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7 In the Classroom at hand. Second, to develop the student's ability v0 that the my solve future problems by hint ‘Eapetience shows thatthe questions and suggestions of four Tat, appropriately sed, very frequently help the Cradene ‘They have two common characteristic, comraon Senne acd generality As they proceed from plain common feove they very often come naturally; Uhey eould have Secured to the student himself. As they are general, they help unobtrusively they jost indicate « general direction dnd Teave plenty forthe student todo. “put the two aims we mentioned before are closely com ‘nected: if he student succeeds in solving the problem at hhand he adds a litle to his ability to solve problems Then, we should not forget that our questions are gear erst applicable in many eases. Ifthe same question fs repeatedly helpful, che student will scarcely fil to notice rind he wil be induced to ak the question by bimselt in asia situation, Asking the question repeatedly, he nay succeed once in eliciting the right idea. By such a cee, he discovers the right way of using the question, td then he has realy assimilated it "The student nay absorb a few questions of our Hist so well that he js nally able to pat (0 himself che right {question inthe right moment and so perform the corre ponding mental operation naturally and_ vigorously. ‘Such a student has cervainly derived the grestest possible prof from our list, What ean the teacher do io order € ‘brain this bet possible resule? ‘Solving. problems is a practical stil ike, let ws «ay, swimming. We acquire any practical skill by Wd practice, ‘Trying to eka, you imitate what otber people do with their bands and fet co keep their heads Thowe water, até, finally, you learn to swim by prac: Teing sutimming. Trying to rolve problems. you have tO ‘Gorerve and to imitate what other people do when sol 6 Four Phase : ing problems and, nally, you lear to do problema fig hem ule ee "The teacher who wits to develop ble stdent ails we do probe mat fil ome terest for robes tei minds nd ive them plenty of opportunity for ination and prac Ifthe teacher wie to develop in his sudente the mental operstions whic coespond to che quesaons and suggests of our the pus tere Ghesons and sggetons to the sudnis alten 3 be an do vo aturaly. Moremer, when the tcacher sles {probiem bnfore the cas, he should dramatise hi ideas 2 fide and be should porto nul! the me questions Shih ne se when helping the aden, Thanks to ich eidance the widen will eventually dover the ight She of these queions and suggestions, and doing so he equiv smething that more important an the Thowedge of any parila mathemati [ac MAIN DIVISIONS, MAIN QUESTIONS 6 Four pes Tying fd hea, we my xe peey ange str ola iw omy dng wethe blon, hve tosh cu pon agin nd tea ur coca of te roa ike be ttc’ eye oe evr mi a fl lea ne he al em pres pi tien ne we ve snc saad solution. a In are to group cone he qin ard sep fesen fa ae Sl tng poe Eerie weave otra the pres we i'w ec wht po we ae tx how he ate ems comet ho te un Tom's tke ote dr, ner saa ce titel emake on Threw you oe6 In the Clessroom plan. Fourth, we Took back at the completed solutions Wwe review and discuss it ‘Each of Gheve phases has its importance. ¢ may ap- pen that a student hits upon an exceptionally bright Flea and jomping all preparations blusts out wich the olution. Sul hacky ideas, of course, are most desirable, ‘put something very undesirable and veforwimate may resul if the student Jeaves out any of the four phases ‘without having a good idea. The worst may happen if the student embarks upon computations of consiric wos without having understood the problem. It is generally wees 10 cary ovt details without having seen fhe main connection, of having made x sort of plan. ‘Many mistakes can be avoided if, carrying out his plan, the student checks each sep. Some of the best eects may be lose if the student fails to reexarane and to reconsider the completed solution 7. Understanding the problem. It is foolish to answer « question that you do not understand. 