4. The categories of left and right modules of a semigroup

Let S be a semigroup. An action of S on the left of a set X is a map

aX : S × X X;(s,x)s.x such that s.(t.x) = (st).x  (8)

for all s,t S. Often we shall write sx for s.x if there is no ambiguity. A left S-set (or a left S-module) is a pair (X,aX) where X is a set and aX is an action of S on the left of X. A [left] S map α : X Y from a left S-set X to a left S-set Y (or a left S-morphism) is a map α : X Y that satisfy the condition

(sx)α = s(xα) for alls S, andx X. (9)

We denote by ModS the category with S-modules as objects and S-maps as morphisms. There is a functor U : ModS Set sending every X 𝖛ModS to the underlying set U(X) and every morphism f : X Y ModS to the map U(f) determined by f. In the sequal, we refer to U as the foregetful functor for left-modules. A subobject relation on 𝖛ModS is defined as follows: For X 𝖛ModS let aX : S × U(X) U(X) denote the action of S on U(X) given by the module structure on X. For X,Y 𝖛ModS write

X Y U(X) U(Y ) andaX = aY (S × U(X)). (10)

The relation on 𝖛ModS defined above is a subobject relation on ModS. In fact, it is easy to check that the J = ȷYX is a morphism in ModS if and only if X and Y satisfy Equation (10). The class of all these morphisms form a choice of subobjects in ModS (see Equation 2.1.2). Notice that, if X and Y are submodules of Z ModS then it follows from Equation (10) that X Y if and only if U(X) U(Y ).

It is clear that the foregetful functor U : ModS Set is faithful and preserve images. It follows that (ModS,U) satisfies axiom (Sb:b) of Definition 3.1. If X,Y 𝖛ModS and x U(X) U(Y ), then D = {sx : s S1 is a lefmodule such that x D X and x D Y . Therefore (ModS,U) satisfies axiom (Sb:c) also. If 𝒞ModS is a small subcategory of ModS with imagages and if U0 = U𝒞 then (𝒞,U0) clearly satisfy axioms (Sb:b) and (Sb:c). Further, if the partially ordered set 𝖛𝒞 of leftmodules has a largest member Z, then it follows from Equation (10) that the functor U0 is an embedding. In this case 𝒞 is an 𝒮-category with respect to U0. For convenience of using these facts in the sequal we state as a propositionm below.

Proposition 4.1. Let ModS be the category with subobjects in which objects are left S modules, morphisms are left module morphisms and the subobject relation is defined by Equation (10). The category ModS has images and the foregetful functor U : ModS Set is faithful and preserve subobjects. Moreover, the pair (ModS,U) satisfy axioms (Sb:b) and (Sb:c) of Definition 3.1. If 𝒞ModS is a small subcategory of ModS with images and if U0 = U𝒞 then (𝒞,U0) satisfy axioms (Sb:b) and (Sb:c). Further, if there is a leftmodule Z 𝖛ModS which is an upperbound of partially ordered subset 𝖛𝒞 of 𝖛ModS then, by Equation (10), the functor U0 is an embedding. In this case 𝒞 is an 𝒮-category with respect to U0.

Proof. The first part of the statement is routine to verify. Equation (10) shows that the natural forgetful functor preserves subobjects. To show that ModS has images, let f : X Y ModS. Then U(f) is a mapping from U(X) to U(Y ) and h = U(f) is a surjective map of U(X) onto Y = codh = imU(f) U(Y ). If s S and y Y there is x X with

sy = s(h(x)) = s(f(x) = f(sx) Y .

Hence there is a module X Y with U(X) = Y and a morphism f : X X with U(f) = h. Since h is surjjecive, f is an epimorphism. Also

U(f) = hȷU(Y ) Y = U(f)ȷU(Y ) U(X) = U(fȷY X).

Since U is faithful, f = fȷYX . Thus f = fȷYX is a canonical factorization of f. Let f = gȷYX be any canonical factorization of f. Then

U(f)ȷU(Y ) U(X) = U(f) = U(g)ȷU(Y ) U(X).

Since the left-hand side of equation above is the image factoprization of U(f) in Set, we have U(X) U(X). Since X and X are submodules of Y we must have X X. Therefore X = imf by the definition of images (see Equation 2.2). Moreover, we have

U(imf) = U(X) = imU(f).

If 𝒞ModS is a small subcategory with images then it is clear that 𝒞 satisfy axioms (Sb:b) and (Sb:c) with respect to U0 = U𝒞. Furthermore if there exists a largest module Z for the partially ordered set 𝖛𝒞 then by Equation (10) 𝖛U0 is an orderembedding of 𝖛𝒞 into 𝖛Set and so, (𝒞,U0) satisfies axiom (Sb:a). Therefore 𝒞 is ab 𝒮-category withrespect to U0. This completes the proof.

