0% found this document useful (0 votes)
128 views9 pages

Solution of The Cubic: A Simple Version of Formula

This document discusses the history of solving cubic equations. It explains that while the Babylonians were familiar with solving quadratic equations, it took until the 16th century for mathematicians like Cardano to develop a general solution for cubics. The key events were that Scipione del Ferro first solved certain cases of the cubic equation around 1510, but kept his method secret. He passed the solution along to his student Antonio Fior on his deathbed. Fior later shared the solution with Cardano, who published the first general solution for cubics in 1545 using modern algebraic notation and symbols.

Uploaded by

Shubham Sharma
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
128 views9 pages

Solution of The Cubic: A Simple Version of Formula

This document discusses the history of solving cubic equations. It explains that while the Babylonians were familiar with solving quadratic equations, it took until the 16th century for mathematicians like Cardano to develop a general solution for cubics. The key events were that Scipione del Ferro first solved certain cases of the cubic equation around 1510, but kept his method secret. He passed the solution along to his student Antonio Fior on his deathbed. Fior later shared the solution with Cardano, who published the first general solution for cubics in 1545 using modern algebraic notation and symbols.

Uploaded by

Shubham Sharma
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

GENERAL I ARTICLE

Solution of the Cubic


A Simple Version of Cardano/s Formula

Jasbir S Chahal

1. Introduction

Cardano's formula for solving a cubic is the crowning


achievement of renaissance mathematics. Yet, it does
not receive the same recognition in our curricula as does
the quadratic formula, which was discovered long be-
fore it. It is rather surprising that there have not been
Jasbir S Chahal is
attempts to simplify further the messy formulas of Car- Professor of Mathematics
dano (see ([1],pp.606-616), or ([2], pp.187-189)) to a at Brigham Young
form that would be easier for the students to remember. University, Provo, Utah,
Apparently the messy nature of the formulas for solv- USA. His research
interest is in number
ing the cubic is a reason for the lack of their popularity.
theory.
Another reason could be the Galois theory, which mod-
ern authors use in their exposition of Cardano's formula.
We show that a simple trick, namely a rescaling of the
Part 1. Foundation of Basic Arith-
discriminant, reduces not only the formula to a simpler metic, Resonance, Vol.l1, No.1,
form , but also its verification to a trivial calculation, January 2006.
with no reference to Galois theory. Although Galois Part 2. Foundation of Basic
theory is an indispensable tool in algebra and number Arithmetic, Resonance, Vol.ll,
No .2, February 2006.
theory, it is not necessary to wait until one learns it,
Part 3 . Foundations of Basic
for Cardano's formula. Cardano's formula can be intro- Arithmetic, Resonance, Vol.",
duced in a first course on complex numbers. No.7, July 2006 .

By the celebrated theorem of Abel-Ruffini, a general


equation of degree five or more is not solvable by radi-
cals, whereas solving a quartic equation can be reduced
to solving a cubic equation. Thus Cardano's formula
filled the essential gap in our understanding of the so-
lu tions of polynomial equations. The purpose of this
article is to present a lowbrow exposition of Cardano 's
formula than that found in the literature and to tell the
Keywords
story behind its discovery in order to put the matter in Cubic, discriminant, roots of
a proper historical perspective. unity, radicals .

-RE-S-O-N-A-N-C-E-I-A-U-9-U-st--2-00-6------------~-----------------------------~-
GENERAL I ARTICLE

There are reasons to 2. History


believe that the
There are reasons to believe that the Babylonians of
Babylonians of 2000 2000 BC were familiar with solving quadratic equations,
Be were familiar with albeit neglecting the negative solutions. Much later,
solving quadratic Brahmagupta (c.628 AD) and then al Khwarizmi (780-
equations. 850 AD) described the quadratic formula more or less
as we know it today.

The next step was to solve the cubic. The Arabs and
the Chinese worked out special cases of the cubic numer-
ically. But it took almost a thousand years from Brah-
magupta's time to find a general solution to the cubic,
often attributed to the Italian Cardano. The story of
its discovery is as dramatic as it can be in the world of
mathematics.

