Page 127 - Discrete Mathematics and Its Applications
P. 127

106  1 / The Foundations: Logic and Proofs


























                                                FIGURE 7    Coloring the Squares of the Standard Checkerboard
                                                with Three Colors.


                                                squares. This contradicts the fact that this board contains 20 blue squares, 21 black squares, and
                                                22 white squares. Therefore we cannot tile this board using straight triominoes.  ▲


                                                The Role of Open Problems


                                                Many advances in mathematics have been made by people trying to solve famous unsolved
                                                problems. In the past 20 years, many unsolved problems have finally been resolved, such as the
                                                proof of a conjecture in number theory made more than 300 years ago. This conjecture asserts
                                                the truth of the statement known as Fermat’s last theorem.




                                 THEOREM 1       FERMAT’S LAST THEOREM The equation
                                                      n
                                                           n
                                                     x + y = z  n
                                                 has no solutions in integers x, y, and z with xyz  = 0 whenever n is an integer with n> 2.



                                                                         2
                                                                     2
                                                                              2
                                                Remark: The equation x + y = z has infinitely many solutions in integers x, y, and z; these
                                                solutions are called Pythagorean triples and correspond to the lengths of the sides of right
                                                triangles with integer lengths. See Exercise 32.
                                                    This problem has a fascinating history. In the seventeenth century, Fermat jotted in the
                                                margin of his copy of the works of Diophantus that he had a “wondrous proof” that there are no
                                                                       n
                                                                  n
                                                                            n
                                                integer solutions of x + y = z when n is an integer greater than 2 with xyz  = 0. However,
                                                he never published a proof (Fermat published almost nothing), and no proof could be found in
                                                the papers he left when he died. Mathematicians looked for a proof for three centuries without
                                                success, although many people were convinced that a relatively simple proof could be found.
                                                (Proofs of special cases were found, such as the proof of the case when n = 3 by Euler and the
                                                proof of the n = 4 case by Fermat himself.) Over the years, several established mathematicians
                                                thought that they had proved this theorem. In the nineteenth century, one of these failed attempts
                                                led to the development of the part of number theory called algebraic number theory. A correct
   122   123   124   125   126   127   128   129   130   131   132