Thinking The Unthinkable The Story of C
Thinking The Unthinkable The Story of C
Thinking The Unthinkable The Story of C
Birth
In 1545 Jerome Cardan, an Italian mathemati-
This story begins in 1545. What carne earlier can cian, physician, gambler, and philosopher, pub-
be summarized by the following quotation from lished a book entitled Ars Magna (The great art),
Bhaskara, a twelfth-century Hindu mathemati- in which he described an algebraic method for
cian (Dantzig 1967): solving cubic and quartic equations. This book was
a great event in mathematics. It was the first major
The square of a positive number, also that of a
negative number, is positive; and the square root of achievement in algebra since the time, 3000 years
a positive number is two-fold, positive and negative; earlier, when the Babylonians showed how to solve
there is no square root of a negative number, for a quadratic equations. Cardan, too, dealt with qua-
negative number is not a square. dratics in his book. One of the problems he pro-
posed is the following (Struik 1969):
If some one says to you, divide 10 into two parts, one
Reprinted from Mathematics TeacherS! (Oct., 1988): 583-92; with
permission of the National Council of Teachers of Mathematics.
of which multiplied into the other shall produce ...
The author would like to acknowledge financial assistance from the 40, it is evident that this case or question is impos-
Social Sciences and Humanities Research Council of Canada. sible. Nevertheless, we shall solve it in this fashion.
711
PART VIII/THE SEARCH FOR CERTAINTY
Cardan then applied his algorithm (essentially of x 3 = 15x + 4, found by using Cardan's formula,
the method of completing the square) to x + y = 10 with the solution x = 4, found by inspection. The
and xy = 40 to get the two numbers 5 + ,t=l5 and task was undertaken by the hydraulic engineer
5 - ..J-15. Moreover, "putting aside the mental Rafael Bombelli about thirty years after the publi-
tortures involved" (Burton 1985), Cardan formally cation ofCardan's work.
multiplied 5 + '-1-15 by 5 - '-1-15 and obtained 40. Bombelli had the "wild thought" that since the
He did not pursue the marter but concluded that radicands 2 + ..J-121 and 2 - ..J-121 differ only in
the result was "as subtle as it is useless" (NCTM sign, the same might be true of their cube roots.
1969). Although eventually rejected, this event Thus, he let
was nevertheless historic, since it was the first time
ever that the square root of a negative number was
explicitly written down. And, as Dantzig (1985) and
has observed, "the mere writing down of the im-
possible gave it a symbolic existence."
In the solution of the cubic equation, square and proceeded to solve for a and b by manipulating
roots of negative numbers had to be reckoned these expressions according to the established rules
with. Cardan's solution for the cubic x 3 = ax + b for real variables. He deduced that a = 2 and b = 1
was given as and thereby showed that, indeed,
= (2 + ..J-i) + (2- R) = 4
712
THINKING THE UNTHINKABLE: THE STORY OF COMPLEX NUMBERS (WITH A MORAL)
713
PART VIII/THE SEARCH FOR CERTAINTY
f 1
---dx
x 2 + a2
= f 1
(x + ai)(x- ai)
dx
=--
2ai
f
1 ( - 1- - - - )
1 dx
x+ai x-ai
1
2ai
[log (x + ai) -log (x- ai)]. This use, in tum, raised
questions about the meaning of the logarithm of
complex as well as negative numbers. A heated con- Leonhard Euler (1707-1783)
troversy ensued between Leibniz and Bernoulli. From A Portfolio ofEminent Mathematicians, ed. David Eugene
Leibniz claimed, for example, that log i = 0, arguing Smith (Chicago: Open Court, 1908)
that log ( -1) 2 =log 12, and hence 2 iog ( -1) = 2log correct proof of the fundamental theorem of alge-
1 = 0; thus log ( -1) = 0, and hence 0 =log( -1) =log bra, claimed as late as 1825 that "the true meta-
i' = 2 log i, from which it follows that log i = 0. physics of v=J. is elusive" (Kline 1972).
Bernoulli opted for log i = (1ti)/2; this equation It should be pointed out that the desire for a
follows from Euler's identity e" = - 1, which implies logically satisfactory explanation of complex num-
thatlog ( -1) = 1ti and hence thatlog i = hlog ( -1) = bers became manifest in the latter part of the
(1ti)/2, although this argument is not the one that eighteenth century, on philosophical, if not on
Bernoulli used. The controversy was subsequently utilitarian, grounds. With the advent of the Age of
resolved by Leonhard Euler (Leapfrogs 1978). Reason in the eighteenth century, when math-
Complex numbers were used by Johann Lam- ematics was held up as a model to be followed, not
bert for map projection, by Jean D'Alembert in only in the natural sciences but in philosophy as
hydrodynamics, and by Euler, D'Alembert, and well as political and social thought, the inadequacy
Joseph-Lonis Lagrange in incorrect proofs of the of a rational explanation of complex numbers was
fundamental theorem of algebra. (Euler, by the disturbing.
