<!-- Generated from https://www.teclamairme.com/#group-T20 by readviews.py; never edited by hand. -->

# Glossary cards — Chapter ᚓ(E): The cutting

*Teclam Áirme, the Mathematical Compendium — a copy for AI readers of https://www.teclamairme.com/#group-T20, as of 2026-09-30. Links are relative to https://www.teclamairme.com/fortheaireaders/.*

[Guide for AI readers](index.md) · [The chapter](ch20.md)

*Measure theory and dynamics: the Cantor set, Lebesgue measure, Hausdorff dimension, a paradox, and the logistic map at rest. Each card: its symbol, type, one-line headline, the longer note, the reference entries it cites and the cards it links to.*

### the letter Eadhadh — `ᚓ` {#g-og-edad}

*Ogham letter*

Ogham letter name of unknown meaning, paired with Iodhadh; the medieval glossators read it as the aspen.

Edad is of unknown meaning, and it is one of a pair: edad and idad look made to match each other. The three Old Irish word-oghams, the Bríatharogaim ascribed to Morann mac Moín, to Mac ind Óc, and to Cú Chulainn, give érgnaid fid, “discerning tree”; commaín carat, “exchange of friends”; and bráthair bethi, “brother of birch,” the last uncertain. The Auraicept glosses crand fir no crithach, test-tree or aspen, and reads the name itself as ed uath, “horrible grief”—which is glossing by sound, not by evidence. Modern tree lore and ogham divination descend from the medieval plant glosses by way of Robert Graves’s The White Goddess of 1948, which read the letter order as an ancient seasonal calendar of tree magic. Celtic scholarship rejects that reconstruction; the associations in circulation today are modern, not ancient.

Cites: [2.15](ch02-refs.md#r-2.15), [2.16](ch02-refs.md#r-2.16), [2.17](ch02-refs.md#r-2.17), [2.19](ch02-refs.md#r-2.19)

### measure of the Cantor set — `λ(𝒞)` {#g-lamC}

*Integer or rational*

Zero in Lebesgue measure, even though the set has as many points as the whole line.

The Lebesgue measure of the Cantor set is 0: after n steps the set lies inside 2^n intervals of total length (2/3)^n, which tends to 0. Exactly 0, an integer, despite the set having the cardinality of the continuum. It is the first of the three zeros in the numerator.

Cites: [20.1](ch20-refs.md#r-20.1), [20.2](ch20-refs.md#r-20.2)

See also: [Cantor set](ch20-cards.md#g-cantor), [cardinality of the continuum](ch27-cards.md#g-cont), [Lebesgue measure](ch20-cards.md#g-lebesgue)

### Hausdorff dimension of the Cantor set — `dim_{H} 𝒞` {#g-dimH}

*Proven transcendental*

ln 2 / ln 3, about 0.6309, a transcendental dimension.

Hausdorff introduced fractional dimension in 1918; for a set made of m copies of itself scaled by 1/r the dimension is ln m / ln r, so the Cantor set has dimension ln 2 / ln 3 = log₃ 2 ≈ 0.6309. It is irrational because no power of 2 is a power of 3, and then transcendental by the Gelfond–Schneider theorem (1934), since 3 raised to an algebraic irrational cannot equal 2.

Cites: [20.7](ch20-refs.md#r-20.7)

See also: [Cantor set](ch20-cards.md#g-cantor), [the Gelfond–Schneider theorem](ch04-cards.md#g-gelfschn)

### the removed length — `Σ 2^{n−1}/3^{n}` {#g-removed}

*Integer or rational*

The middle thirds sum to exactly 1, which is why nothing is left to weigh.

The lengths removed at each step, 1/3 + 2/9 + 4/27 + … = Σ 2^(n−1)/3^n, form a geometric series with sum exactly 1. The whole length of the interval is gone and uncountably many points remain. Exactly 1, an integer.

Cites: [20.2](ch20-refs.md#r-20.2)

### the Cantor function at one half — `c(½)` {#g-cantorfn}

*Integer or rational*

The devil’s staircase, flat almost everywhere in Lebesgue measure, reaches ½ at ½.

The Cantor function, introduced by Cantor in 1884 and nicknamed the devil’s staircase: continuous, non-decreasing from 0 to 1, constant on every removed interval, so its derivative is zero almost everywhere. By symmetry c(½) = ½ exactly. Exactly ½, an integer-or-rational.

