This article was downloaded by: [Stanford University Libraries]

On: 16 May 2012, At: 16:12

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Communications in Algebra
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/lagb20

A funny property of sphere and equations over groups

a
Anton A. Klyachko
a
Department of Mathematics and Mechanics, Moscow State Univesity, Moscow, 119899

Available online: 27 Jun 2007

To cite this article: Anton A. Klyachko (1993): A funny property of sphere and equations over groups, Communications in
Algebra, 21:7, 2555-2575

To link to this article: http://dx.doi.org/10.1080/00927879308824692

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to
anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should
be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,
proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in
connection with or arising out of the use of this material.
 COMMUNICATIONS IN ALGEBRA, 21(7), 2555-2575 (1993)

I
A FUNNY PROPERTY OF SPHERE AND EQUATIONS OVER GROUPS.

Anton A. Klyachko
Department of Mathematics and M e c h a n i c s

Mo-TOW St ate Uni verz-i t y

Mo-TOW 4 d 9899

I
Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

I
I
I
I
, In t h i s paper we p r e s e n t and prove a simple but
I
, non-obvious t o p o l o g i c a l fact. This fact 1s s o simple
I
I

l
and funny t h a t can be included i n a collection o f
I p u z z l e s o r suggested as a problem f o r a school
I mathematical tournament. The alm o f t h l s article is t o
I
1 p r e s e n t a method by which t h i s fact, In spite of its
I unserious character, can be applied t o produce new
non- t r i v i a l group- t h e o r e t ic r e s u l t s .

2555

Copyright @ 1993 by Marcel Dekker, Inc.

 KLYACHKO

0. Introduction.

The results of this paper concern with the theory

of equations over groups. One can find the basic
definitions and classical results of this theory in
we ll -known mnogrphy C 121. The recent years approaches
and advancementes can be found in 113 - till and in the
literature cited there.
To discribe the present moment situation in few
words one can say that there are three princple
conject.ures concerning the solvabi1i ty of equations
over groups
Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

(
=
i
L Kerud re-Laundebach conject ure.
Any s i n g 1 e e q u a t i o n over any t o r s i on- free g r o u p

i s s o l v a b l e.

A A s~i n g 1 e p o w e r e d e q u a t i o n over- a n y group is

. s v Z uabl e.

None of this conjectres has been settled yet. In

fact., the Kerudre-Laundebach conjecture is the common
title for a few conjectures generalizing one another.
We imply the classical version.
& The group K=cG, t;I=l>
is n o t t r i v i a l , p r o v i d e d the g r o u p G is not t r i u i a2 .
The main result of this paper is the following
Theo~elRA. (;I]. i s t r u e , i f aLZ the c o e f f i c i e n t s of I
have i n f i n i t e o r d e r , i n particular, if G i s t o r s i o n - f r e e .
 FUNNY PROPERTY OF SPHERE AND EQUATIONS OVER GROUPS 2557

This paper contains also some r e s u l t s of another kind,

as the following.
m F f 3 m B. The sol vabi l i t y of s y s t em

o f e q u a t i o n s over group with one unknown and zero

exponent sum v e c t o r is i n v a r i a n t w i t h r e s p e c t t o change

of unknown w i t h n o n - t o r s i o n c o e f f i c i e n t s .

In p a r t i c u l a r i t implies:
Theorem C. The s y d em o f e q u a t i o n s [ ~ ( t ) , =I
over a group G i s -1 vabL e over 6,if every coefficient

o f ~ ( th)a s infinite order Cfor exarnpLe ,if G is

tor-+ on- f r e e ) .

More precise formulations are given in Section 2.

Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

The r e l a t i o n s between Theorem A and conjectures Ci and

C 2 are evident, t h a t is why we l i k e t o s a y more about
t h e other r e s u l t s . T h i s r e s u l t s is a s t e p in direction
t o prove conjecture C2. One can see i t observing t h a t
CZ? is equivalent t o the f'ollowing statement,
(2'. The nat uraL m a p s
A ---- > c A e ; Cv, A34v, W=1> <------ B
are monomorphic, if A and B are torfion- free groups,

and vE(A&)\ (ALIBI.

