What is a Group Ring?

Author(s): D. S. Passman
Source: The American Mathematical Monthly, Vol. 83, No. 3 (Mar., 1976), pp. 173-185
Published by: Mathematical Association of America
Stable URL: http://www.jstor.org/stable/2977018
Accessed: 31-10-2015 23:28 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact [email protected].

Mathematical Association of America is collaborating with JSTOR to digitize, preserve and extend access to The American
Mathematical Monthly.

http://www.jstor.org

This content downloaded from 178.250.250.21 on Sat, 31 Oct 2015 23:28:19 UTC
All use subject to JSTOR Terms and Conditions
 WHAT IS A GROUP RING?

D. S. PASSMAN

1. Introduction.Let K be a field. Supposewe are given some three element set {a, (, y} and we
are asked to form a K-vector space V with this set as a basis. Then certainlywe merelytake V to be
the collection of all formal sums a ca+ b ,(+ c y with a, b, c E K. In the same way if we were
originallygiven four, five or six element sets we would again have no difficultyin performingthis
construction.After awhile, however, as the sets get larger,the plus sign becomes tedious and at that
point we would introducethe E notation.In general,if we are given a finite set S, then the K-vector
space V with basis S consistsof all formalsums ),2s aa *a with coefficientsa,a E K. Finally,there is
no real difficultyin letting S become infinite. We merely restrictthe sums Y2a ca to be finite, by
which we mean that only finitely many nonzero coefficientsaa, can occur.
Of course, addition in V is given by
(E a, ))+ (nba a) = E(aa+ba)b a
and scalar multiplicationis just
b( , a, a) (ba,,) cta.
Moreover,by identifying 3 E S with the element ,B'= ab, a E V,where b, = 1 and b",= 0 for a ,
we see that V does indeed have this copy of S as a basis and our originalproblem is solved.
Now, how do we multiplyelementsof V? Certainlythe coefficient-by-coefficient
multiplication
a, .a) 0 & .
a)=(a,) a
is exceedinglyuninterestingand other than that no naturalchoice seems to exist. So we are stuck.But
suppose finallythat we are told that S is not just any set, but ratherthat S is in fact a multiplicative
group. We would then have a natural multiplicationfor the basis elements and by way of the
distributivelaw this could then be extended to all of V.
Let us now startagain.Let K be a field and let G be a multiplicativegroup,not necessarilyfinite.
Then the group ring K[G] is a K-vector space with basis G and with multiplicationdefined
distributivelyusing the given multiplicationof G. In other words, for the latter we have

x y) E (axby) (xy)- Ecz z,

xeGa.
xE (Eby
y EG x,yEG zeG

where

CZ= axby xbx-lz.

xy=z xEG

Certainlythe associativelaw in G guaranteesthe associativityof multiplicationin K[G] so K[G] is a

ring and in fact a K-algebra. It is clear that these easily defined group rings offer ratherattractive
objects of study. Furthermore,as the name implies, this study is a meeting place for two essentially
differentdisciplinesand indeed the resultsare frequentlya rathernice blendingof grouptheory and
ring theory.
Supposefor a moment that G is finite so that K[G] is a finite dimensionalK-algebra.Since the
studyof finite dimensionalK-algebras(especiallysemisimpleones over algebraicallyclosed fields)is
in far bettershape than the studyof finitegroups,the groupringK[G] has historicallybeen used as a
tool of grouptheory.This is of coursewhatthe ordinaryandmodularcharactertheoryis all about(see
[21for example).On the otherhand,if G is infinitethen neitherthe grouptheorynor the ringtheoryis
particularlyadvancedand what becomes interestinghere is the interplaybetween the two. Our main
concernin this articlewill be with infinitegroupsand in the followingwe shalldiscussand provea few

173

This content downloaded from 178.250.250.21 on Sat, 31 Oct 2015 23:28:19 UTC
All use subject to JSTOR Terms and Conditions
 174 D. S. PASSMAN [March

selected results. These results, however, are by no means representativeof the entire scope of the
subject,but ratherwere chosen becausetheirproofsare both elegantand elementary.Few references
will be given here, but the interested reader is invited to consult the book [5] or the more recent
surveys, [4] and [6].
2. Zero divisors.Before we go any further,two observationsare in order.First,once we identify
the elements of G with a basis of K[G], then the formalsums and productsin , ax x are actually
ordinarysums and products.In particular,we shall drop the dot in this notation. Second, if H is a
subgroupof G then, since H is a subset of the basis G, its K-linear span is clearlyjust K[H]. Thus
K[H] is embedded naturallyin K[G].
Now let x be a nonidentityelement of G so that G contains the cyclic group (x ). Then K[G]
contains K[(x)] and we brieflyconsider the latter group ring. Suppose first that x has finite order
n > 1. Then 1, x,_ , xn-' are distinct powers of x and the equation

