3 History in mathematics education

Transcription

1 3 History in mathematics education 3.1 Introduction Over the years mathematicians, educators and historians have wondered whether mathematics learning and teaching might profit from integrating elements of history of mathematics. It is clear that mathematics education does not succeed to reach its aims for all students, and that it is therefore worthwhile to investigate whether history can help to improve the situation. The interest in using history and the belief in its value for learning have grown remarkably in recent years judging from the number of research groups in this field in Italy and France, and the History in Mathematics Education (HIMED) movement founded after a number of successful conferences and if we go back in time we find that acknowledged mathematicians from the eighteenth and nineteenth century held the same point of view. Joseph Louis Lagrange ( ) wrote in one of his lectures for trainee school teachers: Since the calculation of logarithms is now a thing of the past, except in isolated instances, it may be thought that the details into which we have entered are devoid of value. We may, however, justly be curious to know the trying and tortuous paths which the great inventors have trodden, the different steps which they have taken to attain their goal, and the extent to which we are indebted to these veritable benefactors of the human race. Such knowledge, moreover, is not a matter of idle curiosity. It can afford us guidance in similar inquiries and sheds an increased light on the subjects with which we are employed (cited in Fauvel & Van Maanen, 2000, p. 35). And in one of his notebooks Niels Henrik Abel ( ) remarked: It appears to me that if one wants to make progress in mathematics one should study the masters (cited in Fauvel & Van Maanen, 2000, p. 35). Only recently there has been a stronger call for methodological and theoretical foundations for the role of history in mathematics education. The report History in Mathematics Education: The ICMI Study (Fauvel & Van Maanen, 2000) has made a valuable contribution in this respect by collecting theories, results, experiences and ideas of implementing history in mathematics education from around the world. The purpose of this chapter is to explain why and how history of mathematics might play a role in the learning and teaching of early algebra. After a brief account of the historical development of algebra we describe how history has instigated some essential ideas for the experimental learning strand. 3.2 Arguments for using history We would not plead for the use of history of mathematics in mathematics education if we did not believe that history can make a difference. Incorporating history in mathematics education can be beneficial for students, teachers, curriculum developers and researchers in different ways. We give a number of arguments frequently 35

2 History in mathematics education math as a growing discipline teacher profits value for the designer mentioned to illustrate this (Radford, 1995, 1996ab, 1997; Fauvel, 1991; Fauvel & Van Maanen, 2000). Students can experience the subject as a human activity, discovered, invented, changed and extended under the influence of people over time. Instead of seeing mathematics as a ready-made product, they can see that mathematics is a continuously changing and growing body of knowledge to which they can contribute themselves. Learners will acquire a notion of processes and progress and learn about social and cultural influences. Moreover, history accentuates the links between mathematical topics and the role of mathematics in other disciplines, which will help to place mathematics in a broader perspective and thus deepen students understanding. History of mathematics provides opportunities for getting a better view of what mathematics is. When a teacher s own perception and understanding of mathematics changes, it affects the way mathematics is taught and consequently the way students perceive it. Teachers may find that information on the development of a mathematical topic makes it easier to explain or give an example to students. For instance, heuristic approaches provided by history can be contrasted with more formal, contemporary methods. In addition it is believed that historical knowledge gives the teacher more insight in different stages of learning and typical learning difficulties. On a more personal level, history also helps to sustain the teacher s interest in mathematics. Not only the mathematics teacher but also the educational developer or researcher can profit from history in studying subject matter and learning processes. It provides teachers and developers with an abundance of interesting mathematical problems, sources and methods which can be used either implicitly or explicitly. Historical developments can help the researcher to think through a suitable learning trajectory prior to a teaching experiment, but it can also bring new perspectives to the analysis of student work. In the context of using history to study learning processes we mention the so-called Biogenetic Law popular at the beginning of the 20 th century. The Biogenetic Law states that mathematical learning in the individual (philogenesis) follows the same course as the historical development of mathematics itself (ontogenesis). However, it has become more and more clear since then that such a strong statement cannot be sustained. A short study of mathematical history is sufficient to conclude that its development is not as consistent as this law would require. Freudenthal explains what he understands by guided reinvention : Urging that ideas are taught genetically does not mean that they should be presented in the order in which they arose, not even with all the deadlocks closed and all the detours cut out. What the blind invented and discovered, the sighted afterwards can tell how it should have been discovered if there had been teachers who had known what we know now. (...) It is not the historical footprints of the inventor we should follow but an improved and better guided course of history (Freudenthal, 1973, p. 101, 103). 36

