The World of Numbers

This chapter tells the story of how the number system grew, layer by layer, to meet human need, and then teaches you to work precisely within each layer.

• The historical arc. Tally marks give natural numbers; Brahmagupta's work in India gives zero and negative numbers, producing the integers.
• Rational numbers ℚ. Their definition as p/q, the rules of equality and arithmetic, closure, and density on the number line.
• Locating numbers exactly. Placing any rational number, and even irrational lengths such as √2, at an exact point using ruler-and-compass construction.
• Irrational numbers. What they are, why √2 cannot be a fraction (proof by contradiction), and why π needs an infinite series.
• Decimal expansions. Why rational decimals terminate or repeat, how to predict which without long division, and how to reverse a repeating decimal back to p/q.
• The complete real number line ℝ, formed by uniting the rational and irrational numbers, and a glimpse of what lies beyond it.

From counting to zero: the birth of integers

The natural numbers ℕ = {1, 2, 3, 4, …} are the numbers you count with; they came from the physical act of one-to-one correspondence, matching a pebble to a cow, a notch to a day. For centuries no symbol existed for "nothing." Brahmagupta, in his 628 CE treatise the Brāhmasphuṭasiddhānta, defined zero explicitly as the result of subtracting a number from itself, and gave it working rules of arithmetic.

where: $a$ stands for any number. These three rules look obvious today only because Brahmagupta made zero a genuine number rather than a placeholder or a blank. Brahmagupta then asked what happens when a smaller number is subtracted from a larger one, and answered by grounding the question in commerce: he called positive numbers dhana (fortunes) and negative numbers ṛiṇa (debts). Combining the positive whole numbers, their negative counterparts, and zero gives the integers, written ℤ (from the German Zahlen, numbers). Brahmagupta's rules for combining fortunes and debts are exactly the sign rules used today:

where: $a, b$ are positive numbers, so $-a$ and $-b$ represent debts. The last rule — a debt times a debt is a fortune — is the one students find hardest to accept. Brahmagupta's own reasoning was economic: if someone removes four debts of ₹3 each from your account, you become ₹12 richer, so $(-3) \times (-4) = +12$. Removing a negative action is a positive outcome.

Rational numbers: definition, equality and closure

A rational number is any number that can be written as $\dfrac{p}{q}$, with $p, q$ integers and $q \neq 0$. The set is denoted ℚ (for quotient).

where: $p$ is the numerator and $q$ is the denominator; every natural number and every integer is also rational, since $5 = \dfrac{5}{1}$, so ℕ ⊂ ℤ ⊂ ℚ. A single rational number has infinitely many equivalent fraction names — $\dfrac{2}{3} = \dfrac{4}{6} = \dfrac{10}{15} = \cdots$ — so we need a rule to check whether two of them represent the same value:

where: cross-multiplying and comparing $ad$ and $bc$ decides equality without a common denominator. By convention, "the rational number $\dfrac{p}{q}$" takes $p$ and $q$ co-prime (no common factor other than 1) with $q > 0$, so each rational number has exactly one simplest representative. Brahmagupta also fixed the arithmetic of fractions, still taught exactly as he stated it:

where: the addition rule works for any denominators, because $\dfrac{a}{b}$ and $\dfrac{c}{d}$ are first rewritten over the common denominator $bd$. Under these rules ℚ is closed under addition, subtraction, multiplication, and division whenever the divisor is not zero: combining two rationals this way always produces another rational.

Locating rationals and the density of ℚ

To place $\dfrac{p}{q}$ on the number line, divide the unit interval between two consecutive integers into $q$ equal parts, then count $p$ of those parts from 0 (rightward if positive, leftward if negative). The figure below marks $\dfrac{3}{4}$ and $\dfrac{9}{4} = 2\dfrac{1}{4}$ this way.

The absolute value of a rational number $x$, written $|x|$, is its distance from 0 on the number line:

where: $|x|$ is always non-negative. More generally, the distance between any two rational numbers $a$ and $b$ on the line is $|a - b|$. Rational numbers are dense: between any two distinct rationals there is always another one, and therefore infinitely many. The simplest proof of this uses their average.

where: $\dfrac{a+b}{2}$ is itself a ratio of integers whenever $a$ and $b$ are, so it is automatically rational, and it sits exactly halfway between $a$ and $b$, strictly inside the interval.

