You are here
Home ›complemented lattice
Primary tabs
complemented lattice
Let be a bounded lattice (with and ), and . A complement of is an element such that
and .
Remark. Complements may not exist. If is a non-trivial chain, then no element (other than and ) has a complement. This also shows that if is a complement of a non-trivial element , then and form an antichain.
An element in a bounded lattice is complemented if it has a complement. A complemented lattice is a bounded lattice in which every element is complemented.
Remarks.
-
In a complemented lattice, there may be more than one complement corresponding to each element. Two elements are said to be related, or perspective if they have a common complement. For example, the following lattice is complemented.
Note that none of the non-trivial elements have unique complements. Any two non-trivial elements are related via the third.
-
If a complemented lattice is a distributive lattice, then is uniquely complemented (in fact, a Boolean lattice). For if and are two complements of , then
Similarly, . So .
Mathematics Subject Classification
06C15 Complemented lattices, orthocomplemented lattices and posets06B05 Structure theory
- Forums
- Planetary Bugs
- HS/Secondary
- University/Tertiary
- Graduate/Advanced
- Industry/Practice
- Research Topics
- LaTeX help
- Math Comptetitions
- Math History
- Math Humor
- PlanetMath Comments
- PlanetMath System Updates and News
- PlanetMath help
- PlanetMath.ORG
- Strategic Communications Development
- The Math Pub
- Testing messages (ignore)
- Other useful stuff
Recent Activity
new question: Sorry to steal a few minutes of your time for this question, but i honestly don't know what else to do. by Whrazithar
new question: equality of the determinants of submatrices of an orthogonal matrix by ismayli
Jun 11
new correction: Typo by suitangi
Jun 2
new question: Creating another set with same cardinality. by hkkass
Jun 1
new image: ProblemOneRevised by unlord
new Education: Chapter II by rspuzio
May 31
new collection: The Calculus by Davis and Brenke by rspuzio
new question: Proofs by weixifan
new question: Summation Integration Question by trevor.nickle
May 27
new correction: typo+finite measure hypothesis by Filipe



Comments
Lattices and power sets
Hi!
A small question:
Which additional properties are necessary/sufficient to make a lattice $(L, <)$ isomorphic to a power set (\Pot(M), \subsetneq)?
Ok, the following properties are necessary:
- complete (every subset has an infimum and a supremum)
- complemented
- distributive
But what set of properties is sufficient?
Re: Lattices and power sets
A complemented distributive lattice is a Boolean algebra. A complete Boolean algebra A is lattice isomorphic to the power set of some set S if A is atomic (every element of A has an atom below it). You can show that S is in fact the set of all atoms of A. Check out the book Intro to Lattices and Order by Davey and Priestley.
Re: Lattices and power sets
Ok, so this would mean, a lattice that is
- complete and
- complemented and
- distributive and
- atomic
is order-isomorphic to a power set. Any such isomorphism will identify the atoms of the lattice with the singletons of the power set.
Are all the four requirements are necessary? Don't have the book at hand...
Re: Lattices and power sets
I think so. I saw a counter-example in the book if, say, the atomic condition is dropped. I'll find out.
Re: Lattices and power sets
I found the counter-example:
Consider the power set of the reals R. Consider the Boolean subalgebra A of R generated by the following intervals
(-oo, a)
[b,c}
[d,oo)
empty set
where a,b,c,d are reals.
Then A is complete and atomless, and A is not lattice isomorphic to any power set of a set.
Re: Lattices and power sets
Hi, thanks, but..
Sorry I don't get it!
Do you mean a < b < c < d ?
Do you mean A \subseteq R or A \subseteq \Pot(R) ?
With (-oo, a), [b, c), [d, oo) \in A ?
or (-oo, a), [b, c), [d, oo) \subseteq A ?
Maybe a sup-generated subset of the power set \Pot(R) ??
Regards, Schneemann.
Re: Lattices and power sets
Sorry for the sloppiness. First, a, b, c, d are arbitrary real numbers, there are no specific orderings on these numbers. Second, the Boolean subalgebra A I am refering to is the set of all finite unions of those intervals that I mentioned. You can show that A is indeed a Boolean algebra under union and intersection and complementation, etc... Furthermore, it is complete. You can derive that A is atomless by using a proof of contradiction.
Re: Lattices and power sets
At the beginning I thought you mean a, b, c, d to be fixed numbers. I'm still not sure, but now I think you mean this set:
S := { (oo,b) | b in RR }
cup { [a, b) | a, b in RR }
cup { [a,oo) | a in RR }
cup { emptyset }
Now A is generated from S by building finite unions.. right? Ok, I'll think about that. Thanks!
Regards. Schneemann