3 History of algebra: a summary learning trajectory In other words, we can find history helpful in designing a hypothetical learning trajectory and use parts of it as a guideline. For instance, Harper (1987) argues that algebra students pass through different stages of equation solving, using more sophisticated strategies as they become older, in a progression similar to the historical evolvement of equation solving. Harper pleads for more awareness of these levels of algebraic formality in algebra teaching. 3.3 History of algebra: a summary It is generally accepted to distinguish three periods in the development of algebra oversimplifying, of course, the complex history in doing so according to the different forms of notation: rhetorical, syncopated and symbolic, as shown in table 3.1 (see, for example, Boyer & Merzbach, 1989). written form of the problem written form in the solution method representation of the unknown representation of given numbers rhetoric syncopated symbolic only words words and numbers words and numbers only words words and numbers; abbreviations and mathematical symbols for operations and exponents word symbol or letter letter specific numbers specific numbers letters words and numbers; abbreviations and mathematical symbols for operations and exponents table 3.1: characteristics of the 3 types of algebraic notation rhetorical phase The rhetorical phase lasted from ancient times until around 250 AD, where the problem itself and the solution process were written in only words. In this period early algebra was a more or less sophisticated way of solving word problems. A typical rule used by the Egyptians and Babylonians for solving problems on proportions is the regula tri or Rule of Three: given three numbers, find the fourth proportionate number (see also Kool, 1999). In modern notations this means: given the numbers a, b and c, find d such that a : b = c : d. The Rule of Three also specifies in which order the numbers in the problem must be written down and then manipulated. An example of such a problem is problem 69 in Rhind Papyrus (ca BC), which says: With 3 half-pecks of flour 80 loafs of bread can be made. How much flour is needed for 1 loaf? How many loafs can be made from 1 half-peck of flour? (note: 1 half-peck 4.8 liters) (Tropfke, 1980, p. 359). Indian mathematicians (7 th and 11 th century) extended the Rule of Three to 5, 7, 9 and 11 numbers. Such problems are commonly classified as arithmetical, but in cases where numbers do not represent specific concrete objects and operations are required on unknown quantities, we can speak of algebraic thinking. 37

4 History in mathematics education unknown Rule of False Position syncopated algebra Depending on the number concept of each civilization as well as the mathematical problem, the unknown was a magnitude denoted by words like heap (Egyptian), length or area (Babylonian, Greek), thing or root (Arabic), cosa, res or ding(k) (Western European). The solution was given in terms of instructions and calculations, with no explanation or mention of rules. These calculations indicate that unknowns were treated as if they were known and reasoning about an undetermined quantity apparently did not form a conceptual barrier. For instance, in the case x of problems that we would nowadays represent by linear equations of type x + -- = a, n the unknown quantity x was conveniently split up into n equal parts. The Babylonians also used linear scaling to solve for the unknown in the equation ax = b, much like the regula falsi or Rule of False Position first used systematically by Diophantus (Tropfke, 1980). According to this rule one is to assume a certain value for the solution, perform the operations stated in the problem, and depending on the error in the answer, adjust the initial value using proportions. For example, an old Babylonian problem goes: The width of a rectangle is three quarters its length, the diagonals are 40. What are length and width? Choose 1 as length, 0;45 as width. (Tropfke, 1980, p. 368, transl.). In the sexagesimal number system, the calculations show the diagonal to be equal to ; ; that is, 1;15. Since the diagonals have to 1 be 40 instead of 1;15, the length is then adjusted to The Chinese (second 115 ; century BC) knew a rule based on the same principle that uses two estimates: the Rule of Double False Position. Although the Rule of False Position is generally not said to be an algebraic algorithm, its wide acceptance and perseverance even after the invention of symbolic algebra indicate it was and can still be a very effective problem solving tool. The phase of syncopated algebra began around 250 AD when Diophantus introduced shortened notations which enabled him to rewrite a mathematical problem into an equation (in curtailed form). He systematically used abbreviations for powers of numbers and for relations and operations. In his equations Diophantus used the symbol ς to denote the unknown and additional unknowns were derived from this symbol (although they were not used in the calculations). Tropfke (1980) explains that this change from representing the unknown by words to symbols really persevered only once the symbols were also used in the calculations. He gives two arguments to indicate that Diophantus appears to have been the first mathematician to do so. First, Diophantus performed arithmetic operations on powers of the unknowns, carrying out additions and subtractions of like terms self-evidently, without explicitly stating any rules. And second, he explained the method and purpose of adding and subtracting like terms on both sides of an equation (Tropfke, 1980, p. 378). 38

