Mathematics · Book 2 · Grades 10–12

High School Mathematics

High School Mathematics · Grades 10–12

1Numbers and Sets of Numbers

Mathematics begins with numbers, and not all numbers are of the same kind: counting numbers, negative numbers, fractions, and numbers like 2\sqrt 2 or π\pi that no fraction can express. This chapter organizes them into nested families, introduces intervals to describe portions of the number line, and uses the absolute value to measure distances between numbers.

1.1 The families of numbers

Definition 1.1 (Sets of numbers)

  • N\N is the set of natural numbers: 0,1,2,3,0, 1, 2, 3, \dots
  • Z\Z is the set of integers: ,2,1,0,1,2,\dots, -2, -1, 0, 1, 2, \dots
  • Q\Q is the set of rational numbers: all quotients pq\frac{p}{q} with pZp \in \Z, qNq \in \N and q0q \neq 0.
  • R\R is the set of real numbers: all the numbers that can be placed on the number line.

Notation 1.2

The symbol \in reads “belongs to”: 3N3 \in \N, 5Z-5 \in \Z, 23Q\frac23 \in \Q. The symbol \subset reads “is included in”: every natural number is an integer, every integer is a rational number (e.g. 5=51-5 = \frac{-5}{1}), and every rational number is real, so

NZQR.\N \subset \Z \subset \Q \subset \R .

A slash negates a symbol: 12Z\frac12 \notin \Z.

The nested families of numbers: ℕ ⊂ ℤ ⊂ ℚ ⊂ ℝ. Each ring contains numbers that do not belong to the smaller ones.
The nested families of numbers: NZQR\N \subset \Z \subset \Q \subset \R. Each ring contains numbers that do not belong to the smaller ones.

Example 1.3

Let us place a few numbers in the smallest family that contains them.

  • 284=7\frac{28}{4} = 7, so 284N\frac{28}{4} \in \N even though it is written as a fraction: always simplify first.
  • 3.5=3510=72-3.5 = -\frac{35}{10} = -\frac72 is rational but not an integer.
  • 0.3330.333\dots (the digit 33 repeating forever) equals 13\frac13, a rational number.
  • 2\sqrt 2 and π\pi are real but not rational, as we are about to see for 2\sqrt 2; such numbers are called irrational.

Theorem 1.4 (Irrationality of 2\sqrt 2)

The number 2\sqrt 2 is not rational: no fraction of integers has square 22.

Proof. We reason by contradiction, in small steps.

  1. Suppose 2=pq\sqrt 2 = \frac{p}{q} where pp and qq are positive integers, and the fraction is fully simplified, so pp and qq are not both even.
  2. Squaring both sides gives 2=p2q22 = \frac{p^2}{q^2}, that is p2=2q2p^2 = 2q^2. So p2p^2 is even.
  3. If pp were odd, say p=2k+1p = 2k+1, then p2=4k2+4k+1p^2 = 4k^2 + 4k + 1 would be odd. Since p2p^2 is even, pp must be even: p=2kp = 2k for some integer kk.
  4. Substituting: (2k)2=2q2(2k)^2 = 2q^2, so 4k2=2q24k^2 = 2q^2, so q2=2k2q^2 = 2k^2. By the same argument as in step 3, qq is even.
  5. Now pp and qq are both even, contradicting step 1. The assumption was impossible: 2\sqrt 2 is irrational.

Remark 1.5

Rational numbers are exactly the numbers whose decimal expansion either stops (like 72=3.5\frac72 = 3.5) or eventually repeats the same block forever (like 13=0.333\frac13 = 0.333\dots or 17=0.142857142857\frac{1}{7} = 0.142857\,142857\dots). Irrational numbers like 2=1.41421356\sqrt2 = 1.41421356\dots have decimal digits that never fall into a repeating pattern. We admit this characterization at this level.

1.2 The number line and intervals

Real numbers fill a line: choosing an origin 00 and a unit of length, every real number corresponds to exactly one point.

