An ultrafilter is a consistent way of deciding every question at once. They are the points of the Stone space, the indices of ultraproducts, and the reason compactness works.
Engineering · Mathematics11 min readKV-MATH-0234
Learning objectives
Define filters and ideals and describe the duality between them.
Distinguish principal from free filters.
State the three equivalent characterisations of an ultrafilter.
Construct the filter generated by a set with the finite intersection property.
Explain why free ultrafilters require a choice principle.
Preview the role of ultrafilters in ultraproducts and Stone duality.
01Filters and ideals
A filter on a Boolean algebra B is a non-empty subset F closed upward and closed under meet; it is proper when 0 ∉ F. An ideal is the order dual.
Filters and ideals as duals
Filter F
Ideal I
Contains
1
0
Closed under
finite meets
finite joins
Closed
upward
downward
Proper when
0 ∉ F
1 ∉ I
Complement map
{ x′ : x ∈ F } is an ideal
{ x′ : x ∈ I } is a filter
Congruence class
the 1-class
the 0-class
Complementation exchanges the two notions exactly, so any theorem about filters has a dual about ideals and neither needs separate proof. The lattice of filters is dually isomorphic to the lattice of ideals, and both are isomorphic to Con B.
02Principal and free filters
The principal filter generated by an element a is the set of elements above a. A filter is free when it is not principal.
Principal
Generated by one element
F = { x : x ≥ a } for some a. On a finite Boolean algebra every filter is principal, because the meet of all its members lies in it.
Free
No least element
The intersection of all members is not itself a member. Requires the algebra to be infinite, and — for ultrafilters — requires a choice principle to exhibit.
The standard example of a free filter is the Fréchet filter on the power set of an infinite set: the cofinite subsets. It is a proper filter, its members have empty intersection, and it is contained in many free ultrafilters — though exhibiting even one of those needs Zorn's lemma.
03Ultrafilters
An ultrafilter is a maximal proper filter. Three characterisations coincide, and each is used in practice.
Key resultThree equivalent definitions
For a proper filter U on a Boolean algebra B, the following are equivalent: (i) U is maximal among proper filters; (ii) for every a ∈ B, exactly one of a and a′ lies in U; (iii) whenever a ∨ b ∈ U, either a ∈ U or b ∈ U.
Maximality ⟹ decisiveness
If neither a nor a′ were in U, adjoining a would give a larger proper filter, contradicting maximality. Both cannot be in U since their meet is 0.
Decisiveness ⟹ primeness
If a ∨ b ∈ U and a ∉ U then a′ ∈ U, so b ≥ (a ∨ b) ∧ a′ is in U by upward closure and meet closure.
Primeness ⟹ maximality
A proper filter properly containing a prime filter would have to contain some a with a′ already inside, forcing 0 into the filter.
The quotient reading
U is an ultrafilter exactly when B/U ≅ 2 — the corresponding congruence is a coatom of Con B. This connects to maximal ideals in the ring picture.
Characterisation (ii) is the one to carry: an ultrafilter decides every question. For every element it commits to either that element or its complement, consistently. That is what makes ultrafilters usable as a device for taking limits.
04Existence and the finite intersection property
A family with the finite intersection property — every finite subfamily has non-zero meet — generates a proper filter, and Zorn's lemma extends it to an ultrafilter.
ProcedureExtending a family to an ultrafilter
in: family with the FIP → out: an ultrafilter containing it
input: family S ⊆ B with the finite intersection property
F₀ := { x ∈ B : x ≥ s₁ ∧ ⋯ ∧ sₙ for some finite s₁,…,sₙ ∈ S }
F₀ is a proper filter, since finite meets from S are non-zero
consider the poset of proper filters containing F₀, ordered by inclusion
every chain has an upper bound: the union, which is proper
(0 lies in no member, so 0 lies in no union)
by Zorn's lemma there is a maximal element U
U is an ultrafilter containing S
Correctness: unions of chains of proper filters are proper, which is the hypothesis Zorn needs. Caveat: the construction is non-constructive. No free ultrafilter on an infinite set can be exhibited explicitly, and their existence is strictly weaker than full choice.
CautionFree ultrafilters cannot be written down
The existence of a free ultrafilter on the natural numbers is not provable in ZF alone. It follows from the Boolean Prime Ideal Theorem, which is strictly weaker than the axiom of choice but not a theorem of ZF. Any argument claiming to construct one explicitly is mistaken, and any theorem relying on one carries that choice principle as a hypothesis.
