Brahmagupta's formula, and Heron's as the case where a side vanishes

What to assume they know

• Heron's formula: area from the three sides alone — Heron's formula, the semi-perimeter, and why three sides determine a triangle's area
• From rectangle to parallelogram: area survives rearrangement — that a parallelogram's sides do not determine its area
• Area of a trapezium as the mean of the two parallel sides multiplied by the height
• Baudhāyana–Pythagoras theorem, for extracting a trapezium's height
• That three points not in a line lie on exactly one circle
• Substituting a value into an algebraic identity to obtain a special case

What they should be able to do

• Demonstrate that four given side lengths do not determine a quadrilateral's area, by hinging or from the chapter's own three figures
• List the kinds of extra information that would make the area determinate
• Define a cyclic 4-gon and state Brahmagupta's formula for it
• Compare Brahmagupta's formula with Heron's and identify the structural resemblance
• Verify Brahmagupta's formula in the rectangle case
• Verify it in the isosceles trapezium case, and reconcile the result with the standard trapezium formula
• Explain what "special case" and "generalisation" mean, using the chapter's own examples
• Derive Heron's formula from Brahmagupta's by setting the fourth side to zero
• Explain why that derivation is a consistency check rather than a proof
• State which quadrilateral of given side lengths Brahmagupta's formula actually measures, and check it against the chapter's own rhombus figures

Where it usually goes wrong

• "Brahmagupta's formula works for any quadrilateral." It works for cyclic ones. Fed a non-cyclic 4-gon it returns a number, and the number is wrong — it is the area of a different 4-gon with the same four sides. This is the most consequential error available in the topic and the chapter's Fig. 6.27 is there to forestall it.
• "Four sides ought to determine the area, since three do for a triangle." Triangles are rigid and quadrilaterals hinge. Build the hinge from four strips of card; nothing else convinces as fast.
• "The rhombus figures have different areas because they were drawn badly." They were drawn in GeoGebra and measured by it (p. 137), and all three genuinely have sides 3, 3, 3, 3. The differing areas are the mathematics, not a drawing error.
• "Heron's formula is a corollary of Brahmagupta's, so Brahmagupta proves Heron." The substitution shows the two are consistent and explains the resemblance. It is not a proof of Heron, and the chapter does not claim it is — its word is that Brahmagupta's may be viewed as a generalisation.
• **"In the trapezium example, s − a is one of the brackets."** The sides there are 2a and 2b, so the brackets are s − 2a and s − 2b. The doubled letters are chosen to keep the height clean and they set exactly this trap.
• "Generalisation means making something vaguer." It means removing a condition, so that the result covers more cases. The chapter's panel on p. 139 makes the point with three worked pairs and it is worth borrowing all three.
• "Any four lengths can be arranged into a cyclic 4-gon." They can, provided each is shorter than the other three together — which is exactly the condition that keeps all four Brahmagupta brackets positive. The parallel with Heron's three brackets is exact and neither is stated in the book.

Questions to check understanding

• Area of a cyclic 4-gon from its four sides
• Verify Brahmagupta's formula on a named special case: rectangle, square, isosceles trapezium
• Given four side lengths, decide whether they can bound a 4-gon at all
• Explain why four side lengths do not determine a 4-gon's area, with an example
• Recover Heron's formula from Brahmagupta's, and state what the recovery does and does not establish
• Identify a general result and its special case, and name the condition that separates them
• Prove that joining the midpoints of a 4-gon's sides halves the area

Examples worth working on the board

Inputs, not answers, except where the chapter itself prints the result. Values marked Verified are worked out here; the chapter prints no answers and this volume has no appended answer key.