Definition 1.6 (Interval)

Let aa and bb be real numbers with a<ba < b. An interval is the set of all real numbers between two bounds. A square bracket means the bound is included, a parenthesis means it is excluded:

notationdescription
[a,b]\intcc{a}{b}axba \leq x \leq b (both ends included)
(a,b)\intoo{a}{b}a<x<ba < x < b (both ends excluded)
[a,b)\intco{a}{b}ax<ba \leq x < b
(a,b]\intoc{a}{b}a<xba < x \leq b
[a,+)\intco{a}{+\infty}xax \geq a
(,b)\intoo{-\infty}{b}x<bx < b

The symbols -\infty and ++\infty (“infinity”) are not numbers, only a way of saying that the interval continues forever; the bracket next to them is always a parenthesis. The whole line R\R is the interval (,+)\intoo{-\infty}{+\infty}.

Intervals on the number line: a filled dot for an included bound, a hollow dot for an excluded one.
Intervals on the number line: a filled dot for an included bound, a hollow dot for an excluded one.

Definition 1.7 (Intersection and union)

Let II and JJ be two sets of real numbers. The intersection IJI \cap J (“II and JJ”) is the set of numbers belonging to both; the union IJI \cup J (“II or JJ”) is the set of numbers belonging to at least one of them.

Example 1.8

Take I=[1,3]I = \intcc{-1}{3} and J=(1,5)J = \intoo{1}{5}. Draw both on the same line: they overlap between 11 and 33. Therefore

IJ=(1,3],IJ=[1,5).I \cap J = \intoc{1}{3}, \qquad I \cup J = \intco{-1}{5}.

Note the brackets: 1J1 \notin J so 1IJ1 \notin I \cap J, but 3I3 \in I and 3J3 \in J, so 3IJ3 \in I \cap J.

Method 1.9 (Working with intervals)

To find the intersection or union of two intervals:

  1. draw the number line and mark the four bounds;
  2. shade the first interval above the line and the second below it;
  3. the intersection is where the shadings overlap, the union is where at least one shading is present;
  4. decide each bracket by testing whether the bound itself belongs to both sets (intersection) or to at least one (union).

1.3 Absolute value and distance

Definition 1.10 (Absolute value)

The absolute value of a real number xx is

x={xif x0,xif x<0.\abs{x} = \begin{cases} x & \text{if } x \geq 0, \\ -x & \text{if } x < 0. \end{cases}

For instance 7=7\abs{7} = 7 and 4=4\abs{-4} = 4. In every case x0\abs{x} \geq 0.

Proposition 1.11 (Distance on the line)

For all real numbers aa and bb, the distance between the points aa and bb on the number line is ba\abs{b - a}. In particular x\abs{x} is the distance from xx to 00.

Proof. If bab \geq a, the distance from aa to bb is ba0b - a \geq 0, which equals ba\abs{b-a}. If b<ab < a, the distance is ab=(ba)>0a - b = -(b - a) > 0, which is again ba\abs{b - a} by definition of the absolute value.

Proposition 1.12 (Absolute value and intervals)

Let aa be a real number and r>0r > 0. Then

xarexactly whenx[ar,a+r].\abs{x - a} \leq r \quad\text{exactly when}\quad x \in \intcc{a - r}{a + r}.

Proof. xar\abs{x-a} \leq r says that the distance from xx to aa is at most rr, i.e. that xx lies no further than rr from aa on either side. The numbers satisfying this are exactly those between ara - r and a+ra + r, bounds included.

The inequality |x - a| ≤ r describes the interval of numbers at distance at most r from a.
The inequality xar\abs{x - a} \leq r describes the interval of numbers at distance at most rr from aa.

Example 1.13

Solve x32\abs{x - 3} \leq 2. The solutions are the numbers at distance at most 22 from 33: the interval [1,5]\intcc{1}{5}. Conversely, the interval [1,7]\intcc{-1}{7} has center 1+72=3\frac{-1+7}{2} = 3 and radius 7(1)2=4\frac{7-(-1)}{2} = 4, so it is described by x34\abs{x - 3} \leq 4.

1.4 Approximations

Irrational numbers, and even most fractions, cannot be written exactly with finitely many decimal digits, so in practice we approximate them.

Definition 1.14 (Approximation to a given accuracy)

A number dd is an approximation of xx to within 10n10^{-n} when xd10n\abs{x - d} \leq 10^{-n}. Truncating or rounding the decimal expansion after the nn-th digit both give such approximations.