way, was the first to designate ..J=1 by i.) The problem of the logical justification of the
Euler, who made fundamental use of complex laws of operation with negative and complex num-
numbers in linking the exponential and trigono- bers also hearne a pressing pedagogical issue at,
metric functions by the formula eix =cos x + i sin x, among other places, Cambridge University at the
expressed himself about them in the following way tum of the nineteenth century. Since mathematics
(Kline 1972): was viewed by the educational institutions as a
Because all conceivable numbers are either greater paradigm of rational thought, the glaring inad-
than zero, less than zero or equal to zero, then it is equacies in the logical justification of the opera-
clear that the square root of negative numbers can- tions with negative and complex numbers became
not be included among the possible numbers .... untenable. Such questions as, 'Why does 2 X i + i =
And this circumstance leads us to the concept of
2?" and "Is ...[;;;; = ,Ja ...fb true for negative a and b?"
such numbers, which by their nature are impossible
and ordinarily are called imaginary or fancied num- received no satisfactory answers. Ill fact, Euler, in
bers, because they exist only in the imagination. his text of the 1760s on algebra, claimed
...J=f '1/-4 = ...f4 = + 2
as a possible result. Robert
Even the great Carl Friedrich Gauss, who in Woodhouse opined in 1802 that since imaginary
his doctoral thesis of1797 gave the first essentially numbers lead to right conclusions, they must have
714
THINKING THE UNTHINKABLE: THE STORY OF COMPLEX NUMBERS (WITH A MORAL)
Maturity
By the latter part of the nineteenth century all
vestiges of mystery and distrust of complex num-
bers could be said to have disappeared, although a
lack of confidence in them persisted among some
textbook writers well into the twentieth century.
These authors would often supplement proofs us-
Karl Friedrich Gauss (1777-1855} ing imaginary numbers with proofs that did not
From A Portfolio afEminent Mathematicians, ed. David Eugene involve them. Complex numbers could now be
Smith (Chicago: Open Court, 1908) viewed in the following ways:
1. Points or vectors in the plane
a logic. Around 1830 George Peacock and others
2. Ordered pairs of real numbers
at Cambridge set for themselves the task of deter-
3. Operators (i.e., rotations ofvectors in the plane)
mining that logic by codifYing the laws of opera-
4. Numbers of the form a + bi, with a and b
tion with numbers. Although their endeavor did
real numbers
not satisfactorily resolve the problem of the com-
5. Polynomials with real coefficients modulo
plex numbers, it was perhaps the earliest instance
x2 + 1
of"axiomatics" in algebra.
6. Matrices of the form
By 1831 Gauss had overcome his scruples con-
cerning complex numbers and, in connection with
a work on number theory, published his results on
[_~ !1
with a and b real numbers
the geometric representation of complex numbers
7. An algebraically closed, complete field
as points in the plane. Similar representations by
(This is an early twentieth-century view.)
the Norwegian surveyor Caspar Wessel in 1797
and by the Swiss clerk Jean-Robert Argand in Although the preceding various ways of view-
1806 went largely unnoticed. The geometric rep- ing the complex numbers might seem confusing
resentation, given Gauss's stamp of approval, dis- rather than enlightening, it is of course common-
pelled much of the mystery surrounding complex place in mathematics to gain a better understand-
numbers. In the next two decades further develop- ing of a given concept, result, or theory by viewing
ment took place. In 1833 William Rowan Hamil- it in as many contexts and from as many points of
ton gave an essentially rigorous algebraic defini- view as possible.
tion of complex numbers as pairs of real numbers. The foregoing descriptions of complex num-
(To Hamilton the complex number (a, b) con- bers are not the end of the story. Various develop-
sisted of a pair of"moments of time," since he had ments in mathematics in the nineteenth century
earlier defined real numbers, under Immanuel enable us to gain a deeper insight into the role of
715
--------------------------------......
PART VIII/THE SEARCH FOR CERTAINTY
716
THINKING THE UNTHINKABLE: THE STORY OF COMPLEX NUMBERS (WITH A MORAL)
717
PART VIII/THE SEARCH FOR CERTAINTY
718
THINKING THE UNTHINKABLE: THE STORY OF COMPLEX NUMBERS (WITH A MORAL)
Some Suggestions for the Teacher (5) We must, of course, supply the student
with "internal relevance" when introducing a given
Let me conclude with some comments on, and concept or result. This point brings us to the im-
suggestions for, the use of the history of math- portant and difficult issue of motivation. To some
ematics in the teaching of mathematics, in par- students the applications of a theorem are appeal-
ticular with reference to complex numbers. Many ing; to others, the appeal is in the inner logical
of the points are implicit in the preceding story. structure of the theorem. A third factor, useful but
(1) I first want to reiterate what I view as the often neglected, is the source of the theorem: How
major contribution of this story for the teacher. did it arise? What motivated mathematicians to
P6lya (1962) puts it very well: introduce it? With complex numbers, their origin
in the solution of the cubic, rather than the qua-
To teach effectively a teacher must develop a feeling dratic, should be stressed. Cardan's artempted di-
for his subject; he cannot make his students sense its vision of ten into two parts whose product is forty
vitality if he does not sense it himself. He cannot
share his enthusiasm when he has no enthusiasm to
reinforces this point. How mUch further one con-
share. How he makes his point may be as important tinues with the historical account is a decision
as the point he makes; he must personally feel it to better made by the teacher in the classroom, bear-
be important. ing in mind the lessons that should be conveyed
through this or similar historical material.