5 History of algebra: a summary Adapted from an unretrievable source figure 3.1: progress in symbolizing equations progressive symbolizing After Diophantus there were other practitioners of syncopated algebra. Figure 3.1 illustrates a collection of (semi-)symbolic equations throughout time, where each example is followed by the modern representation on the line below. In India (7 th century AD) words for the unknown and its powers which were extended in a systematic way were abbreviated to the first or the first two letters of the word. Additional unknowns were named after different colors. In Arabic algebra (9 th century AD) powers of the unknown were also built up consecutively, using the terms for the second and third power of the unknown as base. In abbreviated form, the first letter of 39

6 History in mathematics education these words was written above the coefficient. In Western Europe (13 th century) there were minor differences in the technical terms between Italy and Germany, and only in the second half of the 14 th century the words res and cosa were shortened to r and s respectively. In the middle of the 16 th century Stifel introduced consecutive letters for unknowns and stated arithmetical rules using these letters. From there Recorde, Buteo, Bombelli, Stevin, and many others each developed a system to symbolize powers of unknowns and formulate equations (Tropfke, 1980, p ). Recorde introduced the equal-sign in print, saying: And to avoid the tedious repetition of these words: is equal to: I will set as I do often in work use, a pair of parallels, or Gemowe lines of one length, thus:, because no 2 things, can be more equal. (Recorde, 1557, as cited in Eagle, 1995, p. 82, modernized spelling of the text in figure 3.2). Date: Recorde (1557), The Whetstone of Witte Source: Eagle (1995), Exploring Mathematics through History, p. 82 figure 3.2: algebraic notation of Recorde and introduction of the equal-sign general methods In the rhetorical and syncopated periods we see a certain degree of standardization. Problem solving procedures were demonstrated with one numerical example, which the reader could then easily repeat for new problems by simply replacing the numbers. Diophantus, the Arabs and the mathematicians in Western Europe contributed a variety of general methods of solving indeterminate, quadratic and cubic equations. But with the lack of a suitable language to represent the given numbers in the problem, it was still a difficult task to write the procedures down legibly. In a few isolated cases geometrical identities were expressed algebraically (with variables instead of numbers) but nonetheless written in full sentences. Syncopated notation did 40

7 History of algebra: a summary not (yet) enable mathematicians to take algebra to a higher level: the level of generality. It is important that students experience this limitation themselves in order to appreciate the value and power of modern mathematical notation. Date: Viète (1593), Zeteticorum Liri Quinque Source, left: Latin text from F. van Schooten s edition, p.42 (Leiden, 1646, reprint in Hofman, 1970); source, right: English translation in Witmer (1983), p figure 3.3: symbolic algebra symbolic period The development of algebraic notation in the 16 th century was a process still instigated by problem solving (see also Radford, 1995, 1996a). In 1591 Viète introduced a system for denoting the unknown as well as given numbers by capital letters, resulting in a new number concept, the algebraic number concept (Harper, 1987). The signs and symbols became separated from that what they represent (a contextbound number) and symbolic algebra became a mathematical object in its own right. For a solution to a typical Diophantine problem in the style of Viète, see figure 3.3. A few decades later Descartes proposed the use of lower case letters as we do nowadays: letters early in the alphabet for given numbers, and letters at the end of the alphabet for unknowns. With the creation of this new language system, earlier notions of the unknown had to be adjusted. The first objective had always been to un- 41