1s sad 10 wotk for an end that you do pot desce. Such foolish and sad things often happen, in and out of scoot, bu dhe reacher Should ty to prevent them from happening in his clas ‘The wwuden. shouki understand the problem. Bt he should not only understand it, he should also desire foluion. If the student i lacking in understanding ot ie Snterest it is not always bis fault; the problem should be ‘well chosen, not too difielt and not too easy, natural TIN interesting, and some time should be allowed for ‘natural and interesting presentation. “Frnt ofall, the verbal statement of the probe must bbe understood, The teacher can check this, up t0 3 «et Textent he asks the student t9 repeat the statement, ind the stedent should be able to sate the problen Ruently, The student should also be able to point out the priadpal parts of the problem, the unknown, the 8. Example ; Ata, the condition Henc, Ue teacher can seldom allord to inthe questions What i he snknownt het ane the dtc What i the eondtion? “The sadent should conser the principal pare ofthe problem atenivey, repetedly and from sao side reve age comes nth the poole he show row grt and point oot On the takrown an the dats, Ii ik newer t give eames to the ober he ‘ould introduce uate notaton; devoting tome sien, ton 1o the appropriate choice of sigs, he ged to consider the objects for which the sgn have to be chon ‘There is anower quesion which may be wal in tis Preparsory nage provided that we do-not expect 2 Active smewer bat juts provional anwer, 2 gure: Lit pouible to sts the condition? {athe exposition of Pat It [ps] “Underianding the probe aude itm sages “Geng se ausimed™ and "Working for better. uadersaning 1. Example, Let us lustte some of the pots ex Plaind inthe foregoing scion. We tae the felling Sinple problem: Fite dingnal ofa nectngatar part ipo which the tre ty nde hg order o dics thie problem profitably, the stents smut he familar wih the there at Pythagoras, and With sone of applications in plane geometry, Bu hey tay have very lite sjtematie Knowle in goon txry. The teacher may rey het pon te sven phe amy with sal elem 3 tacher cat make the problem interesting making tconcce The clasfoom ya Tectangol path ltepped whose dimernons could be mente, an can be enimated, the student hove © find, to “measure inteerty" the diagonal ofthe Sagmoe, "Te teacher pein oot ee len she wtih and te hight tthe8 In the Clasr00r8 lasroom, indicates the diagonal with a gesture, and See Hi gure, drawn om the blackbose, by tlesing repeatedly othe soon. repre og Derween he tacher and tbe stuens nay tare as follows Syphat is the wnkown™ Me lenge of te agonal of parallelepiped.” “What ove te data” sot eng the wid, aa the height fe paral piped” ivoducesuileble notation. Which leer should Je pote the unkown “Rehien Jeers would you choo for the length the witha the Beght” “asboe” Shar the conto, inking 8 3802 CISL hisgont of the parallelepiped of WBC 0B, anc ane the nga ce with a the eight” wav vouonable problem? T mea, 1 the contin suffien to determine the wikncon®™ Heer ee tne ov 6, we krow the parallel piped tthe paraeleiped is determined, the diagooal ieetermined Devng a plan. We havea plan when we brow, oF eae e outine, whic exeaatons, computa teow ot CGauructons we have to perform in order fo Teta unknowns ‘The way fom understanding te oot concfing a pla maybe Jong a torn PRE pe maim achieve sm the oon of probe Tose omeatve the isa of plan. ‘Tis iden may sree gradually, Or afer apparently sawoecesal tat re of heaton, may ocurstidenly in 8 ran re nighe en.” The bet thatthe eacer can do pee dent isto procure for him, by snaberasve 9. Devising « Plan 2 Ielp a bright de. ‘The question and suggestions we ae Folie tedacn end oprovle dha er Tiare tobe abl fo see sont postion, ce teacher shoul unk of hi own eee aH il tole ces noting prbion oe ee steve lite Knowle of tesubjete and iopasibie tiv inet on Genta be opt experince and forme acid knowledge Mes rnessbering b nt enough for goo ies, bu we Gaonot hae tay pod tie wibout nalling sane portal, mater alone ne oot enogh fr cor Ffucting a house But we cnnec cutod a Boe wi tr alcng the necoarymatras The: mara es rage ene Tar tems of tar lormeryseqered mathemati Towldge, as former soe problems ot formes evel tnt i en spate at ost wi the ucts Dr jo Brow a relate eet 7 a let "Thc fit dha here ae wall are sally oo many prob- lens whic are ome elated ot Peta prem, tha hve sone olin common wit How cn we Choos the oe or he em whic are vey fl Thee io mgpenon tha pate or inget oma one eon tou poine Look othe wnlnon nd tte tink of Fier roto hag mo immo we cee in eli Torey ned role wc cet red to fur prevent problees we ae iy sly de ch sw teee by ening Te ere problem need to Yours and toed before Could you sei? "Te negoingqueon,welf understand eis comidced vey ten bcp to sac the righ tan of ido but they cannct help alway they anno worewe In the Clessoom magic. If hey do not work, we must look around for some biber