While Theorem C is equivalent to the following
statement. The natural m a F
A ---- > <A*; Cv,M=I> <------ B
are monomorphic, i f every factor of V has i n f i n i t e

o r d e r , and V C (A*B)\(AU B) .
 2558 KLYACHKO

On the other hand, our results are dual in some

sense to conjecture C3 which is equivalent to the
following
C3'. The soLvabiiity of equation of the type
k
( ~ ( t ) )= I , k>4
is a property invariant with respect t o nonconstant

change o f unknowns.

While Theorem C is equivalent to the following.

T h e s o l v a b i Z i t y o f s y s t e m o f egucrt i ons o f the t y p e

r~(t),a=i
over a g r o u p G is a p r o p e r t y i n v a r i a n t wi t h r e s p e c t to

nonconstant change of unknow with non-torsion

Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

coefficients.

Finitely, we remark that the condition of

infiniteness of coefficients orders is actual.The
f'ollowing example due to R. C. Lyndon E l l 3 demonstrates
it.The equation atbl=l is not solvable over the
group ca, b;a3 =l,bab-' =a2>.
The basic topological proposition will be
presented and proved in Section 1. Theorem A will be
proved in Section 2, Theorem B and C - - in Section 3.
Sections 2 and 3 are mutually independent, however the
last is some more difficult.

1. A topological proposition

Let there be a simply connected domain i n a n

o r i e n t e d s u r f ace and a. moving point (a car) on the

 FUNNY PROPERTY OF SPHERE AND EQUATIONS OVER GROUPS 2559

boundary of the domain. We w i l l s a y t h e c a r moves

pro.!XTly a r o u ~ i d t h e d o ~ ~ l a i ni f it moves along tlhe

boundary in the positve direction cont iniously,
perpetualy, w i t h no s t o p s , no r e v e r s e s and visiting

e v e r y p o i n t o f t h e b o u n d a r y i n f i n i t e i y many t i m e s .

-prO~osition.Let. there be an oriented 2-sphere and

a fi ni t e cor?nected graph on i t which di vicies the s p k r e
i nto a fini t e nua3e.r o.F si mple-comected d 0 1 ~ ins.
? Let,
t'kre be a car. on the L?ou~iaryof each do~,mi11 and l e t
t ,he cars m v e properi j/ around i t s clon~i
.~s.
Then inere exist. at least two poi 11ts or^ the s p , k r e
Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

i n whi ch conpl e t e col 1 i sior) happa?. A col1 i si on is

called complete, if i ! ) a point of rmltipiicity k
col l i cle k cars sirml tani ousl y.
Remark 1.; t is easy to denionstrate that the

conditions "with no stops" and "with nc, reVercesa' a y e

actual.
Remark2 One can demonstrate, tliat under

conditions of the proposition and moreover, under

additional condition o f p e r i o d i c i t y of t h e m o t i o n the
totdl number o f p o i n t s of c o m p l e t e collisions can be

arbitrary exept 0 and 1. If we r e p l a c e s p h e r e b y a n y

o t h e r o r i e n t e d compact surface, this n u r ~ b e r can be

a r b i t r a r y i n c l u d i n g 0 and 1.