Evidently S1 (see Clifford and Preston [1961], page 4), the semigroup obtained by adjoining identity to S is a left [as well as right] S-module where the action is the multiplication by elements of S on the left [right] of elements in S1 . Similarly any left ideal Λ [righti deal I] of S is a left [right] S-module and is a proper submodule of S1. Let 𝕃(S) denote the full subcategory of ModS whose objects are left ideals of S1; that is, if X 𝖛𝕃(S) then either X = S1 or X is a left ideal in S. Morphisms in this category are called right translations. It may be observed that, while S1 is a faithful left (as well as right) module, a left [right] ideal may not be faithful as a module. For a S, the map ρa : ssa is a morphism in 𝕃(S) from S1 onto the principal left ideal L(a) = S1a. This map ρa is called the principal right translation of S determined by a.

Since, for every X 𝖛𝕃(S), the action of S on X is the restriction of the product in S to S × U(X), 𝖛U is an order embedding of 𝖛𝕃(S) into 𝖛Set. Thus the natural foregetful functor U : 𝕃(S) Set is an embedding of 𝕃(S) into Set. Therefore the pair (𝕃(S),U) satisfies axiom (Sb:a) of Definition 3.1. Axioms (Sb:b) and (Sb:c) hold by Proposition 4.1. Thus we have:

Proposition 4.2. Let 𝕃(S) denote the full subcategory of ModS whose objects are left ideals of S1 and let U : 𝕃(S) Set be the natural forgetful functor. Then 𝕃(S) is an 𝒮-category with respect to U.

In the following, for brevity, we shall avoid explicit reference to the forgetful functor U as far as possible. Instead, we shall assume that the given 𝒮-category has been identified with its image in Set (see Proposition 3.3) so that U reduces to the inclusion.

We can dualise the definitions and results above. Thus a right S module is a set on which S acts on the right in the usual sense. A morphism f : X Y of right modules is a mapping that preserve rightr actions. There is a category ModS in which objects are right S-modules and morphisms are morphisms of right modules. We can define a subobject relation in ModS by dualising Equation (10). The category ModS with subobjects so defined has images and the foregetful functor U : ModS Set (that forgets right action) preserve subobjects and images of morphisms.

Remark 4.1: Notice that a left S-module X uniquely determines a dual (contravariant) representation λ : sλs where λs denote the map on the set U(X) defined by

λs(x) = s.x for allx X.

Conversely every contravariant representation of S by functions on the set X uniquely determines a left S-module structure on X. Let X and Y be left S-modules giving representations ρ and λ respectively. A map α : X Y is a morphism of S-modules if and only if α commutes with ρ and λ:

λ α = α ρ.

Thus in the following we may replace modules by representations and mprphisms by maps commuting with representations. We shall say a representation ρ (by functions on a set X) or the associated module Mρ is faithful if the function ρ : S 𝒯X is injective. Dual remarks are valid for right modules of S and representationys by functions on a set.

For each a S, L(a) = S1a and R(a) = aS1 are principal left and right ideals of S respectively. Notice that the natural left action of S on L(a) can be extended to an action of S1 so that L(a) is a left S1-set which is generated by a and thus L(a) is a cyclic S1-set. [see E. Krishnan2000, §2.5,2.6]. A morphism ρ : L(a) L(b) of left ideals is a mapping which satisfies the condition: there exists s S1 such that uρ = us for all u L(a). Thus ρ is the restriction of the right translation ρs : xxs to L(a) and is a morphism of the corresponding left S1-set. We shall refer to a triplet (a,s,b) S × S1 × S as left-admissible triple in S if it satisfy the condition

as = tb for somet S1. (11)

Given a left admissible triple (a,s,b), it is clear that the map

ρ(a,s,b) = ρsL(a) : uus (12)

is a morphism in 𝕃(S) induced by the principasl right translation ρs. Conversely every morphism ρ : L(a) L(b) 𝕃(S) between principal ideals in S indused by a principal right translation ρs determines a left admissible triple (a,s,b) such that ρ = ρ(a,s,b). Notice that these representations of certain morphisms in 𝕃(S) as left-admissible triples are not, in general, unique. In fact, we can see that

ρ(a,s,b) = ρ(a,s,b)L(a) = L(a),L(b) = L(b) andas = as( oras = as). (13)

