Chinese Mathematics

3.1 The Nine Chapters on the Mathematical Art

The Jiuzhang Suanshu (Nine Chapters on the Mathematical Art) is the most important and influential mathematical text in the history of Chinese mathematics. Compiled from earlier sources over the period c. 200 BCE to 200 CE, it stands alongside Euclid's Elements as one of the two great mathematical classics of the ancient world — though their approaches could hardly be more different. Where Euclid proceeds axiomatically from definitions and postulates to theorems, the Nine Chaptersproceeds algorithmically from problems to solutions.

The text contains 246 problems organized into nine chapters, each devoted to a specific area of practical mathematics. The problems are presented in a consistent format: a statement of the problem, the answer, and then the method (algorithm) for arriving at the answer. Here is a summary of the nine chapters:

Several features of the Nine Chapters are remarkable:

• Negative numbers: Chapter 8 introduces red and black counting rods to represent positive and negative numbers — the earliest systematic use of negative numbers in any mathematical tradition. Rules for adding and subtracting signed numbers are given explicitly.
• Fractions: The text works fluently with fractions, including addition, subtraction, multiplication, and division, as well as reduction to lowest terms using what we would call the Euclidean algorithm for GCD.
• The Pythagorean theorem: Chapter 9 (Gougu) applies the right-triangle relationship extensively, calling the two shorter sides gou (hook) and gu (leg), and the hypotenuse xian (bowstring).
• Square and cube roots: Chapter 4 gives algorithms for extracting square and cube roots that are essentially the same as the methods used before calculators — digit-by-digit extraction.

In 263 CE, Liu Hui wrote a brilliant commentary on the Nine Chapters, providing mathematical justifications (proofs) for the algorithms, correcting errors, and adding his own results. Liu Hui's commentary transformed the Nine Chapters from a practical handbook into a work of genuine mathematical depth.

3.2 Rod Numerals and Counting Board Arithmetic

The computational engine of Chinese mathematics was the counting board (suanpan or chouma), a flat surface on which bamboo or ivory rods were arranged to represent numbers and perform calculations. Unlike the abacus (which was a later development in China, appearing around the 2nd century CE and becoming dominant by the 14th century), the counting board allowed much more sophisticated manipulations, including the representation of polynomials and matrices.

The rod numeral system was fully positional and decimal. Each column on the counting board represented a power of 10, and the alternating vertical/horizontal convention provided a visual separator between adjacent digits. For example, the number 60,390 would be laid out as:

Red rods represented positive numbers and black rods represented negative numbers — a convention introduced in Chapter 8 of the Nine Chapters. This is the earliest known systematic use of negative numbers in any civilization, predating Indian and European recognition of negative numbers by centuries.

The counting board enabled all four arithmetic operations with remarkable efficiency. Addition and subtraction were performed by physically adding or removing rods. Multiplication used a process similar to our long multiplication. Division was performed by a "galley" method that is essentially identical to modern long division. The rod calculus also enabled the extraction of square roots and cube roots through digit-by-digit algorithms that are equivalent to the modern methods.

The physical nature of the counting board had important mathematical consequences. Unlike written algorithms (which preserve intermediate steps), the counting board was constantly overwritten during computation — each step erased the previous configuration. This meant that Chinese mathematicians developed a deeply procedural approach to mathematics: the emphasis was on the algorithm (the sequence of operations) rather than on the static representation of intermediate results.

3.3 Gaussian Elimination — The Fangcheng Method

Chapter 8 of the Nine Chapters, titled Fangcheng ("rectangular arrays"), contains what is perhaps the most remarkable mathematical anticipation in the ancient world: a complete method for solving systems of simultaneous linear equations using matrix operations — essentially identical to what is now called Gaussian elimination. This method was developed in China approximately 2,000 years before Carl Friedrich Gauss (1777–1855) rediscovered it in Europe.

The method works by arranging the coefficients of a system of equations in a rectangular array (matrix) on the counting board, and then performing row operations to reduce the system to triangular form, from which the solution can be read off by back-substitution.

The Fangcheng procedure then eliminates variables systematically:

The solution is $x = 9\frac{1}{4}$, $y = 4\frac{1}{4}$, $z = 2\frac{3}{4}$ dou per bundle. We can verify:

$$3(9.25) + 2(4.25) + 1(2.75) = 27.75 + 8.5 + 2.75 = 39 \;\;\checkmark$$

Chapter 8 contains 18 problems of this type, involving systems of up to 5 equations in 5 unknowns. The method handles both well-determined and underdetermined systems (the latter yielding families of solutions). It also handles systems with negative coefficients, using the red/black rod convention for signs.

The rules for operating with positive and negative numbers are stated explicitly in theNine Chapters:

3.4 The Chinese Remainder Theorem

The Chinese Remainder Theorem (CRT) is one of the fundamental results of number theory, with applications ranging from ancient calendar computation to modern cryptography (it is a key component of the RSA algorithm). Its earliest known statement appears in the Sunzi Suanjing (Sun Zi's Mathematical Manual, c. 3rd–5th century CE).