Cites: [20.20](ch20-refs.md#r-20.20), [20.6](ch20-refs.md#r-20.6)

See also: [almost everywhere](ch20-cards.md#g-aeverywhere), [Lebesgue measure](ch20-cards.md#g-lebesgue)

### the measure of the rationals — `λ(ℚ ∩ [0, 1])` {#g-lamQ}

*Integer or rational*

Exactly 0: a countable set can be covered by intervals of total length as small as you like, so the rationals weigh nothing.

Cover the n-th rational by an interval of length ε/2ⁿ; the total is ε, and ε was arbitrary, so the Lebesgue measure of the rationals in [0, 1] is 0 even though they are dense. Countable sets are null: the reason almost every real number is irrational, transcendental and, in the sense of Chapter ᚏ(R), undefinable. Exactly 0, an integer.

Cites: [20.22](ch20-refs.md#r-20.22), [20.5](ch20-refs.md#r-20.5)

See also: [dense and nowhere dense](ch20-cards.md#g-dense), [Lebesgue measure](ch20-cards.md#g-lebesgue)

### the derivative of the Cantor function at one half — `c′(½)` {#g-cantorderiv}

*Integer or rational*

Exactly 0: the devil’s staircase is flat across the whole middle third (⅓, ⅔), where it sits at height ½, so its slope at ½ is 0; it is flat almost everywhere in Lebesgue measure, yet it climbs from 0 to 1.

The Cantor function is constant on each removed middle third, so its derivative is 0 on a set of Lebesgue measure 1; it rises only on the Cantor set, which has Lebesgue measure 0. One half lies inside the very first removed interval, (⅓, ⅔), where the staircase is flat at height ½, so c′(½) is exactly 0 and not merely 0 almost everywhere. The function is not absolutely continuous, which is the hypothesis the fundamental theorem of calculus needs if a function’s whole rise is to be recovered by integrating its derivative. Cantor introduced the function in 1884 and reported Scheeffer’s point that it is a counterexample to an extension of the fundamental theorem of calculus claimed by Harnack; Scheeffer (1884), Lebesgue (1904), and Vitali (1905) discussed and popularized it. Exactly 0, an integer.

Cites: [20.19](ch20-refs.md#r-20.19), [20.20](ch20-refs.md#r-20.20), [20.6](ch20-refs.md#r-20.6)

See also: [absolute continuity](ch20-cards.md#g-abscont), [almost everywhere](ch20-cards.md#g-aeverywhere), [Cantor set](ch20-cards.md#g-cantor), [the Cantor function at one half](ch20-cards.md#g-cantorfn), [the fundamental theorem of calculus, for x²](ch11-cards.md#g-ftc), [Lebesgue measure](ch20-cards.md#g-lebesgue)

### the fat Cantor set — `λ(SVC)` {#g-fatcantor}

*Integer or rational*

Exactly ½: remove middle quarters, then sixteenths, and so on, and what is left is still nowhere dense but weighs half the interval.

The Smith–Volterra–Cantor set (Smith 1874, Volterra 1881, Cantor) removes from [0, 1] the middle 1/4, then an interval of length 1/16 from the middle of each remaining piece, then 1/64, and so on, so the removed lengths sum to ½ and the remainder has Lebesgue measure ½ while containing no interval at all. It shows that measure and topological size are different things. Exactly ½, an integer-or-rational.

Cites: [20.24](ch20-refs.md#r-20.24), [20.8](ch20-refs.md#r-20.8)

See also: [Cantor set](ch20-cards.md#g-cantor), [dense and nowhere dense](ch20-cards.md#g-dense), [Lebesgue measure](ch20-cards.md#g-lebesgue)

### the dimension of the Koch curve — `dim_{H} Koch` {#g-koch}

*Proven transcendental*

ln 4 / ln 3, about 1.2619: four copies at one third the size, transcendental by Gelfond–Schneider.

Helge von Koch’s 1904 curve, a continuous curve with no tangent anywhere, replaces each segment by four of one third the length, so its Hausdorff dimension is ln 4 / ln 3 ≈ 1.2619 and its length is infinite; the snowflake it bounds has area 8/5 of the starting triangle. The dimension is irrational because no power of 4 is a power of 3, and then transcendental by Gelfond–Schneider (1934).

Cites: [20.24](ch20-refs.md#r-20.24), [20.7](ch20-refs.md#r-20.7), [20.9](ch20-refs.md#r-20.9)

See also: [the box-counting dimension](ch20-cards.md#g-boxdim), [the Gelfond–Schneider theorem](ch04-cards.md#g-gelfschn), [one third](ch04-cards.md#g-third)

### the dimension of the Sierpiński triangle — `dim_{H} Sierpiński` {#g-sierp}

*Proven transcendental*

ln 3 / ln 2, about 1.585: three copies at half the size.

