Hero and the tradition of the circle segment (Q6171825)

This paper presents in great detail some answers to several questions raised by the four procedures for finding the approximate area of a circular segment Hero provides in his \textit{Metrica}. In particular, regarding the formula (what Hero refers to as ``the Ancient method''): \[ \frac{1}{2}(b+h)h,\tag{\(\ast\)} \] where \(b\) denotes the base of the segment and \(h\) its height, the author establishes its use in Uruk, as documented by the late 5th century BC tablet W 23291-x, a tablet that is the background for the same method mentioned in the 3rd century BC Demotic Egyptian treatise assembled by \textit{R. A. Parker} [Demotic mathematical papyri. Providence, RI: Brown University Press; London: Lund Humphries (1972; Zbl 0283.01001)] from P. Cairo 89127-30, 89137-43. He also presents ``some grounds for why it would be seen as plausible'', given that the formula is very far from good for small values of \(h\), when compared with \(b\). The author also refers to ``some startling coincidences'' (p.\ 452). An algorithm in the Old Babylonian tablet BM 85194 is also considered relevant. There are also attempts at reconstructing the exploratory path by which the procedures might have been arrived at. They are summarized in two theorems. With \textit{segment$_1$} and \textit{segment$_2$} standing for the areas of adjacent segments on a rectangle inscribed in a circle, $h_1$ and $b_1$ standing for the height and base of \textit{segment$_1$} and $h_2$ and $b_2$ for the height and base of \textit{segment$_2$}, it is shown that, (i) with $\pi=3$, \textit{segment$_1$}$+$ \textit{segment$_2$} $= \frac{1}{2}(b_1 + h_1)\cdot h_1 +\frac{1}{2}{b_2 + h_2}\cdot h_2$, and that (ii) \textit{segment$_1$}$+$ \textit{segment$_2$} $= \frac{1}{2}(b_1 + h_1)\cdot h_1 + (\frac{b_1}{2})^2 \cdot \frac{\pi-3}{2}+ \frac{1}{2}(b_2 + h_2)\cdot h_2 + (\frac{b_1}{2})^2 \cdot \frac{\pi-3}{2}$. By taking $\pi$ to be $3\frac{1}{7}$, (ii) becomes \textit{segment$_1$}$+$ \textit{segment$_2$} $= \frac{1}{2}(b_1 + h_1)\cdot h_1+ \frac{(\frac{b_1}{2})^2}{14}+ \frac{1}{2}(b_2 + h_2)\cdot h_2 + \frac{(\frac{b_2}{2})^2}{14}$. This allows the author to vindicate Hero, for he was right in stating that the Ancient method is tied to taking $\pi=3$ and that the Revised method to taking $\pi = 3\frac{1}{7}$. ``As to the Ancient method for finding the area of the segment, well, it works perfectly on a semicircle, perfectly on a side of a square, and fairly well on one figure in between, the equilateral triangle. Therefore, it works fairly well.'' (p.\ 469) This positive tone is at odds with the language used by Van der Waerden, for whom the fact of the same \textit{inaccurate} formula appears in the Chinese \textit{Nine chapters}, in Hero's \textit{Metrica}, and in Egypt is assigned a great amount of significance for the hypothesis, put forward by \textit{A. Seidenberg} [Arch. Hist. Exact Sci. 18, 301--342 (1978; Zbl 0392.01002)], of a single origin for mathematics. We find, on page 40 of [\textit{B. L. van der Waerden}, Geometry and algebra in ancient civilizations. Berlin: Springer-Verlag (1983; Zbl 0534.01001)], the following (more on the matter of other area formulas found in the \textit{Metrica} can be found on pages 184--186 of the same book): ``We find it very curious that the same inaccurate formula \((\ast)\) also occurs in the ``Metrica'' of Heron of Alexandria [\(\ldots\)] and in a papyrus from Cairo written in the third century B.C. and published by R. A. Parker. [\(\ldots\)] As a general principle, if one finds one and the same \textit{correct} rule of computation in several civilizations, one always has to take into account the possibility of independent invention, but if the rule is \textit{incorrect}, independent invention is next to impossible. Therefore we are bound to suppose that the Chinese formula [\(\ldots\)] and the equivalent formula \((\ast)\) used in the Cairo papyrus and mentioned by Heron, were derived from a common origin. According to Parker, the Demotic papyrus shows clear traces of Babylonian influence. The problems and the solutions in the papyrus are just of the same kind as those we find in Babylonian problem texts. However, the false rule \((\ast)\) is not found in any extant Babylonian text, as far as I know.'' In this sense, the current paper answers the question left open by Van der Waerden, by detecting a Babylonian text in which \((\ast)\) can be found. This had not been noted before. \textit{J. Friberg} [Unexpected links between Egyptian and Babylonian mathematics. Hackensack, NJ: World Scientific (2005; Zbl 1128.01002)] found, not long ago, on page 133, that ``In Babylonian mathematics, the use of the rule is not documented''.