-f Of the
.-P ~t i s s u f f i c i e n t t o
prove, that there are at least two points of SOm
 KLYACHKO

FIG.?
Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

FIG. 2

collisions, because one can eliminate every

non-complete collision by a local v a r i a t i n g of t h e

m o t i o n ( F i g . 1). T h e c a r s pass t h e v e r t e x i n i n d i c a t i n g

order, and move in the same maner o u t s i d e a small

n e i g h b o u r h o o d of t h e v e r t e x .

i t is s u f f i c i e n t t o p r o v e , t h a t t h e r e a r e a t least

one p o i n t of u o l l i s i o r r . Indeed, i f we h a v e a sphere

w i t h a s i n g l e c o l l i s i o n p o i n t , we c o u l d t a k e two c o p i e s
of s u c h a s p h e r e , cut out a small neighborhood of

collision points conriect the spheres by a pipe

-tesselatedN into a single domain by the simplest

manner and d e f i n e p r o p e r l y t h e m o t i o n on t h e b o u n d a r y
 FUNNY PROPERTY OF SPHERE AND EQUATIONS OVER GROUPS 2561

; I

FIG. 3

of t h i s doulain ( E i g . 2 ) . Tlius we would u b t a i n a sphere

w i t h no c o l l i s i o n p o i n t s .
S i n c e one c a n o b t a i n no a d d i t i o n a l c o l l i s i o n p o i n t s

by a l i t l e v a r i a t i n g of t h e m o t i o n , we may a s s u m e t h a t
Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

i n e a c h moment e v e r y car, e x e p t m y b e o n e , is o u t s i d e

the vertices.

L e t e v e r y c a p b e o u t s i d e t h e v e r t i c e s i n t h e moment
t . Let us consider the dual tesselation of the sphere,
select on the dual tesselation the edges dual to the
edges having cars on them in the m m n t t , and orient
this selected edges on the right with respect to the
correspond i g cars ( Fi g. 3).
Let us consider a minimal oriented circle on the
dual tesselation. Such a circle exists, because every
vertex has an edge exiting from it (since every domain
has a car on the boundary mving around in the positive
direction), and the number of the vertices is finite.
Such a circle is illustrated in Fig.4a. The oriented
 FIG. 4a

FIG. 4b
Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

FIG.5

edges in t h e i n t e r i o r of the c i r c l e form a forest TO

complete t h e proof one has t o not ice t h a t such a c i r c l e
is c o n t r a c t i n g as t h e time i s going.For example, as i t
i s i 1l u s t r a t e d i n Fig. 4: when t h e c a r corresponding t o
t h e edge a pass t h e next vertex, t h e edge a disappears,
and the edge a' appears. F i n e t l y , we w i l l have a c i r c l e
of' t h e form presented i n t h e Fig. 5a o r 5b. The first
means a presence of two c a r s the same edge, t h e second
 FUNNY PROPERTY OF SPHERE AND EQUATIONS OVER GROUPS 2563

means a presence of a car on a dead-ended edge.Either

implies a complete collision (in future or in past) of
multiplicity 2 or 1 respectively.
If the initial circle with a forest inside has the
opposite orientation, one has to replace the direct
course of time by the converse one in the reasoning
above.

2. Equations over groujx.

We WI11 use the following lemma which is an easy

to grove collorary of Vm-Campen's lemma.
Lem 1 . i ~C w3. ... , w , , I c A 4 3 , MW*, ... % >>/) A
Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

$ I l l , then t h e r e exists a tesselation of ,?-Mere with

oriented c o r n e r s labelled by elements of AU B ( t h u s one

can con-ider v e r t i c e s and f a c e s of the tesselation as

aLso labelled up to conjugacy as usualy) and t h e

following conditions hold:

- the label of the distinguished vertex Vo(the

~ X k r i 0 rvertex, is a non-trivial element of A;

- the Label o f e a c h other v e r t e x is i d e n t i t y ( i n A

or B>;
- the ZabL of each face is SO- W i Cup to

conjugancy and inversion>.

TheOl%lll 1. The natural

A---> <Ad3;l~,Al=I>
 2564 KLYACHKO

FIG. 6

I
-1 .
coLZision time = t : d = q if mt<n, d=C& f nctc2n.

FIG. 7
Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

is m n o m o r p h i c , if W e ( A e ) \ A , and each f a c t o r o f W has

i nf i n i t e order.