Wacław Sierpiński described the triangle in 1915, three copies of itself at half scale, so its Hausdorff dimension is ln 3 / ln 2 ≈ 1.5850; the same pattern appears in Pascal’s triangle taken modulo 2 and in the Tower of Hanoi’s state graph. Transcendental by Gelfond–Schneider, since 2^x = 3 has no algebraic irrational solution.

Cites: [20.10](ch20-refs.md#r-20.10), [20.7](ch20-refs.md#r-20.7)

See also: [Hausdorff dimension of the Cantor set](ch20-cards.md#g-dimH), [the Gelfond–Schneider theorem](ch04-cards.md#g-gelfschn)

### the dimension of the Menger sponge — `dim_{H} Menger` {#g-menger}

*Proven transcendental*

ln 20 / ln 3, about 2.727: twenty copies at one third the size, with zero volume and infinite surface area, yet topologically only a curve.

Karl Menger’s 1926 sponge keeps twenty of the twenty-seven sub-cubes at each step, so its Hausdorff dimension is ln 20 / ln 3 ≈ 2.7268, its volume is 0, and its surface area infinite; it is a universal curve: every curve, meaning every compact metric space of covering dimension one, is homeomorphic to a subset of it. Transcendental by Gelfond–Schneider.

Cites: [20.11](ch20-refs.md#r-20.11), [20.24](ch20-refs.md#r-20.24), [20.7](ch20-refs.md#r-20.7)

See also: [Hausdorff dimension of the Cantor set](ch20-cards.md#g-dimH), [the Gelfond–Schneider theorem](ch04-cards.md#g-gelfschn), [homeomorphism](ch12-cards.md#g-homeo), [one third](ch04-cards.md#g-third)

### the pieces of the Banach–Tarski paradox — `5` {#g-banachtarski}

*Integer or rational*

A solid ball can be cut into five pieces—built with the axiom of choice — that reassemble into two balls of the same size; some of them have no measure at all, which is why measure theory is not contradicted.

Banach and Tarski proved in 1924, using the axiom of choice, that a ball in ℝ³ can be decomposed into finitely many pieces and reassembled by rotations and translations into two balls each the size of the original; Raphael Robinson showed in 1947 that five pieces suffice and four do not. Some of the pieces are non-measurable, which is why measure theory is not contradicted: it simply refuses to weigh them. An integer.

Cites: [20.12](ch20-refs.md#r-20.12)

See also: [the axiom of choice](ch27-cards.md#g-ac), [a non-measurable set](ch20-cards.md#g-vitali)

### almost everywhere — `a.e.` {#g-aeverywhere}

*Function or operator*

True except on a set of Lebesgue measure zero—which is weaker than everywhere, because a null set can still be uncountable.

A property holds almost everywhere if the set on which it fails is measurable with Lebesgue measure 0. The gap between that and everywhere is exactly the subject of this chapter: the Cantor function’s derivative is 0 almost everywhere, and yet the function climbs from 0 to 1, because all of its rise happens on the Cantor set, which is uncountable and has measure 0. A qualifier on a statement, so uncolored.

Cites: [20.2](ch20-refs.md#r-20.2), [20.23](ch20-refs.md#r-20.23), [20.6](ch20-refs.md#r-20.6)

See also: [Cantor set](ch20-cards.md#g-cantor), [the derivative of the Cantor function at one half](ch20-cards.md#g-cantorderiv), [Lebesgue measure](ch20-cards.md#g-lebesgue)

### absolute continuity — `Σ|Δf| < ε` {#g-abscont}

*Function or operator*

The hypothesis the Cantor function lacks: for every ε there is a δ such that any finite set of disjoint intervals of total length under δ moves the function by less than ε in total.

A function on an interval is absolutely continuous when, for every ε > 0, there is a δ > 0 such that every finite collection of pairwise disjoint subintervals whose lengths add to less than δ has Σ|f(bᵢ) − f(aᵢ)| < ε. It is stronger than uniform continuity, and it is exactly the condition under which the fundamental theorem of calculus survives the move to the Lebesgue integral: f is absolutely continuous precisely when f(x) = f(a) + ∫ₐˣ f′, an equivalence known as the fundamental theorem of Lebesgue integral calculus. The Cantor function is the standard witness that differentiability almost everywhere is not enough—it is continuous, nondecreasing, and of bounded variation, its derivative is 0 almost everywhere, and it still climbs from 0 to 1. The Banach–Zaretsky theorem says the missing ingredient is the Luzin N property: absolute continuity is continuity plus bounded variation plus sending every null set to a null set. A hypothesis, uncolored.