• Fig. 6.27 (p. 137). Three 4-gons in a row, all with side lengths 3, 3, 3, 3 — so all rhombuses — and all labelled A, D across the top and B, C below. The first is upright and its printed area is 9. The second leans and reads 8.01. The third leans further and reads 5.41. The chapter says it drew them in GeoGebra and measured them with the software's area tool, and invites the reader to repeat the experiment or to build the figures from four rods joined at their ends.
• The reading the chapter does not spell out. Verified: the first figure, at area 9, is the square. So the three pictures are one hinge caught at three openings, and the area falls monotonically as it closes.
• What extra information would suffice (p. 137). The chapter lists one angle of the 4-gon, or how long one of the diagonals is, or how steeply the two diagonals cross — and then adds that a geometric property of the figure could serve instead.
• Fig. 6.28 (p. 138). A cyclic 4-gon ABCD inscribed in a circle, shaded, with D at the top, A on the left, B at the lower left and C at the right, and the four sides labelled a = AB, b = BC, c = CD and d = DA.
• Brahmagupta's formula as printed (p. 138). With the four sides a, b, c, d and s half their total, the area is the square root of the product of the four quantities s − a, s − b, s − c and s − d. Brahmagupta is dated 628 CE for the discovery, with his lifetime given as 598–668 CE.
• Brahmagupta beside Heron. Heron has s out in front and three brackets; Brahmagupta has four brackets and no free s. Showing them one above the other is the whole of section 5.
• Example 6 (p. 138), the rectangle. All rectangles are cyclic, so the formula must apply. Sides a, b, so s = a + b. Verified: each bracket is either a or b, two of each, so the product is a²b² and the area is ab. The chapter then invites the reader to try other special cases.
• Example 7 (pp. 138–139), the isosceles trapezium. Fig. 6.29 shows the trapezium with the shorter parallel side AB = 2a on top, the longer DC = 2b below, both slanting sides c, and the height h marked at the right by a dashed double arrow. All isosceles trapezia are cyclic, so the formula applies. Perimeter 2a + 2b + 2c gives s = a + b + c. Verified: the four brackets are c + b − a, c + a − b, a + b and a + b, so the area is (a + b)√((c + b − a)(c + a − b)), which the chapter writes as (a + b)√(c² − (b − a)²).
• The reconciliation (p. 139). Dropping a perpendicular from B to the base gives height √(c² − (b − a)²), so the formula reads (a + b)h; and half the sum averaging the two parallel sides and multiplying by the height gives the same thing. Verified.
• A numerical instance, not in the book. Isosceles trapezium with parallel sides 6 and 14 and both slant sides 5. In the chapter's letters that is a = 3, b = 7, c = 5, because the parallel sides are written 2a and 2b. Verified: s = 15; the four brackets are s − 2a = 9, s − 2b = 1 and s − c = 10 twice; the product is 900 and the area is 30. Cross-check by the trapezium route: the overhang at each end is 4, the height is √(25 − 16) = 3, and half of 6 + 14 times 3 is again 30.
• The trap to walk into on purpose. A student who forgets that the parallel sides are 2a and 2b will use s − a = 12 and s − b = 8, get a product of 9600, and report an area of about 98 for a trapezium that plainly cannot hold more than 14 × 3. Verified. Show the wrong answer, show that it fails a sanity check against the bounding rectangle, and then fix it. The doubled labels exist to keep the height clean and they set exactly this trap.
• The boxed panel on special cases (p. 139). Its printed examples: a square as a special case of a rectangle, so the square's area formula comes from the rectangle's by making the two sides equal, and likewise for the perimeters; an isosceles right-angled triangle as a special case of a right-angled triangle, so that the Baudhāyana–Pythagoras relation reduces to the hypotenuse being √2 times a leg; and the square of a two-term sum as the c = 0 case of the square of a three-term sum.
• The vanishing side (p. 140). A triangle of sides a, b, c is treated as a 4-gon ABCD whose fourth side has length zero, so that A and D coincide. Any triangle is cyclic, because three points not in a line lie on exactly one circle, so Brahmagupta's formula applies. Verified: with d = 0 the semi-perimeter is the triangle's own, the fourth bracket becomes s, and the formula is Heron's. The chapter's own verdict is that Brahmagupta's result can be read as the wider one, and it adds a colliding-planets aside about the two coinciding vertices.
• Why that is a check and not a proof, which the chapter does not say. Brahmagupta's formula was established for genuine cyclic quadrilaterals. A figure with a zero-length side is not one, so the substitution d = 0 is a limiting statement, and it is legitimate only because the area varies continuously as d shrinks. That is true and it is not a triviality, and a student who is shown the substitution as a derivation has been shown a pattern rather than an argument.
• Section 12, the bonus, and it is worth the whole video. Verified: feed 3, 3, 3, 3 into Brahmagupta's formula. Then s = 6, every bracket is 3, the product is 81 and the area is 9 — exactly the first of the chapter's three printed rhombus areas, and the largest of the three. So the cyclic member of that hinged family is the square, and Brahmagupta's formula is returning the maximum area the four sides can enclose. The chapter prints all the numbers needed to see this and never connects them. This connection is added here.
• Exercise Set 6.2 Q8 (p. 142), with its answer deferred by the book itself. Join the midpoints of a 4-gon's sides in order and prove the parallelogram so formed has half the original's area. The question adds, in brackets, that whether the new figure is always a parallelogram is settled in a later chapter on quadrilaterals. Worth including as the chapter's own honest signpost.