Proof. Let, w = c, ...cn - the normal f'orni Ci #l,for

i=$ .....n-4. ci6A,i 15 : ci6B.i 6 - ~ : n€ .Each
face of the tesselation is of the type illustrated in Fig.6.
Let u s chose t h e regim of the m t i o n of the car along
the boundary of such a face as following. In t h e o r i g i n
tnoment the car is i n the point marked with * , and t h e
velocity of the motion is one edge i n the u n i t of t i m e .
l t is easy to see t h a t the c o l l i s i o n s can happen
o n l y in the v e r t i c e s (and in the integer moments of
time) . the proposition guarantees a complete c o l l i s i o n
in an i n t e r i o r v e r t i x V.
 FUNNY PROPERTY OF SPHERE AND EQUATIONS OVER GROUPS 2565

FIG. 8

The motion is p e r i o d i c with the period 2n, hence

there is a collision in an i nter ior vertex in an
integer rnonlent t , 0 $, t 4 2n. If t # 0, n, then
some factor of w has a finite order ( F i g . 7 ) .
If t=O (t=n), then at.the point V all adjointing
Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

faces are jointing by the points * !*) (fig.8). In

this case on the boundary of the union of this faces we
have a label equalled to identity in A*B, and we can
reduce the tesselation in the usual way.
Theorem 2. ( A generalized HNN-extension). k t H
and H 9 be two i s o m o r p h i c s u b g r o u p s of a g r o u p A

Then for a n y B and f o r any W 6 (A*)\A with the factors

o f i n f i n i t e order, t h e n a t u r a L map

A----> <A*; iwdhw=hq, h H)> is m n o m o r p h i c.

Proof. Let A' be the HNN-extension of the group A

b. = <A, t; ?t HGH 9>
 KLYACHKO

then by Theorem 1 t h e n a t u r a l map

A - - <A?&; [A' ,t-Aw1-lr =G
is monomorphic. While i n t h e subgroup of' G generated by
A and B we have t h e e q u a l i t y
w'lhw = w'%t-lhtfl w = w'lth't** w = hB
f o r each h H. Q. E. D.
Theorem 3. ~ e {wi(t)=l,
t it11 (1)
be a s-yste m of equations over a group A, e r e the

e x p o n e n t sum of t i n ench Wi is zero. Let u(t)k(A*t>)

\A, and each c o e f f i c i e n t of U has i n f i n i t e order, then

the syys e m

{ ~ ~ ( ~ ( t ) ) =i €i ,11 (2)
Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

is ,wLvable over A i f f the s y s t e m ( 1 1 i s solvable.

Proof. I t i:; s u f f i c i e n t t o prove t h a t solvability

of (11 i m p l i e s s o l v a b i l i t y of ( 2 ) . Let x be a s o l u t i o n
of ( 1) x C A'= A, wi(x)=l. Let G =gp( ix-l AX! i€x)
c A'.
G has t h e automorphism g: ---> x-'gx. By Theorem 2 the
n a t u r a l tnap
ACG--> <G, t; (u(t))-2 ~ u ( t ) = x - % ~ = ~
i s monomorphic. Thus t o complete t h e proof we have j u s t
t o observe that vi (u(t))=l i n H.

3. Cervaire-Laundebach conjecture.

We need t o generalize Propsition 1 in a l i t t l e .

hopcsition la k t U LL condi t i ons of

Propsi t i on .i be fuL fi L L ed, but -5op me a1 L aced.

 FUNNY PROPERTY OF SPHERE AND EQUATIONS OVER GROUPS 2567

Then, as well. there are at Lea+ ttea pointsof

compl et e coL 1 i si o n s , jrovi ded t he fol 2 owi n g condi t i on.

If i n a vertex V cars Ct , C2 ,... , Ch ( i n the

anticLockwi.ss order) stop, then cars C j - d CitfC~d n>

Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

are n e v e r s i t u a t e d i n V , i m u L t a n i o ~ - Z y < F i g . 903.

