Regula falsi

The regula falsi, also known as the method of false position, is a family of mathematical algorithms. These algorithms are used to solve linear equations or to find approximate solutions for nonlinear equations. In mathematics, solving an equation means finding a single unknown value that makes the statement true. This value is often called a root or a zero. While simple algebra can solve many problems directly, complex equations often require a more iterative approach. The regula falsi method replaces random trial and error with a structured way to correct guesses.

Historically, the method began as a way to handle direct proportions. This is known as simple false position. In this version, a person makes an initial guess, which we can call $x_1$. They then calculate the resulting output, $y_1$, through multiplication. If the output is not the desired target, the user calculates a proportional adjustment. For example, the Rhind papyrus contains a problem where one must find a value to reach a specific result. If a guess produces a value three times too small, the user simply multiplies the guess by three to find the correct answer. This turns a single mistake into a direct path to the truth.

As mathematics progressed, the more complex double false position emerged. This version is mathematically equivalent to linear interpolation. Instead of one guess, it uses a pair of test inputs, $x_1$ and $x_2$, along with their corresponding outputs, $y_1$ and $y_2$. The algorithm calculates a new estimate, $x_3$, by finding where a straight line connecting these two points crosses the horizontal axis. The formula for this estimate is $x_3 = x_1 - y_1 \cdot \frac{x_2 - x_1}{y_2 - y_1}$. For an affine linear function, this method provides the exact solution immediately. For nonlinear functions, it provides an approximation that can be improved through iteration.

The history of these methods spans many ancient civilizations. Simple false position appears in cuneiform tablets from Babylon and in Egyptian papyri. Double false position developed later as a purely arithmetical algorithm. The ancient Chinese text, The Nine Chapters on the Mathematical Art, dated between 200 BC and AD 100, devoted much of Chapter 7 to this procedure. Chinese mathematicians applied it to diverse story problems, including those involving conic sections. One famous example is a "joint purchase" problem. It describes a situation where a group of people contributes different amounts of coins, resulting in an excess or a deficit. By using the method, they determined there were 7 people and the item price was 53.

In the Middle East, the method was known as "reckoning by two errors," or *hisāb al-khaṭāʾayn*. The Egyptian mathematician Abu Kamil wrote a treatise called the Book of the Two Errors. Later, in the 10th century, the Lebanese mathematician Qusta ibn Luqa provided a formal geometric proof for the technique. This method was used for centuries to solve practical legal and commercial questions, such as partitioning estates. It was even memorized using mnemonics and verses. In 1202, the Italian mathematician Fibonacci included the method in his book *Liber Abaci*. He called it *regulis elchatayn* after learning it from Arab sources.

In modern numerical analysis, the method is used as a root-finding algorithm. When an equation is difficult, we move all terms to one side to create a function, $f(x)$. We seek the value where $f(x) = 0$. If we find two points where the function has opposite signs, the intermediate value theorem guarantees a root exists between them. This is called bracketing. The regula falsi method is a two-point bracketing method. It creates a sequence of shrinking intervals that contain the root. This is different from the bisection method, which simply picks the midpoint of an interval. Instead, regula falsi uses a secant line to pick a more informed estimate.

While powerful, the method has specific behaviors depending on the function's shape. For nonlinear functions, the method uses a secant or chord to approximate the curve. If the second derivative of the function has a constant sign, one endpoint of the interval may stay fixed while the other moves toward the root. This can sometimes make the method slower than the bisection method. However, it remains a fundamental concept in the study of iterative numerical approximation. It serves as a bridge between simple arithmetic and the complex computational tools used to solve the world's most difficult equations.