The objective of my story, then, is to give the (6) Historical projects deriving from this story
teacher some feeling for complex numbers, to im- about complex numbers can be given to able students
bue him or her with some enthusiasm for complex as topics for research and presentation to, say, a
numbers. mathematics club. Possible topics are the following:
When it comes to suggestions for classroom use, (a) The logarithms of negative and
it cannot be overemphasized that these are only complex numbers.
suggestions. The teacher, of course, can better judge (b) What is a number? That is, discuss the
when and how, at what level, and in what context to evolution of various number systems and the evo-
introduce and relate historical material to the discus- lution of our conception of what a number is.
sion at hand. The introduction ofhistorical material (c) Hypercomplex numbers (e.g., the
can, however, convey to the studen~ the following quaternions). Their discovery is another fascinating
important lessons, which are usually not imparted story.
through the standard curriculum. (d) Gauss's congruences of integers and
(2) Mathematics is far from a static, lifeless Cauchy's congruences of polynomials. The latter
discipline. It is dynamic, constandy evolving, full lead to a new definition (description) of complex
of failures as well as achievements. numbers.
(3) Observation, analogy, induction, and in- (e) An axiomatic characterization of
tuition are the initial and often the more natural complex numbers (see (7) under the heading
ways of acquiring mathematical knowledge. Rigor "Maturity''). In this connection we ought to discuss
and proof usually come at the end of the process. the notion of characterizing a mathematical system,
(4) Mathematicians usually create their subject and thus the concept of isomorphism. (Cf. the
without thought of practical applications. The lat- various equivalent descriptions ofcomplex numbers
ter, if any, come later, sometimes centuries later. discussed previously.)
This point relates to "immediate relevance" and to (7) Many elementary and interesting illustra-
"instant gratification," which students often seek tions of Hadamard's comment demonstrate that
from any given topic presented in class. "the shortest path between t:vvo truths in the real
719
PART VIII/THE SEARCH FOR CERTAINTY
domain passes through the complex domain." We Jones, Phillip S. "Complex Numbers: An Example"of
are referring to elementary results from various Recurring Themes in the Development of Math-
ematics-I-III." Mathematics Teacher47 (February,
branches of mathematics, results whose statements
April, May 1954): 106-14,257-63, 34G-45.
do not contain complex numbers but whose "best"
Kline, Morris. Mathematical Thought from Ancient to
proofs often use complex numbers. One such ex-
Modern Tt'mes. New York: Oxford University Press,
ample was given previously. Some others by Cell 1972.
(1950),Jones (1954), and the NCTM (1969) can
Leapfrogs: Imaginary Logarithms. Fordham, Ely, Cambs.,
be found in the Bibliography. England: E. G. Mann &Son, 1978.
Leapfrogs: Complex Numbers. Fordham, Ely, Cambs.,
England: E. G. Mann & Son, 1980.
BIBLIOGRAPHY
McClenon, R. B. "A Contribution of Leibniz to the
Birkhoff, Garrett. A Source Book in Classical Analysis. History of Complex Numbers." American Math-
Cambridge, Mass.: Harvard University Press, 1973. ematical Manthly30 (November 1923): 369-74.
Burton, David M. The History ofMathematics. Boston: , Nagel, Ernest. "Impossible Numbers: A Chapter in the
Allyn &Bacon, 1985. History of Modem Logic." Studies in the History of
Cell, John W. "Imaginary Numbers." Mathematics ldeas3 (1935): 429-74.
Teacher43 (December 1950):39+-96. National Council ofTeachers ofMathematics. Histori-
Crossley, John N. The Emergence ofNumber. Vtctoria, cal Topics for the Mathematics Classroom. Thirty-first
Australia: Upside Down A Book Co., 1980. Yearbook. Washington, D.C.: The Council, 1969.
Dantzig, Tobias,Numb~theLanguageofScience. New POlya, George. Mathematical Discovery. New York: John
York: The Free Press, 1930, 1967. Wiley & Sons, 1962.
Davis, Philip]. The Mathematics ofMatrices. Waltham, Sondheimer, Ernest, and Alan Rogerson. Numbers and
Mass.: Blaisdell Publishing Co., 1965. Infinity-an HistoricalAccount ofMathematical Con-
Ebbinghaus, Heinz-Dieter, et al. Zahlen. Heidelberg: cepts. New York: Cambridge University Press, 1981.
Springer-Verlag, 1983. Stroik, Dirk}. A Source Book in Mathematics, 120D-1800.
Edwards, Charles H. The Historical Development ofthe Cambridge, Mass.: Harvard University Press, 1969.
Calculus. New York: Springer-Verlag, 1974. Windred, G. "History of the Theory of Imaginary and
F1egg, Gtaham. Numbers-Their History and Meaning. Complex Qyantities." Mathematical Gazette 14
London: Andre Deutsch, 1983. (1930?): 533-41.
720