How To Solve It A New Aspect of Mathematical Method G. POLYA Stanford University Princeton University Press Princeton, New JerseyCopyright 1945 by Princeton University Press Copyright © renewed 1973 by Princeton University Press Second Euition Copyright © 1957 by G. Polya AIL Rights Reserved LO. Cand: 7960544 ISBN 0.691.02956-5 (paperback edn) ISHN 0-691-08097-6 (hardcover cde.) Fit Princeton Paperback Printing, 1971 Second Printing, 2975 This book is sold subject to the condition that it shall not, by way of trade, be lent, resold, hired out, or otherwise disposed of ‘without the publisher's consent, in any form of bind ing or cover other than that in which it iv published. Printed in the United States of America by Princeton University Press, Princeton, New Jersey From the Preface to the First Printing A great discovery solves a grent problem but there is a rain of discovery in the solution of any problem. Your [Problem may be modest: But if i challenges your eutiox ify and brings into play your inventive faculties, and if you solve i By your own means, you may experience the tension and enjoy the triumph of discovery. Such experi- fences at a susceptible age may cate a tase for mental ‘work and leave their imprint on mind and character for lifetime. Thus, a teacher of mathematics has a great epport nity. If he fills his alloted time with drilling his students in routine operations he kille chelr interes, hampers their intellectual development, and misuse hid oppor: tunity. But if he challenges the curiosity of tis students by setting them problems proportionate co their know! ‘edge, and helps them to solve their problema with stmu- lating questions, he may give them a taste for, and some smeans of, independent thinking ‘Alo a student whote college curriculum includes some mathematics bas a singular opportunicy. This oppor: nity is lost, of course, if be regards mathematics 26 a subject in which he has to earn x0 and so much eredit ‘and which he should forget alter the final examination au quickly as posible, The opportunity may be lost even if the student has tome natural talent for mathematics because he, 24 everybody else, must discover his talen's and tastes; he cannot know that he likes raspberry pie if be has never tasted raspberry pie, He may manage to find ‘out, however, that 2 mathematics problem may be as ‘much fun a3 erossword pezze, o that vigorous mentalvi From the Prelace to the First Printing work may be an exercise as desirable at a fat game of tennis, Having tasted the pleasure in mathematic he will, not forget it easly and then there isa good chance that mathematics will becomne something, for him; a hobby, or 4 tool of his profesion, or his profession, ot a great ambition, ‘The author remembers the time when he wat a student ‘himself, a somewhat ambitious student, eager to und stand 2 lide mathematics and physics. He listened t0 leetures, reid books, tried to take in che solutions and facts presented, but tere was a question that dsrurbed Ihim again and again: "Yes, the sokution sews o work, it appears to be corset: but how is it possible to invent such 2 solution? Yes, this experiment stems to wok, this appears to be a fact; but how can people discover such facts? And how could I invent or discover sch things by myself?” Today the author is teaching mathematics in a ‘university; he thinks or hopes tha some of his mare cager students ask similar questions and he tiesto satisfy their curiosity. Trying to understand not only the solution of this or that problem but alo the mo:ives and procedures ‘of the solution, and trying to explain these motives and procedures to others, he was Enally Jed to write the present book. He hopes that it will be useful to teachers who wish to develop their students’ ability o solve prob- Jems, and ro students who ae keen on developing theit ‘own abilities. Although the present book pays special attention to the requirements of students and teachers of mathematics, it should interest anyboly concerned with the ways and ‘means of invention and discovery. Such interest may be more widespread than one would assume without rele tion, The space devoted by popular newspapers and magazines to crossword puzzles and other riddles seems tw show that people spend some vime in solving unprac From the Preface to the Fizst Printing il steal problems, Behind the desize to solve this or that problem that confers no material advantage, there may Fea deeper curiosity, adesive 19 understand the ways and the motiver and procedures, of solution. TThe following pages are writen somewbat concisely, Dat 2+ simply as possible, and are based on a Yong and serious study of methods of solution. This sort of study, talled hewrstie by some