To solve any polynomial equation it suffices to take the


leading coefficient equal to 1. Moreover, the so called
Viete substitution X = x - A/3 reduces the cubic equa-
tion
x 3 + AX2 + BX + C = 0
to one of the form x 3 + ax + b = o.
Thus there is no loss of generality in assuming that the
general cubic equation has no square term.

Hindu, Islamic or even the Italian algebra of Cardano's


time was entirely rhetorical. There were no symbols for
an unknown or its powers. Everything was communi-
cated in words, and to facilitate memorization, formulas
were stated as verses. For example, here is part of a
verse (see [3J,p.36) for the equation x 3 + px = q:

Squeaxno, adtwix
Noesquax, adsub
Axesquno, subadsub

-~----------------------------~------------R-ES-O-N-A-N-C-E-I-A-U-9-U-st--20-0-6
GENERAL I ARTICLE

It was only after Cardano had published the solution of We owe our tradition
the cubic in 1545 that Francois Viete (1540-1603) intro- of using earlier
duced, in his book The Analytic Art, our present usage letters a, b, C, ... for
of letters to represent unknown quantities. He used vow- constants and the
els for variables and consonants for constants. However, later ones x, y, z, ...
we owe our tradition of using earlier letters a, b, c, for variables to
for constants and the later ones x, y, z, for variables
Descartes.
to Descartes. Viete had no symbol for equality. It was
Robert Recorde who introduced the symbol = for equal-
ity in 1557. The signs + and - appeared for the first
time in Germany at the end of the fifteenth century as
symbols for surplus and deficit in business records. In
1514, the Dutch mathematician Vander Hoecke became
the first to use them in algebraic expressions. Thomas
Harriot was the first (in 1631) to use a dot for multipli-
cation, and he is also responsible for the inequality signs
< and >. In the same year (1631) William Oughtred in-
troduced the cross sign x for multiplication. The square
root symbol V was invented by Christoff Rudolff (1510-
1558), though some historians dispute it. In 1655, John
Wallis was the first to use the symbol 00 for infinity,
probably suggested by the late Roman symbol 00 for a
millennium. For more, see [4].

Leibniz used these symbols in his calculus which was


popularized by the Bernoullis. The Bernoulli family had
great influence on Euler. Finally, it was Euler who uti- Thus the
lized these symbols throughout his writings and made mathematical
them the language of mathematics. Thus the mathe- symbols, which
matical symbols, which look very intimidating to many look very
people, are very recent phenomena. But they facilitated
intimidating to
great advances in mathematics.
many people, are
Arabs, and the early Europeans who were to take off very recent
from where the Arabs had left, did not consider negative phenomena. But
coefficients. Thus there were dozens of cases of the cubic they facilitated
equation to be considered. For example, the so-called great advances in
depressed form alone, with square term absent, was split mathematics.

-R-ES-O-N-A-N-C-E-I-A-U-9-US-t--20-0-6-----------~-----------------------------~
GENERAL I ARTICLE

But the academic into three cases:


life in sixteenth 3 3 3
centu ry Italy was
x + px + q = 0, x = px + q and x + px = q (1)
quite different. The with p, q > O.
most common way
for a professor to It was Scipione del Ferro (1465-1526), a professor at
stay in his position the University of Bologna, who was the first to find a
was to win public method of solving equations (1) somewhere around 1510.
contests. To insure priority, modern professors announce results
even before they have fully checked their proofs. But
the academic life in sixteenth century Italy was quite
different. There was no tenure. University appointments
were mostly temporary, subject to periodic renewal. The
most common way for a professor to stay in his position
was to win public contests. A new contender for his job
would exchange with him a list of problems to be solved
by the other in a specified amount of time. It was re-
quired that a solution to every problem submitted must
exist. Sometime later they would meet each other in
a public forum to present their solutions; so it was a
good strategy for professors to keep their discoveries se-
cret and use them for these public contests. Professor
del Ferro never had the occasion to use his solution for
such a contest and just before his death in 1526, secretly
passed it on to his student Antonio Fiore, as well as his
successor Professor della Nave (1500-1558) at the Uni-
versity of Bologna. Even though they never publicized
the solution, the news that someone had found a solution
to the cubic started to circulate among Italian mathe-
maticians. Another Italian Nicolo Tartaglia (1500-1557)
from Brescia boasted to have the solution. This was too
much for Fiore to take, so he challenged Tartaglia to a
public contest. All of Fiore's word problems required the
knowledge of a solution to the cubic equation. Having
no solution of the cubic yet, Tartaglia was thus trapped,
but during the time set aside he worked day and night
and just before the contest on the night of 12 February
1535, found the solution to the cubic. Having worked