Proof. Let us replace each vertex in which some

cars stop by a small face, as it is illustrated in
Fig.9b, and let car C i be moving along
each
corresponding segment of the boundary of new face D f'or
the t i m it was staying at V in the original
tesselation. As it i s easy to see, one can organize a
mtion of a new car d around D with no stops as in
Proposition 1 and with no collisions, provided the
condition of Proposition la CFig.lO).Finetly, we will
a obtain a mtion satisfying Proposition 1. The proof is
over.
Now, we need another version (due to J. Howie C 7 l )
of Van-Camwn' s lemma.
 KLYACHKO

t h e graphic of a traffic i n the boundary of D.

Exi ,ct ence of i n t e r v a l s T.i,i+j C accordi ng t o t he condi t i on>

a l l o w s t h e car d t o avoid coLLisions.

Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

FIG. 10

be not monomrphic. Then t h e r e @ x i - s s a t e s s e l a t i o n of

2-sphere such t hat

- t h e e d g e s are o r i e n t e d ;

- t h e c o r n e r s are Z abel Z e d wi t h e l e m e n t s of H;
- t h e e x t e r i o r v e r t e x h a s non-t ri v i a l Z abeL ;

-each i n t e r i r i o r v e r t e x h a s t r i u i a L L a b e l :

- t he l abel of each face i s equaZ t o -wme W i up to

permutation and i nver,i on.

Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

FIG. 11
FUNNY PROPERTY OF SPHERE AND EQUAmONS OVER GROUPS
2569
 KLYACHKO

i s -roL vabl e over H.

Proof. Proving by contradiction, It. is sufficient
to organize a motion satisfying the condition of
Proposition 13 on arbitrary tesselation satisfying the
conditions of Lemma la with fewer than two collision
points. Such a tesselat.ion has three kinds of faces
( Fig.11). Let each car m v i ng around a face of k-ind 1
Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

( 2 ) be situated in the origin mment at the corner be

(6:) and move with the velosity one edge in the unit of
t i m , but at the corner c (6%)
where it use to be
staying for 2k-d units of t.ime. As for cars mving
around faces of kind 3, let them move without stops
with the same constant velocity and pass the corner p*
in the origin moment.
It is easy to see that. cars of different kinds
never stav simultaniously.
It is evident also that during the intervals
i 0 . 1 > + 2 ~ .~ ~ . z ncaro moves in positive direction
with respect to orientation of' edge ,and during
intervals ( d , 2 3 + 2 ~ ,z EZno car moves in nggative
direction: in part.icular, there can be no collision
outside vertices.
 FUNNY PROPERTY OF SPHERE AND EQUATIONS OVER GROUPS 257 1

source sink

FIG. 1 2
Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

corners c and c'%l ternaies. and there are no stops

in t.he other cnrers.

FIG. 13

If there is a collision in an interior source-vertex

(or sink-vertex) (Fig. 12), it mans a relation in a

gp(ai,P) (or gp(bi ,PO)). This relation has to be
trivial, and the tesselation is reducable as usually.
As for the vertices of other kinds, it is evident
that the schedule of staying of cars in such a vertex
satisfy the condition of Proposition la (Fig.13)
Thus, the only point which can be a point of
complete collision is the exterior vertex.
 2572 KLYACHKO

Contradiction to Proposit.ion la.

LetG be a group. We will use the following

notat i ons.
GLi',t'iGti c G*t>
P, zgp( {G'~', 0, i<ml) R, =gp( {G
t il, I <i<ml)

: P - R the natural isomsrphism p: ---><Ip t

Lenolra 3. Let wE(;*ct> , a n d t h e exponent sum of t in

W is equal t o one. T h e n W is conjugat ed t o a product

Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

d > each ai, b i , c t P m

2> each ai$ Pm-l
3) each hip R W .
Proof. It is evident, that v is conjugated to a product
v t , V € ~ , V + R for
~ some s. Let us
consider this form of w, such that
1) s is minimal;
2)the length of v in sense of Rr =P, is minimal
Rs-i
f o r t h i s s.
It is evident, that if we change each R, -factor h in
the normal form of v by t-I @-'(h)t, we wi 11 obtain a
product (1) satisfying the requaired conditions for m=s-l.
 FUNNY PROPERTY OF SPHERE AND EQUATIONS OVER GROUPS 2573

Theorem 4. T h e group K=<G,t; w=l>

i s n o t trivial. if G is not trirjiaZ, and each

c o e f f i c i e n t of W has i n f i n i t e order.