(1 -X) (1 + x + * + Xn1) = 1-Xn = O

shows that K[(x)], and hence K[G], has a properdivisorof zero. On the other hand,if x has infinite
orderthen all powersof x are distinctand K[(x)] consistsof all finite sumsof the form E aix. Thus
this groupring looks somethinglike the polynomialring K[x] and indeed every element of K[(x)] is
just a polynomialin x dividedby some sufficientlyhigh power of x. Thus K[(x)] is containedin the
rationalfunction field K(x) and is therefore an integraldomain.
Now what we have shown above is the following.If G has a nonidentityelement of finite order,a
torsion element, then K[G] has a nontrivialdivisorof zero, but if G has no nonidentityelement of
finite order, then there are at least no obvious zero divisors.Because of this, and with franklyvery
little additionalsupportingevidence,it was conjecturedthat G is torsionfree if and only if K[G] has
no zero divisors. Remarkablythis conjecturehas held up for over twenty-fiveyears.
We still know very little about this problem.In fact all we know is that the conjectureis true for
some rathersimpleclassesof groupsas, for example,free groups,or abeliangroups.In the lattercase
the proof is even quite easy. Thus, supposethat G is torsionfree abelian,and let a, P be elementsof
K[G] with a/3 = 0. Then, clearly, a and ,B belong to K[H] for some finitely generated subgroup
H C G and, by the fundamentaltheoremof abeliangroups,H is just the directproductof the infinite
cyclic groups (x,), (x2), - and (x,). It is then quite easy to see that K[H] is contained naturally
between the polynomialring K[xX,x **, x,] and the rationalfunctionfield K(x1,x2, ** , x,). Indeed,
K[H] is just the set of all elements in K(x1,X2,* * , x,) which can be written as a polynomialin
xI, x2, , x, divided by some sufficientlyhigh power of (xI x2... x,). Thus K[H] is certainly an
integraldomain and hence, clearly, either a = 0 or / = 0. Actually the best result to date on this
conjectureconcernssupersolvablegroups. Here for the first time nontrivialring theory comes into
play but the proof is unfortunatelytoo complicatedto give here.
On a more positive note, there is a variant of the zero divisor problem which we can handle
effectively.A ring R is said to be primeif aR/ = 0 for a, : E R impliesthat a = 0 or : = 0. Clearly
this agrees with the usual definitionin the commutativecase. In addition,this concept makes more
sense arddis more importantfrom a ring-theoreticpoint of view than the very much more stringent
conditionof no zero divisors.For example, the matrixring Knis alwaysprime, even though it does
have zero divisors for n -' 2. The main result of interest here is as follows.
THEOREM I. The group ring K[G] is prime if and only if G has no nonidentityfinite normal
subgroup.
The proof in one directionis quite easy. SupposeH is a nonidentityfinite normalsubgroupof G
and set a = ElhEH h, the sum of the finitelymanyelementsof H. If h E H then hH = H so ha = a and
thus
a 2
ha=IHIa.
hEH

This content downloaded from 178.250.250.21 on Sat, 31 Oct 2015 23:28:19 UTC
All use subject to JSTOR Terms and Conditions
1976] WHAT IS A GROUP RING? 175

In particular,if : = IH I1 - a then afp= 0. Furthermore,since H is normalin G, we have for all

x E G, Hx = xH, and hence ax = xa. Thereforea commuteswith a basisof K[G], so it is centraland
we have
aK[G]f = K[G]af = 0.
Finally, since a and,, are clearly not zero we see that K[G] is not prime.
The conversedirectionis much more difficultand as a starterwe must ask ourselveshow we are
going to find a finite normalsubgroupin G. Observethat if H is such a subgroupand if h E H, then
certainly h has finite order, and furthermore,for all x E G we have
hX = x-1hx E x-1Hx = H.
Thus all G-conjugatesof h are containedin H and so certainlythere are only finitelymanydistinct
ones. This, therefore, leads us to define two interestingsubsets of G, namely,
A(G) = {x E 0G x has only finitely many conjugatesin G}
and
Al(G) = {x E G x has only finitely many conjugates
in G and x has finite order}.

At this point a certainamountof grouptheoryobviouslycomes into play. Whilethe proofsare by

no means difficultit does seem inappropriateto offer them here. Therefore we just tabulate the
necessaryfacts below. See [5, Lemma 19.3] for details.
LEMMA 1. Let G be a groupand let A(G) and Al(G) be definedas above. Then
(i) A(G) and Al(G) are normalsubgroupsof G.
(ii) A(G) : AI(G) and the quotientgroup A(G)IA+(G) is torsionfree abelian.
(iii) I+(0) # (1) if and only if G has a nonidentityfinite normalsubgroup.
Part(iii) above certainlyseems to answerourfirstquestion.Now observethat if x is any elementof
G, then the G-conjugatesof x are in a naturalone-to-one correspondencewith the right cosets of
CG(x), the centralizer of x. Thus x E A(G) if and only if [G: CG(x)] < x and we shall need the
following two basic propertiesof subgroupsof finite index ([5, Lemmas 1.1 and 1.21).
LEMMA2. Let G be a groupand let H,, H2, , Hn be a finite collectionof subgroupsof G.
(i) If [G: H,] is finite for all i, then [:G n H,] is finite.
(ii) If G is the set theoreticunion G = U ,Hig, of finitelymany rightcosetsof the subgroupsHi,
thenfor some i we have [G : Hi] finite.
We can now proceed with the remainderof the proof of Theorem I. If a = XaxxE K[G], let us
define the support of a, Suppa, to be the set of all group elements which occur with nonzero
coefficientin this expressionfor a. Thus Suppa is a finitesubset of G whichis empty preciselywhen
a = 0. Now suppose that K[G] is not prime and choose nonzerogroup ring elements a and : with
aK[G]p = 0. If x E Suppa and y E Supp( then certainly(x-'a)K[G](Py-') = 0, and furthermore,
1 E Suppx -'a, 1 E Supp 3y-'. Thuswithoutloss of generality,we may assumethat a and ( have 1 in
their supports.
Let us now write a = ao + a, and f = go + f3 where Supp ao, Suppgo C A(G) and Supp a1,
Supp S3Q G - A(G). In other words, we have split a and f into two partial sums, the first one
containingthe groupelements in A(G) and the second one containingthe rest. Since 1 is containedin
the supportsof a and : we see that ao and go are nonzeroelements of K[A(G)]. Ourgoal here is to
prove that aofo = 0. Once this is done the theorem will follow easily. Let us suppose by way of
contradiction that aofo $ 0. Then clearly ao( = aofo + aofl is not zero since Supp aofo C A(G) and