appropriate polneof contact, and explore the vati- ‘un aspects of our problem; We have fo vary, to transform, to modify the problem. Could you restate the problem? Some of the questions of our list hine specie means to vary the problem, 28 generalization, specialization, use of {rnslogy, dropping a part ofthe condition, and so on; the Geils are important but we eanot go into them now. Variation of the problem may lead to some appropriate owls problem: If you cannot solve the proposed problem try to solve fist some related problem. ‘Trying «0 apply various known problems or theorems, considering various modifications, experimenting. with ‘rious wuailiary problems, we may stray so far fom our ‘original problem that we are in danger of losing it alto. eiher. Yet there is a good question that may bring us back to it: Did you wae all the date? Did you wse the tahole condition? 10, Example. We return to the example considered in section 8 AS we left it, che students just succeeded in Tnderseanding the problem and showed some mild inter tat in i. They could now have some ideas of their own, ome initiatives If the teacher, having watched sharply, ‘Cannot detect any sign of such initiative he has 9 resume Carefully his dialogue with the students, He must be pro pared (o repeat with some modification the questions Dinich the stedents do not answer. He must be prepared {fo meet often with the disconcerting silence of the fudents (which will be indicated by dot «++ 3) “Do you hnow a related problem?” ‘Leak at the unknown! Do you know « problem hav- ing the same unknown?” Swell, what i the unknown?” so, Example " “The diagonal ofa parallelepiped" "Do you know any prablem with the seme unknown?” "No. We have not had any problem yet about the agonal of a parallelepiped.” “Do you know any problem with a similar unknown? ou see, the diagonal ea amen, de segment of & traight line. Did you never ave # problem wow own wa the ei of line™ Ot courte ave saved such problems Frese, tofinda se ofa righ angle” “Gott Hee te probiem related lo youre ond tole before: Could you se” “You were lacky enough to remember 2 problem which is related to your present one and which you solved before. Would you like to use it? Could you introduce tome auxiliary element in order to make ifs use posible?” “Lock here, the problem you remembered is about a twimngle, Have you any triangle in your igure?" Let us hope that the list hint was explicit enough 10 provoke the idea of the solution which is to introduce 4 right triangle, {empbasized in Fig. a) of which thea In the Classroom rol diagonal isthe hypotenuse. Yer the teacher TRIuM be prepared forthe cine tat even thi ashy iid hist ioslicent co wake the torpor of he 8 hs and so he should be prepared 10 tse 2 whole grmutofmore and more explic ins Rola you ike to havea range nthe fie” “At sgt of eiange would you ike to have in che figure?” you eansot find yet the diagonal: ut you sid shat you cold nd he sde ofa tangle. Now, what wil you ie oud you Gnd dhe diagonal, iff were side of 8 crangle? ‘hen, evemualy, with more orl lp he students ote in toduelng the deci ailiary element, the gir clang euposzed in Fig the teacher should rence melt hat the wcidents sce sufiiently far ‘Thead before encouraging them to go into ctl cael Think hae i ws a goo den to draw tha vino ‘You have now a tangle ba have ou the unkwown?™ “Eye unkown the Bypotenuse ofthe trenle: We can eleatte ty the theorem of Pythagorn” "ou ca if bth leg are oven but ae they? “One legis genet se. And the oer 1 think, # not dita eG. Yeu the ote leg ithe hypotenuse of tomther ght range” very ght Nom Tac that you have plan.” TN cazeing oat the plan, To devie 2 plan, to cote ele the en the soloion i not esp Te ake 20 mh Sleds Torment nequired Knowledge, good ental ene Stcemation upon the purpox, and one mare TRREY oon Teck. To carry out dhe plan fs meh easier: hat we need aint patience tpn ives. a general eating; we have to convince Example ie ourselves that the detils fit into the ovine, and so we Ihave to examine the details one ater the other, patiently, sill everything is perfectly clear, and no obscure cornet remmalns in which an error could ie hidden TF the student hav really conceived a plan, the teacher hhes now a selaively peaceful time. The main danger is that the stadent forgets his plan. This may eaily happen if the student received hi pla from outside, and ac- ‘cepted it on the authority of the teacher; but if he worked for ic himself, even with some help, and conceived the final idea with