Proof, We can assume that the exponent sum of t in

w is equal to one, otherwise group K has a non-trivial
cyclic quotient. By Lemma 3 we can assume that w has
form ( 1). Let H=G,*. ..G-, the free product of mtl copies
of 6.
The equation w-1 ( in the form ( I ) )
can be considwed as an equation over H by mans of the
natural isomorphism H 2 P,, G i = G ri? , P = ,R By
Lemnla 2 the system of equations over H
Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

w=1
<t-1pt=pq, p 6 PI
IV N
has a solution t in an overgroup HH. (To verify
, the conditions of L e m 2 one need to observe the
following simple fact. If u 6 (Ad3)\A, and each factor of
u has infinite order,t.hengp(A,u)=Ajrcw@.)
u N
It is evident ,that w ( t ) = l . Thus, H is an overgroup
of G containing a solution of ( Z ) , hence the natural
map G ---? K is a mnomorphism, and K is non-trivial.
Q.E. D.

ACKNOWLEGMENT
The author thanks A. Yu. Ol'shanski i for a lot of
useful conversations and remarks.
 2574 KLYACHKO

REFERENCES

Cil W. Bogley & J. Pride.

S. Aspherical relative
presentat ions. Proc. Ed i nburgh Math. Soc.
C 21 S.Brodsk i . Equat ions over groups and groups w i th a
single defining relation. Russian Math. Surveys. 35
( 1980) no. 4.
C33 M. Edjvet. The solution of certain sets of equations
over groups. In Groups-%. Andrews. 1989 ( eds.
C. M. Campbell & E. F. Robi nsonj. vol. 1. London Math.
Soc. Lecture Notes Series no 159 (Cambridge
University Press l99lj.
C 4J U Ed jvet. Equations over groups and a theorem of
Downloaded by [Stanford University Libraries] at 16:12 16 May 2012

Hi g n m , Neuman and Neumn. Proc. London Math. Soc.

Y. 62 ( 1991), no.3
[!XI MEdjvet & J.Howie. The solutions of length four
equations over groups. Trans. American Math. Soc.
V. 326 (1991), no.1
C 6 l S.M. Gersten. Reduciable diagram and equatons over
groups. In Essays in Group Theory. (ed.
S.M. Gersten). EXR I pub1 i cati ons vol. 8
( Spr i nger-Velgrad 1987)

C?l J. Howie. The solutions of length three equations

over groups. Proc. Edinburgh Math. Soc. 26 (1983)
t81 J. Howie. The quotient of a free product by a single
high-powered relator. I. Pictures. Fifth and higher
powers. Proc. London Math. Soc. 59 (1989)
 FUNNY PROPERTY OF SPHERE AND EQUATIONS OVER GROUPS 2575

C91 J.Howie. The quotient of a free product by a single

hi gh-powered re lator. I I. Fourth power:;. Proc.
London Math. Soc. 61 (1991)
C101 J.Howie. Nonsingular system of two length three
e q u a t i o n s over groups. Math. Proc. Camb. Phi 1. Soc.
110 (1991) ,nol.
[ill R. C.Lyndon. Equations i n groups. Bol. Soc. Brasi 1.
Mat. 11 (1980),no 1.
Cl2l R. C.Lyndon & P. E. Shupp. Combi nator i a1 group
theory. (Springer. Berlin. 1977')'

Received: June 1992

Downloaded by [Stanford University Libraries] at 16:12 16 May 2012