripts of the di;j 's is
understood modulo 3 in the rst position, and modulo 6 in the second position.</p>
      <p>As before we write a formula 'grid that captures several properties of the
intended expansion of the N N grid as shown in Figure 2. There the unary
predicates l and f mark the elements in the left column, respectively, bottom
row of the grid. The predicate e marks elements on the main diagonal and the
predicate e0 marks elements with coordinates (k; k + 1). The predicates ci;j and
di;j together de ne a partition of the universe as follows: an element (k; k0) with
k0 &gt; k satis es ci;j with i = k mod 6, j = k0 mod 3, and an element (k; k0)
with k k0 satis es di;j with i = k mod 3; j = k0 mod 6. Transitive relations
connect elements that are 'close' in the grid; more precisely, paths of the same
transitive relation have length at most 7 and they follow one of four designed
patterns.</p>
      <p>
        The formula 'grid comprises several conjuncts. The most complex group of
conjuncts describes the requirement that each element of a certain colour has
a successor of an appropriate colour as shown in Figure 3. Conjuncts from this
group have the following form:
8x colour(x) ^ diag(x) ^ border(x) ! 9y(t(x; y) ^ colour0(y)) ;
(
        <xref ref-type="bibr" rid="ref1">1</xref>
        )
where colour and colour0 stands for one of the predicates letters ci;j or di;j ,
diag(x) means one of the literals e0(x), :e0(x), e(x), :e(x) or &gt; (i.e. the logical
constant true), border(x) means one of the literals l(x), :l(x), f (x), :f (x) or
&gt;, and t stands for one of the transitive predicate letters b, r or g. The precise
combinations of the literals and predicate letters in these conjuncts can be read
from Figure 3 and they are de ned in Table 1. The remaining conjuncts to 'grid
express the following properties:
(
        <xref ref-type="bibr" rid="ref2">2</xref>
        ) there is an initial element: 9x(d00(x) ^ e(x) ^ l(x) ^ f (x)).
(
        <xref ref-type="bibr" rid="ref3">3</xref>
        ) the predicates ci;j together with di;j enforce a partition of the universe.
(
        <xref ref-type="bibr" rid="ref4">4</xref>
        ) the predicates l, f , e and e0 propagate as intended; e.g.
      </p>
      <p>^ 8x c0;j (x) ^ l(x) ! 8y(c0;j+1(y) ^ (b(x; y) _ g(x; y)) ! l(y))
0 j 5</p>
      <p>
        ^ 8x di;j (x) ^ e(x) ! 8y((b(x; y) _ g(x; y) _ r(x; y)) ^ di+1;j+1(y) ! e(y)) :
0 i 2
0 j 5
(
        <xref ref-type="bibr" rid="ref5">5</xref>
        ) some pairs of elements connected by one transitive relation are also connected
by another one as intended in Figure 2; for instance:
8x(c11(x) !8y(b(x; y) ^ c10(y) ! r(x; y)))
8x(c20(x) !8y(g(x; y) ^ c30(y) ! r(x; y)))
      </p>
      <p>The structure depicted in Figure 2 is a model of 'grid. We show the following
Claim. Every model of 'grid is grid-like.</p>
      <p>
        Proof. Let A j= 'grid. We de ne an embedding h of the standard grid GN on
N N into A as follows. Let next : N N 7! N N be a successor function de ned
on GN as depicted by the thick path in Figure 3 starting in (0; 0) (ignoring any
colours). Let a 2 A be an element such that A j= d00(a) ^ e(a) ^ l(a) ^ f (a) that
exists by condition (
        <xref ref-type="bibr" rid="ref2">2</xref>
        ). De ne h(0; 0) = a. Now, we proceed inductively: suppose
h(k; k0) has already be de ned and h(k; k0) = a. Let a0 be the witness of a for
the appropriate conjunct from the group (
        <xref ref-type="bibr" rid="ref1">1</xref>
        ), i.e. where the monadic literals for
x agree with the monadic literals satis ed by a in A. De ne h(next(k; k0)) = a0.
      </p>
      <p>
        Correctness of the embedding can be shown by induction considering various
cases depending on the colours of the elements. Interested readers are invited
to do the proof by drawing: print Figure 3 on a sheet of paper, use appropriate
colours to draw the edges induced by transitivity of b, g and r, apply universal
conjuncts from the group (
        <xref ref-type="bibr" rid="ref5">5</xref>
        ) and compare the resulting picture with Figure 2
(when they agree the proof is nished!).
      </p>
      <p>We have one more obstacle to overcome. De ning the embedding we moved
through the grid in four directions, so in order to appropriately assign tiles to
nodes in models of 'grid we need to respect the four directions as we have xed
order of variables. The idea is to de ne four binary relations rt(x; y), lt(x; y),
up(x; y) and dn(x; y) with the intention that rt and the inverse of lt together give
the expected horizontal grid successor, and up and the inverse of dn together give
the expected vertical grid successor. For instance, the relation rt(x; y) is de ned
The formula above says that the relation rt connects elements that are connected
by (at least) one of the transitive relation and satisfy one of the possible
combination of colours: in the second line the combinations for crossing the diagonal
from left to right are listed, in the third line the left big disjunction describes
combinations when both elements are located above the diagonal, and in the
right big disjunction|when both elements are located on and below the
diagonal. The de nition of lt(x; y) is written complementing the de nition of rt:
lt(x; y) :=(b(x; y) _ g(x; y) _ r(x; y)) ^
(d00(x) ^ c50(y)) _ (d14(x) ^ c31(y)) _ (d22(x) ^ c12(y))
_ _</p>
      <p>(cij (x) ^ ci 1;j (y))
(i;j)2f(1;2);(2;2);(3;1);(4;1);(5;0);(0;0)g
_ (dij (x) ^ di 1;j (y))
(i;j)2f(0;1);(2;3);(1;5)g
_
Analogously one de nes the relations up and dn. Now the counterpart of the
formula (9a) from the previous subsection can be written as follows:
^ 8x(Cx ! 8y(rt(x; y) !
