Lecture Notes On The Theory of Didactic

Lecture notes
on the Theory of Didactic Situations in mathematics
by Anna Sierpinska
Concordia University, Montreal, Canada
prepared for the course MATH 645 Topics in Mathematics Education Research
given in year 2003
in the frame of the program
Master in the Teaching of Mathematics
at Concordia University
© 2003 Anna Sierpinska

MATH 645: Theory of Situations/ Lecture 1 Instructor: Anna Sierpinska
LECTURE 1
INTRODUCTION TO THE CONCEPT OF DIDACTIC SITUATION AND THE DISTINCTION BETWEEN
SITUATIONS OF ACTION, FORMULATION AND VALIDATION
Note: All references to page numbers in brackets refer to the textbook. Other references are
given in the footnotes.
1. WHAT IS A DIDACTIC SITUATION?
The notion of didactic situation is grounded in certain assumptions about learning and teaching.
These assumptions can be formulated using the metaphor of game (p. 40).
* the teacher is a player faced with a system composed of a student and a didactic milieu
* the student is himself a player in a game of him/herself with a didactic milieu
* in the student’s game with the didactic milieu, knowledge is the means of understanding the
ground rules and strategies and later, the means of elaborating winning strategies
* the teacher’s aim is to engage the student in such a game; aiming at a particular
mathematical knowledge, the teacher will try to set the student-milieu system so that, indeed,
this knowledge would appear as the best means available for the understanding of the rules of
the game and elaborating the winning strategy
What is a milieu?
‘Milieu’ should perhaps be understood in an ecological sense, as in ‘water is the natural milieu of
fish’. Thus the ‘didactic milieu is the natural milieu of students’. A person, a human being,
normally lives in several different milieus and plays different roles in them. In a family milieu
one can be a child, a mother, a father, etc. In a sports milieu, one can be the player, the coach,
etc. Other possible roles could be played in a workplace milieu, social milieu, etc. In the school
milieu, one can be a student, a teacher or an administrator. In each course, the student has to cope
with a specific milieu, and there are even more specific milieus for each class in a course. To
‘survive’ (‘to win’) in a milieu one has to get to know the ‘rules of the game’ and develop
strategies for winning the game.
Difference between this approach to teaching and learning and the traditional point of
view
Learning is not reduced to the result of a transmission of information from teacher to students.
Learning is understood more as sense making of situations in a milieu, and developing ways of
1
coping with them. Teaching of a knowledge K consists in organizing the didactic milieu in such
a way that knowledge K becomes necessary for the student to survive in it. If the situations in a
mathematics class are such that a certain type of social behavior is sufficient for survival in them,
without any use of mathematical knowledge, then it is the social behavior, not the mathematical
knowledge that the students will learn. If the teacher solves the problems for the students and
only asks them to reproduce the solutions, they will learn how to reproduce teacher’s solutions,
not how to solve problems. In this sense, the kind of game the student has to play with the
milieu, to survive in it, determines the kind of knowledge that he or she will acquire. Thus, in the
theory of situations, ‘knowledge is [understood as] the outcome of the interactions between the
student and a specific milieu organized by the teacher in the framework of a didactic situation’
(Balacheff, 19931, p. 133).
The context of the didactic situation
A didactic situation (DS) does not exist in a void. The persons who are the main actors in the DS,
playing the roles of the teacher and the student, may look at the DS from the outside, not as actors
but those who plan an action in view of some far away goals. The student will see the DS as a
means to reach personal life goals (e.g. becoming an engineer, obtaining a high school diploma in
order to get a better paid job, etc.). The teacher will look at the DS as an educational designer, or
even as a researcher, and will design the DS in view of certain professional objectives, e.g.
curricular objectives, assessment objectives, or research objectives. Each person takes the
existence and characteristics of the other into account in the planning.
In the diagram on page 248, the situation in which the persons who are only planning to
act the roles of the teacher and the student (P1, S1) is called a ‘meta-didactic situation’. When
these persons act as teacher and student (P2, S2), and interact about the learning situation of the
student (e.g. the teacher gives a problem to the student and the student inquires if he or she
understands well the conditions of the problem) then they are in a ‘didactic situation’. When the
teacher withdraws from the scene, and the student engages in solving the problem for the sake of
learning something, she is in a learning situation (S3). As the student endorses the problem as her
own, she acts as a problem-solver (S4). In the problem, there may be a real or an imagined story
with a material milieu with which some persons have to deal (S5) (e.g. persons purchasing some
1
Balacheff, N. (1993): Artificial Intelligence and Real Teaching. In C. Keitel and K. Ruthven (Eds.), Learning from
Computers: Mathematics Education and Technology. Berlin: Springer-Verlag (pp. 131-158)
2
goods and wanting to get a good deal). Sometimes the problem-solver identifies herself with
these persons and solves their problem.
2. SITUATIONS OF ACTION, FORMULATION, VALIDATION AND INSTITUTIONALIZATION
There are different types of didactic situations, depending on the kind of ‘game’ that the teacher
plays with the student-milieu system (p. 161-2). For the description of the types of situations, I
decided to go from situations in which the teacher is the most authoritarian figure (the most
common traditional classroom situations) to situations in which there are almost no teacher
interventions.
Situation of institutionalization. The teacher plays the role of a representative of the official
curriculum, the official mathematics as represented by the school institution, the textbooks
officially approved by the ministry, and the official culture. He informs the students about the
officially accepted terminology, definitions, theorems considered important from the institution’s
points of view. For the students, the milieu thus obtains the explicit features of an institution,
with clear assumptions and rules. Knowledge acquires the features of a law, rather than of an
answer to scientific inquiry: it is validated and justified through the authority of the institution
rather than through criteria such as internal, logical consistency and relevance for the solution of
scientific or technological problems.
Situation of validation. The teacher takes on the role of the theoretician evaluating the
productions of other theoreticians, whose role, in the classroom, is played by the students. The
students try to explain some phenomenon, or to verify a theoretical conjecture. The teacher acts
as the chair of a scientific debate: s/he intervenes only to put some order in the debate among
students, draw their attention to possible inconsistencies, and encourage them to be more precise
and systematic in the use of concepts. For the students, the milieu resembles that of an academic
seminar rather than that of a lecture room. Knowledge has the dynamic features of a theory in the
making, not of a finished, institutionalized theory.
Situation of formulation. The milieu for the students is developed on the basis of some
previously shared experience or activity: The students exchange and compare observations
between themselves. They may not have the language to formulate their observations, so their