Example 1.15

From π=3.1415926\pi = 3.14159\,26\dots: the truncation 3.1413.141 and the rounding 3.1423.142 are both approximations of π\pi to within 10310^{-3}. The rounding is at distance less than 12×103\frac12 \times 10^{-3}, the truncation only guarantees 10310^{-3}. Writing 3.141π3.1423.141 \leq \pi \leq 3.142 frames π\pi between two decimal bounds.

1.5 Exercises

Exercise 1.1

For each number, give the smallest of the sets N\N, Z\Z, Q\Q, R\R to which it belongs:

153,7,227,9,10,2.4,0.\frac{15}{3}, \qquad -7, \qquad \frac{22}{7}, \qquad \sqrt{9}, \qquad \sqrt{10}, \qquad -2.4, \qquad 0 .
Solution

Solution of Exercise 1.1.

153=5N\frac{15}{3} = 5 \in \N. 7Z-7 \in \Z. 227Q\frac{22}{7} \in \Q (it is not an integer: 22=7×3+122 = 7 \times 3 + 1). 9=3N\sqrt 9 = 3 \in \N. 10R\sqrt{10} \in \R (irrational, since 1010 is not the square of a rational — admitted here, in the spirit of Theorem 1.4). 2.4=2410Q-2.4 = -\frac{24}{10} \in \Q. 0N0 \in \N.

Exercise 1.2

Write each statement with interval notation, then draw it on a number line: (a) 2x<5-2 \leq x < 5; (b) x>3x > 3; (c) x1x \leq -1; (d) the distance from xx to 22 is at most 33.

Solution

Solution of Exercise 1.2.

(a) [2,5)\intco{-2}{5}: filled dot at 2-2, hollow dot at 55. (b) (3,+)\intoo{3}{+\infty}: hollow dot at 33, shading to the right. (c) (,1]\intoc{-\infty}{-1}: shading from the left up to a filled dot at 1-1. (d) “distance from xx to 22 at most 33” means x23\abs{x - 2} \leq 3, i.e. x[1,5]x \in \intcc{-1}{5}: filled dots at 1-1 and 55.

Exercise 1.3

Compute IJI \cap J and IJI \cup J for

(a) I=[3,2], J=[0,4);(b) I=(,1), J=[2,+).\text{(a) } I = \intcc{-3}{2},\ J = \intco{0}{4}; \qquad \text{(b) } I = \intoo{-\infty}{1},\ J = \intco{-2}{+\infty}.
Solution

Solution of Exercise 1.3.

(a) The two intervals overlap between 00 and 22: IJ=[0,2]I \cap J = \intcc{0}{2} (0J0 \in J and 2I2 \in I, and both belong to the other set as well), and IJ=[3,4)I \cup J = \intco{-3}{4}.

(b) IJI \cap J is the set of xx with 2x-2 \leq x and x<1x < 1: [2,1)\intco{-2}{1}. The union covers everything: IJ=RI \cup J = \R.

Exercise 1.4

Compute without a calculator:

6,49,35,21,12.\abs{-6}, \qquad \abs{4 - 9}, \qquad \abs{-3 - 5}, \qquad \abs{\sqrt 2 - 1}, \qquad \abs{1 - \sqrt 2}.
Solution

Solution of Exercise 1.4.

6=6\abs{-6} = 6; 49=5=5\abs{4 - 9} = \abs{-5} = 5; 35=8=8\abs{-3 - 5} = \abs{-8} = 8; 2>1\sqrt 2 > 1 so 21=21\abs{\sqrt2 - 1} = \sqrt2 - 1; 12<01 - \sqrt2 < 0 so 12=21\abs{1 - \sqrt2} = \sqrt2 - 1 as well (a number and its opposite have the same absolute value).

Exercise 1.5

Solve the equations and inequalities and give the solution sets:

x=5,x1=3,x41,x+2<3.\abs{x} = 5, \qquad \abs{x - 1} = 3, \qquad \abs{x - 4} \leq 1, \qquad \abs{x + 2} < 3 .

(Note that x+2=x(2)\abs{x+2} = \abs{x - (-2)} is a distance to 2-2.)