Figures to have open

• A physical four-strip hinged rhombus, sides pinned at equal length, opening and closing with a live area readout. The chapter suggests four rods joined at their ends (p. 137) and this is the figure the whole topic turns on.
• Fig. 6.27 (p. 137), redrawn as the three positions of that one hinge with the printed areas 9, 8.01 and 5.41 attached. The printed figure gives three separate drawings and the areas are set inside the artwork. Presenting them as one hinge rather than three unrelated figures is the explanation's improvement and it is a large one.
• Fig. 6.28 (p. 138), redrawn: a cyclic 4-gon with its circumcircle and the four sides labelled.
• Fig. 6.29 (p. 138), redrawn with the doubled labels 2a and 2b kept, the equal slant sides c marked, the height h dropped from a top vertex to the base, and the overhang b − a explicitly marked. The overhang is not labelled in the printed figure and it is what the surd is made of.
• A side-by-side of Heron's and Brahmagupta's formulas with the fourth bracket highlighted as it turns into s. Not in the book.
• An accuracy-free comparison panel for section 12: the hinge's area against the opening angle, with the maximum marked where the figure is cyclic. Not in the book, and it can be schematic.

Where this sits in the book

• NCERT Ganita Manjari, Class 9 Mathematics (NCF-SE 2023), Chapter 6, the unnumbered block "Brahmagupta's Formula for the Area of a Cyclic 4-gon" (pp. 137–139) with Examples 6 and 7, and the following unnumbered block "Brahmagupta's Formula Generalises Heron's Formula" (p. 140).
• The boxed panel "Special Cases and Generalisation in Mathematics" (p. 139) in full.
• Figures 6.27 (p. 137), 6.28 and 6.29 (p. 138).
• Exercise Set 6.2 Q8 (pp. 142–143), with the chapter's own forward reference to a later chapter on quadrilaterals.
• Chapter Summary (p. 154), final bullet, which states Brahmagupta's formula and uses the word quadrilateral where the body of the chapter says 4-gon.
• Heron's formula and the semi-perimeter are §6.8.1 (pp. 134–137), handled in Heron's formula: area from the three sides alone; the parallelogram version of the same sides-do-not-determine-area point is §6.7 (pp. 131–132), handled in From rectangle to parallelogram: area survives rearrangement.
• Brahmagupta appears earlier in this chapter as the source of √10 as a value for π (p. 122), and elsewhere in this book in Chapter 3's rules for zero. Both are the same mathematician, dated 628 CE in both places.