Proposition 4.3 (The category of principal left ideals). To every semigroup S there corresponds a category 𝕃0(S) specified as follows:

i)
𝖛𝕃0(S) = {L(a) : a S}.
ii)
𝕃0(S)(L(a),L(b)) = {ρ(a,s,b) = ρsL(a) : s S1 withas L(b)}.
iii)
Composition in 𝕃0(S) is given by the rule
ρ(a,s,b)ρ(c,t,d) = ρ(a,st,d) if L(b) = L(c); undefined  if L(b) L(c).
iv)
ρ(a,s,b) = ȷL(b) L(a) if and only if ρ(a,s,b) = ρ(a,1,b). Moreover {ρ(a,1,b) : a L(b)}is a choice of subobjects in 𝕃0(S) and the foregetful functor U : 𝕃0(S) Set is an embedding category 𝕃0(S) (with choice of subobjects above) into Set.
v)
If ρ(a,s,b) is a morphism in 𝕃0(S), then
imρ(a,s,b) = L(as) andρ(a,s,b) = ρ(a,s,as))ρ(as,1,b) (14)

gives the image-factorization of ρ(a,s,b) in 𝕃0(S).

Furthermore, 𝕃0(S) is an 𝒮-category is an 𝒮-subcategory of 𝕃(S).

Proof. By definition, for each left admissible triplet (a,s,b), the map ρ(a,s,b) = ρsL(a) is a morphism in 𝕃(S). Hence

𝕃0(S) = {𝕃0(S)(L(a),L(b)) : a,b S}

is a subset of the set of morphisms of 𝕃(S). Also morphisms ρ(a,s,b) and ρ(c,t,d) are composable iff L(b) = L(c). In this case, the triplet (a,st,d) is left admissible and we haveith respect to

ρ(a,s,b)ρ(c,t,d) = (ρsL(a))(ρtL(c)) = ρsρtL(a) = ρstL(a) = ρ(a,st,d).

Therefore 𝕃0(S) is closed with respect to composition of maps. It follows that the composition defined by the rule given in iii) coincides with composition of the corresponding maps. Now, for any a S, (a,1,a) is left-admissible and ρ(a,1,a) = 1L(a). Hence 𝕃0(S) is the morphism set of a subcategory of 𝕃(S). Clearly, the vertex set of 𝕃0(S) is the set of all principal left ideals of S.

It is obvious that L(a) L(b) if and only if (a,1,b) is an admissible triple and ρ(a,1,b) = ȷL(b) L(a). Therefore inclusions between vertices of 𝕃0(S) are morphisms of the form ρ(a,1,b) with a,b S. It is easy to verify that the set {ρ(a,1,b) : a L(b)} is a choice of subobjects for 𝕃0(S). Further, for any ρ = ρ(a,s,b), we can see that the function ρ maps the set L(a) onto the set L(as). Therefore L(as) is the image of ρ in 𝕃0(S) as well as in 𝕃(S). Since 𝕃(S) is an 𝒮-category, by Corollary 3.2 𝕃0(S) is an 𝒮-subcategory of 𝕃(S).

Let ϕ : S T be a homomorphism of semigroups. If (a,s,b) is a left admissible triple in S, by Equation (11) as = tb for some t S1. Then aϕ,bϕ T, sϕ,tϕ T1 and (aϕ)(sϕ) = (tϕ)(bϕ). Therefore (aϕ,sϕ,bϕ) is an admissible triple in T. We see that the equations

𝕃0(ϕ) L(a) = L(aϕ) 𝕃0(ϕ) ρ(a,s,b) = ρ(aϕ,sϕ,bϕ) (15)

for each a S and admissible triple (a,s,b) in S is an inclusion-preserving functor 𝕃0(ϕ) : 𝕃0(S) 𝕃0(T). Furthermore the assignments

𝕃0(S) : S𝕃0(S),ϕ𝕃0(ϕ)

is a functor 𝕃0 from the category 𝔖 of semigroups to the category 𝔖𝔠𝔞𝔱 of small 𝒮-categories.

Dually there is a full subcategory (S) ModS of all right ideals of S1 in which morphisms are called left-translations. The morphism

λa : tat;S1 R(a)

in (S) is called the principal left translation of S determined by a.

Dually we can define the category 0(S) of principal right ideals. A morphism in 0(S) from R(a) to R(b), denoted by λ(a,s,b) : R(a) R(b) is a left translation λsR(a) with s S1 and sa R(b). We shall say that a triplet (a,s,b) S × S1 × S satisfying the condition sa R(b) is right admissible. Notice that (a,s,b) is right admissible if and only if sa = bt for some t S1. Moreover all definition and results for 𝕃0(S) holds for 0(S) with appropriate modifications. Proof of the following statement (the dual of Proposition 4.3) is obtained by dualising (that is, replacing left ideals by right ideals, left translations by right translations, etc.) the proof of Proposition 4.3.