satisfaction, he will not lo this idea ‘easly. Yet dhe teacher must insist thatthe student should check each step. ‘We may convince ourselves ofthe correctnes of 2 step fn our reasoning either “intuitively” or “formally.” We may concentrate upon the point in question till we see itso clearly and distinely thar we have no doubt that the sep fs correct; or we may derive che point in ques tom xccording to formal sles. (The difference between “inaight™ and "fgems? proof” is clear enough in many important eases; we may Ieave further discussion 10 philosophers) “The main poin: is thatthe student should be honestly convinced of the correcincts of each step, In certain cases, the teacher may emphasize the dfforence between “see ing” and “proving”: Car you see cleorly that the step is correct? Bua can you also prove that the step is correct? 12, Example. Let us resume our work at ibe point where we let Cat the end of section 10, The student, a last, las got the idea of the solution. He sees the right triangle of which the unknown » is the hypotenuse and the given height ¢ is one of the legs the other Jeg is the diagonal of a Tace. The student must, posibly, be urged to introduce suitable notation. He should choose to de rote that other leg, the diagomal ofthe face whose sides4 In the Classroom are a and b, Thus, he may see more clearly the ides of the olution which is to introduce an auxiliary problem Sehote unknown is >, Finally, working at one right tt Tngle after the other, he may obtain. (Gee Fig. ») nas yoete xo Vetere TS Tae SE sa wae 0H oe 13, Looking Back date their knowledge and develop their ability to solve problems. A good teicher should understand and impress fm his students the view that no problem whatever is com pletely exiausted There remains slways something to do: with susicient study and penetration, we could improve any solution, and, in any case, we can abways improve our tunderstinding of te solution "The student has now carved through his plan. He has ‘written down the solution, checking exch step. Thus, he should have good reasons to believe that his solution i torrec. Nevertheles, rors are always posible, especially if the argument i long and involved. Hence, verifications are desirable. Especially, if there is some rapid and in itive procedure ta tes. etber the resul or the argument, it should not be overlooked. Gan you check the resuli? Can you check the argument? In order ta convince outelves of the presence or of the ‘quality of an object, we like to see and to touch it. And 5 we prefer perception through two different senses, 10 we preter conviction by to different procts: Gam you de: ve the remlt diferentiy? We prefer, of course,» short and intwtive argument to along and heavy one: Can you fevit eta glanced ‘One of the frst and foremoet duties of the teacher is not to give hs studencs the impresion that mathematical problems have little connection wich each other, and no Connection at al with anything else. We have a natural ‘Opportunity to investigate the connections of « problem ‘when looking back at ie solution. The students will find looking back at the solution really interesting if they Ihave made an hones effor:, and have the consciousness fof having dove well. Then they are eager to see what else they could accomplish with chat elfore, and how they ‘could do equally well another time. The texcher should encourage the students to imagine cases in which they6 In the Classroom could utilize again the procedure used, or apply the re- ult obtained, Can yow tse the remlt, or the nicthod, Jor dome other problem? TA. Example, In section 12, the students finally ob- tained he solution: HE the ehrce edges of a rectangular pirallelogram, ised from the same comer, are a,b, 6 the diagonal is IPOS Gan you check the result? The teacher cannot expect 2 good answer to this question from inexperienced sta Gents. The students, however, should acquire fairly early the experience that problems "in le«ers” have a great advantage over purely numerical problems; if the prob- Jem is given "im letters its result is accesible to several {esis to which a problem “in umber isnot susceprible a all, Our example, although fairly simple, is suficient to show this. The weacher can ask several questions about the result which the students may readily answer with ‘r¥ex"; bat an answer "No" would show a serious lw in the result pid you use all the deta? Do all the data a,b, € appear in your formula for the diagonal?” ‘Length, widch, and height play the same role in our