Cites: [20.25](ch20-refs.md#r-20.25)

See also: [almost everywhere](ch20-cards.md#g-aeverywhere), [the derivative of the Cantor function at one half](ch20-cards.md#g-cantorderiv), [the fundamental theorem of calculus, for x²](ch11-cards.md#g-ftc), [the Lebesgue integral](ch20-cards.md#g-lebesgueint)

### dense and nowhere dense — `cl A = X` {#g-dense}

*Function or operator*

Dense: every interval, however small, contains one of its points, as the rationals do in the reals. Nowhere dense: not even its closure contains an interval, as with the Cantor set. Neither says anything about length.

A subset is dense when its closure is the whole space—equivalently, when the interior of its complement is empty—and the rationals are the standard example in the reals. A subset is nowhere dense when the closure has empty interior, equivalently when the complement of the closure is dense. The two words describe where a set sits, not how much of the line it takes up, which is the whole point of this chapter: the rationals are dense and have measure zero, the Cantor set is nowhere dense and also has measure zero, and Volterra’s nowhere-dense set of 1881 is nowhere dense with positive measure. The Baire category theorem, which says that in a complete metric space the complement of a countable union of nowhere dense sets is never empty, is what makes nowhere dense the topological word for small. A property of sets, uncolored.

Cites: [20.26](ch20-refs.md#r-20.26), [20.27](ch20-refs.md#r-20.27)

See also: [Cantor set](ch20-cards.md#g-cantor), [the fat Cantor set](ch20-cards.md#g-fatcantor), [the real numbers](ch01-cards.md#g-reals), [union](ch01-cards.md#g-union)

### a logistic-map fixed point — `f_{2}(1/2)−1/2, f_{r}(x)=rx(1−x)` {#g-logisticfixed}

*Integer or rational*

Exactly 0: one half is a fixed point of the logistic map at r=2.

For the logistic family f_{r}(x)=rx(1−x), f_{2}(1/2)=2·(1/2)·(1/2)=1/2. The residual is therefore 0. The family is a standard model of iteration, bifurcation, and deterministic chaos; its period-doubling route is governed by the Feigenbaum constants in Chapter ᚃ(F). A rational computation carrying dynamical systems.

Cites: [20.13](ch20-refs.md#r-20.13), [20.14](ch20-refs.md#r-20.14)

See also: [the logistic map](ch20-cards.md#g-logisticmap), [the logistic parameter](ch20-cards.md#g-logisticr)

### Lebesgue measure — `λ` {#g-lebesgue}

*Function or operator*

Length, made rigorous enough to measure very strange sets.

Lebesgue measure, from Henri Lebesgue’s 1902 thesis, extends length from intervals to a vast class of sets while keeping countable additivity: the measure of a countable disjoint union is the sum of the measures. It is the foundation of modern integration and of probability. A function on sets, so uncolored.

Cites: [20.24](ch20-refs.md#r-20.24), [20.5](ch20-refs.md#r-20.5)

See also: [the Lebesgue integral](ch20-cards.md#g-lebesgueint)

### Cantor set — `𝒞` {#g-cantor}

*Non-real object*

What remains of [0, 1] after removing middle thirds forever: uncountable, yet weightless.

The middle-thirds set, published by Cantor in 1883 (Henry John Stephen Smith had described such sets in 1874): the numbers in [0, 1] whose base-3 expansion uses only the digits 0 and 2. Closed, nowhere dense, uncountable, and self-similar. It is a set rather than a number, so it is colored as a non-real object.

Cites: [20.1](ch20-refs.md#r-20.1), [20.2](ch20-refs.md#r-20.2), [20.3](ch20-refs.md#r-20.3)

See also: [dense and nowhere dense](ch20-cards.md#g-dense)

### the Lebesgue integral — `∫ f dλ` {#g-lebesgueint}

*Function or operator*

Integration by slicing the range instead of the domain, which is why it can integrate functions Riemann’s integral cannot and, under the convergence theorems, passes limits inside the integral.

Lebesgue’s 1902 thesis defines the integral by measuring the sets on which f takes each value, rather than by summing over intervals of the domain; the indicator function of the rationals, which has no Riemann integral, has Lebesgue integral 0. The dominated and monotone convergence theorems make it the integral of modern analysis and probability, and Chapter ᚋ(M)’s probabilities are Lebesgue integrals. A tool, uncolored.