8 History in mathematics education simultaneous linear equations cover the value of the unknown but in the new symbolic algebra the unknown served a higher purpose, namely to express generality, since x could stand for an arbitrary number. Algebra as generalized arithmetic was a fact, and in its new role algebra detached itself from arithmetic. The historical development of algebra shows that, no matter how revolutionary symbolic algebra was, it was not a necessary condition for the existence of equations. As a matter of fact, linear equations were very common in Egypt and the Babylonians knew how to solve linear, quadratic and specific cases of cubic equations. Babylonian problems in two unknowns concerning sum and difference were frequently x + y solved using the rule (in modern notation): if = S and = D then x = S + D 2 2 and y = S D (Tropfke, 1980, p. 389). The Babylonians were also familiar with the subtraction (elimination) method. In order to solve the system of equations (in modern notation) 1 x + -- y = 7 4 x + y = 10 x y general method in China the first equation was multiplied by 4 and the second equation was then subtracted from the first, which gave 3x = 18. Hence x = 6, and from the second equation it followed that y = 4. In a very different part of the world a systematic treatment of solving equations developed in ancient China. Just like the ancient civilizations, the Chinese lacked a notational system of writing problems down in terms of the unknowns, but the computational facilities of the rod numeral system enabled them to surpass the rest of the world in equation solving. The Jiu zhang suanshu or Nine Chapters on the Mathematical Art was a very influential text, composed around the beginning of the Han dynasty (206 BC to 220 AD). It is an anonymous collection of 246 problems on socio-economic life, including numerical answers and algorithmic rules. The book discusses the Rule of Three, which originated in the barter trade, the Rule of Double False Position and several other methods of solving linear equations (Lam Lay- Yong & Shen Kangshen, 1989). More significantly, it is the oldest book known until now that contains a method of solving any system of n simultaneous linear equations with n unknowns, with worked-out examples for n = 2, 3, 4 and 5. This was done using the method fang cheng (calculation by tabulation), writing the coefficients down or organizing them on the counting board in a tabular form and then performing column operations on it (much like the Gauss elimination method of a matrix). Figure 3.4 shows an example of such a problem and two representations in tabular form. The general application of the fang cheng method led quite naturally to negative numbers and some rules on how to deal with them, which is in great contrast with the late acceptance of negative numbers in other parts of the world. 42

9 History of algebra: a summary Three bundles of high quality rice, two bundles of medium quality and one bundle of low quality rice yield 39 dou; two bundles of high quality, three bundles of medium and one bundle of low quality yield 34 dou; one bundle of high quality, two bundles of medium quality and three bundles of low quality rice make 26 dou. How many dou are there in a bundle of high quality, medium and low quality rice respectively? 34 Using modern notation to express the information of this problem, a system of three linear equations in three unknows emerges; Numerical data is entered onto the board by use of rods working from right to left and from top to bottom. Resulting rod configurations for the set of equations are shown in Figure 3(a). Following rod algorithm, elementary column operations reduce a column to two entries: a variable's coefficient and an absolute term. See first column in Figure 3(b). In this reduced matrix form, a solution for one variable is obtained (36z = 99) and back substitution supplies the remaining required values. For the given problem, z = 2 -- dou, y = 4 -- dou and x = dou x + 2y + z = 39 2x + 3y + z = 34 x + 2y + 3z = 26 FIGURE 3a FIGURE 3b Date: Nine Chapters, 200 BC AD Source: Calinger (Ed.) (1996), p figure 3.4: fang cheng method Diophantus Diophantus certainly demonstrated a pursuit of generality of method, but his first concern was to find a (single) solution for each problem. His major work, the Arithmetica (ca. 250 AD), is a collection of about 150 specific numerical problems that exemplify a variety of techniques for problem solving. Diophantus distinguished different categories and systematically worked through all the possibilities, reducing each problem to a standard form. Negative solutions were not accepted, and if there was more than one solution, only one was stated. He solved linear equations in one unknown by expressing the unknown and the given numbers in terms of their sum, difference and proportion. If a problem contained several unknowns, he expressed all the unknowns in terms of only one of them, thereby dealing with successive instead of simultaneous conditions. Consider for example the type: To split up a given number in two parts that have a given difference (Tropfke, 1980). The Babylonians solved such a problem with unknowns using a standard rule involving two unknowns, as mentioned previously in this section. Diophantus, on the other hand, defined one unknown in the problem, represented it by a symbol and combined the two conditions into one single equation. For example, he assumed the smaller of the two numbers to be ς; the other number then had to be ς + d and the sum 2ς + d had to be equal to n. Diophantus is also known for his treatment of indeterminate equations: equations of the second degree and higher with an unlimited amount of rational solutions. Once 43