Sun Zi's method was more heuristic than the general algorithm above. He gave the rule as a verse (paraphrased): "If the remainder on dividing by 3 is $a_1$, multiply by 70. If the remainder on dividing by 5 is $a_2$, multiply by 21. If the remainder on dividing by 7 is $a_3$, multiply by 15. Add these and subtract 105 as many times as needed." That is:

$$x \equiv 70a_1 + 21a_2 + 15a_3 \pmod{105}$$

Note that $70 = 2 \times 35 = M_1 y_1 \cdot a_1/a_1$ (the magic multiplier for the mod-3 condition), and similarly $21 = M_2 y_2$ and $15 = M_3 y_3$.

The CRT was later generalized by Qin Jiushao (1202–1261) in his Shushu Jiuzhang (Mathematical Treatise in Nine Sections, 1247), where he gave a systematic method (the "Great Extension" or Dayan method) for solving CRT problems with moduli that are not necessarily pairwise coprime. Qin Jiushao's algorithm includes a method equivalent to the extended Euclidean algorithm for finding multiplicative inverses. This was the most sophisticated number-theoretic algorithm anywhere in the world until Euler's work in the 18th century.

3.5 Liu Hui's Algorithm for Pi (263 CE)

Liu Hui (c. 225–295 CE) is one of the greatest mathematicians of the ancient world. His commentary on the Nine Chapters (263 CE) transformed Chinese mathematics by providing rigorous justifications for the algorithms in the text and by developing new methods of his own. His most celebrated achievement is his algorithm for computing $\pi$ using inscribed regular polygons — a method known in the West as the method of exhaustion, attributed to Archimedes (c. 250 BCE) but developed independently by Liu Hui.

The critical formula for doubling the number of polygon sides involves computing the side length of a $2n$-gon from the side length of an $n$-gon. If $s_n$ is the side length of a regular $n$-gon inscribed in a circle of radius $r$, then:

Liu Hui also developed an ingenious acceleration technique. He noticed that the errors in successive polygon approximations decrease by approximately a factor of 4 with each doubling. If $A_n$ and $A_{2n}$ are the areas of the $n$-gon and $2n$-gon, then:

$$\pi \approx A_{2n} + \frac{A_{2n} - A_n}{3}$$

This is a form of Richardson extrapolation (developed in the West in 1911), and it allowed Liu Hui to extract additional digits of accuracy from the same computational effort. Applied to the 96-gon and 192-gon approximations, this gives $\pi \approx 3.141592$ — six correct digits from polygons that individually give only four or five.

Liu Hui expressed his final result as: $\pi \approx 3.1416$, and he also gave the simplified approximation $\pi \approx \frac{157}{50} = 3.14$ for everyday calculations.

3.6 Zu Chongzhi (429–500 CE)

Zu Chongzhi was one of the greatest mathematicians and astronomers of ancient China. Building on Liu Hui's method, he pushed the computation of $\pi$ to an extraordinary seven decimal places:

$$3.1415926 < \pi < 3.1415927$$

This accuracy was not surpassed anywhere in the world for nearly a thousand years — until al-Kashi in Samarkand computed $\pi$ to 16 decimal places in 1424.

Zu Chongzhi must have used polygons with at least 12,288 ($= 3 \times 2^{12}$) or perhaps 24,576 sides, requiring hundreds of steps of the side-doubling formula — each step involving square root extractions to many digits. The sheer computational labor, performed entirely with counting rods on a counting board, is staggering.

In Europe, this fraction was not discovered until Adriaan Metius published it in 1585 (and it is sometimes called "Metius's fraction"), more than a thousand years after Zu Chongzhi.

Cavalieri's Principle — 1,100 Years Early

Zu Chongzhi and his son Zu Geng (Zu Gengzhi) are also credited with discovering what is known in the West as Cavalieri's Principle (stated by Bonaventura Cavalieri in 1635). Their version states:

Zu Geng used this principle to derive the volume of a sphere. The key idea is to compare a hemisphere of radius $r$ with a cylinder of radius $r$ and height $r$ from which an inverted cone has been removed. At height $h$ above the base:

• The hemisphere has a circular cross-section of radius $\sqrt{r^2 - h^2}$, so its area is $\pi(r^2 - h^2)$.
• The cylinder-minus-cone has a circular annular cross-section with outer radius $r$ and inner radius $h$, so its area is $\pi(r^2 - h^2)$.

Since the cross-sections are equal at every height, the volumes are equal. The volume of the cylinder-minus-cone is:

$$\pi r^3 - \frac{1}{3}\pi r^3 = \frac{2}{3}\pi r^3$$

This is the volume of a hemisphere, so the volume of a full sphere is:

$$V = \frac{4}{3}\pi r^3$$

This derivation is identical in logic to the one Cavalieri and his European successors would use more than eleven centuries later. It represents a profound application of proto-integral reasoning — comparing volumes by comparing infinitely many cross-sectional slices.