This content downloaded from 178.250.250.21 on Sat, 31 Oct 2015 23:28:19 UTC
All use subject to JSTOR Terms and Conditions
176 D. S. PASSMAN [March

SuppaoP3 C G - A(G) so there can be no cancellationbetween these two summands.Thus we can

choose z E G to be a fixed element in Suppaof.
We observe now that if u E Suppao then [G :CG(U)] is finite and thus, by Lemma 2 (i),
H= nuueSuppaoCG(U) has finite index in G. Moreover, if h E H, then h centralizes all elements in
Suppao and hence h centralizes ao. Let us now write Suppat = {x1, x2, ,x, } and Suppf =
{yi, Y2, ,ys}. Furthermore,if x, is conjugateto zy7 in G for some i,j, then we choose some fixed
gi, E G with g-' x,g = zy-'.
Now let h E H and recallthat aK[G]P = 0. Then since ahf = 0 and h centralizesao, we have
0 = h-'ahf = h-'(ao+ al)h . ( = ao( + h-'a,h3.
Thus z occurs in the support of h-'a, hp= - ao( so there must exist i, j with z = h-'xihyj or in other
words h-'xih = zy7'. But then this says that xi and zy7' are conjugatein G, so by definitionof gijwe
have
= zy-' g-' xgij,
h-'xih
and thus hg-' centralizesxi. We have thereforeshownthat for each h E H there exists an appropriate
i, j with h E CG(xi)gi,and hence

HC U CG(x,)gij.
I,)

Now [G H] is finite, so G = UkHwk is a finiteunion of rightcosets of H and we concludefrom

the above that

G = U CG(X,)giWk.

We have therefore shown that G can be written as a finite set theoretic union of cosets of the
subgroupsCG(xi)and hence we deduce from Lemma2 (ii) that for some i, [G CG(xi)] is finite. But
this says that xi E A(G) and this is the required contradiction,since by definition xi E Suppa1 C
G - A(G). Thus aofo = 0.
We can now completethe proof quite quicklyand easily. Since ao and gfoare non zero elementsof
K[A(G)] with ao(o = 0 we see that K[A(G)] has nontrivialzero divisors.Hence A(G) cannotpossibly
be torsion free abelian. On the other hand, accordingto Lemma 1 (ii), A(G)/A+(G) is torsion free
abelian so we must have A+(G)$ (1). Thus finally we deduce from Lemma 1 (iii) that G has a
nonidentityfinite normal subgroupand the theorem is proved.
In conclusion,we remarkthat TheoremI has an amusingapplicationto the zero divisorproblem.
Namely, we can show that if G is a torsion free group then K[G] has nontrivialzero divisorsif and
only if it has nonzero elements of square zero. Of course if a E K[G], a X0 with a2 = 0, then
certainlyK[G] has zero divisors,so this directionis reallytrivial.In the otherdirectionlet a and ( be
nonzeroelements of K[G] with a(p = 0. Since G is torsionfree, TheoremI impliesimmediatelythat
K[G] is prime so we have PK[G]aO0. But
(K[G]a .
PK[G]a = PK[G] (a:) K[G]a = 0
since ap = 0 and hence we see that every element of (K[G]a has square zero.

3. Idempotents.Let us returnagain to the group ring of a finite group and considerits regular
representation.That is, we view V= K[G] as a K-vector space on which K[G] acts as linear
transformationsby rightmultiplication.In particular,since V is finite dimensionalhere, each choice
of a basis for V gives rise to a certainmatrixrepresentationfor K[G]. More precisely,for each such

This content downloaded from 178.250.250.21 on Sat, 31 Oct 2015 23:28:19 UTC
All use subject to JSTOR Terms and Conditions
1976] WHAT IS A GROUP RING? 177

basiswe obtainan appropriatefaithfulhomomorphismr: K[G]-> KnfromK[G] into Kn,the ringof

n x n matricesover K. Clearly n = dim V = IG 1.
Supposefirstthat r correspondsto the naturalbasis for V, namely G itself. Then for each x E G
rightmultiplicationby x merelypermutesthe basiselements. Thus r(x) is a permutationmatrix,that
is a matrixof 0's and l's havingpreciselyone 1 in each row and column.Moreover,if x # 1, then gxk g
for all g E G so r(x) has only 0 entries on its main diagonal.Hence clearly trace r(x) = 0. On the
other hand, r(1) is the identitymatrix,so trace r(1) = n = 1. Now matrixtrace maps are K-linear
so for a = >a,x E K[G] we obtain finally

tracer(a) = E ax trace r(x) = a, IG

In otherwords,the traceof r(a) is just a fixedscalarmultipleof a,, the identitycoefficientof a.

We are thereforeled, for arbitrarygroupsG, to definea maptr: K[G] -> K, calledthe trace,by

tr(Iax xy = a,.

Moreover,it seems reasonableto expect that tr should have certain trace-likebehaviorand that it
should prove to be an interestingobject for study. Indeed, we observe immediatelythat tr is a
K-linear functionalon K[G], and furthermore,that for a = E axx, 1 = E b,y we have

trag = E
xy = I
axby

and this is symmetricin a and : since xy = 1 if and only if yx = 1. Thus tr a} = tr (a.