10 History in mathematics education Arabic standard equations again the general method involved reducing the problem to one unknown and finding a single solution. Often the method he used for determining this single solution was displayed in such a manner that the reader is able to repeat it and find an arbitrary number of other solutions. Much later the Arabs also played an important role in the historical development of equation solving. Although the boundaries of my research have been set at (systems of) linear equations, their achievements on quadratic and cubic equations deserve to be mentioned. A well-known book on Arabic algebra is al-khwarizmi s Hisab aljabr w'al-muqabalah, written around 825 AD. It contains a clear exposition of the solutions of six standard equations (in modern notation, where a, b and c are positive numbers): bx=c, ax 2 =bx, ax 2 = c, ax 2 = bx + c, ax 2 +c=bxand ax 2 + bx=c, followed by a collection of problems to illustrate how all linear and quadratic equations could be reduced to these standard forms. Al-Khwarizmi also gave geometric proofs and rules for operations on expressions, including those for signed numbers, even though negative solutions were not accepted at that time. But as far as the difficulty of the problems and the notations are concerned, the book remained behind compared to the work of Diophantus; everything was written in words, even the numbers. The Arabs did not succeed at solving cubic equations algebraically, but in the 11 th century AD. Omar Khayyam presented a well-known yet incomplete treatise on solving cubic equations with geometric means. Arabic algebra became known in the Western world in the twelfth century, when al- Khwarizmi s work was translated by Robert of Chester. By the fourteenth century, mathematical textbooks on arithmetic and algebra were very common in certain parts of Europe, and equation solving (even of the third and fourth degree) had become a regular subject in the Italian abbacus schools. In 1545 Cardano presented the solution of the general cubic equations by means of radicals. After the invention of symbolic algebra, equation solving developed very rapidly and soon found new applications in other areas of mathematics. 3.4 Implementing historical elements of algebra There are different ways of implementing history in educational design. First, it can be used as a designer guide. Milestones in the development of mathematics are indications of conceptual obstacles. We can learn from the ways in which these obstacles were conquered, sometimes by attempting to travel the same course but at other times by deliberately using a different approach. Reinvention does not mean following this path blindly. On the contrary, it means that developers need to be selective and should attempt to set out a learning trajectory in which learning obstacles and smooth progress are in balance. History can set an example but also a non-example. And second, we can choose between a direct and an indirect approach, bringing history into the open or not. Learning material can be greatly enriched by integrating historical solution methods and pictures and fragments taken from original integrating history in teaching 44

11 Implementing historical elements of algebra research objectives sources, but in some situations it may be more appropriate that only the teacher knows the historical information. Having decided to use history of mathematics as a source of inspiration and an educational tool for both the researcher and the students, it is one of the main objectives in this project to determine the role of history in the experimental lesson series: How does the historical development of algebra compare with the individual learning process of the student following the proposed learning program? Do historical problems and texts indeed help students to learn algebraic problem solving skills? How do students react to historical elements in their mathematics lessons? Reinvention of early algebra Prior to the start of the design process, we turned to the history of algebra for possible signs of appropriate and inappropriate activities. Rojano extracted some lessons from history, two of which agree with Realistic Mathematics Education theory, which we discuss in section (Rojano, 1996). First, Rojano observes that since problem solving has always proven to bestow meaning on new knowledge in the past, we must be aware of the risk of teaching symbolic manipulation in advance of situations where symbolism is meaningful. And second, Rojano points out that we should not deny students informal knowledge and methods but use them to give meaning to the new area of knowledge, in particular symbolic algebra. After all, when Viète introduced symbolic algebra which formed a rupture with previous algebraic thinking he used the skills and knowledge of classical Greek mathematics to achieve his aims (ibid.). historical accesses to early algebra In order to facilitate the reinvention of early algebra in the classroom, we need to investigate where the historical development of algebra indicates accesses from arithmetic into algebra. First, word problems and informal algebraic methods form an obvious link between arithmetic and algebra. In its early days, algebra was considered to be advanced arithmetic. For many centuries algebra was intended for fluent arithmeticians as a problem solving tool; only in recent centuries algebra evolved as an axiomatic study of relations and structures. Although algebra has made it much simpler to solve word problems in general, it is remarkable how well specific cases of such mathematical problems were dealt with before the invention of algebra, using arithmetical procedures. Some types of problems are even more easily solved without algebra! As the science of restoration and opposition, which is the literal translation of the title of al-khwarizmi s algebra treatise Hisab al-jabr w'al-muqabalah, algebra was founded on arithmetical techniques like the Rule of Three and the Rule of False Position. Both these techniques provide opportunities for reasoning with unknown or variable quantities within an arithmetical context (see also section 5.3.3). Another possible access is based on the use of notation, for 45

