Find the mechanism before the formula
Nielsen theory counts essential fixed-point classes of maps. Reidemeister theory reaches behind those classes to the induced endomorphism of a group, where ordinary conjugacy is replaced by twisted conjugacy. Iterating the endomorphism produces a sequence of Reidemeister numbers, and a zeta function packages their growth when every number is finite.
We will compute twisted classes for multiplication on the integers, derive the Reidemeister zeta function from its logarithm, change the multiplier, explain the R-infinity obstruction, and keep the mapping-torus torsion connection inside the theorem that actually permits it.
This guide belongs to the Gold Standard of Topology, the Gold Standard of Education and the How X Works hub.
Choose a reading route
Start with the question closest to the decision you need to make. Every route returns to the same worked evidence and transfer checks.
| Route | Chapters | Use it when |
|---|---|---|
| 1 | 1–3 | You need twisted conjugacy, iterates and the zeta definition. |
| 2 | 4–5 | You want the complete multiplication-by-two calculation. |
| 3 | 6–8 | You need transfer, growth and the Nielsen comparison. |
| 4 | 9–10 | You must compute safely and respect R-infinity and torsion limits. |
| 5 | 11–12 | You want FAQs and a reproducible Reidemeister-zeta audit. |
Chapter contents
Open the group you need, then use the chapter links. Teaching sections remain open on the page.
Build the model
Test the mechanism
1. Twisted conjugacy is the algebraic equivalence relation
For an endomorphism φ:G→G, elements x and y are φ-twisted conjugate when there is g in G with y=gxφ(g)−1. The number of equivalence classes is the Reidemeister number R(φ), which can be finite or infinite.
For additive G=Z, the relation becomes y=x+g−φ(g). If φ(g)=dg, then y−x lies in (1−d)Z. The twisted classes are therefore cosets of (1−d)Z, giving R(φ)=|1−d| when d is not one.
- State the group and endomorphism.
- Define the twisted relation with one convention.
- Check whether the number of classes is finite.
- Repeat the finiteness check for every iterate used.
2. Iteration creates a twisted-conjugacy growth sequence
The nth iterate of multiplication by d is multiplication by dn. Hence R(φn)=|1−dn| whenever this index is finite. The sequence measures algebraic fixed-class growth under iteration, not a count of all geometric periodic points.
For d greater than one the absolute value simplifies to dn−1. For d=1, every iterate has infinitely many twisted classes, so the standard finite Reidemeister zeta function cannot be formed.
3. The zeta function weights every iterate through its logarithm
When all R(φn) are finite, define Rφ(z)=exp(Σ from n≥1 of R(φn)zn/n). The 1/n weight is part of the definition and is what allows determinant, product and torsion formulas in suitable settings.
The ordinary Taylor coefficients of Rφ(z) are not the Reidemeister numbers. To check a proposed closed form, take its logarithm and compare the coefficient of zn with R(φn)/n.
| Object | Built from | Question answered |
|---|---|---|
| R(φ) | φ-twisted classes | How many algebraic fixed classes occur? |
| R(φⁿ) | classes of each iterate | How does class complexity grow? |
| Rφ(z) | weighted exponential series | Can the whole sequence be packaged? |
| torsion value | chain-level mapping-torus data | Does a theorem identify a special value? |
4. Worked example: multiplication by two on the integers
Let φ(n)=2n. Then R(φn)=|1−2n|=2n−1. The first values are 1, 3, 7 and 15. Substitute them into the logarithmic definition.
Using Σanzn/n=−log(1−az), we obtain log Rφ(z)=−log(1−2z)+log(1−z). Exponentiating gives Rφ(z)=(1−z)/(1−2z).
5. Worked check: recover the iterate counts from the rational form
Take the logarithm of (1−z)/(1−2z). Expanding gives Σ(2n−1)zn/n. Multiplying the coefficient of zn by n recovers 1, 3, 7 and 15 for n=1 through four.
The nearest pole is z=1/2, matching exponential growth rate log 2 for the Reidemeister sequence. This analytic growth statement does not by itself prove a count of primitive geometric periodic orbits.
6. Transfer check: change multiplication by two to three
Let ψ(n)=3n. Then R(ψn)=3n−1, with first values 2, 8, 26 and 80. The same calculation gives Rψ(z)=(1−z)/(1−3z).
The pole moves from one half to one third and the exponential growth changes from log 2 to log 3. What transfers is the quotient calculation on Z. It should not be reused for a nonabelian group without computing its twisted classes or invoking an appropriate theorem.