Now one of the most interestingquestionsabouttr concernsidempotentse E K[G] andthe values
taken on by tr e. To see what we mightexpect these valuesto be, let us againassumethat G is finite.
Since the trace of a linear transformationis independentof the choice of basis, we compute
IG t tr e = trace r(e)
by taking a basis more appropriatefor e. Namely we write V as the vector space direct sum
V = Ve + V(1 - e) and then we choose as a basis for V the union of ones for Ve and for V(1 - e).
Since e acts like one on the first set and like zero on the second, we conclude therefore that

IG Itr e = trace r(e) = dim Ve.

Thusif we are able to divideby IG t in K (thatis, if the characteristicof K does not divideIG I) then
tre = (dim Ve)/!!G .
In other words, we see that tr e is contained in the prime subfield of K, i.e., the rationals Q if
charK = 0 or the Galois field GF(p) if charK = p. Furthermore,in characteristic0 since 0_
dim Ve |IGI we have 0?tre - 1
Now tt turnsout that these two propertiesof tr e are indeed true in general,but we shouldobserve
that there is a basic distinctionbetween them. The fact that tr e is containedin the primesubfieldis
clearlyan algebraicpropertywhile the inequality0 c tr e _ 1 is in some sense analyticin nature.With
this basic dichotomyin mind, we now proceed to consider some proofs.
THEOREMII. Let e E K[G] be an idempotent.Then tr e is containedin the primesubfieldof K.
We firstconsiderfields K of characteristicp > 0, and as it turnsout, the propertyof e we use here
is eP = e. The reason for this is that in algebrasover fields of characteristicp the pth power map is
fairlywell behaved. Indeed the identity(a + b)P= aP + bP alwaysholds in commutativeK-algebras
and for noncommutativeones an appropriategeneralizationexists. To be more precise let A be a

This content downloaded from 178.250.250.21 on Sat, 31 Oct 2015 23:28:19 UTC
All use subject to JSTOR Terms and Conditions
178 D. S. PASSMAN [March

K-algebraand define [A, AJ, the commutatorsubspaceof A, to be the subspacegeneratedby all Lie
products[a, b] = ab - ba with a, b E A. Then the resultwe are alludingto is the followingand a proof
can be found in [5, Lemma 3.4].
LEMMA3. Let A be an algebraover a field K of characteristicp > 0. If a,, a2, "l am E A and if
n > 0 is a given integerthen thereexists an elementsb E [A, A with
(a, + a2+ + **+ a.)Pn
= a P1"+ a2Pn + ..+ a
pn
+ b.
With this fact we can now prove the firstpart of TheoremII. Let e = I a,x be an idempotentin
K[G] and let S denote the subset of Suppe consistingof all those elementsof ordera powerof p, the
characteristicof K. Then since S is finite, there exists an appropriatepth power, say pS, with xP' _ 1
for all x E S. Now let n be any integerwith n ? s and we applyLemma3 to ePr = e. Thusthereexists
an element y in the commutatorsubspace of K[GJ with
e =e pn=E (ax) x pn+ Y,

and we proceed to computethe tracesof both sides of this equation. Observefirstthat try = 0 since
for any a, ,BE K[GJ we have tr a: = trO3aand hence tr [a, PIJ 0. Also since n _ s it followsthat for
any x E Suppe we have xP"= I if and only if x E S. Thus clearly

tre = (a.)P
=
(E a,X)
xes xE=S

Now this equationholdsfor all integersn ? s andin particularif we take n = s and s + 1 we obtain

(tr e )P -(Eax ) -(E ax) -tr e .

Thereforetr e is an element k E K whichsatisfiesk P = k and since all such elements k are contained
in GF(p), the theorem is proved in characteristicp > 0.
We now proceed to consider the characteristic0 case and here we will need the following (see
[7, Assertions IV, Y, VI]).
LEMMA4. Let A = Z[ai, a2,, ar] be an integraldomain in characteristic0 which is finitely
generatedas a ringoverthe integersZ and let b E A be an elementnotcontainedin therationalnumbers
Q. Thenthereexistsa maximal ideal M of A such that F = AIM is a field of characteristic p > 0 for
some primep and such that the image of b in F is not containedin GF (p).
We remarkthat this fact is an easy consequence of the Extension Theorem for Places if b is
transcendentalover Q. But if b is assumedalgebraic,then its possible imagesin the fields A/M are
greatlyrestricted.Fortunatelyin this case we can applythe FrobeniusDensityTheorem,a resultfrom
algebraicnumbertheory, to prove the lemma.
Now let K have characteristic0 and let e = E axx be an idempotent in K[GJ. If A =
Z[ax Ix E Suppe], then A is clearlyan integraldomainin characteristic0 and A is finitelygenerated
as a ringover the integersZ. Furthermore,e is an idempotentin A [GI, wherethe latteris the subring
of K[GI consistingof all elementswith coefficientsin A. Now let M be any maximalideal of A with
F = AIM a field of characteristicp >0. Then under the naturalhomomorphism

A[G1- *(AIM)[G] = F[G]

the imageof e is an idempotentand hence by the firstpartof TheoremII the imageof tr e is contained
in GF (p). The second part of Theorem II now follows immediatelyfrom Lemma 4 with b = tr e.
Thuswe have provedthe firstconjecturedpropertyof tr e, namelythat it is alwayscontainedin the
primesubfieldof K and it remainsto considerthe second property,namelythat 0 ? tr e ' 1 if K has