main effort in this situation goes into creating such a language and agree on some common
meanings. The teacher chairs the exchanges (in order to avoid chaos) and highlights (repeats
louder, writes on the board) some formulations of the students, in case they may not have been
heard by other students. Knowledge, in this situation, appears as a result of a personal
3
experience, which needs to be communicated, and thus slightly de-personalized and de-
contextualized, in order to be understood by others.
Situation of action. The teacher organizes a milieu for the students to engage with but then
completely withdraws from the scene. The milieu for the students is that of a problem so chosen
and formulated that (a) the students are willing to adopt it as their own, and are interested in
solving it to satisfy their own curiosity or ambition; (b) the students have the means to construct
the solution by themselves, either by inventing a new procedure or choosing one among those
they know, without, however, the teacher suggesting which one to choose. In this situation,
knowledge appears as a means for solving a problem or a class of problems.
In many mathematics classrooms, and certainly in most university lecture rooms, the
institutionalization situations enjoy an absolute reign. Other kinds of situations do not appear, or
they appear in degenerate forms. A degenerate form of a situation of validation is one where the
students solve the ‘proof’ problems, e.g. ‘Prove that, in a right angled triangle, if one the acute
angles is 45° then the triangle is isoceles’, without the statement having been formulated by the
students as their own conjecture, and where the style of making the validating argument is
prescribed by the teacher. A degenerate from of a situation of formulation can be one where the
teacher asks the students to formulate definitions and theorems and sanctions the formulations
with approval or disapproval (‘correct’, ‘incorrect’). A degenerate form of a situation of action
could be one in which the teacher gives the students a problem to solve but then constantly gives
them hints and suggestions about what to do and which method to use. But scientific knowledge
does not grow that way: everything starts with a problem, tentative solutions, communication of
the results, their justification, revision of the results in the wake of the criticisms and queries from
the scientific community. Thus the ‘natural order’ of the growth of scientific knowledge is from
action, through formulation and validation, to institutionalization. But in school we are not
inventing new knowledge; we are teaching institutionalized knowledge most of the time. So we
think that it is a waste of time to reproduce, in the classroom, the tortuous path of its production,
and we find that it is more economical to teach directly the results. But, in actual fact, we do not
gain anything, because, by teaching only the results of scientific inquiry, we are teaching not
scientific knowledge but law. So we are completely missing the point of our teaching.
Of course, we have to be realistic and admit that we cannot afford the time for teaching
every single bit of the curriculum in this ‘genetic’ way. Therefore, we have to reflect on the
knowledge we aim to teach. What parts of this knowledge can be and which cannot be sacrificed
to direct institutionalization? Does it make sense to have the students re-invent decimal notation,
4
long division rules, algebraic notation? Is it possible to create situations in the classroom through
which the students, in one term, would have re-invented the differential calculus? Probably not.
What is possible, then?
This is one of the questions posed by the theory of situations.
3. EXAMPLE OF A SEQUENCE OF CLASSROOM SITUATIONS: THE RACE TO 20
In order to get a better sense of the different types of situations, let us engage in a series of
classroom activities that will exemplify these situations. These activities have been experimented
by a team of researchers conducted by Brousseau in 1970s and described in a paper published in
1978 (pp. 4-18).
The lesson (in grade 5 or 6) was divided into several phases.
Phase 1: The teacher introduces a game to the students (5 mn)
Teacher: Today we’ll play a game with numbers. It is called ‘A race to 20’. Two players play
the game. One player says ‘1’ or ‘2’. The other can add 1 or 2 to the number of the
opponent and says the result. He who first says ‘20’ is the winner. Let me play this game
with one of you. Anyone wants to volunteer?
The teacher starts playing the game with a student on the board. Both she and the student write
their numbers on the board. After a few steps she relinquishes her place to another student. The
record of the game could be, for example:
T/S2 S1
2 3
5 7
8 9
11 13
14 16
17 19
20
Phase 2: The students play the game in pairs (4 rounds, 10 mn)
Teacher: Now, sit in pairs and play up to 4 rounds of the game, keeping a record of the game on
paper.
It is expected that students will find that saying numbers at random is not the best strategy. Some
will find that saying ‘17’ is a sure guarantee of winning the game.
Phase 3: The students play the game in teams (6-8 rounds, 20 mn).
5
Teacher: Now divide in two groups and play the game in teams. For each round, one student
should be chosen as a representative of the team to play at the board. The teams can discuss
their strategies between rounds. But do not interfere with the representative at the board.
The teacher keeps the record of the results of the teams on the board.
Phase 4: Game of discovery: Formulation of propositions
Teacher: Now, I am inviting each team to formulate the strategies that they think allowed them to
win. The other team then verifies the statement. If the statement turns out to be true, the
team wins a point. If the statement turns out to be false, the team that proved it false receives
3 points.
If the game of discovery grinds to a halt, the teams can return to playing the game.
It is expected that the students will discover that playing 2, 5, 8, 11, 14, 17, 20 leads to winning
the game, and that they will prove this statement by playing the game from each number on
(action proofs).
Now, in the case of a class of MTM students, one should expect some more sophisticated
proofs. For example, the students might notice that the winning numbers are of the form 20-3k,
k=0, 1, 2, 3, 4, 5, 6. This could be noticed right away, or after a phase in which the teacher would
propose a change of the rules of the game: adding by steps of not 1 or 2 but of 1 or 2 or 3, and
racing to, say, 25. The discovery game could then be generalized: the students would be expected
to formulate propositions about the winning numbers in the general case of a race to n and steps
1 through m. The students would be expected to find out and prove that
if the race is to n and the steps are 1, 2, ..., m, (m<n) then the winning numbers are
a(k) = n - (m+1)k, where k = 0, 1, 2, ..., floor(n/(m+1).
Some students could try to prove the proposition by induction, for example, like this:
Proof, by induction on k:
If k = 0 the number a(k) = n, which is a winning number by definition.
Lets assume that a(k) is a winning number and try to deduce from that that a(k+1) is also a
winning number.
If I say a(k+1) then my opponent can say any number a(k+1) + j where 1 ≤ j ≤ m. But
then I can add y = a(k) - a(k+1) - j, which equals a(k), a winning number, so I can win.
I can add the number y because this is a number between 1 and m; proof: y = a(k) -
a(k+1) - j = n - (m+1)k - n + (m+1)(k+1) - j = (m+1) - j. Since 1 ≤ j ≤ m, then -1 ≥ -j
≥ -m and (m+1) - 1 ≥ (m+1) - j ≥ (m+1) - m, i.e. m ≥ (m+1) - j ≥ 1, i.e. 1 ≤ y ≤ m, and
the rule of the step from 1 to m is satisfied.
Hence, by induction, the numbers a(k) are winning numbers.
6
4. ANALYSIS OF THE EXAMPLE
Let us reflect on the following questions:

• What type of situation does each of the phases represent?
• What is the knowledge learned in each phase?
• What is the milieu in each phase?
• What are the rules of the game between the student and the milieu in each phase? What is the
relation of the teacher to these rules?
Phase 1: In this phase we are in a didactic situation but not in a learning situation. The teacher
and the students are in the roles T2 and S2, with the teacher explicitly setting up a milieu for the
students and explaining to the students the rules of the game with this milieu. The milieu is an
actual game - a game with numbers. The students, as students, know that they will be playing
that game not just for fun but also to learn some mathematics; they do not know what
mathematics they will be learning, but they promise themselves to be able to soon find out. This
phase could be classified as part of a situation of institutionalization, which is spread through the
whole activity, in periods when the teacher comes back and gives more instructions. The students
obey the instructions because they accept the teacher’s authority as a representative of the school
institution: ‘I don’t know why I should play that game, but I trust the teacher is doing it for some
purpose which is useful for me, so I’ll play it’. When they think that way, the students are in the
role of S3, and they see themselves in a learning situation within a didactic situation.
Phase 2: In this phase the students are in the role of S4; they are problem-solvers, coping with the
game, wanting to win, and they forget, for a moment, that they are students. The milieu, in that
phase, is not didactic; it is the milieu of playing games and wanting to win. The students arrive at
some intuitions about the winning strategies; for example, they may find out that it is good to say
‘17’ because they won several times by saying that number. This is a situation of action: the
teacher is out of the scene, the students have endorsed the problem as their own; it is the situation
itself that provides feedback and allows them to keep in control of the validity of their solutions.
This way the students start developing some personal, still implicit, knowledge.
Phase 3: The students are still in the roles of problem-solvers and the milieu is still that of a game,
but the situation gets slowly transformed into a situation of formulation. By having to
communicate with other members of the team and thinking out some common strategies, the
intuitive personal knowledge gets de-personalized and de-contextualized. There may not be
7
much of justification at that moment, because things are happening too fast, and social rather than
mathematical factors may impact on whether a suggestion of a strategy is taken or not in a team.
The students may find out more winning numbers, and even describe, in general terms, their
pattern (an arithmetic sequence starting from 20 with a difference of -3).
Phase 4: The students are now in the roles of S3: looking at their problem solving actions from
the outside and judging them, as well as submitting them for judgment by other students. With
the teacher only chairing the session, this situation classifies as a situation of validation. The
knowledge developed in this phase can be that of argumentation. This argumentation may not
necessarily have the features of mathematical proof. The argumentation may be of a social nature,
with some students using persuasion to convince the others, rather than some kind of objective
and unemotional reasoning.
The class ends in discussing the question: What kind of intervention should the teacher
make in closing the activity, in a situation of institutionalization? What mathematical terminology
and theory would she introduce? What was the mathematical knowledge that was aimed at in the
activity?
8
LECTURE 2
THEORY OF SITUATIONS AS A MEANS TO OVERCOME THE ‘PROCEDURES VS UNDERSTANDING’
DILEMMA IN MATHEMATICS TEACHING
In the ‘Introduction’ to Brousseau’s book, the descriptions and explanations of the situations of
action, formulation and validation are based on the notions of ‘feedback’ and ‘dialectic’, not the
metaphor of ‘game’ which I used last week to introduce the notion of ‘didactic situation’. In fact,
the metaphor of game appeared in Brousseau’s theory only around 1985-6, while the text of the
‘introduction’ was based on a research done in the early seventies when the theory was still in
diapers. It is nevertheless worthwhile trying to understand this older version of the theory
because it contains seeds of many later developments, and gives us a better sense of where the
theory comes from.
I’ll start, therefore, by trying to explain the meaning of the notions of ‘feedback’ and
‘dialectic’ in the context of the theory of situations. This discussion will lead me to showing how
the theory could be seen as an emergent of the dialectic between the fatalism of those claiming
that nothing but procedures can be realistically taught in mathematics classes and the optimism of
those who claim that teaching understanding of mathematics is just a matter of good will of
teachers and of good teaching materials. I’ll conclude by stating that one of the main claims of
the theory is that ‘mathematical meaning’ - this object of all our didactic desires - is not
something given or absolute: it has to be researched, studied. The study of the meanings of the
mathematical contents of our teaching is an absolutely fundamental element of research in
mathematics education or ‘didactics of mathematics’, which permeates all and every single
question in this domain. Questions related to the teaching and learning of mathematics which
can be resolved without the study of the mathematical content of this teaching and/or learning are
simply not part of ‘didactics of mathematics’; they could be legitimate questions in pedagogy,
psychology, sociology, anthropology, communication studies, educational technology, etc.
This class will close with an activity in which you will be invited to experience a phase of
such ‘study of meaning’. The mathematical concept under scrutiny will be the operation of
division.
1
PART I: THE NOTIONS OF ‘FEEDBACK’ AND ‘DIALECTIC’ IN THE EARLY VERSIONS OF THE THEORY
OF SITUATIONS
1. Feedback
The term ‘feedback’ comes from cybernetics, which is defined as the study of communication
and manipulation of information in service of the control and guidance of systems (biological,
physical, chemical, cognitive, etc.).
Suppose we have two systems, A and B, and system A makes an action aimed at system
B. This action results in a feed forward of information from A to B. System B may re-act to
this information by sending a feedback of information to system A. On the basis of this
information, the system A may change its way of acting on system B. If the system B does not
send feedback to A or A does not react to any information that may come from B, then the A-B
super system is called an ‘open-loop’ system) In the opposite case, the system A-B is called a
‘closed-loop’ system. (An example of a closed-loop system is placing a heater in a closed room
with a thermostat). If the system A acts on B in order to attain some goal, and reacts to its
action by attempting to minimize the difference between the goal and the output of B, then the
feedback is called negative feedback. If the feedback amplifies the difference, the feedback is
called positive feedback.
Example 1
When I present a project of a research paper at a conference, I feed forward some information to
the potential readership of the finished paper if published. During the discussion period after my
talk I receive feedback from the audience. In the course of the discussion, the differences
between my intentions and the audience’s interpretations are minimized. This makes me revise
my paper so as to better match the interests and ways of understanding of the readers of the paper.
In this example, the systems A and B were both cognitive systems, and the feedback was negative