3.7 The Pascal Triangle in China

The triangular array of binomial coefficients known in the West as "Pascal's Triangle" (after Blaise Pascal, who published it in 1665) was known in China centuries earlier. It was first described by Jia Xian (c. 1010–1070 CE) and later popularized by Yang Hui (c. 1238–1298 CE), after whom it is called the Yang Hui Triangle (Yang Hui Sanjiao) in Chinese mathematics.

Jia Xian used the triangle in connection with his method for extracting higher-order roots. To find the $n$-th root of a number, one needs the binomial expansion:

$$(a + b)^n = \sum_{k=0}^{n} \binom{n}{k} a^{n-k} b^k = \binom{n}{0}a^n + \binom{n}{1}a^{n-1}b + \cdots + \binom{n}{n}b^n$$

Yang Hui, in his Xiangjie Jiuzhang Suanfa (Detailed Analysis of the Nine Chapters, 1261), published a diagram of the triangle through row 6 and explicitly attributed it to Jia Xian. Yang Hui also applied the triangle to problems in combinatorics and permutations.

The triangle was independently discovered in multiple civilizations:

The culmination of the Chinese algebraic tradition came with Zhu Shijie (c. 1249–1314), whose masterwork Siyuan Yujian (Precious Mirror of the Four Elements, 1303) dealt with systems of polynomial equations in up to four unknowns. Zhu Shijie published the triangle through row 8 and used it systematically in his work on polynomial algebra. He also discovered formulas for sums of series, including:

$$\sum_{k=1}^{n} k = \frac{n(n+1)}{2}, \qquad \sum_{k=1}^{n} k^2 = \frac{n(n+1)(2n+1)}{6}, \qquad \sum_{k=1}^{n} k^3 = \left(\frac{n(n+1)}{2}\right)^2$$

3.8 Legacy and Comparative Perspective

Chinese mathematics developed an approach fundamentally different from the Greek deductive tradition. Where Greek mathematics prized proof from axioms, Chinese mathematics prized algorithm and computation. Where Euclid sought to demonstrate eternal truths about abstract geometric objects, the Nine Chapters sought to provide effective methods for solving practical problems. Both approaches are essential to modern mathematics, and the Chinese algorithmic tradition has proven especially relevant to the computer age.

Major Chinese Innovations and Their Western Counterparts

The pattern is striking: in algorithm design and computational mathematics, Chinese mathematicians consistently anticipated Western discoveries by centuries and sometimes millennia. This is not coincidental — it reflects the algorithmic orientation of the Chinese tradition. When your primary goal is to compute answers rather than prove theorems, you naturally develop sophisticated computational methods.

Influence on East Asian Mathematics

Chinese mathematical works were transmitted throughout East Asia. The Nine Chapters was studied in Korea and Japan, where it formed the basis of mathematical education for centuries. The Japanese tradition of wasan (Japanese mathematics), which flourished in the 17th–19th centuries, has its roots in Chinese mathematical texts and methods.

Korean mathematics similarly drew heavily on Chinese sources. The civil service examination system in both Korea and Japan included mathematics as a subject, with the Nine Chapters and its commentaries as the core curriculum — a pattern modeled on the Chinese imperial examination system, which had included mathematical testing since the Tang Dynasty (618–907 CE).

Why Chinese Mathematics Was Underappreciated

For much of the 20th century, the history of mathematics was written from a predominantly European perspective, and Chinese contributions were undervalued or ignored. Several factors contributed to this:

• Language barriers: The primary sources are in classical Chinese, which few Western historians of mathematics could read.
• Different standards of rigor: Chinese mathematics did not employ the axiomatic-deductive framework that Western historians considered the hallmark of "real" mathematics. (Liu Hui's proofs, while rigorous, were not presented in the Euclidean format.)
• Lack of continuity: Chinese mathematics experienced periods of decline (particularly during the Qing Dynasty, when the examination system de-emphasized mathematics), which obscured its earlier achievements.
• Transmission gaps: Unlike Indian mathematics, which was transmitted to Europe through the Islamic world, Chinese mathematical ideas did not have a clear transmission pathway to the West. The independent discoveries were therefore not recognized as prior.

The work of Joseph Needham (Science and Civilisation in China, 1959), Jean-Claude Martzloff (A History of Chinese Mathematics, 1987/1997), and Karine Chemla (commentary on the Nine Chapters, 2004) has done much to correct this imbalance. Today, Chinese mathematics is recognized as one of the great independent mathematical traditions of the ancient world, distinguished by its algorithmic sophistication, its early use of matrix methods and negative numbers, its pioneering work in number theory, and its remarkable computational achievements in approximating $\pi$.

The Chinese mathematical tradition demonstrates a profound truth: mathematics is not a single stream flowing from Greece to Rome to Europe to the modern world. It is a vast river with many tributaries, and the Chinese tributary — with its emphasis on algorithm, computation, and practical problem-solving — is one of the mightiest.