This content downloaded from 178.250.250.21 on Sat, 31 Oct 2015 23:28:19 UTC
All use subject to JSTOR Terms and Conditions
1976] WHAT IS A GROUP RING? 179

characteristic0. Here the proof is muchmore technicalin nature,it is analyticratherthan algebraic,

and we shall present some basic reductionsand then indicate why the proof should work.
Let K be a field of characteristic0 and let e = E aXx be an idempotent in K[G]. If F=
Q(a. Ix E Suppe) then F is a finitelygeneratedfield extensionof Q and e is an idempotentin F[GJ.
Furthermore,F can be embeddedin the complex numbersC and hence again e is an idempotentin
C[G]. Thus,as a firstreduction,we mayclearlyassumethat K = C is the fieldof complexnumbers.A
second minorobservationis that we need only show that tr e - 0. The reason for this is that if e is an
idempotent,then so is 1 - e and then 1 - tr e = tr(1- e) >0O yields tr e - 1.
Let us now considerthe complex groupring C[G]. As is to be expected when one combinesthe
structureof group ringswith the richnessof the complex numbers,many nice additionalproperties
emerge. For example, if a = E axx and : = E b,x are elements of C[G] we set

(ae,:)=E abxg

and

where a is the complex conjugate of a and Ia I is its absolute value. Then clearly (, ) defines a
Hermitianinnerproducton C[G] with the groupelements as an orthonormalbasis and with jj jjthe
usual associatednorm. Furthermorelet a* be given by

ar* = aXx-1.

Then the map * is easily seen to satisfy

(a + f)* = a* + a*, (ap)* =P*a*, a** = a,

so that * is a ringantiautomorphism
of C[G] of order2, that is, an involution.Now observethat
(a,,)= tra/3* = trf3*a
and hence we deduce easily that if y is a third element of C[G], then

(a, 3y) = (ay*,f3) = (13*a,y).

In other words, * is the adjoint map with respect to this inner product for both right and left
multiplication.
We can now use the above machineryto obtainan alternateproof of the assertiontr e _ 0 at least
when G is finite.Let I = eC[G] be the rightideal of C[GJ generatedby the idempotente and let I'
be its orthogonalcomplement.Then since C[G] is a finite dimensionalvector space, we know that
I + IP = C[G] is a directsumdecomposition.But IP is not just a subspaceof C[G], it is in fact also a
rightideal. To see this let a E I, f E I' and -yE C[G]. Thensince I is a rightideal aRy*E I andhence

(a, 3y) = (ay*, /3)= 0.

Thus f3y is orthogonalto all a E I and we conclude that fy E IP.
We have thereforeshown that I + I' = C[G] is a decompositionof C[G] as a direct sum of two
rightideals and we let f + f' = 1 be the correspondingdecompositionof 1. As is well known,f and f'
are then both idempotentswith I= fC[G] and I' = f'C[G]. Now f is orthogonalto f'C[G] =
(1-f)C[G] so for all a E C[G] we have
0 = (f, (1 - f)a) = ((1 - f)*f, a)

This content downloaded from 178.250.250.21 on Sat, 31 Oct 2015 23:28:19 UTC
All use subject to JSTOR Terms and Conditions
180 D. S. PASSMAN [March

and hence (1- f)*f E C[G] =O. Thus f = f*f so

f* =(f*f)* =f*f=f
and we see that f is a self-adjointidempotent,a projection.It is now an easy matterto complete the
proof. Since eC[G] = I = fC[G] both e andf are left identitiesfor this ideal and hence we have
tre = trfe = tref = trf.
Furthermoref = f*f so
tre = trf = trf*f= - IfI2=0
and the result follows.
It is amusingto observe that this proof yields only tr e 0Oand not tr e E Q, but it does have the
redeeming virtue of being extendable to infinite groups. Of course the finiteness of G was used
cruciallyhere. In fact it was used to deduce that I + I' = C[G] since such decompositionsare not in
generaltrue for infinitedimensionalinnerproductspaces.But the proof reallyrestsupon the auxiliary
element f and there is an alternatecharacterizationof this element whichwe can use. Let a E I and
consider the distance between a and 1 E C[G]. Then by definition
d(a,l)2 = -a-112 =(a - 1,a-1)
and since 1 = f + f' and (a - f,f) = O we have
d(a, 1)2= (a -f-f', a -f-f ) a-f 11f2+fff '2.
Thus d (a, 1) f?L f' and equalityoccursif and only if a = f. In otherwords,f is the uniqueelement of
I which is closest to 1.
Now let G be an arbitrarygroup, let e be an idempotentin C[G] and set I = eC[G]. Then we
define the distance from I to 1 to be

d = inf d(a, 1)- inf 11a-1 1

Since C[G] is no longercompleteif G is infinite,there is no reasonto believe that some elementof I

existswhichis closest to 1. But therecertainlyexists a sequenceof elementsof I whose corresponding
distances approach d. As it turns out, it is convenient to choose a sequence fl,f2, of
fn.,

elements of I with
d2<Ifn -12<d2 + /n4.

Then this sequence plays the role of the auxiliaryelement f of the finite case and indeed the final
conclusionmirrorsthe originalformulatr e = lft112obtainedabove. Namelyafter a certainamountof
workwith inequalitiesand approximations,whichwe will not give here (see [5, Section22] for details),
we deduce finally that

tre = limlInfI2n0.

This completes the proof.