veriters, is not in fashion now. fadays but has a long past and, perhaps, some future. Studying the methods of solving problems, we perceive another face of mathematic. Yer, mathematics has two faces, it i the vigorous science of Euclid but it is also something cli, Mathematics prevented in the Budlidean sway appenrs 28a systematic, deductive seience; but mathe: mnaties in the making appears 2s an experimental, in ductive science, Both aspects ae as old asthe science of mathematics tell. But the second aspect is new in one respect; mathematics “in statu nascendi,” in the process fof being invented, has never before been presented in {quite this manner to the student, or to the teacher him: tell, of to the gencral public. The subject of heuristic has manifold connections; mathematicians, logicians, psychologists, educationalists, teen philosophers may clatn Various parts of i as belong fing to their special domains. The author, well aware of the posibilty of criticism from opposite quarters and Aeenly conscious of his Jimitations, bas one claim to make: he has some experience in solving problems and in teaching mathematics on various level “The subject is more fully dealt with in a more exten- sive book by the author which is on the way to com: pletion. Stanford University, August 5, 1044From the Preface to the Seventh Printing From the Preface to the Seventh Printing ‘Lam glad to say that Thave now succeeded in falfiling, a least in pare, 2 promise given in the preface to the first princing: The two volumes Induction and Analogy in Mathematics and Patterns of Plausibie inference whieh ‘omsticuce my recemt work Mathematics and. Plansible Reasoning continue the line of thinking begun in How to Sotve Mt Zurich, August 30, 1954 Preface to the Second Biition ie Preface to the Second Fsition ‘The preient second edition adds, besides a few minor Smprovernents, 2 new fourth part, "Problems, Hints, Solutions.” ‘As this edition was being prepared for print, a study appeared (Educational Testing Service, Princeton, N.}: of. Time, June 38, 1936) which seems to have formu lated a fev pertinent observations=they are not new to the people in the knov, but it was high time to formu- late them forthe general public: mathematics has the dubious honor of being che lease popular subject ia the curriculum .. . Futare teachers pus through the elementary schools learning to detest mathematics ‘They return to the elementary scboal to teach a new generation to detest i T hope that the present edition, designed for wider difusion, will convince some of its seadars dhat mathe matics, besides being a necessary avenue to engineeting jobs and scientific knowiedge, may be fu and may 210 ‘open up a vista of mental activity on the highest level Zurich, Fune 30, 1956Contents From the Preface 9 the First Printing From the Prelace to the Seventh Printing Preface to the Second Edicion. “How ‘Lo Solve It" list Introduction PART I, IN THE CLASSROOM Purpose ‘Helping the seudent 2. Questions, recommendations, mental operations Gonerality ‘Common sense ‘Teacher and student. imitation and practice te Mein divisions, main questions Four phases Understanding the problest Example Devising a plan 0, Example 31, Carrying out the plan awxii Contents 22 Example Looking back Pxample Various approaches ‘The teacher's method of questioning Good questions and bad questions More examples 18. A problem of construction 19. A problem to prove 20. A rate problem PART II. HOW TO SOLVE IT A dialogue PART IM, SHORT DICTIONARY OF HEURISTIC Analogy ‘Ausiliary lements Auxiliary problem Bokano Bright idea (Can you check the result? Can you derive the result diferent? Can you use the result? Canrying out a3 4 6 5 % 9 38 2 46 se Ea 8 6 6 58 Contents Condition Contradicioryt Gorollary Gould you derive something useful from the data? Gould you restate the problem} Decorposing and recombining Definition Descartes Dete Diagnosis Dia you use all the data? Do you kaow a related problem? rave a figuret Examine your guess iguree Generalization Have you seen it before? ‘Here isa problem related to yours and solved before Rew Rewristic rezoning {Bt you cannot soe the propose problem veal fndetion ation, bope, success ion and mat “Aoventors paradox Fic possible to sats the condition? ibita Lerma “ oomtins niy rseatereec, 8 103 08 5 m4 a 195 125xv Contents Look at the unknown Modern heuristic Notation Pappus ‘Pedantry and mastery Practical problems Problems to find, problems to prove Progress and achievement Puzrles ‘Reducto ad absurdum and indirect proof Redundant Routine problem Rules of discovery Rules of style Rules of teaching Separate the various parts of the condition Secting up equations Signs of progress Specialization Subconscious