Solution

Solution of Exercise 1.5.

x=5\abs{x} = 5: distance 55 from 00, so x=5x = 5 or x=5x = -5.

x1=3\abs{x - 1} = 3: distance 33 from 11, so x=4x = 4 or x=2x = -2.

x41\abs{x - 4} \leq 1: distance at most 11 from 44, so x[3,5]x \in \intcc{3}{5}.

x+2<3\abs{x + 2} < 3: distance strictly less than 33 from 2-2, so x(5,1)x \in \intoo{-5}{1}.

Exercise 1.6 ★★

Describe each interval by an inequality of the form xar\abs{x - a} \leq r or xa<r\abs{x - a} < r:

[2,8],(5,1),[7,3].\intcc{2}{8}, \qquad \intoo{-5}{1}, \qquad \intcc{-7}{-3}.
Solution

Solution of Exercise 1.6.

Each interval is described by its center aa (midpoint of the bounds) and radius rr (half the length).

[2,8]\intcc{2}{8}: a=5a = 5, r=3r = 3, so x53\abs{x - 5} \leq 3.

(5,1)\intoo{-5}{1}: a=2a = -2, r=3r = 3, open bounds, so x+2<3\abs{x + 2} < 3.

[7,3]\intcc{-7}{-3}: a=5a = -5, r=2r = 2, so x+52\abs{x + 5} \leq 2.

Exercise 1.7 ★★

Show that 0.2727270.272727\dots (the block 2727 repeating forever) is rational. (Hint: call it xx and compute 100xx100x - x.)

Solution

Solution of Exercise 1.7.

Let x=0.272727x = 0.272727\dots Then 100x=27.2727100x = 27.2727\dots, and subtracting:

100xx=27.27270.2727=27,100x - x = 27.2727\dots - 0.2727\dots = 27,

so 99x=2799x = 27 and x=2799=311x = \frac{27}{99} = \frac{3}{11}, a quotient of integers: xx is rational.

Exercise 1.8 ★★

True or false? Justify each answer with an argument or a counterexample.

  1. The sum of two integers is an integer.
  2. The quotient of two integers is an integer.
  3. The sum of two rational numbers is rational.
  4. The sum of a rational number and an irrational number is irrational.
Solution

Solution of Exercise 1.8.

1. True: adding integers (positive or negative whole numbers) always produces an integer.

2. False: 12\frac{1}{2} is a quotient of the integers 11 and 22 and is not an integer.

3. True: pq+pq=pq+pqqq\frac pq + \frac{p'}{q'} = \frac{pq' + p'q}{qq'} is again a quotient of integers (with nonzero denominator).

4. True: suppose rr is rational, tt irrational, and r+t=sr + t = s were rational. Then t=srt = s - r would be a difference of two rationals, hence rational (by 3, applied with r-r) — contradiction. So r+tr + t is irrational.

Exercise 1.9 ★★

Using 1.41421.4151.414 \leq \sqrt 2 \leq 1.415, frame the numbers 222\sqrt2, 2+3\sqrt2 + 3 and 2-\sqrt2 between two decimal bounds.

Solution

Solution of Exercise 1.9.

Multiplying 1.41421.4151.414 \leq \sqrt2 \leq 1.415 by 2>02 > 0: 2.828222.8302.828 \leq 2\sqrt2 \leq 2.830.

Adding 33: 4.4142+34.4154.414 \leq \sqrt2 + 3 \leq 4.415.

Multiplying by 1<0-1 < 0 reverses the inequalities: 1.41521.414-1.415 \leq -\sqrt2 \leq -1.414.

Exercise 1.10 ★★★

Adapt the proof of Theorem 1.4 to show that 3\sqrt 3 is irrational. (Replace “even” by “multiple of 33”: first check that if p2p^2 is a multiple of 33, then so is pp, by examining the remainders 00, 11, 22 of pp upon division by 33.)

Solution

Solution of Exercise 1.10.

First the auxiliary fact. Divide pp by 33: the remainder is 00, 11 or 22, i.e. p=3kp = 3k, p=3k+1p = 3k+1 or p=3k+2p = 3k+2. Squaring:

(3k)2=3(3k2),(3k+1)2=3(3k2+2k)+1,(3k+2)2=3(3k2+4k+1)+1.(3k)^2 = 3(3k^2), \quad (3k+1)^2 = 3(3k^2 + 2k) + 1, \quad (3k+2)^2 = 3(3k^2 + 4k + 1) + 1 .