4. Semisimplicity.We close with one final problem, the semisimplicityproblem. The amusing
thing about this one is that while it has been workedon for over twenty-fiveyears, it is only recently
that a viableconjecturehas been formulated.In this sectionwe shalldiscussthe conjecturealongwith
some of the early work in characteristic0.
We firstoffer some definitions.If R is a ring, then an R-module V is an additiveabeliangroup
which admits right multiplicationby the elements of R. More precisely, we are given a ring

This content downloaded from 178.250.250.21 on Sat, 31 Oct 2015 23:28:19 UTC
All use subject to JSTOR Terms and Conditions
1976] WHAT IS A GROUP RING? 181

homomorphismR -* End V from R to the endomorphismringof V ahd by way of this mapthere is a

naturalactionof R on V whichwe denote by rightmultiplication.Thus in essence, V is an R-vector
space. Furthermorewe say that V is an irreducibleR-module if there are no R-submodulesof V
other than 0 or V itself. For example,if R = F is a field then the irreducibleF-modules are precisely
the one-dimensionalF-vector spaces.
Now what we would like to do is study R in termsof its irreduciblemodulesbut there is a natural
obstructionhere. It is quite possible that we cannoteven tell the elements of R apartin this manner,
that is, there could exist two distinctelements r, s E R such that for all irreducibleR -modules V and
for all v E V we have vr = vs. Of course if the above occurs then v(r - s) = 0 and hence we cannot
distinguishthe element r - s from 0. Thus what is of interest here is
JR = {r E R I Vr = 0 for all irreducibleR -modules V},
the Jacobsonradicalof R. This is easily seen to be a two sided ideal of R which can be alternately
characterizedas follows (see [3, Theorem 1. 2. 3]).
LEMMA5. Let R be a ring with 1. Then
JR = JrE Rf 1 - rsis invertiblefor all s (} R.
Now a ring is said to be semisimpleif its Jacobson radical is zero and this then leads to two
problems in group rings. The first is to characterizethose fields K and groups G with K[G]
semisimpleand then the second, more ambitiousproblem,is to characterizein generalthe Jacobson
radicalJK[G]. The initial work here concernedfields of characteristic0 since for G finite it was a
classicalfact that K[G] is semisimple.PresumablyK[G] is alwayssemisimplein characteristic0 and
the first result here on infinite groups concernedthe field C of complex numbers.
THEOREMIII. For all groupsG, C[G] is semisimple.
For our proof, whichclosely mirrorsthe original,we revertto the notationof the previoussection
and we furthermoreintroducean auxiliarynorm on C[G] by defining Ia I= E| ax I if a = , ajx.
Clearly Ia + I' Ia I+ Ijf and I a/3 I' a lI I, l . Now let a be a fixed element of JC[GJ. Then for all
complex numbers4, 1 - ;a is invertibleby Lemma 5 and we consider the map
f (;)= tr (1 - a)-' .
This is a complex function of the complex variable ; and we shall show that f is in fact an entire
function and we will find its Taylor series about the origin.
For conveniencewe set g(;) = (1 - ;a)-' so that f(;) = tr g(;). Furthermore,it is clear that all
g(;) C C[G] commute and hence we have the basic identity

g( g(ij) = (1 - {a) -_ (1 - 7a)-'

=[(1 - qa) - (1 - {afl(l - (a)-'(l - -qa)-'

=~
(- B)a g (;)g (q).)
We first show that |g(rq)l is bounded in a neighborhood of ;. Now by the above g(q) =
gso lg(i1)f lg(;)f+f-l fag(;l Ig(7)f and hence
Ig(-q) '{ -,q1 lag(;)l}=g(;) l
In particular,if we choose i1 sufficientlyclose to g then we can make the factor{.*. } largerthan 1/2
and thus we deduce that q - - implies Ig(iq)IJ?21g(;)I.
Next we show that f(t) is an entire function.To do this we firstplug the above formulafor g(Qq)
into the right hand side of the basic identity to obtain

g(;) - g(71)= (; - 7)ag(;) {g(') - (a-71)ag(;)g(71) }

This content downloaded from 178.250.250.21 on Sat, 31 Oct 2015 23:28:19 UTC
All use subject to JSTOR Terms and Conditions
182 D. S. PASSMAN [March

Then we divide this equation by ; - -q and take traces to obtain

f )0- f( ) - tr ag(;)2 = _(-7)tr a2g(;)2g(-).

Finallysince Itr y y Iwe concludefromthe boundednessof Ig(77)Iin a neighborhoodof ; that

lim = tr ag ()2

Hence f(;) is an entire function with f(;) given by f'(;) = tr ag(;)2.

Now we computethe Taylorseries for f aboutthe origin.Sincef(;) = tr(1 - a )-1 whatwe clearly
expect here is that for ; small we can write (1 - {a)-1 as the sum of an appropriategeometricseries
and then obtain f by taking traces. To be more precise and rigorousset
n

S = E tr a.
1=0
Then

f ;)-Sn(;) = tr g(;) - ot

= trg( ){1-(1- a) iai}

tr
tg(;)gn+ 'an+1

and thus

If(; Sn ()|_||n+1 I a In +1 I g(;

Now by our previousremarksIg(;) is boundedin a neighborhoodof zero andhence for sufficiently

small we deduce that

lim sn(')
n -r
= f(0.
We have therefore shown that

t(0) E; tr a
. =0

is the Taylor series expansionfor f(;) in a neighborhoodof the origin. Furthermore,f is an entire
functionand hence we can invoke a well-knowntheoremfrom complex analysis([1, Theorem3, pg.
142])to deducethat the above seriesdescribesf(;) andconvergesfor all ;. In particularwe have

lim tra n = 0

and this holds for all a E JC[G].