05Ultrafilters on a set
The most-used case is B = the power set of a set I. Here an ultrafilter is a family of subsets of I deciding, for each subset, whether it is 'large'.
The two kinds of ultrafilter on a set
Kind
Description
Exists?
Principal at i
all subsets containing the fixed point i
always, explicitly
Free
contains all cofinite sets, no finite set
only via BPI; never explicit
On a finite set
principal only
no free ultrafilters exist
The distinction matters immediately for ultraproducts: an ultraproduct over a principal ultrafilter collapses to a single factor and gives nothing new, whereas an ultraproduct over a free ultrafilter is the construction that yields compactness, non-standard models and Jónsson's lemma. Every interesting ultraproduct uses a free ultrafilter.
06Where ultrafilters go next
Stone duality
Points of the dual space
The Stone space of a Boolean algebra has the ultrafilters as its points, topologised so that the algebra is recovered as the clopen sets.
Ultraproducts
Indices for the quotient
The ultraproduct of a family of algebras is the direct product modulo the congruence determined by an ultrafilter on the index set. Łoś's theorem transfers first-order properties.
Compactness
The proof mechanism
The compactness theorem for first-order logic follows from the ultraproduct construction applied to a family of models of finite subsets of a theory.
Jónsson's lemma
Locating subdirect irreducibles
Ultraproducts appear in the statement precisely because they capture 'approximable by members of the generating class'.
Boolean products
Sheaf-like representations
The stalks of a Boolean product are indexed by the Stone space, hence by ultrafilters.
Boolean prime ideal theorem
The choice principle
The next page treats BPI itself, its equivalents and its strength relative to choice.
Frequently asked
Is every filter contained in an ultrafilter?
Every proper filter is, by the Zorn's lemma argument. The improper filter — the whole algebra — is not, since ultrafilters are proper by definition. The extension result is exactly the Boolean Prime Ideal Theorem in filter form.
Why does an ultraproduct over a principal ultrafilter collapse?
Because the ultrafilter concentrates all its attention on a single index. Two elements of the product are identified when they agree on a member of the ultrafilter, and every member contains the distinguished point, so agreement at that one coordinate suffices. The ultraproduct is therefore isomorphic to that single factor.
Do ultrafilters exist on every Boolean algebra?
Proper ultrafilters exist on every non-trivial Boolean algebra, by extending the principal filter of any non-zero element. On the one-element algebra there are none, since every filter contains 0. The interesting question is whether free ultrafilters exist, which requires the algebra to be infinite and requires BPI.
Sources and further reading
S. Burris and H. P. Sankappanavar, A Course in Universal Algebra, Millennium Edition (a corrected re-typesetting of Springer GTM 78, 1981).
G. Grätzer, Universal Algebra, 2nd edition, Springer.
R. McKenzie, G. McNulty and W. Taylor, Algebras, Lattices, Varieties, Volume I.
Original KEVOS® explanatory article. Written from the topic map of the cited works; no text is reproduced from them.
Handbook application: from concept to controlled practice
Purpose. This expanded section turns the original page into a practical handbook. It preserves the supplied material and adds a repeatable way to apply, check and review Filters, Ideals and Ultrafilters. It does not replace a contract, legislation, a controlled standard, competent engineering judgement or specialist advice.
The operating aim is to turn a compact mathematical statement into a usable chain of definitions, claims, examples and checks. Read the original explanation first, then use the workflow and checks below to convert knowledge into evidence.
Treat Filters, Ideals and Ultrafilters as a network of definitions and implications, not as a list of formulas. The working vocabulary on this page—ultrafilters, filters, ideals, algebra, principal—should be made explicit before any proof or computation begins. Record the ambient set or structure, the permitted operations and the equality or equivalence relation in use. A compact theorem often changes meaning when the base field, finiteness condition, commutativity assumption or direction of an action changes.
For a proof, write the hypotheses as a checklist and mark the line at which each one is used. For a computation, state the representation of the input, the arithmetic model, the termination condition and the output invariant. For a classification problem, distinguish existence from uniqueness and distinguish an object from its representation. These separations prevent a correct local calculation from being mistaken for the general result.
A useful worked example should be small enough to inspect completely but rich enough to exercise the main mechanism. Compute the result in two ways where practical: symbolically and by substitution, structurally and numerically, or directly and through a normal form. Then include one near-miss example in which a hypothesis fails. The contrast explains why the theorem is shaped as it is and gives the reader a diagnostic pattern for later problems.