Only the first is a multiple of 33: if p2p^2 is a multiple of 33, then pp is too.

Now suppose 3=pq\sqrt3 = \frac pq fully simplified. Squaring gives p2=3q2p^2 = 3q^2, so p2p^2 is a multiple of 33, so p=3kp = 3k. Then 9k2=3q29k^2 = 3q^2, so q2=3k2q^2 = 3k^2 and qq is also a multiple of 33 — but then the fraction pq\frac pq was not fully simplified: contradiction. So 3\sqrt3 is irrational.

1.6 Problem: Between any two numbers

Problem 1.1

Weekend problem — rationals and irrationals interlace: every interval, however small, contains infinitely many of each, and no measurement can ever tell them apart

The rationals look like a crowd (all the fractions!) and the irrationals like exotic exceptions (2\sqrt2, π\pi). This problem reveals the true picture: the two families interlace so finely that every interval of the number line, however microscopic, contains infinitely many of each — with strange consequences, such as this: no physical measurement, however precise, can ever decide whether a length is rational.

Part I — The four kingdoms.

  1. For each number, name the smallest of the sets N\N, Z\Z, Q\Q, R\R containing it (Definition 1.1): 7-7; 134\frac{13}{4}; 16\sqrt{16}; 0.1212120.121212\ldots (recall the repeating-decimals weekend problem of the Middle School volume); 8\sqrt8; π\pi (admit its irrationality — proved only in the university volumes).
  2. Prove that Q\Q is stable under addition and multiplication: if x=pqx = \frac pq and y=rsy = \frac rs are rational, write x+yx + y and xyxy as single fractions.
  3. Deduce by contradiction: (a) the sum of a rational and an irrational is irrational; (b) the product of a nonzero rational and an irrational is irrational.
  4. Show that the set of irrationals is stable under neither operation: exhibit two irrationals whose sum is rational, and two whose product is rational.
  5. Locate 2+8\sqrt2 + \sqrt8 and 2×8\sqrt2 \times \sqrt8 in the four kingdoms (simplify 8\sqrt8 first, the Middle School volume’s method).

Part II — The rationals are dense.

  1. Find a rational strictly between 3.473.47 and 3.483.48; then a second one; then describe how to produce as many as desired (the zoom of the no-next-number weekend problem of the Middle School volume, now with a proof in sight).
  2. The general theorem. Let a<ba < b be any two reals, with gap g=ba>0g = b - a > 0. Choose nn with 10n<g10^{-n} < g, and consider the multiples of 10n10^{-n} (the decimal grid of step 10n10^{-n}). Explain why at least one grid point falls strictly between aa and bb, and conclude: every interval of positive length contains a rational number.
  3. Upgrade the conclusion: every such interval contains infinitely many rationals. (Apply question 7 again, inside a smaller interval.)
  4. Now the irrationals: given a<ba < b, take a rational rr strictly inside (question 7) and consider the numbers r+210kr + \frac{\sqrt2}{10^k}. Using question 3, show they are irrational, and that for kk large enough they still lie in the interval: every interval also contains infinitely many irrationals.
  5. Two classic riddles settled: is there a smallest positive real number? A real number “just after 33”? Answer both with question 7’s theorem, and salute the childhood version (the no-next-number weekend problem of the Middle School volume).

Part III — Absolute value, the geometry of R\R.

  1. Solve, and express the solutions as intervals or unions (Proposition 1.12): x5=2\abs{x - 5} = 2; x5<2\abs{x - 5} < 2; x+13\abs{x + 1} \geq 3.
  2. A piston must be manufactured to 8080 mm within a tolerance of 0.050.05 mm. Write the requirement with an absolute value, then as an interval. Two pistons measure 79.9779.97 and 80.0680.06 mm: verdicts?
  3. The triangle inequality on the line: a+ba+b\abs{a + b} \leq \abs a + \abs b. Check it on (a,b)=(3,5)(a, b) = (3, -5) and (2,7)(-2, -7), prove it when aa and bb have the same sign and when they have opposite signs, and state exactly when equality holds.
  4. Solve x2=x+4\abs{x - 2} = \abs{x + 4} by reading it as an equality of distances on the line. Which point of the two-mirrors weekend problem of the Middle School volume have you just computed, one dimension down?
  5. Simplify x2\sqrt{x^2} — carefully. Test your formula on x=3x = 3 and x=3x = -3, then use it to solve x2<9x^2 < 9 with an absolute value.