question; our problem i symmetric with respect 2 Bethe expression you obtained for the diagonal sym trie in a, 6, ef Does it remain unchanged when #, © are interchanged” ‘Our problem is 4 problem of solid geometry: to find the diagonal of a parallelepiped with given dimensions 4, b,c. Our problem is analogous to a problem of plane feometry: to Gnd the diagonal of a rectangle with given ‘Simensions a, bs the result of our ‘sotid” problem anal ‘ogous to the result of the plane’ problem” TE the height ¢ decreases, and nally vanishes, tbe 24. Example oy parallelepiped becomes a parallelogram. I you pute ~ 0 flyer ome >You bn te ce aria or agosl ofthe rectangular parlors "ft ight eicreney he agonal Ines. Does your formula tow this? "iLalt cree mesures a, 5, ofthe parallelepiped Sn crease in the me proportion, the diagnal alo increta in Se sae opr yon fmol yo tite axa bye fora be respetively, the expresion of the diagonal, owing co tht subttaon, sot sso be tulip by voto" “if, b, care mearired in et. your formula gies the diagonal card inet ov: bu yon change a et fs Int Ince the Soma should remain cored. I at Phe tq ea gunn Thee quenions have several god ee. Fn, an n- seTigent steers cannot be beng pred by the face thatthe formula pac to many tte was canine telore thatthe formule ore Beare e derived it {Sty But now het more conned ani gain fSnidence comes rom a deren sone, fa 0 3 torr of experimen evidence” Then thanks to the foregoing queons, the dais of the formula aelre ew tigiftance, and ave inked up with varow fee "The tora far tertove a better chance of being fe smembered, the knowlege ofthe audet i comaidsed Finally the quesocs ean been ranted to si lar probes. After some expeiene with sinilar prob Jems an nelignt sade may pers the undering fener dente of al selena, sviaton of toe Sei symmetry, anor. Hb got taka the Babe of Girecing hie avention to sach port, hai solve roles ay detely prot.8 In the Classroom ‘Can you check the argument? To recheck the argument sep by step may be necesary in dificult and important tates. Unualy, itis enough te pick out "touehy" points for rechecking, In our case, It may be advisable to discuss retrompectively the question which was less advisable 10 discust ab the solution was not yet attained: Can you prove that the triangle with sides x,y, ¢ is a right tr langle? (See the end of section 12) ran you ute the sevult or the method for some other problem? Widh a litte encouragement, and after one or ‘iso examples, the students easly find applications whiel, ‘consist enenially in giving some concrete interpretation to the abstract mathematical elements of the problem. ‘The teacher hime used such a conczeteinterpresation fs he took the room in which the discussion takes place forthe parallelepiped of the problem. A dull student may Propose, as application, to ealeulate the diagonal of the ‘Cafeteria instead of the diagonal of the classroom. Ifthe stadents do not volunteer more imaginative remarks, the teacher himgelf may put a slighty diferent problem, for instance: "Being given the length, the width, and the height of a rectangular parallelepiped, ind the distance ofthe center irom one ofthe corner.” ‘The students may use the result of the problem they just solved, observing that the distance required is one half of the diagonal they just calealated. Or they may vse the method, introducing suitable right triangles (the later alternative ix less obvious and somewhat mare clumsy inthe present case). ‘Aiter this application, the teacher may discuss the con- figuration of the four diagonals of the parallelepiped, And the six pyramids of which the six faces are che buses, the center the common vertex, and the semidiagonels the lateral edges When the geometric imagination of the ‘students i saficiently enlivened, the teacher should come 15. Various Approaches 19 back to his question: Can you use the result, or the method, for some other problem? Now there isa better ‘chance that the students may find some more interesting concrete interpretation, for instance, the following: "In the center of the fat reciangular top ofa building which is 21 yards long and 16 yards wide, a flagpole isco be erected, 8 yards high. To support the pole, we noed four equal cables. The cables should start from the same point, # yards under the top of the pols, and end at the four corners ofthe top of the building. Hlow long is each cable?” ‘The students may use the method of the problem they solved in detail incoducing a right eriangle in a vertical plane, and anther one in x horizontal plane. Or they ‘may use the result, imagining a rectangulst paralcle piped of which the diagonal, x, is one of the four cables and the edges are ars ba8 ems. ‘By Straightforward application of the formula, x = 1445 For more examples, see CAN You WE THE RESULT? 