work Symmetry Terms, old and new Test by dimension “The future mathematician ‘The intelligent problem-solver ‘The intttigent reader ‘The traditional mathematics profesor ‘concn nly antec 123 19 134 Mat i M9 18 37 Contents Variation of the problem ‘What is the unknown? Why prools ‘Wisdom of proverbs Working backwards PART IV, PROBLEMS, HINTS, SOLUTIONS Problems Bins. Solutions 209 a4 251 258 242How To Solve It fod 1291 08 ue 27 3 20 ‘aoe ® omen, GO AHL ONIAaKVASWIaNA LI FATOS OL MOH ine "wDIRord Dove sor May sun oF fa woqgond pasodond yp SMS ee RON pmege wenn nog sed sno go funy om. MORTEM mG UY ERD os honnag wont 397 pane “pues -aoygaud any pumpirpun 07 94 WORIntroduction ‘The following considerations are grouped around the preceding list of questions and suggestions entitled "How fo Solve It." Any question or suggestion quozed from it will be printed in italics, and the whole list will be referred to simply as “the lst" or a5 “our Hist The following pages will discuss the purpose of the lise, illustrate its practieal use by examples, and explain the underlying notions and mental operations. By way of prcliminary explanation, this much may be said: If ting dhem properly, you address these questions and suggestions to yourself, they may help you to solve your problem. Jf, using them properly, you address the same questions and suggestions to one of your students, you may help him co solve his problem. The book is divided into four parts. The ttle of the first part ie "Ta the Classroom." It contains twenty sections. Exch section will be quoted by its number im heavy type at, for instanee, “section 7." Sections 1 to 5 discuss the “Purpose” of our lst in gen- eral terms. Sections 6 to 17 explain what a Divisions, Main Questions” of the lit, and Practical example. Sections 18, 19, 20 add "More Ex- amples” ‘The title of the very short second part is “How to Solve 1. Te is written in dialogue; a somewhat idealized deacher answers short questions of a somewhat idealized sfudent . The third and most extensive part isa "Short Diction- of Heuristic"; we shall relzr to it asthe "Dietionana Introduction 1 contains sixtyseven articles arranged alphabeticaly, For example, the meaning of the term steunientc. (set in small capitals) is explained in an article with this tite fon page 11a. When the title of such an article is referred to within the text ie will bese: in sovalleapitals. Cera paragraphs ofa few articles are more technical; they ate enclosed in square brackets. Some articles are fairly closely connected with the fist part to which they add further ilustrations and score specific comments. Other articles go somewhat beyond the aim of the fist part of whieh they explain the background. ‘There is 2 key: article on NoDIAN aezuausTic. 1c explains the connection ‘of the main articles and the plan underlying the Diction 217; ft containe also directions how to find information about particular items of the list. It must be emphasized that here i «common plan and a certain unity, becasse the articles of the Dictionary show the greatest outward variety. There ate a fev longer artiles devoted to the systematic though condensed discussion of some generat theme; others contain more specific comments, still others ross references, oF historical data, or quotations, or aphorisms, or even jokes. ‘The Dictionary should no: be read too quickly its text is often condensed, and now and then somewhat subsle- ‘The reader may reler to the Dictionary for information about particular points If these points come from hit experience with his own problems or his own students, the reading hat a much bevter chance to be profitable ‘The tle of Uhe fourth partis “Problems, Hfins, Solu tons." Te proposes a few problems to the more ambitious reader, Each problem is followed (in proper distance) by fa “hint” wast may reveal a way to the result which it explained inthe “tlution.” ‘We have mentioned repeatedly the “student” and the “teacher” and we thallzeler to and again. It Introduction at may be good to observe thatthe “student” may be a high shoo! student, or a college student, or anyone else who is studying mathematics Also the “teacher” may be a high school teacher, or x eallege insizetor, ar anyone interested in the weckinique of tesching mathematic, The ‘author looks at the situation eometimes from the poi of view of the student and sometimes from that of the teacher (the latter cate is preponderant in the fist pai). Yet most of the time (especially in the third part) the point of view is that of a person who is neither teacher for scudent but anxious to tolve the problem belore him.How To Solve ItPART I, IN THE CLASSROOM PURPOSE, 1, Helping the stadent. One of the most important tasks of the