Part IV — What no measurement can decide.

  1. A physicist measures a rod: 1.41421±0.000011.41421 \pm 0.00001 m. Can any such measurement — this one or a future, more precise one — ever prove that the rod’s length is irrational? Or that it is rational? Use questions 7 and 9 on the tolerance interval, and explain why only proof (as for the diagonal, the irrationality weekend problem of the Middle School volume) can settle irrationality.
  2. The truncations 1.41.4, 1.411.41, 1.4141.414, 1.4142,1.4142,\dots of 2\sqrt2 are all rational. What do they show about how close Q\Q presses against every irrational? Formulate the general fact.
  3. Interval arithmetic: measured values a=2.5±0.1a = 2.5 \pm 0.1 and b=1.2±0.1b = 1.2 \pm 0.1. Bracket a+ba + b and a×ba \times b between certain bounds. Which operation degrades the precision more?
  4. An engineer’s rule says errors of independent measurements “add” for sums: express with the triangle inequality (question 13) why the error of a+ba + b is at most the sum of the errors — the inequality is the engineering rule.
  5. Finale, in a short paragraph: assemble the picture of the real line established by this problem and its ancestors — no next number (the no-next-number weekend problem of the Middle School volume), rationals == repeating decimals (the repeating-decimals weekend problem of the Middle School volume), both families dense (questions 8 and 9), measurement forever undecided (question 16). End with the teaser this chapter cannot yet prove: in a precise sense there are vastly more irrationals than rationals — the university volumes count the infinities.
Solution

Solution of Problem 1.1.

1. 7Z-7 \in \Z; 134Q\frac{13}{4} \in \Q; 16=4N\sqrt{16} = 4 \in \N; 0.121212=1299=433Q0.121212\ldots = \frac{12}{99} = \frac{4}{33} \in \Q (the repeating-decimals weekend problem of the Middle School volume); 8=22\sqrt8 = 2\sqrt2 is irrational: smallest set R\R; π\pi: R\R.

2. pq+rs=ps+qrqs\frac pq + \frac rs = \frac{ps + qr}{qs} and pq×rs=prqs\frac pq \times \frac rs = \frac{pr}{qs}: quotients of integers with nonzero denominators — rational.

3. (a) If rr is rational, xx irrational and r+x=qr + x = q were rational, then x=qrx = q - r would be a difference of rationals, hence rational (question 2): contradiction. (b) If r0r \neq 0 and rx=qr x = q rational, then x=qrx = \frac qr is rational: contradiction.

4. 2\sqrt2 and 2-\sqrt2 are irrational with sum 00; 2\sqrt2 and 2\sqrt2 have product 22. Rational results from irrational ingredients: the irrationals form no stable kingdom.

5. 8=22\sqrt8 = 2\sqrt2, so 2+8=32\sqrt2 + \sqrt8 = 3\sqrt2: irrational (question 3b, multiplier 33). And 2×8=16=4N\sqrt2 \times \sqrt8 = \sqrt{16} = 4 \in \N: the product of two irrationals lands in the smallest kingdom of all.

6. 3.4753.475, then 3.4713.471 (or 3.47013.4701, 3.470013.47001, …): appending decimal digits produces fresh rationals between the two forever.

7. The multiples of 10n10^{-n} march along the line in steps of 10n<g10^{-n} < g. The first multiple strictly greater than aa — it exists, since the multiples eventually exceed aa — lies at most one step beyond aa, hence before a+g=ba + g = b: strictly between aa and bb. A grid point is a decimal, hence a rational: every interval of positive length contains one.

8. Between aa and the rational r1r_1 found in question 7 there is (question 7 again) a rational r2r_2; between aa and r2r_2 a rational r3r_3; and so on: infinitely many, all distinct, all in the original interval.

9. r+210kr + \frac{\sqrt2}{10^k} is the sum of a rational and an irrational (210k\frac{\sqrt2}{10^k} is irrational by question 3b), hence irrational. Since 210k<210k\frac{\sqrt2}{10^k} < \frac{2}{10^k} shrinks below any bound, for kk large enough r+210kr + \frac{\sqrt2}{10^k} still lies before bb: an irrational inside the interval — and varying kk gives infinitely many.