feedback. If, in the course of the discussion the differences between my intentions and the
audience’s interpretations are amplified, the feedback was positive, and I produce a paper which
is even less understandable than its previous version.
Let us take some examples from the first phase of the class on ‘Race to 20’.
Example 2
Let A be the cognitive system of the high school teacher (Hteacher). Let B be the system
composed of the students and the didactic milieu of the moment. The central element of this
milieu is the game ‘race to 20’. The system A feeds forward the information about the rules of
2
the game ‘race to 20’ in a verbal form. In the aim of controlling the accuracy of the transmission
of meaning, A ‘puts B in motion’, i.e. the Hteacher makes a student play the game. By acting,
system B sends feedback to system A. This feedback carries information back to A on whether
the rules of the game have been understood in the intended way. System A re-acts by more
verbal feed forward in case B does not act in the intended way (negative feedback). Da capo.
Example 3
If we now enter inside the student-milieu system, we may look at the cognitive system of the
student as one system, S, and the system of the game played by two players as the other system,
G. G can be seen as a set of all possible two-column tables obtained in playing the game, some
of them classified as ‘left player wins’ and other by ‘right player wins’. By entering the game as
a player, say, a ‘left player’, S feeds forward some information into G, and obtains a feedback in
the form of the ‘wins’ or ‘loses’ verdict. S’s goal is to always get a ‘wins’ verdict, and thus aims
at controlling the system B so that it always produces such a verdict. The feedback, in this case,
is a negative feedback.
Example 4
Let us now use the notions of positive and negative feedback to compare the effects of traditional
and constructivist styles of communication between teacher and the student-milieu systems. In
the traditional style of communication, the feedback of the teacher (direct corrections, pointing
out of errors, hints) is a negative feedback, aiming at minimizing the differences between the
expected output of the student-milieu system and the actual output. In the constructivist style of
communication, the teacher’s feedback is a positive feedback: in an attempt to understand the
student’s way of thinking the teacher will make the student-milieu system focus on the
development of a knowledge that may have little to do with the knowledge intended by the
teacher: the difference will be amplified.
2. What does the term ‘dialectic’ refer to?
2.1 Meanings of ‘dialectic’ in the history of philosophy
In the history of philosophy, the term ‘dialectic’ has had many meanings. For example, in Plato’s
Republic, ‘dialectics’ was synonymous with what we call ‘philosophy’ today, i.e. a systematized
intellectual reflection on the nature and genesis of being (i.e. what is and what is not and how can
we distinguish between one and the other). In the time of Plato, ‘philosophy’ had a much larger
meaning of ‘all knowledge’. In the Middle Ages, the term referred to logic, in the context of a
3
classification of the so called Liberal Arts into trivium and quadrivium. Trivium contained three
domains of knowledge about language: grammar, rhetoric, and logic, called ‘dialectic’ at that
time. All three ‘sciences’ provided a technical knowledge necessary for conducting debates.
Grammar was the basis for constructing correct sentences and statements; rhetoric served the
purpose of persuading the potential opponent that a given statement is true; dialectic or logic was
meant to guide the opponents in examining their statements for consistency and truth.
In modern times, the term ‘dialectic’ in philosophy has been associated with Hegel’s
dialectic method of discussing and solving the various ‘dualisms’ such as the mind/body,
freedom/determinism, universal/particular, the state / the individual dualisms. He claimed that
these apparent oppositions can be proved as compatible with each other if seen from the
perspective of a third, more general concept. In fact, he saw reality in its evolution as a
continuous fight between opposing tendencies which are resolved through a more general
tendency, of which the basic two can be thought of as particular cases. In formulating a
philosophical argument aiming at resolving a duality he would use the pattern of THESIS-
ANTITHESIS-SYNTHESIS. In the ‘thesis’ he would argue in positive terms for one of the
points of view. In the ‘antithesis’ he would argue in favor of the opposite point of view, stressing
the contradictions with the first. In the ‘synthesis’ he would propose a point of view which would
bring the former two together as complementary in the frame of a more general conceptual
framework. For example, everyone can see the contradictions between morality from an
individual’s point of view and morality from the society’s point of view. What is good and
pleasant for an individual is not necessarily good and pleasant for the rest of the society. Doing
only what is good for the society and not what we would really like to do limits our free will, to
which we think we are entitled if we are not slaves. For Hegel, these contradictions are overcome
by the concept of ethical life, which refers to modern institutions such as the family, the civil
society and the state. These institutions are a realization of our individual free will. As a
consequence, abiding by their rules does not constrain our free will. Philosophers have criticized
this resolution of the morality dualisms, but this fact does not abolish Hegel’s theory of a dialectic
character of the evolution of ideas. According to Hegel, there is no stop after a synthesis has been
proposed. A synthesis becomes a thesis which can be subject to a critique leading to an
antithesis. The resolution of the opposition gives rise to a new synthesis etc.
4
2.2 The meaning of the term ‘dialectic’ in the distinctions between the types of
didactic situations
In the theory of situations, the term ‘dialectic’ refers to the method used by a cognitive system
(teacher, student) to manage the contradictions between its expectations concerning the output
from the system it attempts to control (the student-milieu system, the milieu, resp.) and the
feedback. Feedback is just communication of information. The process of dialectic turns this
information into knowledge: out of the contradiction, something positive is coming out, that
explains the contradiction and generates ways of avoiding it in the future.
In the situation of action, a student may expect to win by playing 7 on the basis of her
belief that 7 is a lucky number. If she loses, she may resolve the contradiction between her
expectations and the outcome of the game by concluding that 7 may be a lucky number but not in
this game, and starting to notice the properties of the numbers 1-20, specific to the game. She
may notice, for example, that when she played 12 she won, so next time around she’ll try to play
12. But, with a smart player, she’ll lose, and this will be again a contradiction with her previous
strategy or ‘theory’. Overcoming the contradiction by means of a new ‘theory’, and continuing
this ‘dialogue’ with the situation, the student will teach herself a method for playing the game so
that she wins or is able to predict the outcome before the end. However, in the situation of
action, there is no need for these ‘theories’ or ‘rules for action’ to be verbalized; they may thus
remain largely implicit and unconscious for the student.
The situation changes dramatically in the situation of formulation, where the necessity to
communicate forces the students to bring the ‘theories’ or ‘rules for action’ to the level of
consciousness. An explicitation of a strategy by a student may enter in contradiction with the
milieu in two ways:
- via feedback with respect to the form: other students may consider the formulation as unclear;
- via feedback with respect to the validity: the strategy may prove to be ineffective in a game, or
may be rejected by an argument of another students.
The resolution of the contradiction in each case brings about some positive new
knowledge about the situations: a better way of expressing one’s ideas or an improved strategy.
Concerning the validity of the statements of the students, a situation of formulation does
not force them to distinguish between validity of a statement and the efficacy of a strategy, nor
between a convincing, or authoritative, or forceful statement and its truth value. The minds of the
disputants are geared towards action and effective action in the situation of the game actually
played or to be played in the future; all arguments are subordinate to this goal.
5
The situation of validation changes the milieu with which the students play in that
respect. The objects manipulated now are no longer moves in the game, but statements about the
moves in the game. A student’s ‘theory’ may fall into a contradiction with another student’s
‘theory’. In the situation of validation the students will work on deciding which theory is ‘true’,
but the outcome of this work may be a third theory, a clearer, more precise statement. If, in a
didactic situation, the object of the students’ attention is the validity of certain mathematical
statements, then it is considered a ‘situation of validation’, even if the arguments used to prove or
disprove the validity of these statement are not, properly speaking, mathematical proofs. They
can be mathematical arguments like ‘If I play 17, I win because my opponent can only play 18 or
19 according to the rules of the game, and in each case I can then play 20’. But they can also be
empirical arguments like ‘If I say 15 I lose because each time I played 15 I lost’, or ‘Each time I
play 14 I win; proof, let’s do it!’.
In the situation of institutionalization, the students have to overcome the contradictions or
just differences between their own ways of playing the game, speaking about it and justifying
their strategies and the teacher’s ways of doing those things. In the ideal case, a student is able to
resolve the contradictions and bring his or her own knowledge to a higher level of generality.
2.3 Theory of situations as the result of applying the ‘dialectic method’ to resolve
the duality between ‘procedural teaching’ and ‘teaching for understanding’ in
mathematics education.
Theory of situations was first created as a synthesis aimed at overcoming the opposition, in the
traditional teaching of mathematics, between procedural and explicit verbal knowledge (PK) on
the one hand, and meaning and understanding (MK), on the other (pp. 128-1311). In the
traditional teaching of mathematics only the former was the object of the teacher’s didactic
concern and action. The latter was left to happen by itself, as a function of students’ intelligence
and practice in the application of the procedures, definitions and theorems in solving exercises
and problems. The two types of knowledge were opposed by several features:
- PK is taught, MK is not;
- yet MK is necessary for the acquisition of PK; students who do not understand fail at the
examinations;
1
see also: Brousseau, G. (1988): Représentation et didactique du sens de la division. In: G. Vergnaud, G. Brousseau,
M. Hulin (Eds.), Didactique et Acquisition des Connaissances Scientifiques. Actes du Colloque de Sèvres, mai 1987.
Grenoble: La Pensée Sauvage éditions, pp. 47-64.
6
- but they are not taught MK, so they are not to be blamed for their failure in PK;
- teaching methods and curricula are to be blamed for students’ failure;
- hence, it is necessary to change the curricula and teaching methods: let’s reform the system!
This opposition, unresolved, produces ever new reforms. The slogans of ‘teaching for
understanding’ are at the start of almost every reform, but, somehow, inevitably, the
institutionalized teaching of mathematics converges towards the traditional teaching of PK and
abandoning of the MK to the students’ own devices.
A question for the research in mathematics education is therefore: What are the objective
causes of this convergence?
Is it the laziness of the teachers that is to blame? Or, rather, the fact that MK requires
more time for preparing classes, less manageable classroom situations, a lot more reading of more
voluminous students’ work? The workload of the teacher may increase exponentially with
respect to the traditional teaching to the point where it becomes unmanageable. Moreover,
statistically, the results on the official final examinations of the students subjected to the teaching
of MK on top of PK do not normally show a spectacular improvement with respect to those
taught only PK. Actually, a teaching focused on MK, with its openness to all kinds of
interpretations and understandings, may leave some students with conceptions contradictory with
the more official meanings of terms and lead to errors. Not only that, but some of these
conceptions may become so entrenched that they become real mental obstacles to understanding
new knowledge. Therefore, teachers lose motivation to put more work: they see it does not pay
off proportionally to their efforts.
Is it the teachers’ lack of mathematical and didactic knowledge that is to blame? Or
rather the lack of such knowledge in the society and culture? This knowledge needs systematic
research, invention and experimentation, and its development should not be left to teachers who
have other matters to attend to. There exists a body of mathematics education knowledge, but it
does not always translate easily or well into practical knowledge for teachers and their
classrooms.
Is it the students’ lack of interest in mathematics or lack of intelligence that is to blame?
Or is it rather that the existing teaching methods and curricula disregard completely the basic
laws of human learning?
These are the naïve speculations and common answers to the traditional opposition
between procedures and understanding. These answers are rather fatalistic: they lead to accepting
the necessity of the predominance of the PK in teaching mathematics.
7
Theory of situations is an attempt to find a synthesis via a dialectic between the opposing
terms instead of resigning to having to reject one of the terms. It uses the dialectic method on two
levels
- the development of the didactic theory (research on mathematics teaching and learning) as a
synthesis out of the opposition between the teaching methods focused on PK and those focused
on MK,
and
- the development of mathematical knowledge in students as a synthesis of oppositions between
- implicit expectations of the results of an action on a milieu and the actual feedback from the
milieu,
- formulation of these expectations and the feedback from the objective and the social milieus,
- the expectations and the validity of these expectations viewed as mathematical statements
- the students’ own strategies, interpretations, formulations and arguments and the official
mathematical algorithms, definitions, terminology, notation and proof methods.
The didactic theory tries to identify and explain the phenomena of teaching and learning of
mathematics; in particular those that are responsible for the convergence of the didactic system
towards the PK focused teaching. It tries to capture these phenomena using concepts such as
‘didactic contract’, ‘epistemological obstacle’, ‘didactic obstacle’, ‘the Dienes effect’, ‘the
Jourdain effect’ or ‘the Topaze effect’, ‘the metacognitive shift’, ‘didactic memory’. Attempts
are also made to apply this theory to ‘engineer’ didactic milieus which would be less likely to
degenerate into PK focused teaching. We shall be seeing both the theoretical and the engineering
aspects of the theory in this course.
PART II: AN EXERCISE IN THE STUDY OF MEANING OF A MATHEMATICAL CONCEPT: THE