12 History in mathematics education instance by comparing the historical progress in symbolization and schematization with that of modern students. And third, in order to anticipate possible objectives that students might have to the algebraic approach, educational developers and teachers could study old textbooks on early algebra in order to learn more about how algebra was understood and applied just after it became accepted Historical influences on the experimental learning strand The historical development of algebra indicates certain courses of evolution that the individual learner can reinvent. Ideally, the student will acquire a new attitude towards problem solving by developing certain (pre-)algebraic abilities: a good understanding of the basic operations and their inverses, an open mind to what letters and symbols mean in different situations, and the ability to reason about (un)known quantities. In this section we discuss only a few general themes and topics selected for the program; a more detailed description of how certain historical problems and methods have been integrated in the experimental teaching units is elaborated in section word problems rhetorical algebra barter equations Word or story-problems seem to constitute a suitable foundation for a learning strand on early algebra. This type of problem offers ample opportunity for mathematizing activities. Babylonian, Egyptian, Chinese and early Western algebra was primarily concerned with solving problems from every day life, although one also showed interest in mathematical riddles and recreational problems. Fair exchange, money, mathematical riddles and recreational puzzles have shown to be rich contexts for developing convenient solution methods and notation systems and are also appealing and meaningful for students. The early rhetorical phase of algebra finds itself in between arithmetic and algebra, so to speak: an algebraic way of thinking about unknowns combined with an arithmetic conception of numbers and operations. The natural preference and aptitude for solving word problems arithmetically form the basis for the first half of the learning strand, where students own informal strategies will be adequately fit in. The transfer to a more algebraic approach is instigated by the guided development of algebraic notation in particular the change from rhetorical to syncopated notation as well as a more algebraic way of thinking. One of the study s aims is to establish whether or not the evolvement of intuitive notations used by the learner shows similarities with the historical development of algebraic notation. The barter context in particular appears to be a natural, suitable setting to develop (pre-)algebraic notations and tools such as a good understanding of the basic operations and their inverses, an open mind to what letters and symbols mean in different situations, and the ability to reason about (un)known quantities. The following Chinese barter problem from Nine Chapters on the Mathematical Art taken from Vredenduin (1991) has inspired us to use the context of barter as a natural and historically-founded starting-point for the teaching of linear equations: 46

13 Implementing historical elements of algebra By selling 2 buffaloes and 5 wethers and buying 13 pigs, 1000 qian remains. One can buy 9 wethers by selling 3 buffaloes and 3 pigs. By selling 6 wethers and 8 pigs one can buy 5 buffaloes and is short of 600 qian. How much do a buffalo, a wether and a pig cost? In modern notation we can write the following system: 2b + 5w = 13p (1) 3b + 3p = 9w (2) 6w + 8p = 5b (3) where the unknowns b, w and p stand for the price of a buffalo, a wether and a pig respectively. From the perspective of mathematical phenomenology, Streefland and Van Amerom posed a number of questions regarding the origin, meaning and purpose of (systems of) equations (1996, p. 140). The example given above is interesting especially when looking at the second equation, where no number of qian is present. In this barter equation the unknowns b, w and p can also represent the animals themselves, instead of their money value. The introduction of an isolated number in the equations (1) and (3) therefore changes not only the medium of the equation (from number of animals to money) but also the meaning of the unknowns (from object-related to quality-of-object-related). Several historical texts have been integrated in the instructional materials to illustrate the inconvenience of syncopated notations and the value of modern symbolism. Other authentic sources are used to let students compare ancient solution methods like the Rule of False Position and the Rule of Three with contemporary techniques (see also the appendix) Expectations It is our belief that history of mathematics can play a positive role in mathematics education and educational design, as mentioned in section 3.2. The historical development of algebra indicates sufficient possibilities of integrating old problems and methods in a pre-algebra program of problem solving. Another aspect concerns student reactions to learning history in a mathematics lesson. In this regard we anticipate that most students will enjoy the change of scenery. The historical contexts are different from the common ones in contemporary mathematics text books and yet they show a clear relation with everyday life events. For instance, some problems involve old currencies, exchange rates for trading goods, or proportions of ingredients for baking bread. Tasks that involve acting out a story, such as the hermit problem (see section 6.7.6), provide opportunities for a creative classroom activity like drama. We assume a number of students at secondary school level having a higher intellectual capacity on average than the primary school pupils to be interested in learning about the history of mathematics for its own sake: to see where 47

14 History in mathematics education mathematics comes from, how people learned mathematics centuries ago, and the types of problems people encountered then. They can experience mathematics as a human activity: mathematics produced, changed, and improved by man. 48