15. Various approaches. Let us stil yetan, for a while, the problem we considered in the foregoing vections 8, 10, 12, 14. The main work, the discovery of the plan, wat escribed in section 10. Let us observe that the teacher ‘could have proceeded differently. Starting from the same pint asm section 10, he could have followed a somewhat ‘iferent line, asking the following questions: “Do you know any related problem?” “Do you know an analogous problem?” "You seth prope problem ia problem of wld geometry. Could you think of a skupler analogous prob Tem of plane geometry?” ze 2 "You see, the proposed problem is about a figure in space, it in concerned with the diagonal of a zectangular2 In the Classroom parallelepiped. What might be an analogous problem About a figuze in the plane? It should be concerned with the dixgonal—of—a rectangular ‘Parallelogram.” ‘The students, even if they are very slow and indiffer- ‘ent, and were not able to guess anything bei ‘obliged finally co contribute at lest a minute part of the dea. Besides, if che studenss are so slow, the teacher should not take up the present problem about the para: Jelepiped without having discussed before, in order 10 prepare the students, the analogous problem about the parallelogram. Then, ke can go on now as follows: “Here is a problem related to yours end solued before. an you use i" “Should you introduce some auxiliary element in order tomate its use posible?” Eventually, she teacher may succeed in suggesting to the students the desirable idea. It consists in conceiving the diagonal of the given parallelepiped as che diagonal of a suitable parallelogram vehich rust be introduced Into the gure (as intersection ofthe parallelepiped with 2 plane passing through two opposite edges). The idea is festentially the same as before (section 10) but the ap- proach is diferent. In section 10, the contact with she Available knowledge of the students was established ‘through the unknown; « formerly solved problem was recollected because its unknown was the same 25 that of ‘the proposed problem. in the present section analogy provides the contact with the idea of the solution. 16. The teacher's method of questioning chown in the foregoing sections 8, 10, 12, 14, 15 is ewentially thi Begin with a general question oF suggestion of our list, 1, Hf necessary, come down gradually to more specific and concrete questions or suggestions Hl you reach one which elicis s rexpomse in the student's mind. If you 16, The Teacher's Method of Questioning a1 have to help the student exploit his iden, start agen, if pouible, from a general question or suggestion contained and return again to some more special one if ‘and 10 on, Of course, our lst is just a fast List of this kind; it seems to be suficient for the majority of simple cats, but ‘there is no doubs ha it eould be perfected I is impor tant, however, that the suggestions from which we start should be simple, natural, ad genera, and that chit list should be shor The suggestions must be simple and natural because ‘otherwise they cannot be smobirsve, ‘The suggestions must be general, applicable not only to the present problem but to problems of all sorts, if they are to help develop the ability ofthe student and not justa special technique. ‘The list must be short in order that the questions may be often repeated, unartificially, and under varying ci- cumstances; thusy there is a chance that they will be ‘eventually aimilated by the student and will contribute (o the development ofa mental habit, Tt i neccesary to come down gradually to specific ug: ‘gestions, in order that the student may have as great @ share ofthe wark 3s posible. This method of questioning is not a rigid one; for tunately 50, because, In those matters, any rigid, mechan ‘al, pedantical procedure is necossarily bad. Our method. admits 2 certain elasticity and variation, it admits various approaches (section 18), it can be and should be so applied that questions asked by the teacher could have occurred to the student hirset fa reader wishes to try the method here proposed in Is dass he should, of coursc, proceed with cuution. He should study carefully the example introduced in section 8, and the following examples in sections 18,19, 20. He