OPERATION OF DIVISION IN GRADE 6
In order to teach not only the procedures but also the meaning of mathematical concepts to the
students one has to study this meaning. The meaning of mathematical concepts is not something
absolute and given once for all. It changes in time, and it changes in function of the contexts in
which it is used, and the purposes for which it is used. A concept like, for example, function,
may have a different meaning for an algebraist, for a geometer, for an engineer and for a teacher.
Thus the meaning of mathematical concepts is not a given in research in mathematics education:
it is, rather, a problem.
8
Thus, studies of the meaning (or meanings, aspects, etc.) of mathematical concepts are an
important part of the theory of didactic situations.
Let us engage in such a study, taking, for example, the notion of division.
We can pose the following questions:
(1). What is the meaning of the operation of division for a research mathematician or for a
university teacher?
(2). What is the meaning of the operation of division for an elementary school teacher?
1. Meaning of the operation of division for a research mathematician
For a research mathematician, the operation of division appears in the context not of arithmetic
but the study of number and algebraic structures. The question, for them, is not How to divide?,
but Is it possible to divide in this particular structure?
Division is not discussed without the operation of multiplication. The operation of
multiplication is thought of as any binary operation defined in a set of elements that need not
necessarily be numbers. So one must first have a set of elements with an operation defined in it,
called ‘multiplication’, or something else, it does not matter (you can call it ‘operation star’). In
order to be able to speak about division, one must have the notion of the identity element, or an
element e such that a*e = e*a = a for any element a in the set. Then one needs the notion of
inverse: if a is any element, then it has an inverse if there exists a unique element b such that
a*b=e. The inverse of a, if it exists, is denoted by a-1. Having all these notions we can now
define division: to divide a by b means to multiply a by the inverse of b; or a/b = a*b-1. This
means that division a/b is defined only if b has an inverse.
In the ordinary field of real numbers, the identity is the number 1, and all numbers
except for zero have inverses. In the ring of integers only 1 and -1 have inverses. In the algebra
of matrices, multiplication is defined in the well known way (i’th row times j’th column), but
there is no unique identity matrix for all matrices. In fact one has to take only square matrices of
a given dimension in order to be able to speak about the identity matrix. Only very special
matrices have inverses (those with non-zero determinants) and then to divide a matrix by another
matrix is to multiply it by the inverse of that matrix.
As another non-numerical example, let us take the set of four elements, denoted
{0, 1, x, 1+x} and let’s define the operation * in it by the following table:
9
* 0 1 x 1+x
0 0 0 0 0
1 0 1 x 1+x
x 0 x 1+x 1
1+x 0 1+x 1 x
From this table we can see that all elements except for 0 have inverses. The inverse of 1 is 1,
the inverse of x is 1+x, the inverse of 1+x is x. Dividing x by 1+x we have to multiply x by
the inverse of 1+ x, which is x. But x * x = 1+x so, if we denote division by #, x # (1+x) =
1+x.
2. Meaning of the operation of division for a grade 6 teacher in 1936 and 1998.
With respect to the question of the meaning of division for an elementary teacher, one could start
by looking at textbooks, old and modern, classifying the school ‘problems on division’ into some
types, according to some features that would have to be established.
In this class we shall look at samples of division problems from two textbooks, one
American textbook from 1936, and one Polish textbook from 1998, both addressed to 6-graders.
The activity will proceed in several phases:
Phase 1: Individual work. Each student received the two samples and classifies each sample
according to some criteria chosen by him- or herself.
Phase 2: Small group work. The class is divided into 4 small groups. Each group agrees on a
common classification and writes down the classification criteria explicitly on a sheet of paper.
Phase 3: Presentations: Each small group presents their criteria of classification and gives
examples of problems from each category. Every next group stresses what is different in their
classification with respect to the previous group.
Phase 4: Whole class discussion on (a) the criteria of classification (b) the differences between
the classifications obtained for the two textbooks (c) the differences between the meanings of
division between the two textbooks (d) the differences between the meanings of division in the
6th grade and the meaning of division in academic mathematics.
The text that follows will be available to the students only after class.
10
2.1 Information about the textbooks
2.1.1 Categorization of division problems used by
Knight, F.B., Studebaker, J.W., Ruch, G.M. (1936): Study Arithmetics. Grade Six.
Chicago: Scott, Foresman and Co.
In this textbook the classification of problems on division is done along one main variable: the
kind of entities being divided: numbers, such as fractions, mixed numbers, decimals, or
magnitudes or measures. Within each category, subcategories are defined: e.g. in dividing
fractions, both elements can be fractions, one of the elements can be a whole number, the divisor
can be a fraction with numerator equal to 1.
Titles of chapters and sections related to division