We conclude the proof by showing that if JC[G] 0 then there exists an element a E JC[G]
which does not satisfy the above. Indeed, suppose is a nonzero element of JC[G] and set
a = f3 */1 /3112. Then a E JC[G] since the Jacobsonradicalis an ideal. Furthermorewe have a = a *
and
tr a = 11 11-2 tr 3P* = 1 :
1-2K 112 = 1.

Now the powers of a are also symmetricunder * so for all m ?0 we have

tra am+1 = tr a2m( 2m)* 11a2m ?>(tra2m)2.

This content downloaded from 178.250.250.21 on Sat, 31 Oct 2015 23:28:19 UTC
All use subject to JSTOR Terms and Conditions
1976] WHAT IS A GROUP RING? 183

Hence by induction tr a"2' = 1 for all m -0 and this contradicts the fact that tr a' ->0. Thus
JC[G] = 0 and the result follows.
Needless-to-saythis analyticproof both excited and annoyed the algebraistswho rightlyviewed
group rings are an algebraicsubject. Later algebraicproofs were indeed given and the best result to
date in characteristic0 is as follows. Let K be a field of characteristic0 whichis not algebraicover the
rationalsQ. Then for all groups G, K[G] is semisimple.Furthermore,for all the remainingfields the
semisimplicityquestion is equivalent to that of the rational group ring (see[5], Theorem 18.3).
We shouldmentionat this point that there is an amusingploy to try to extend the above argument
to Q[G]. Namely, we again considerf(;)= tr(1 - ;a)-, this time as a map from Q to Q, and we
observe as before that for | small

f()= 2 tra'.
0

Now the righthand side here describesan analyticfunctionin a neighborhoodof the origin and this
functionhas the ratherstrangepropertythat it takes rationalsto rationals.The questionthen is: must
such a function necessarily be a polynomial? Unfortunately this is not the case and a simple
counterexampleis as follows. Let r,, r2,, rn,... be an enumerationof the nonnegativerationalsand
define

-( d)
1 (ri-
(2 2 _ ... (r- _2)
n=1 n! (r +) (r2
22)
+1)(r+1)

Then h is easilyseen to be an entirefunctionwhichtakes Q to Q andwhichis not a polynomial.

Actually, it now seems that the semisimplicityproblem for Q[G] will be settled by purely ring
theoreticmeanssince the resultwill follow froman appropriatenoncommutativeanalogof the Hilbert
Nullstellensatz.To be more precise, it has been conjecturedthat the Jacobson radicalof a finitely
generatedalgebrais alwaysa nil ideal. This conjecture,along with the knownfact ([5] Theorem18.5)
that Q[G] has no nonzeronil ideals,wouldthen easilyyield JQ[G] = 0. Thuswhatis reallyof interest
is the case of fields of characteristicp > 0 and here we are just beginningto guess at what the answer
might be. Suppose first that G is finite. Then K[G] is semisimplehere if and only if p 4 IG 1. Of
course this latter condition does not make sense for infinite groups, but the equivalent condition,
namely that G has no elements of order p, does. None-the-lessthis is not the right answer.It is not
true that JK[G] # 0 if and only if G has an element of order p. What does seem to be true is that
elements of order p do play a role, providedthat they are suitably well placed in G. To give this
ambiguousidea some more meaningwe startby quotinga simplefact ([5] Lemmas16.9 and 17.6).
LEMMA 6. Let a be an elementof the groupringK[G]. Thena E JK[G] if and only if a E JK[H]
for all finitely generatedsubgroupsH of G with H D Suppa.
Thisof coursesays thatwhatever"wellplaced"meansexactly,it mustsurelydependupon how the
particularelement sits in each finitelygeneratedsubgroupof G. Now we get morespecific.Let H be a
normal subgroupof G. Then we say that H carriesthe radicalof G if
JK[G] = JK[H] K[G]. .

Our goal here is to find an appropriatecarriersubgroupH of G such that the structureof JK[H] is
reasonablywell understood.Admittedlythis is a somewhatvague statementbut we would certainly
insistthatJK[H] be so simplein naturethatwe can at least decide easilywhetheror not it is zero.
The first candidatefor H is based upon the A' subgroupand Lemma 6 above. We define
A+(G) = {x E G Ix E A+(L) for all finitelygeneratedsubgroupsL of G containingx} .
Then A+(G) is a characteristicsubgroupof G and in all of the examplescomputedso far A+(G) does

This content downloaded from 178.250.250.21 on Sat, 31 Oct 2015 23:28:19 UTC
All use subject to JSTOR Terms and Conditions
184 D. S. PASSMAN [March

indeed carry the radical.Furthermore,the ideal

I = JK[A+(G)] K[G] .