7. Reidemeister and Nielsen counts meet only under hypotheses
A self-map f induces an endomorphism f* on the fundamental group. Reidemeister classes can label fixed-point classes, while the Nielsen number retains the essential classes with nonzero index. The two counts can agree in structured settings, but R(f*) can also count inessential classes that Nielsen theory removes.
For the degree-d circle map with d greater than one, both iterate sequences equal dn−1, so the Reidemeister and Nielsen zeta functions have the same rational form. That agreement is an example, not a definition valid for every space and map.
8. Twisted-conjugacy growth records algebraic iteration
The radius of convergence captures the exponential scale of R(φn) when the sequence is finite. Congruences and product expansions can separate primitive algebraic data in favourable cases. Rationality has been proved for several finite, eventually commutative, nilpotent and infra-solvable settings under stated conditions.
Rationality is not automatic for arbitrary groups. A short list of coefficients can always be fitted by many rational functions. Keep the group theorem, finiteness conditions and exact coefficient check beside the claimed formula.
9. Compute quotient indices before fitting a generating function
For finitely generated abelian groups, represent 1−φn by an integer matrix and use Smith normal form when its cokernel is finite. A zero determinant warns that the cokernel may be infinite. For nonabelian groups, class algorithms need group-specific normal forms and termination guarantees.
Use exact integer arithmetic because dn and matrix determinants grow quickly. Check several iterates independently, then verify the logarithmic series of the closed form. Numerical evaluation near a pole is not a substitute for the exact class count.
10. R-infinity and torsion mark theorem boundaries
- If any R(φⁿ) is infinite, the standard finite zeta series is not defined.
- Reidemeister numbers count twisted classes, not automatically essential fixed-point classes.
- A rational fit from a few iterates is not a rationality theorem.
- A pole describes sequence growth, not every geometric orbit.
- Mapping-torus Reidemeister torsion depends on a specified representation and acyclicity conditions.
- A special-value identity cannot be transferred outside the setting of the theorem.
Classical results connect Reidemeister zeta functions with Reidemeister torsion of suitable mapping-torus constructions, including formulations using induced maps on a unitary dual. The exact space, representation, sign and evaluation point matter. State them rather than saying simply that the zeta function “is” the torsion.
11. Frequently asked questions
Why can R(φ) be infinite?
Twisted conjugacy can have infinitely many equivalence classes. On Z with φ equal to the identity, the relation does not identify distinct integers, so R(φ)=∞.
Is this the same as the Nielsen zeta function?
No. The sequences can agree in settings such as expanding circle maps, but Nielsen numbers retain essential geometric fixed-point classes whereas Reidemeister numbers arise from twisted conjugacy.
Does every rational Reidemeister zeta function yield a torsion formula?
No. Torsion identities require their own hypotheses on the mapping-torus construction, representation and chain complex. Rationality alone is not enough.
12. A reproducible Reidemeister-zeta checklist
| Stage | Evidence to retain |
|---|---|
| Algebra | group, endomorphism and twisted relation |
| Finiteness | R(φⁿ) finite for every required iterate |
| Sequence | exact class counts for several n |
| Function | logarithmic definition and coefficient recovery |
| Transfer | changed endomorphism or group with fresh calculation |
| Claims | separate Nielsen, growth and torsion hypotheses |
The gold standard is to derive the rational form from exact quotient classes, expand its logarithm back to the sequence, and stop at R-infinity or a missing torsion hypothesis. Analytic elegance is useful only while the underlying equivalence relation stays visible.
Reading routes · Previous chapter · Contents · Continue reading
Continue through the Gold Standard ecosystem
Return to the Gold Standard of Topology, the Gold Standard of Education parent, or the How X Works hub.
Read next in this subject
- Nielsen Zeta Functions, Iteration Growth and Entropy Bounds
- Reidemeister Trace, Twisted Conjugacy and Fixed-Point Class Detection
- Nielsen Periodic-Point Theory, Boosting and Essential Reducibility
Authoritative reading
- Fel’shtyn and Hill, the Reidemeister zeta function and its torsion connection
- Dekimpe and Tertooy, existence of the Reidemeister zeta function on groups
- Fel’shtyn and Hill, dynamical zeta functions, Nielsen theory and Reidemeister torsion
- Jiang, Lectures on Nielsen Fixed Point Theory, American Mathematical Society