C2C
^ 8x(Cx ! 8y(lt(x; y) !
C2C</p>
      <p>_
and similarly for CV . As before, the formulas can be further rewritten as uted
thanks to the fact that in the de nitions of rt(x; y), lt(x; y), up(x; y) and dn(x; y)
the binary literals have the form t(x; y) (and not t(y; x)).</p>
      <p>As in the previous section all formulas used for the reduction are either
guarded or can easily be rewritten as guarded. Hence we also get the following
Corollary 2. The satis ability problem for the intersection of the uted
fragment without equality with the two-variable guarded fragment is undecidable in
the presence of three transitive relations.</p>
      <p>It is not obvious if the formula 'grid can be modi ed to de ne embeddings of
nite grids. Hence decidability of the corresponding nite satis ability problem
remains open. It also is open whether the full uted fragment, F L, with one or
two transitive relations (without equality) is decidable.</p>
      <p>Acknowledgements The research presented in Section 2 is supported by the
Polish National Science Centre grant No 2016/21/B/ST6/01444.</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          1.
          <string-name>
            <surname>Andreka</surname>
            , H., van Benthem,
            <given-names>J.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Nemeti</surname>
            ,
            <given-names>I.</given-names>
          </string-name>
          :
          <article-title>Modal languages and bounded fragments of predicate logic</article-title>
          .
          <source>Journal of Philosophical Logic</source>
          <volume>27</volume>
          ,
          <issue>217</issue>
          {
          <fpage>274</fpage>
          (
          <year>1998</year>
          ). https://doi.org/10.1023/A:1004275029985
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          2.
          <string-name>
            <surname>Blackburn</surname>
            , P., van Benthem,
            <given-names>J.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Wolter</surname>
            ,
            <given-names>F</given-names>
          </string-name>
          . (eds.):
          <article-title>Handbook of Modal Logic</article-title>
          . Elsevier Science Inc., New York (
          <year>2006</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          3. Borger, E., Gradel,
          <string-name>
            <given-names>E.</given-names>
            ,
            <surname>Gurevich</surname>
          </string-name>
          ,
          <string-name>
            <surname>Y.</surname>
          </string-name>
          :
          <article-title>The classical decision problem</article-title>
          .
          <source>Perspectives in Mathematical Logic</source>
          , Springer (
          <year>1997</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          4.
          <string-name>
            <surname>Kieronski</surname>
            ,
            <given-names>E.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Pratt-Hartmann</surname>
            ,
            <given-names>I.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Tendera</surname>
            ,
            <given-names>L.</given-names>
          </string-name>
          :
          <article-title>Two-variable logics with counting and semantic constraints</article-title>
          .
          <source>SIGLOG News</source>
          <volume>5</volume>
          (
          <issue>3</issue>
          ),
          <volume>22</volume>
          {
          <fpage>43</fpage>
          (
          <year>2018</year>
          ). https://doi.org/10.1145/3242953.3242958
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          5.
          <string-name>
            <surname>Quine</surname>
            ,
            <given-names>W.V.</given-names>
          </string-name>
          :
          <article-title>The variable</article-title>
          .
          <source>In: The Ways of Paradox</source>
          , pp.
          <volume>272</volume>
          {
          <fpage>282</fpage>
          . Harvard University Press, revised and enlarged edn. (
          <year>1976</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          6.
          <string-name>
            <surname>Segou n</surname>
          </string-name>
          , L.,
          <string-name>
            <surname>ten Cate</surname>
            ,
            <given-names>B.</given-names>
          </string-name>
          :
          <article-title>Unary negation</article-title>
          .
          <source>Logical Methods in Computer Science</source>
          <volume>9</volume>
          (
          <issue>3</issue>
          ) (
          <year>2013</year>
          ). https://doi.org/10.2168/LMCS-9(
          <issue>3</issue>
          :25)
          <fpage>2013</fpage>
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          7.
          <string-name>
            <surname>Tendera</surname>
            ,
            <given-names>L.</given-names>
          </string-name>
          :
          <article-title>Finite model reasoning in expressive fragments of rst-order logic</article-title>
          . In: Ghosh,
          <string-name>
            <given-names>S.</given-names>
            ,
            <surname>Ramanujam</surname>
          </string-name>
          ,
          <string-name>
            <surname>R</surname>
          </string-name>
          . (eds.)
          <source>Proceedings of the Ninth Workshop on Methods for Modalities, M4M@ICLA</source>
          <year>2017</year>
          , Indian Institute of Technology, Kanpur, India, 8th to 10th
          <year>January 2017</year>
          . EPTCS, vol.
          <volume>243</volume>
          , pp.
          <volume>43</volume>
          {
          <issue>57</issue>
          (
          <year>2017</year>
          ). https://doi.org/10.4204/EPTCS.243.4
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>