Chapter 3 - Dividing with proper fractions
(Some sections are meant to introduce a concept, to explain; other - to practice a concept, yet
other - to apply the concept to ‘real life’ situations; we arrange the tittles of the sections in
indents, like this:
[Explanatory section]
[Practice section]
[Application section])
Meaning of division by fractions
Knowing what the divisor is
Fraction divided by fraction
Using division of fractions
Mixed numbers in answers
Halloween races
Roman numerals
Divisors with numerator 1
The state fair
Whole number divided by fraction
A Thanksgiving dinner
Fraction divided by whole number
Oyster farming
11
Chapter 4 - Multiplying and dividing with mixed numbers

...
Dividing measures
Making Valentines
Chapter 8 - Multiplying and dividing with decimals

...
Decimal divided by a whole number
Using division of decimals
Remainders in division
Elsie and Bill learn about bees
A new idea in division
Changing fractions to decimals
Dividing a decimal by a decimal
Dividing by .1, .01, and .001
Further work in dividing with decimals
A week in a logging town
Whole number divided by a decimal
Problems using decimals
Final work in dividing decimals
Examples of problems:
1. Alice, Ruth and Mary were the Pop-corn Committee for the Pearson School Halloween party.
The girls bought 3/4 of a quart of popcorn and divided it equally among themselves to pop. Each
girl took what fraction of a quart of corn to pop?
2. Tom and Jimmy were to make a box for a game to be played at the Halloween party. They
needed 4 boards each 3/4 ft. long. The janitor gave the boys a board 3 ft. Long. How many
boards each 3/4 ft. Long could they have cut from the 3-foot board?
3. Henry brought 3/4 of a bushel of walnuts to the party. He divided the nuts into 50 equal
shares. Each share was what fraction of a bushel?
4. The children had a peanut relay race. Each team ran 7/8 of a block, and each pupil on the team
ran 1/8 of a block. How many pupils were on each team?
12
5. Each of the girls on the Refreshment Committee served 1/2 of a pumpkin pie at the party. The
pies had been cut so that each piece was 1/8 of a whole pie. Into how many pieces was each 1/2
pie cut?
6. 9/10 ÷ 15/16
7. 5/8 ÷ 15/16
8. The cookie recipe that Mrs. White planned to use called for 3/8 cup of chocolate. She had only
1/4 cup of chocolate. What fraction of the full recipe could she have made with that amount of
chocolate?
9. Divide and put your answer in simplest form: 9/10 ÷ 3/5.
10. On Halloween the Pine Hill School had some Hard Luck races. The route for the races was in
three laps. The first lap was from the school to Five Corners: 1/4 mile. The second lap was from
Five Corners to Orr’s Sawmill: 7/8 mile. The third lap was from Orr’s Sawmill to the school: 3/4
mile. Helen said that the second lap of the route was 31/2 times as long as the first. Jane said that
it was 33/8 times as long. Which girl was correct?
11. Divide: (a) 76 ) 912 (b) 431 ) 35351
12. Woods family went to the State Fair. Father and Andy drove to the fair in the truck, taking
some cattle to be entered for prizes. Mother and Ruth drove the family car. On the way to and
from the Fair, Father used a total of 24 gallons of gasoline and 5 quarters of oil for the truck. The
gasoline cost 18 cents per gallon, and the oil cost 30 cents per quart. Father drove the truck 107
7/10 miles in going to the fair and 108 3/10 mile in returning. Besides the cost of the gasoline
and oil, the expenses for the truck were $1.00 for repairing a tire. To the nearest cent, what was
the cost per mile for the truck for the round trip?
13. Divide 3/4 by 5/9.
14. Nancy earned her Christmas money making Christmas cards. She bought 2 sheets of
cardboard at 5 cents each, a bottle of drawing ink for 25 cents, and some watercolors for 25 cents.
(A) How much did all these things cost? (B) The cardboard sheets were 22 inches by 28 inches
in size. She cut each sheet into strips 22 inches long and 5 1/2 inches wide. How many of the 5
1/2 inch strips did she cut from the 2 sheets? How many pieces were too narrow for her to use?
15. Sally and Ruth decided to make some valentines, which would be different from those they
could buy in the stores. They bought a sheet of red paper 22 inches by 28 inches. Each girl took
1/2 of it. How many hearts could each girl have cut from her share, if each heart used up to 1
square inch of paper?
15. 2 )155.8
13
16. 32 ) 5.12
17. 6 ) .828
18. During 8 hours on Tuesday there was .96 inch of rainfall. This was an average of what
decimal fraction of an inch per hour.
19. Mr. Burns and his family drove their car and trailer to Arrow Head camp to spend a few
days. They drove 297.5 miles in 8.5 hours in traveling to the camp. How many miles per hour
did they average?
20. 7.8 ) 7581.6
21. Mr. Mills told Ned and Alice that they could sell vegetables during the summer and keep half
of the profits. Mr. Mills helped Ned build a stand. To make the boards below the shelf, they
sawed up some 14-foot boards. How many boards 3.5 ft. Long could they have sawed from each
14-foot board?
22. 1.25 ) 3
2.1.2 Categorization of division problems used by
Zawadowski, W. et al. (1998): Matematyka 2001. Podrecznik do klasy 6 szkoly

podstawowej. Warszawa: WSiP.
The categorization in this textbook goes still along the same variable: ‘type of numbers divided’,
but it is a lot less detailed, and less explicit for the student. A new type of numbers appears:
‘rational numbers’. Decimals are just the positive rational numbers written in decimal notation.
Rational numbers are all numbers which can be written in the form p/q where p and q are
integers and q ≠ 0. The information about a section being related to division (or some other
curricular topic) appears only in the table of contents as an additional information addressed
mainly to the teacher. The curriculum is, in a sense, hidden in the textbook. This may have been
done to avoid compartmentalization of knowledge in students into ‘rubrics’ such as
‘multiplication’, ‘division’, etc.
Titles of sections related to division

5. Instead of dividing… — division of ordinary fractions
…
12. About three princes who shared their gold — division of decimal numbers
13. Minus times minus — multiplication and division of rational
numbers
14
…
15. A difficult choice — operations in rational numbers
…
25. Which way is the best? — operations in rational numbers
…
33. As close as possible! — operations in rational numbers
Examples of problems
1. Look at the two series of operations. How do the divisors and results change? Can you find
the missing results?
8:8=1 3/16 : 8 = 3/128

8:4=2 3/16 : 4 = 3/64
8:2=4 3/16 : 2 = 3/32
8:1=8 3/16 : 2 = 3/16
8 : 1/2 = ? 3/16 : 1/2 = ?
8 : 1/4 = ? 3/16 : 1/4 = ?
8 : 1/8 = ? 3/16 : 1/8 = ?
• Add two further operations to each column.

