up | next | prev | ptail | tail |
Define
(2.2) |
and
Let
(2.3) |
denote a family of partial transformation of such that for each , and are idempotent mappings of and respectively onto . Then is a biordered set if , and satisfy the following axioms.
(2.4) |
Then there exists a subset of such that for every , is universally maximal in . Further imply
(2.5) |
where and .
In a biordered set , we call the quasi-order the right (left) quasi-order, the partial transformations , , the partial right (left) translation (or -translation) corresponding to and the set , the sandwich set for and .
When there is no confusion we denote the biordered set by . Also if and are two biordered sets, for convenience we denote quasi-orders, translations etc. by the same symbols if the context makes the meaning clear. Thus means and are -related in if and they are so related (under this order) in if . We adopt the same convention to translations, sandwich sets etc.
As observed earlier, axioms for biordered sets include as well as their duals. The dual of can be obtained by interchanging and in . It may be noted that axiom is essentially self dual.
By (2.1), is a partial order so that
and consequently
Equality of two partial transformations and of is to be understood in the sense of equality of the associated subsets of . Since this idea is used frequently in this work, we state the necessary and sufficient condition for this equality explicitly as follows.
(2.6) |
Remark 2.1: The structure of a biordered set depends not only its quasi-orders, but also on the set of partial translations. For, it is possible to have two different biordered sets with the same underlying set and the same left and right quasi-orders. This is shown by the following example.
Example 2.1: Let . Define and where . Then and are quasi-orders on such that and
where . Then and are quasi-orders on such that and
It is clear that , and . Let denote the idempotent mappings of onto such that . Define by for . Further let denote identify map on . If
then is a biordered set. Further if denotes another idempotent mapping of onto such that and if
then is also a biordered set. These two differ only with respect to translations.
Remark 2.2: Let be a semilattice. If
(where each denotes identity map on the -ideal of ), then is a biordered set. Since for each , and are identity maps, axioms and their duals are obviously satisfied. Axiom (and ) is satsified with where denotes the product of and in and is satsified with .
On the other hand if a biordered set satsifies the condition that
(2.7) |
then must coincide with defined above. For , , implies that
and is the greatest member of . Thus is a semilattice.
We say that the biordered set satisfying the condition (2.7) is a semilattice. In the following we do not distinguish between the biordered set and the semilattice .
It may be noted that in a biordered set one or more of the quasi-ordered sets , or (but not all) may be semilattices without the biordered set satisfying condition (2.7). This is shown by an example given below. Clearly if is a semilattice, .
Example 2.2: Let be a biordered set and let 1 denote a symbol not representing any element of . Let . Extend and to by defining
for every , and by defining
With these extended quasi-orders and translations, becomes a biordered set.
We say that is the identity (the zero) of the biordered set if and only if for every
Let stands for if has identity, and it denotes the biordered set constructed above if has no identity. Similarly stands for if has a zero. If not, it denotes the biordered set obtained by adjoining to , where , for every .
Example 2.3: Let be a semilattice and denote arbitrary non-null sets. Let and define and in as follows.
Then and are quasi-orders such that
Also
and
Define and by
and
Then it is clear that and are idempotent mappings of and respectively onto . Let
Then is biordered set.
Axioms and their duals are easy to verify. Now
Hence for any we have ,
and consequently
If , then from the fact that and are universally maximal in , we have , and
If , then
and
From this axiom follows, Axiom may be verified similarly.
If then hypothesis of implies that
and hence axiom is satisfied. (In this case, of axiom coincide.)
Note: In this example
and
Hence unless and
Thus in a biordered set, in general, .
Example 2.4: Let denote the biordered set constructed in 2.3. Then if both and contain more than one element, and if contains only one element, .
Let denote the biordered set obtained by adjoining to . Then is a semilattice.
When both and contain more than one element, this gives an example of a biordered set such that and is a semilattice.
Following are certain consequence of the definition of biordered sets. ( stands for a biordered set, , etc. are its elements.)
Proof. If by . Conversely if , and by . Hence by (2.1)
□
Proof. Let . Then by and Lemma 2.3* we have
and thus, . Dually we can show that . □
Proof. In the first case, if
so that . Since , by , . Hence
and therefore is universally maximal in .
In the second case, we have by that and hence
Thus and by , . It now follows that is universally maximal in . □
Proof. For , by axioms and ,
If
then
and
Cosequently is a bijection of onto and
Now let
Then and hence by
Dually we also have . Thus is order preserving and so also is . Consequently is an order isomorphism.
The equality follows from the fact that order isomorphisms preserve universally maximal elements. □
Lemma 2.7. Let be as in . Then we have the following:
Furthermore,
In particular if , there exist and in such that
Proof. Statement (a) follows from (a) and (c) of and statement (b) from (a) of this lemma and (a) of . From statement (b) of this lemma, , and we have
and
We now obtain from (b) of , the statement (c)(i). The second relation in (c) can be proved dually.
(d) If , clearly, and . Conversely assume that . The fact that and together give . In other words and consequently .
If we assume that , we arrive at the same conclusion.
(e) If
Thus and satisfy the hypothesis of , and therefore there exist and in such that
where . By statement (c) of this lemma, we see that and since by (d), ; which proves (e). □
Definition 2.2. A mapping of a biordered set into biordered set is a bimorphism if it satisfies the following axioms and their duals.