Proposition 4.4 (The category of principal right ideals). To every semigroup S there corresponds a category 0(S) specified as follows:

i)
𝖛0(S) = {R(a) : a S}.
ii)
0(S)(R(a),R(b)) = {λ(a,s,b) = λsR(a) : s S1 withsa R(b)}.
iii)
Composition in 0(S) is given by the rule
λ(a,s,b)λ(c,t,d) = λ(a,ts,d) if R(b) = R(c); undefined  if R(b) R(c).
iv)
λ(a,s,b) = ȷR(b) R(a) if and only if λ(a,s,b) = λ(a,1,b). Moreover {λ(a,1,b) : a R(b)}is a choice of subjects in 0(S) and 0(S) is a category with subobjects in which the set-theoretic inclusions among vertexes in 0(S) is the subobject relation.
v)
If λ(a,s,b) is a morphism in 0(S), then
imλ(a,s,b) = R(sa) andλ(a,s,b) = λ(a,s,sa))λ(sa,1,b)

gives the image-factorization of λ(a,s,b) in 0(S). Furthermore, 0(S) is an 𝒮-category and is a subcategory of (S).

The proposition Proposition 4.3 shows that there are close relations between the structure of the category 𝕃0(S) and the semigroup S. For example, the following observations are useful:

Corollary 4.5. Let S be a semoigroup and let ρ = ρ(a,s,b) : L(a) L(b) be a morphism in 𝕃0(S). Then we have

(a)
ρ is surjective if and only if as b;
(b)
ρ is a split injection if and only if there exists s S1 such that (b,s,a) is left admissible and ass = a.
(c)
ρ is a split surjection if and only if if there exists s S1 such that (b,s,a) is left admissible and bss = b.
(d)
ρ is an isomorphism if only if a as b.

Statements dual to the above are valid for morphisms in 0(S).

Proof. The statement (a) is an immediate consequence of the definition of ρ(a,s,b). To prove (b), assume that an element s S1 exists satisfying the requirements. Then σ : ρ(b,s,a) is a morphism in 𝕃0(S) such that

ρσ = ρ(a,s,b)ρ(b,s,a) = ρ(a,ss,a)

by Proposition 4.3. Since ass = a, by (13) we have ρ(a,ss,a) = ρ(a,1,a) = 1L(a). Therefore ρ is a split monomorphism. Conversely if ρ = ρ(a,s,b) is a split monomorphism, then there is σ = ρ(b,s,a) : L(b) L(a) such that ρσ = 1L(a). (b,s,a) is left admissible and the fact that ρ is a split monomorphism gives

ρσ = ρ(a,s,b)ρ(b,s,a) = ρ(a,ss,a) = ρ(b,1,b)

and so, ass = a. Consequently, the element s above satisfies the requirements.

To prove (c), assume that ρ = ρ(a,s,b) : L(a) L(b) is a split epimorphoism so that σρ = 1L(b) for some morphism σ = ρ(b,s,a). Then by the definition of composition in 𝕃0(S) σρ = ρ(b,ss,b) and by (13), we have

σρ = ρ(b,ss,b) = ρ(b,1,b)

This gives bss = b. On the other hand, if s exist satisfying the requirements in (c), then σ = ρ(b,s,a) is a morphism in 𝕃0(S) from L(b) to L(a) and since bss = b, by (13),

σρ = ρ(b,ss,b)ρ(b,1,b) = 1L(b).

Hence ρ is a split epimorphism. The statement (d) follows from (b) and (c).

Remark 4.2: Notice that 𝕃0(S) is a proper subcategory of the category ModS of all left S1-systems. In general, 𝕃0(S) may not be a full subcategory of ModS. Another category which may be of interest in this connection is the full subcategory 𝕃(S) of ModS whose objects are left ideals of S. We have

𝕃0(S) 𝕃(S) ModS.

Suppose that ρ : L(a) L(b) is a morphism in 𝕃(S). If a is a regular element in S there is an idempotent e a. Let s = eρ. Then s = es L(b) and (e,s,b) is anadmissible triple. By Equations (12) and (13) ρ = ρ(e,s,b) = ρ(a,s,b). It follows that the inclusion 𝕃0(S) 𝕃(S) is fully-faithful if S is regular. Dual remarks hold for the category 0(S) of all principal right ideals, the category (S) of all right ideals and the category ModS of all right S1-systems.