-~----------------------------~------------R-ES-O-N-A-N-C-E-I-A-U-9-U-st--20-0-6
GENERAL I ARTICLE

out the solution himself, Tartaglia easily defeated Fiore Cardano no longer
who had inherited the solution from his teacher. felt an obligation to
Tartaglia as he would
At that time Girolamo Cardano (1501-1576) was lectur-
only be publishing the
ing in Milan on algebra. When he heard about Tarta-
glia's solution he wrote to Tartaglia. He wanted to see same solution found
the solution so that it could be included in his lec- independently some
tures on algebra. Tartaglia showed the solution only 25 years earlier by a
after extracting an oath from Cardano that it would mathematician now
not be included in Cardano's forthcoming book, even deceased.
with full credit to him. Tartaglia wanted to publish it
himself. Cardano kept his promise but assisted by his
brilliant student Lodovico Ferrari (1522-1565), started
working on the problem himself. Ferrari even managed
to solve the fourth degree equation. But their solutions
depended on reducing the problem to the cases solved
by Tartaglia.

Tartaglia still had not published anything. Cardano did


not want to break his promise to Tartaglia, but felt a
need to make the solution available to the public. Mean-
while, after hearing the rumor of the original solution by
della Ferro, Cardano and Ferrari visited Professor della
Nave in Bologna who graciously let them verify that
del Ferro indeed had the solution. Cardano no longer
felt an obligation to Tartaglia as he would only be pub-
lishing the same solution found independently some 25
years earlier by a mathematician now deceased. Thus
in 1545, Cardano published his most important work,
Ars Magna mainly devoted to the solution of the cubic.
When the book appeared, Tartaglia was furious, even
though Cardano had mentioned him as one of the orig- When the book Ars
inal discoverers of the solution. To recoup his prestige, Magna appeared,
Tartaglia challenged Ferrari to a public contest, but this Tartaglia was furious,
time he was defeated. To this day the method described even though Cardano
in Ars Magna of solving the cubic equation is called Car- had mentioned him
dano's Formula. We now explain it from a modern point as one of the original
of view, which unifies all the cases into a single formula. discoverers of the
For the original case-by-case discussion, see [3]. solution.

-R-ES-O-N-A-N-C-E-I-A-U-9-US-t--20-0-6-----------~-----------------------------~
GENERAL I ARTICLE

When we say a 3. Roots of a Complex Number


polynomial equation is
When we say a polynomial equation is solvable by rad-
solvable by radicals, icals, we mean the solutions can be found in terms of
we mean the solutions expressions involving the four algebraic operations on
can be found in terms the coefficients of the polynomial, and extracting their
of expressions square roots, cube roots, and so on. So let m > 1 be
involving the four an integer. We now indicate how to extract all the mth
algebraic operations roots of a non-zero complex number.
on the coefficients of
the polynomial, and
Any complex number z = x + iy i= 0 can be represented
geometrically as
extracting their square
roots, cube roots, and z = r ei() : = r (cos () + i sin ()) ,
soon.
J
where its modulus r = x 2 + y2 > 0 and the argument
Arg(z) of z is the angle () = tan- 1 (;). Let \Ii be the
positive real mth root of rand w be the mth root of
uni ty given by
27r 27r
W = cos - + i sin - .
m m

If we put a = \Ii ei()/m, then the mth roots of z are a,


wa, w 2 a, ,wm-1a (which are clearly distinct, hence
account for all of them). In particular, the ratio of any
two of them is an mth root of unity. For example, the
three cube roots of 8i are

for d = 0, 1, 2.

4. Cardano's Formula

Recall that the quadratic equation


2
x + bx + c = 0 (2)

has two solutions


-b ±.JE.
x=
2

~------------------------------~~----------R-E-S-O-N-A-N-C-E-I-A-U-9-US-t-2-0-0-6
GENERAL I ARTICLE

where the quantity .6. = b2 - 4c is called the discrim-


inant. The discriminant discriminates the solutions.
When there is no discriminant, that is, when .6. = 0,
the two roots are equal. In fact, the two roots are equal
if and only if .6. = o.