A bimorphism of in to is a retraction (co-retraction) if it satisfies the following axiom.
is an isomorphism if and only if is both a retraction and a co-retraction. (The fact that is an isomorphism of onto may be expressed by writing .)
Identity map of is evidently an isomorphism. Also all retractions (co-retractions) are surjections (injections) and therefore isomorphisms are bijections. Axiom gives that if is a retraction (co-retraction), the bimorphism associated with is a co-retraction (retraction). If is a one-to-one retraction (onto co-retraction), we can see that so that . Hence in this case both and are isomorphisms and clearly this holds for all isomorphisms.
It is not difficult to verify that if
are bimorphism then
is also a bimorphism.
For, let . Then
and
Thus satisfy is verified dually.
Now let . Thus by , and . That is . Thus satisfies also.
Definition 2.3. Let be a biordered set and . Define
(2.9) |
Then is a biordered subset of if the following conditions are satisfied.
Clearly is a biordered subset of itself. Axiom () implies in particular that for each , and are partial transformations of so that
(2.10) |
and that . Clearly . Hence by (), . If and denote the sets defined by (2.4) for and respectively, then by (2.10), . If and are members of and therefore
Since is universally maximal in it must be universally maximal in also. Thus or
(2.11) |
This, together with (2.5) (of axiom ()), shows that for all in such that ,
(2.12) |
Lemma 2.8. Let , where is a biordered set. Let and be defined by (2.9) and let be a family of partial transformations of such that is a biordered set. Then is a biordered subset of if and only if the identity mapping of into is a bimorphism.
If is biordered subset of and if is a bimorphism, then is bimorphism of into . Further, if is a biordered subset of it is also a biordered subset of .
Proof. Let be a biordered subset of . Then definitions of , , and axiom together imply that the mapping of into satisfies axiom of Definition 2.2. Axiom () is the same as axiom with in plane of . Hence is a bimorphism.
On the other hand if is a biordered set such that and are quasi-orders defined by (2.9) and is a bimorphism, then axiom shows that for every , and
Hence is the same as the family defined by (2.9) and as a consequence satisfies . Statement follows from axiom . Therefore is a biordered subset of .
The remaining statements of Lemma 2.8 are consequences of the fact that if and are bimorphism, then is a bimorphism of into . □
Lemma 2.9. Let be a subset of a biordered set . Then is a biordered subset if and only if the following conditions and their duals are satisfied.
Proof. If is biordered subset, (a) and (c) are the same as (2.10) and (2.12). (b) follows from (2.11).
Now suppose satisfied (a), (b) and (c). Define and by (2.9). Then by (a), is a family of partial transformations of . The fact that are satisfied axioms , , and their duals follows from the definitions of and . For , let and denote quasi-ordered sets (Definition 2.4) of and respectively. Then and by (b), at least one universally maximal member of belongs to . Since the quasi-order of is the restriction of to the subset of the set of universally maximal members of is non-empty and these members are universally maximal in also. Thus is universally maximal in if and only if . Now if and then and therefore
□
Dual statement of (2.5) can be verified similarly. Hence satisfies axioms and . To assert that satisfies , we have to show that satisfies axioms also. Let where and
Then and satisfy this condition in also and therefore by there exist and in satisfying (a), (b) and (c) of . Now it is enough to establish that .
Since , . Choose and suppose that . Then by (a), and as a consequence of Lemmass 2.5 and 2.6,
Hence and therefore . By Lemma 2.7(a) and we can prove similarly that there exists such that . But since , statement (b) implies that
and consequently . Thus and . This leads to the conclusion that . Dually we can show that . Thus we observe that is a biordered set and that it satisfies . For , is the sandwich set of and in and therefore is also satisfied.
Proof. Given , for all , . Further given , if , then . Hence condition (a) of Lemma 2.9 is satisfied. If and then and consequently . Thus . Hence satisfies (b) of Lemma 2.9 also. Statement (c) is obvious, since for where , . This proves that is a biordered subset. □
Similarly using Lemma 2.9 we can prove that and are also biordered subsets of .
Theorem 2.11. Let be a biordered set. Then for , is a bimorphism of onto and if ,
(2.13) |
is an isomorphism of onto .
Proof. Let . Then axioms and show that preserves and . Let . Then and hence by .
Also and therefore . Hence by axiom we get
and hence satisfies axiom of Definition 2.2. Axiom follows from axiom and axiom follows from (2.5).
If , it can be similarly proved that is a bimoprhism of into and is a bimorphism of into . But for by axiom ,
and similarly we can prove that
Hence by the definition of isomorphism, is an isomorphism of onto . □
NOTE. clearly is identity map . Also we get by axiom that for all ,
(2.16) |
Dually is also and the two symbols and may be used according to convenience.
Remark 2.12: We have already observed that (Remark 2.2) the concept of a biordered set is a generalisation of the concept of a semilattice. It can be verified without difficulty that a subset of a semilattice is a subsemilattice of if and only if it is a biordered subset of . Further a mapping of semilattice into semilattice is a homomorphism of into (i.e. a mapping such that for all ) if and only if it is a bimorphism of into .
up | next | prev | ptail | top |