10. No smallest positive real: if s>0s > 0 were it, the interval (0,s)(0, s) would still contain a rational (question 7), positive and smaller than ss. No number just after 33: any candidate c>3c > 3 leaves the interval (3,c)(3, c), which is not empty — the no-next-number weekend problem of the Middle School volume’s game, now a theorem about R\R.

11. x5=2\abs{x - 5} = 2: x=3x = 3 or x=7x = 7. x5<2\abs{x - 5} < 2: x(3,7)x \in \intoo{3}{7}. x+13\abs{x + 1} \geq 3: distance to 1-1 at least 33: x(,4][2,+)x \in \intoc{-\infty}{-4} \cup \intco{2}{+\infty}.

12. d800.05\abs{d - 80} \leq 0.05, i.e. d[79.95,80.05]d \in \intcc{79.95}{80.05}. The piston at 79.9779.97 passes; the one at 80.0680.06 fails by a hundredth of a millimeter.

13. (3,5)(3, -5): 2=28\abs{-2} = 2 \leq 8. (2,7)(-2, -7): 9=9=2+7\abs{-9} = 9 = 2 + 7: equality. Same signs: a+b\abs{a + b} is the sum of the distances, equal to a+b\abs a + \abs b. Opposite signs: the walk doubles back, a+b\abs{a + b} is the difference of the distances, strictly less than their sum (unless one of them is 00). Equality exactly when aa and bb have the same sign or one is zero.

14. The solutions are the points equidistant from 22 and 4-4: the midpoint, x=1x = -1. It is the perpendicular bisector of the two-mirrors weekend problem of the Middle School volume, collapsed to one dimension: a single point.

15. x2=x\sqrt{x^2} = \abs{x}, not xx: for x=3x = -3, 9=3=33\sqrt{9} = 3 = \abs{-3} \neq -3. Then x2<9x^2 < 9 reads x2<3\sqrt{x^2} < 3, i.e. x<3\abs x < 3: x(3,3)x \in \intoo{-3}{3}.

16. The measurement asserts only that the length lies in the interval [1.41420,1.41422]\intcc{1.41420}{1.41422} — and by questions 7 and 9 that interval contains infinitely many rationals and infinitely many irrationals. The same holds for every future tolerance, however small. No measurement can ever separate the two families; only a proof about the exact length — like the one for the diagonal of the unit square (the irrationality weekend problem of the Middle School volume) — can.

17. They approach 2\sqrt2 with errors below 101,102,103,10^{-1}, 10^{-2}, 10^{-3}, \dots: rationals press against 2\sqrt2 at every scale. The general fact: every real number is approached as closely as desired by rationals (its decimal truncations) — density once more, seen from the target’s side.

18. a+b[2.4+1.1,2.6+1.3]=[3.5,3.9]a + b \in \intcc{2.4 + 1.1}{2.6 + 1.3} = \intcc{3.5}{3.9}: uncertainty ±0.2\pm 0.2 (errors added). a×b[2.4×1.1,2.6×1.3]=[2.64,3.38]a \times b \in \intcc{2.4 \times 1.1}{2.6 \times 1.3} = \intcc{2.64}{3.38}: an uncertainty of about ±0.37\pm 0.37 around 33 — relative errors add for products, so precision degrades faster.

19. Write the true values a=2.5+e1a = 2.5 + e_1, b=1.2+e2b = 1.2 + e_2 with e1,e20.1\abs{e_1}, \abs{e_2} \leq 0.1. Then the error of the sum is e1+e2e1+e20.2\abs{e_1 + e_2} \leq \abs{e_1} + \abs{e_2} \leq 0.2: the engineer’s rule is the triangle inequality of question 13.

20. The real line has no gaps and no neighbours: after any number, no “next” one (question 10). Its points split into the rationals — exactly the stopping or repeating decimals (the repeating-decimals weekend problem of the Middle School volume) — and the irrationals, and the two families are interwoven so tightly that every interval holds infinitely many of each (questions 8 and 9), which is why no measurement, only proof, can tell them apart (question 16). And the final surprise, left as a promise: the irrationals outnumber the rationals — not by counting one by one, but in the precise sense of comparing infinities, a theory built in the university volumes.