teacher isto help his students, ‘This cask ie ‘not quite easy; it demands ime, practice, devotion, and sound principe. “The student should acquire as much experience of independent wark as possible. But if he it lft alone with his problem without any help or with insuficient help, bbe may make no progres at all. Ifthe teacher helps 100 much, nothing is left to the student. The teacher shoud help, but not too much avd not 106 Hil, 40 that the student shall have a reavonable share ofthe work If the student ix not able to do much, the teacher should Teave hitn at last some Musion of independent work. In order to do 40, the teacher should help the seadent discreetly, unobiristvey. ‘The best is, however, to help the student naturally. ‘The teacher should pu: himself in the student's place, be should see the student's eas, fe should try to understand what is going on in the student's mind, and ask a qu tion or indieate a step that could have occurred to the student himel 2% Questions, recommendations, mental operations. i t0 help the student effectively but unobtrusively nd naturally, the teacher i Jed to atk the same questions and to indieate the sime steps again and again. Thus, in ‘countless problems, we have to ask the question: What‘i In the Classroom 1 the unknown? We may sary the words, and ask the sine thing in many difexene ways: What jp required? What do you went t0 find? What are you supposed to feck? The aim of hese questions i to focus the students Steention upon the unknown. Sometimes, we obtain the fame cifect more naturally with a suggestion: Look at the unknown? Question and suggestion aim at che same fffect; they tend to provoke the same mental oper ‘seemed to the author tha: it might be worth while © collect and to group questions and suggestions which are typically helpful in discussing problems wieh students “The Sst we study contains questions and suggestions of this sort, carefully choten and arranged: they are equally tseful tothe problenssolver who works by himelt. Ie che teaser i sufieienly acquainted wit the Hist and can se, ‘itn the suggestion, the action suggested, Ie may real ine thatthe list enamneraes,Indiecty. mental operations iypicaly useful for the solution of problems. These ‘Operations are listed im the order im which chey ate most ely to occur. ‘3 Generality fs an important characteristic of the qucations and suggestions contained in our ht. Take the Guestions: What 1 the unkown? What are the data? Wher us the condition? Tlhese questions are generally applicable, we can ask them vith good effect dealing ‘with all sorts of problems. Theit use ix not restricted to ny subjeceamatter. Our problem may be algebraic oF igeametcie mathematical or nenmathematical, theoretical br practical a serio problem or a mete puzzle: ft makes tno iiference, the questions make sense and might help tas to solve the problem, "There is a restriction, in fact, but it has nothing ¢© do with the subjectmatter, Cerlaio questions and cages Tiont ofthe list are applicable co “problems to find” only 5: Tescher and Student. tmitation and Practice 3 01 to "problems to prov." we have a problem ofthe Tatter Rind we must use different questions; sce vom ests 4 Common sense, The guctions and suggestions of cur list are general, but, ence for their generality, they ze acuta simple, obvioun, and proced fiom plain common sen. Take the suggestions Look at the um Anon! And ty to think of familar problem haute the soma or similar unknown. "This sogesion adios jou to do what you would do anyhow, without any Eve, you were secously concerned with your prob tem. Are you hungry? You wish to aban food and you think of familie ways of cbtining food. Have you a problem of geomesrie constriction? You wish 1 eon. Sue tangle and you think of familiar wae of ene structing 4 ingle. Have you » problem of any Kind? You wish to find a erain ‘unknown, and you think of femiar ways of finding such an unknown, or some sim ler unknown. Ifyou do #0 you folow exactly the sg gestion we quoted rom our a And you are onthe right Wack to; the sggetion i good a, sugges o ou 4 procedure whichis ery equentl sce ‘llth questions ed miggeton of or lit ate atu simple, cbsins jun plain common sense: bot they wate Plain common sene general terms. They sages 4 Eertaincordet which comes naturally fo any person who is seriously concerned with his problem and has some comiman ens Bu the peion who behaves the righ way ually does ot cae fo exptes his bebvioe tn leat words and: pssiy, he cnt exprs Ses] ou Ut is twexpres inva 5 ‘Teacher and student. Imiation an practice. There ane a which the teacher may hv view en AMtessing to his students » question ova suggestion of the lst: ir, to help te stant tn salve the problem