Irrational numbers and the real number line

Not every length on the number line is a fraction. Consider a square of side 1 unit; by the Baudhāyana–Pythagoras theorem its diagonal $d$ satisfies

where: $d$ is the diagonal's length. Hippasus, of the Pythagorean school, proved around 400 BCE that $\sqrt{2}$ cannot be written as any fraction $p/q$ — the first derivation below reconstructs his argument in full. Numbers like $\sqrt{2}$, $\sqrt{3}$ and $\pi$ that cannot be expressed as a ratio of integers are called irrational numbers. Remarkably, an irrational length such as $\sqrt{2}$ can still be constructed exactly with ruler and compass and marked precisely on the number line — geometry reaches a point that arithmetic-by-fractions cannot name. The circle constant $\pi$ is irrational too (proved by Johann Lambert in 1761); no single fraction equals it exactly — Āryabhaṭa's famous $\dfrac{3927}{1250} = 3.1416$ was only an asanna (approximation). Centuries later, Mādhava of Sangamagrama found an infinite sum of fractions that closes in on $\pi$ exactly:

where: each added term nudges the running total closer to $\pi$'s true value; no finite number of terms ever reaches it exactly, which is precisely why $\pi$ needs an infinite series rather than a fraction. Uniting all rational numbers with all irrational numbers fills every remaining gap on the number line and produces the real numbers, written ℝ. Every measurable length, temperature or quantity corresponds to exactly one point on this line.

Board level - Converting a general repeating decimal into p over q

Convert $2.4\overline{53}$ (that is, $2.4535353\ldots$, where the digits 5 and 3 repeat) into the form $\dfrac{p}{q}$.

Step 1. Let $x = 2.4\overline{53}$. The non-repeating part after the decimal point is "4" (1 digit), and the repeating block is "53" (2 digits).

Step 2. Multiply by $10^1 = 10$ to move the non-repeating digit before the decimal point: $10x = 24.\overline{53}$.

Step 3. Multiply that result by $10^2 = 100$ to slide one full repeating block across as well: $1000x = 2453.\overline{53}$.

Step 4. Subtract the Step 2 equation from the Step 3 equation, so the infinite tail cancels: $1000x - 10x = 2453.5353\ldots - 24.5353\ldots = 2429$. Hence $990x = 2429$, so $x = \dfrac{2429}{990}$. Checking, $\gcd(2429, 990) = 1$, so the fraction is already in lowest terms. $2.4\overline{53} = \dfrac{2429}{990}$.

Board level - Applying Brahmagupta's rules for integers

Using Brahmagupta's laws for fortunes and debts, evaluate: (i) $(-12) \times 5$, (ii) $(-8) \times (-7)$, (iii) $0 - (-14)$, and (iv) $(-20) \div 4$.

Step 1. For (i), a debt times a fortune is a debt: $(-12) \times 5 = -60$.

Step 2. For (ii), a debt times a debt is a fortune: $(-8) \times (-7) = 56$.

Step 3. For (iii), subtracting a debt equals adding the corresponding fortune (removing a debt of 14 makes you 14 richer): $0 - (-14) = 14$.

Step 4. For (iv), a debt divided by a fortune is a debt: $(-20) \div 4 = -5$. Answers: (i) $-60$, (ii) $56$, (iii) $14$, (iv) $-5$.

Higher-order thinking (HOTS / olympiad-style) - Proving the average always lies strictly between

Let $a$ and $b$ be two distinct rational numbers with $a < b$. Prove rigorously, using algebra rather than a picture, that their average $\dfrac{a+b}{2}$ is a rational number lying strictly between $a$ and $b$. Then explain how this single fact guarantees that infinitely many rational numbers exist between any two rationals, however close together they are.

Step 1. Show rationality first. Since $a = \dfrac{p_1}{q_1}$ and $b = \dfrac{p_2}{q_2}$ for integers $p_1, q_1, p_2, q_2$ with $q_1, q_2 \neq 0$, their sum $a + b = \dfrac{p_1 q_2 + p_2 q_1}{q_1 q_2}$ is a ratio of integers, and dividing that sum by the integer 2 keeps it one. So $\dfrac{a+b}{2}$ is rational.