Verification is part of the mathematics. Check domains and codomains, substitute proposed solutions, test identity and zero cases, compare dimensions or cardinalities, and confirm that maps respect the required operations. In numerical work, report precision, conditioning and a residual rather than digits alone. In algorithmic work, separate mathematical correctness from implementation complexity and resource limits.
Step-by-step operating method
Fix the setting. State the objects, ambient structure, notation and assumptions before manipulating symbols.
Separate claims. Distinguish definitions, hypotheses, conclusions, equivalent conditions and consequences.
Choose a method. Select proof, construction, calculation or algorithm according to the question actually asked.
Work a small case. Use the smallest non-trivial example to expose the mechanism and test edge behaviour.
Verify independently. Substitute back, check invariants, test boundary cases or use an alternative derivation.
Worked-example protocol
Illustrative method—not a source theorem. Start with a small admissible input and list the definitions it must satisfy. Carry out each transformation on a separate line, citing the property that permits it. Preserve exact values until approximation is necessary. At the end, verify the output against the original definition and one invariant such as dimension, degree, determinant, order, norm or residual. Then alter one hypothesis and observe which step ceases to be valid. This protocol creates a reusable example without inventing a theorem-specific numerical answer.
Stage
Record
Quality check
Input
Objects, domain, notation, assumptions
Every symbol is defined
Method
Permitted operation or cited result at each step
All hypotheses hold
Output
Exact result and representation
Correct type, domain and form
Verification
Substitution, invariant or alternative derivation
Independent agreement
Boundary test
Zero, identity, degenerate or failed hypothesis
Scope is understood
Common failure modes and recovery actions
1. Watch for
Using a theorem without checking every hypothesis.
Recovery: Return to the governing definition or requirement and restate the decision in one sentence.
2. Watch for
Treating a suggestive example as a proof of the general case.
Recovery: Separate evidence from assumption, assign an owner and set a date for validation.
3. Watch for
Changing notation or conventions part-way through an argument.
Recovery: Run a small counterexample, boundary test, pilot or independent check before proceeding.
4. Watch for
Hiding a division-by-zero, convergence, finiteness or commutativity assumption.
Recovery: Record the consequence, decision and rationale, then update the controlled baseline.
5. Watch for
Reporting a computed result without a residual, substitution or structural check.
Recovery: Escalate when the issue affects safety, compliance, acceptance, material value or an agreed tolerance.
Review checklist
Can every symbol be traced to a definition or prior result?
Which hypothesis does each major step use?
Does the method cover zero, identity, degenerate and boundary cases?
Can the conclusion be checked by a second representation or calculation?
Are mandatory requirements distinguished from recommendations and illustrative values?
Are sources, assumptions, units, dates and versions recorded closely enough to reproduce the decision?
Have safety, legal, ethical, stakeholder and operational consequences been considered at the appropriate level?
Is there a named owner and a trigger for review, escalation, change or retirement?
Questions for deeper application
What is the most important distinction a practitioner must preserve when applying Filters, Ideals and Ultrafilters?
Answer with a fact or cited source where available. Where evidence is incomplete, record the assumption, consequence, responsible owner and next validation action.
Which assumption about ultrafilters would change the result most if it proved false?
Answer with a fact or cited source where available. Where evidence is incomplete, record the assumption, consequence, responsible owner and next validation action.
What evidence would allow an independent reviewer to reproduce or challenge the conclusion?
Answer with a fact or cited source where available. Where evidence is incomplete, record the assumption, consequence, responsible owner and next validation action.
Which boundary, exception or failure case has not yet been tested?
Answer with a fact or cited source where available. Where evidence is incomplete, record the assumption, consequence, responsible owner and next validation action.
What must be handed over, monitored or reviewed after the immediate work is complete?
Answer with a fact or cited source where available. Where evidence is incomplete, record the assumption, consequence, responsible owner and next validation action.
Authoritative references and use notes
The sources below were selected as institutional or primary guidance for the broader practice. They support the handbook method; they do not imply that every statement or clause in a source applies to every project. Confirm the current edition, jurisdiction, contract and application before treating any requirement as mandatory.
MIT OpenCourseWare — Algebra I — Massachusetts Institute of Technology. Used for groups, vector spaces, linear transformations and linear groups. Accessed 2026-08-13.
The Stacks Project — table of contents — The Stacks Project. Used for commutative algebra, homological algebra, modules and derived categories. Accessed 2026-08-13.