Cites: [20.16](ch20-refs.md#r-20.16), [20.24](ch20-refs.md#r-20.24), [20.5](ch20-refs.md#r-20.5)

### a σ-algebra — `σ-algebra` {#g-sigmaalg}

*Function or operator*

The family of sets that a measure is allowed to weigh: closed under complements and countable unions.

A measure needs a domain of measurable sets, and a σ-algebra is one closed under complement and countable union; on the line the standard ones are the Borel σ-algebra, generated by the open sets and named for Borel (his measure appears in his 1898 Leçons), and its completion, the Lebesgue σ-algebra. Kolmogorov’s 1933 axioms make probability a measure on a σ-algebra of events. A structure, uncolored.

Cites: [20.17](ch20-refs.md#r-20.17), [20.24](ch20-refs.md#r-20.24)

### a non-measurable set — `Vitali set` {#g-vitali}

*Non-real object*

A subset of [0, 1] that Lebesgue measure cannot weigh, built with the axiom of choice by Vitali in 1905.

Pick one representative from each class of reals differing by a rational; the result, Vitali’s set of 1905, cannot have a Lebesgue measure, because countably many rational shifts of it tile [0, 1] up to a null set, and countably many equal measures cannot add up to 1: zero gives 0 and anything positive gives infinity. It is the reason Lebesgue measure lives on a σ-algebra and not on all sets, and one of the precursors of the Banach–Tarski paradox. An object, colored as a non-real object.

Cites: [20.18](ch20-refs.md#r-20.18)

See also: [the axiom of choice](ch27-cards.md#g-ac), [the pieces of the Banach–Tarski paradox](ch20-cards.md#g-banachtarski), [Lebesgue measure](ch20-cards.md#g-lebesgue), [a σ-algebra](ch20-cards.md#g-sigmaalg)

### the box-counting dimension — `dim_{box}` {#g-boxdim}

*Function or operator*

Count the boxes of side ε needed to cover a set; if the count grows like ε^(−d), the dimension is d. Easier to compute than Hausdorff’s, and equal to it for the self-similar sets here, though not for the rationals.

The Minkowski–Bouligand or box-counting dimension, named for Hermann Minkowski and Georges Bouligand, measures how the number N(ε) of ε-boxes covering a set scales: d = lim ln N(ε) / ln(1/ε). For self-similar sets like the Cantor set and the Koch curve it agrees with the Hausdorff dimension; for the rationals it gives 1 while Hausdorff gives 0, so the two can differ; Hausdorff’s is often very hard to compute, and this is the practical one. A quantity, uncolored.

Cites: [20.18](ch20-refs.md#r-20.18)

See also: [Cantor set](ch20-cards.md#g-cantor), [Hausdorff dimension of the Cantor set](ch20-cards.md#g-dimH), [the dimension of the Koch curve](ch20-cards.md#g-koch)

### the logistic parameter — `r` {#g-logisticr}

*Integer or rational*

2: the growth rate of the map f(x) = rx(1 − x) on the page. Between 1 and 3 every start inside the interval settles to a fixed point; above 3 the doubling cascade to chaos begins.

Robert May’s 1976 paper in Nature made the logistic map rx(1 − x) a standard example of simple rules with complicated behavior. For 1 < r < 3 there is one attracting fixed point, 1 − 1/r (here ½); at r = 3 it splits into a cycle of two, then four, then eight, at parameter values whose spacing shrinks by Feigenbaum’s δ ≈ 4.669 of Chapter ᚃ(F), until chaos at r ≈ 3.5699. An integer.

Cites: [20.13](ch20-refs.md#r-20.13), [20.14](ch20-refs.md#r-20.14), [20.15](ch20-refs.md#r-20.15)

See also: [Feigenbaum’s δ](ch05-cards.md#g-feig), [the logistic map](ch20-cards.md#g-logisticmap)

### the logistic map — `f(x) = rx(1 − x)` {#g-logisticmap}

*Function or operator*

One line of arithmetic, iterated: this year’s population as a fraction of its ceiling becomes next year’s.

Robert May’s 1976 paper in Nature made f(x) = rx(1 − x) the standard example of a simple deterministic rule with complicated behavior. For r between 1 and 3 every start strictly inside the interval settles to the fixed point 1 − 1/r; at r = 3 that point splits into a cycle of two, then four, then eight, and the period-doubling cascade reaches chaos near r ≈ 3.5699. This chapter takes it at r = 2, the tame end of the road, and leaves the cascade to Chapter ᚃ(F). A map, so uncolored.

Cites: [20.13](ch20-refs.md#r-20.13), [20.14](ch20-refs.md#r-20.14)

See also: [the logistic parameter](ch20-cards.md#g-logisticr)