has been studiedin generaland it has been found to possess many of the propertiesexpected of the
Jacobsonradical.Thus it now appearsthat A+(G) carriesthe radicalbut we are unfortunatelya long
way from a proof of this fact. Finally we remark that this conjecture has a nice ring theoretic
interpretation.Namely it is equivalentto the assertionthat if G is a finitely generatedgroup, then
JK[G] is a union of nilpotent ideals.
Now given the above conjecture,the next naturalstep is to study A+(G) and JK[A+(G)I and it
soon becomesapparentthat these objects are not as nice as we had hoped for. IndeedA+(G) turnsout
to be just any locally finite group and so the problem of determining JK[A+(G)] is decidedly
nontrivial.We are thereforefaced with the problemof studyinglocally finite groupsin general and
here a new ingredientcomes into play.
Let G be a locallyfinitegroupso that by definitionall finitelygeneratedsubgroupsof G are finite.
Now if A is such a finitesubgroupof G we say that A is locallysubnormalin G if A is subnormalin
all finite groups H with A C H C G. In other words, we demand that each such H has a chain of
subgroups.
A = Ho C H1C C RH= H

for some n with H, normalin H,?1.Then using these locally subnormalsubgroupsas buildingblocks
in a certain technical manner we can define a new and interesting characteristicsubgroupof G
denoted by J(G). Again it turnsout that in all computedexamplesJ(G) carriesthe radicalof G and
that furthermorethe ideal
I = JK[V(G)J . K[G]

possessesin generalmanypropertiesexpected of the Jacobsonradical.Thus it now appearsthat if G

is locally finite then Y(G) carriesthe radicalof G.
The semisimplicityproblemfor groupsin generalhas thereforebeen split into two pieces, namely
the cases of finitely generated groups and of locally finite groups. Moreover, if we combine the
correspondingconjecturesfor each of these cases, then what we expect, or at least hope, is that the
groupY(A+(G)) carriesthe radicalof G. While this groupmay appearto be somewhatcomplicatedin
nature, there are in fact some nice structuretheoremsfor it. Furthermore,and most important,we
have JK[Y(A+(G))]$ 0 if and only if 5(A+(G)) contains an element of order p.
Added in Proof. Hopefully reference [4] will appear soon in translation in the Journal of Soviet Mathematics.
There is a new monograph in Russian by A. A. Bovdi entitled Group Rings. A second volume is anticipated. This
author is presently working on an expanded and updated version of [5] which will be entitled The Algebraic
Structureof Group Rings. It will be published in two volumes by Marcel Dekker, Inc. There has been some exciting
progress recently on the zero divisor problem, in particular in the case of polycyclic-by-finite groups. This can be
found in the papers by K. A. Brown, On zero divisors in group rings, to appear in the Journal of the London Math
Society, and by D. R. Farkas and R. L. Snider, K0 and Noetherian group rings, to appear in the Journal of Algebra.

References
1. L. V. Ahlfors, Complex Analysis, McGraw-Hill, New York, 1953.
2. C. W. Curtis and I. Reiner, Representation Theory of Finite Groups and Associative Algebras,
Interscience, New York, 1962.
3. I. N. Herstein, Noncommutative Rings, Carus Mathematical Monographs, No. 15, Math. Assoc. Amer.,
1968.
4. A. V. Mihalev and A. E. Zalesskii, Group Rings (in Russian), Modern Problems of Mathematics, Vol. 2,
VINITI, Moscow, 1973.

This content downloaded from 178.250.250.21 on Sat, 31 Oct 2015 23:28:19 UTC
All use subject to JSTOR Terms and Conditions
1976] MATHEMATICAL NOTES 185

5. D. S. Passman, Infinite Group Rings, Marcel Dekker, New York, 1971.

6. , Advances in group rings, Israel J. Math., 19 (1974) 67-107.
7. A. E. Zalesskii, On a problem of Kaplansky, Soviet Math., 13 (1972) 449-452.

MATHEMATICS DEPARTMENT, UNIVERSITY OF WISCONSIN, MADISON, WI 53706.

MATHEMATICAL NOTES

EDITED BY RICHARD A. BRUALDI

Material for this Department should be sent to Richard A. Brualdi, Department of Mathematics, University
of Wisconsin, Madison, WI 53706.

A CONNECTED, LOCALLY CONNECTED,

COUNTABLE HAUSDORFF SPACE

GERHARD X. RITTER

In this MONTHLY, 76 (1969) 169-171, A. M. Kirch gave an example of a connected, locally

connected countableHausdorffspace. We give another,geometricallymuch simpler,example. The
example we providewill be a modifiedversion of a connected,but not locally connected,countable
Hausdorffspace constructedby R. H. Bing in [1].
Let X = {(a, b) e Q x Q 1b > O},where Q denotes the set of rationalnumbers,andlet 0 be a fixed
irrationalnumber.The points of our space will be the elements of X. With each (a, b) E X and each
? > 0 we associatetwo sets L. (a, b) and R. (a, b) whichconsistof all pointsof X in the trianglesABC
and A'B'C', respectively, where A = (a - b/0,O), B = (a - b/0 + ?,0), C = (a - b/0 + ?/2, 0?/2),
A'= (a + bl0 - ?,0), B' = (a + b10,O), and C'= (a + bl0 - ?12,0?/2). We now define a basic
?-neighborhoodof (a, b) E X by Ne(a, b) = {(a, b)} U Le(a, b) U Re(a, b).
Since 0 is irrational,no two points of X can lie on a line with slope 0 or - 0. Thus, for any two
distinct points of X, we may choose a sufficientlysmall ? > 0 such that the points lie in disjoint
?-neighborhoods.Therefore, X is Hausdorffand, obviously, a countablespace.

(a, b) / / /

L, (a, b) R,.(a, b

FIG. I

Figure 1 shows that the closure of each basic E-neighborhood N. (a, b) consists of the union of four
infinite strips with slope ? 0 emanating from the bases of the triangles R. (a, b) and L. (a, b). Hence
the closures of each pair of basic neighborhoods intersect. Therefore, X is connected but not regular
and, hence, not metrizable.

This content downloaded from 178.250.250.21 on Sat, 31 Oct 2015 23:28:19 UTC
All use subject to JSTOR Terms and Conditions