Exercise. Solve the quadratic equation (with complex


coefficients)

( 1- V; i) z2 - vis Z + 1 = O.

To solve the cubic we may assume, as has been said


earlier, that a general cubic equation is of the form
3
x + ax + b = O. (3)

Again we can define its discriminant D such that no two


solutions are equal if and only if D =1= O. To aid memory,
we make the formulas for the solutions of (3) resemble
as much as possible that of x 2 + bx + c = 0, which are

-b+~ -b-~
Xl = and X2 = .
2 2
For this, we modify the traditional definition D = - (4a 3 +
27b2 ) of the discriminant of x 3 + ax + b slightly. Our de-
finition of the discriminant .6. of x 3 + ax + b is

.6.=----
4a 3 + 27b 2
(4)
27
Let WI and W2 be the two imaginary cube roots of unity:

27rj 27rj
Wj = cos 3 + i sin 3 (j = 1,2).

It is easy to check that for W = WI or W2, 1 + W + w2 = O.


Moreover,

(5)

-RE-S-O-N-A-N-C-E-I-A-U-9-US-t-2-0-0-6------~---------------5-9
GENERAL I ARTICLE

Now using (4), it is easy to check that

-b + ~ -b - ~= -a 3 /27.
2 2

Choose cube roots

(6)

such that

(7)

Theorem. (del Ferro-Tartaglia-Cardano). The three


solutions of x 3 + ax + b = 0 are

The equations (6), (7) and (8) taken together are called
Cardano's formulas.

Proof. Plug each number from (8) in (3) and use (5),
(6) and (7).

Example. To illustrate Cardano's formula we take one


of the simplest examples, namely

x3 - 1 = o.
Here a = 0 and b = -1, so D = 1. From (6), we get
01 = 1 and 02 = o. It follows from (8) that the three
roots of x 3 - 1 are 1 + 0, WI 1 + W2 0, W2 1 + WI 0,
that is, they are 1, WI, W2. This agrees with what we
already know, that is 1,w,w 2 are the three solutions of
x 3 - 1 = (x - 1)(x 2 + X + 1) = 0, where W = WI or W2 is
a primitive cube root of unity.

Exercise. Solve the following cubic equations:

i) x 3 - 2x +4 = 0, ii) x 3 + x 2 - 2x - 1 = O.

-60----------------------------~------------RE-S-O-N-A-N-CE--I-A-U9-U-st--2-00-6
GENERAL I ARTICLE

Remark 1. Our choice of the cube roots Ql and Q2


is dictated by the proof. However, to find the three
solutions of (3), it is obvious that any choice of Ql and
Q2 will suffice, because the other choices just permute
the three numbers in (8).

Remark 2. If the reader is familiar with the group


structure on the points of the elliptic curve

(9)

with coordinates in C, together with the point 0 at in-


finity, the solutions of (3) are the x-coordinates of points
of order two on (9).

Suggested Reading Address for Correspondence


Jasbir S Chahal
[1] D S Dumit and R M Foote, Abstract A 1gebra, John Wiley, 2004. Professor of Mathematics
[2] B L van der Waerden, Algebra, Vol. I, Frederick Unger, 1970. Brigham Young University
[3] Girolamo Cardano,Ars Magna, Dover, 1993. USA
[4] F Cajori, A History ofMathematical Notations, Dover, 1993. Email:[email protected]
[5] D J Struik,A Concise History ofMathematics, Dover, 1948.

Errata
Resonance, Vol.lI, No 7, July 2006

Page 6: second paragraph, 4th line


.. , crescent shaped infected blood cells should read,
... crescent shaped parasites inside the infected blood cells ...

Page 42: The first two sentences should read,


The intensity of solar radiation in the Earth's direction from the Sun is
approximately 1.353 kw/m2, a number also called the 'Solar Constant'.
Accordingly, it is estimated that Earth receives about 96 billion kw
from the Sun constantly.'

Page 50: The author's email id:[email protected];


sh [email protected]

-RE-S-O-N-A-N-C-E--I-A-U-9-Us-t--2-00-6-------------~-------------------------------6-1

You might also like