Step 2. Prove $a < \dfrac{a+b}{2}$. Start from the given $a < b$. Add $a$ to both sides: $a + a < a + b$, that is, $2a < a + b$. Divide both sides by the positive number 2 (which preserves the inequality direction): $a < \dfrac{a+b}{2}$.

Step 3. Prove $\dfrac{a+b}{2} < b$ by a mirror argument. Starting again from $a < b$, add $b$ to both sides: $a + b < b + b = 2b$. Divide both sides by 2: $\dfrac{a+b}{2} < b$. Combining Steps 2 and 3 gives $a < \dfrac{a+b}{2} < b$, exactly as required.

Step 4. Extend to infinitely many. Let $m_1 = \dfrac{a+b}{2}$; by Steps 1 to 3 it is a new rational strictly between $a$ and $b$. Apply the same argument to the pair $a, m_1$ to get $m_2 = \dfrac{a+m_1}{2}$, strictly between $a$ and $m_1$, hence between $a$ and $b$ too. Repeating endlessly, each new average sits strictly closer to $a$ than the last, so no value ever repeats — this generates an infinite list of distinct rationals between $a$ and $b$. The average of two distinct rationals is a new rational strictly between them, and repeating the averaging process forever produces infinitely many such numbers. Timing note: in a board exam, budget about 5 minutes for Steps 1 to 3 (the core inequality proof) and only a brief sentence for Step 4 — full HOTS or olympiad questions may ask you to write out the extension argument in detail, which takes closer to 8 to 10 minutes to justify carefully.

• Writing $\sqrt{2}$ as a "long decimal fraction" and calling it exact — any finite decimal is a terminating (hence rational) approximation, never the irrational value itself.
• Believing $(-3) \times (-4) = -12$ by habit — Brahmagupta's rule is that a debt removed is a fortune gained, so two negatives multiply to a positive.
• Forgetting the condition $q \neq 0$ in $p/q$ and treating $\dfrac{5}{0}$ as some special number rather than undefined.
• Assuming every non-terminating decimal is irrational — a non-terminating but repeating decimal such as $0.\overline{3}$ is perfectly rational ($= \dfrac{1}{3}$); only non-terminating and non-repeating decimals are irrational.
• Predicting a terminating decimal by inspecting the numerator instead of the fully reduced denominator — $\dfrac{7}{20}$ terminates because $20 = 2^2 \times 5$, not because of anything about 7.
• Multiplying by the wrong power of 10 when converting a general repeating decimal — the non-repeating shift and the repeating-block shift are two separate multiplications, not one combined guess.
• Thinking $0.\overline{9}$ is "just below" 1 — algebraically it is exactly equal to 1, as $10x - x = 9 \Rightarrow x = 1$ shows.
• Losing the co-prime condition when comparing fractions — $\dfrac{2}{4}$ and $\dfrac{3}{6}$ look "different" only until you notice both reduce to $\dfrac{1}{2}$.

• State the definitions of ℕ, ℤ, ℚ and irrational numbers, and place any given number into the smallest set that contains it.
• Apply Brahmagupta's rules for zero and for fortunes and debts, including why a debt times a debt is a fortune.
• Test whether two fractions are equal using $ad = bc$, and simplify a fraction to its co-prime, lowest-terms form.
• Add, subtract, multiply and divide rational numbers, and explain why ℚ is closed under these operations (except division by zero).
• Locate any rational number exactly on the number line by subdividing the unit interval into $q$ equal parts.
• Compute the absolute value of a rational number and the distance $|a-b|$ between two rationals on the line.
• Prove that the average of two distinct rational numbers lies strictly between them, and use this to argue ℚ is dense.
• Reproduce, step by step, the proof by contradiction that $\sqrt{2}$ is irrational, and adapt it to $\sqrt{3}$, $\sqrt{5}$ or $\sqrt{7}$.
• Predict, from the prime factors of a reduced denominator alone, whether a decimal expansion terminates or repeats.
• Convert any terminating or repeating decimal (pure or general) back into the exact form $p/q$ using multiply-and-subtract.

Ask the tutor: if you are stuck on proving a number like √3 or √5 is irrational, or want another worked example of converting a tricky repeating decimal into p/q, ask the tutor to walk through it with you step by step.