• By what number should 3/16 be multiplied in order to obtain 3/128?
• By what number should it be multiplied to obtain 3/64?
• What operations could replace each of these divisions? Do you see a rule?
• Write a similar series of operations and give the results.
2. Mom said to Jack: I bought 6 liters of honey. We’ll pour it into 1/2 liter jars. Bring the jars
from the cellar. (A) How many jars should Jack bring? (B) How many jars of 1/4 l would he
have to bring? And how many jars of 3/4 liter?
3. The quotient is equal to the divisor and it is 4 times larger than the dividend. What is the
dividend?
4. Find a number which is 4 times larger than the quotient of the numbers 3 1/2 and 2 4/5
increased by 1.
5. 2 1/3 + 3/4 : 1/2
6. -12,8 x (-0,2)
7. 3 1/3 : (-5/6) : (-2)
15
8. Decide which product is less expensive

(a) Margarine sold in cups of 250 g for 1,32 zl vs margarine sold in cups of 500 g for 2,49 zl.
(b) Yogurt sold in cups of 150 g for 0,93 zl vs yogurt sold in cups of 500 g for 2,60 zl.
2.2 Brousseau’s criteria of classification of division problems
In his already mentioned paper on the ‘didactics of the meaning of division’2, Brousseau
identified two sets of criteria for the classification of division problems: one related to contextual
variables of the problems, and one related to the concepts involved in the solution of the
problems.
(1) Contextual variables
Group 1: type of numbers involved in the division
natural numbers, decimal numbers, rational numbers, real numbers, etc.
representation of numbers (fractions, decimals)
size value of the numbers (<1, >1, small numbers, big numbers)
the mathematical function of the numbers (cardinal numbers, measures, scalars,
linear
transformations)
Group 2: type of magnitudes

physical magnitudes
dimensions
definition mode: magnitudes defined as products of magnitudes,
quotients of magnitudes, etc.
Group 3: type of didactic situation
Group 4: previously taught techniques of calculation

sharing manipulations
repeated subtraction
factoring
systematic approximations from above and from below
2
Brousseau (1988), ibid.
16
reduction to operations in natural numbers

ways of presenting the computations
(2) Conceptions of division

Group 1: Sharing (finding the number of parts, finding the value of a part)
Group 2: Finding the unknown term of a product
Group 3: Fractioning (fractioning of a unit, commensuration, decimal approximation of a
fraction)
Group 4: Linear transformation (finding the value corresponding to 1, ratio of measures,
a function ‘divide by’, etc.)
Group 4: Composition of linear transformations
17
LECTURE 3
1999 MATH 645 CONCORDIA UNIVERSITY
LECTURE NOTES ON THE THEORY OF DIDACTIC SITUATIONS – ANNA SIERPINSKA
LECTURE 3
THE NOTION OF ‘DIDACTIC CONTRACT’
PART I - IDENTIFICATION OF A PHENOMENON: THE ‘DIDACTIC CONTRACT’, AND ITS IMPACT ON
THE MEANING OF MATHEMATICAL CONCEPTS TAUGHT AT SCHOOL
We have assumed that the didactic situation can be described as a game between a (person in the
role or position of the) teacher and the student-milieu system. Every game has its rules and
strategies. The rules and strategies of the game between the teacher and the student-milieu
system, which are specific of the knowledge taught, are called the ‘didactic contract’ (p. 41). The
assumption that the rules taken into account are pertinent from the point of view of the knowledge
taught or aimed at is essential for the definition of the didactic contract. If the rules taken into
account have nothing or little to do with this knowledge, but are relevant from the point of view
of, say, classroom management, or political correctness, or the general culture, then we might
speak of other types of contract, maybe (e.g. pedagogical contract) but not of a didactic contract.
It is assumed in the Theory of Didactic Situations that ‘didactics’ refers to the problems of
teaching and learning of a particular knowledge, not any knowledge in general.
Contrary to games such as chess or bridge, the rules of the didactic contract are not
explicit and they can be slightly different from classroom to classroom, culture to culture, and
they can even change in the history of a single classroom with the same teacher and the same
students. The fact remains, nevertheless, that in every didactic situation there is a didactic
contract, and that across different cultures, classrooms and time some rules remain constant, such
as, for example, that the teacher is expected to perform teaching actions such as giving the
students tasks that are specific to the knowledge he or she aims at, and the student is expected to
attend to the tasks given by the teacher.
The rules of the didactic contract are implicit: the teacher and the students do not sign a
chart of ‘rights and obligations’ upon entering into a didactic relation. But they are there and we
know that they are there when they are broken. Suppose that one day, in the middle of a class, I
suddenly sit down, open a newspaper and start to read. What would you say? That I am not
doing my job! I was supposed to teach you something, start speaking to you, assign tasks, etc.,
and here I am, sitting, and reading a newspaper! You would immediately report on my strange
behavior to the chair of the department.
11/4/03 1
LECTURE 3
Let us look at another situation: The teacher gives the students the task to write the
numbers 38, 24, 49, 46, 51 in an ascending order. The students are left to solve the problem
individually, and then the teacher writes on the board the solution to this exercise. Then she turns
to the class and asks: ‘Why did we put 46 and then 49?’ A student answers: ‘Because if we only
had the other numbers, it would have been too easy’. The teacher answers with some anger in her
voice: ‘This is not what you are asked for!’ (‘Ce n’est pas ce qu’on te demande!’). Another
student says: ‘Because 46 is smaller than 49’. Teacher: ‘Yes, very good’ (and she writes the
symbol ‘<‘ between the two numbers). We can see that the first student, in her intervention,
stepped out of her role of student: she expressed her view on the didactic reasons in designing the
task, thus entering the role reserved for the teacher. The teacher felt angry with the ‘usurper’.
The historico-epistemological context of the emergence of the notion of didactic contract

in the theory of didactic situations
The intellectual background of the idea of ‘didactic contract’ has been identified and analyzed by
Bernard Sarrazy in (Sarrazy, 19951).
‘The concept of didactic contract has been introduced by Brousseau in 1978, as a possible
cause of the so-called ‘elective failure’ in mathematics (some students appear to have difficulty
only in mathematics at school, while succeeding reasonably well in other subjects). He used it to
explain the case of a young boy, Gaël, who was repeating his first grade because of difficulties in
mathematics. The observation of the behavior of the boy during the remedial sessions with a
tutor showed that, for Gaël, knowing something meant only to be able to repeat certain ritualized
actions, modeled by the teacher. He was putting so much effort into finding out what actions the
teacher expected him to perform, or into ‘uncovering the implicit contract’, that he was not able
to engage with unraveling the meaning of the mathematical knowledge involved in the tasks
posed to him. When asked questions like ‘why did you add these two numbers?’ he would
invariably answer: ‘because this is what the teacher said that we have to do’, ‘this is how I was
taught’, etc. The boy was not retarded or unintelligent but simply had a notion of the didactic
contract that did not allow him to learn any mathematics. The solution of the boy’s problems
came with changing his notion of the contract by creating didactic situations for him in which the
contract was obviously not the same as with his schoolteacher. The boy was no longer told ‘Do
as I told you!’. Instead the tutor would play a game with Gaël, betting on the possible outcomes
of a calculation, and then finding out who was right.
1
Sarrazy, B. (1995): Le contrat didactique. Revue Française de Pédagogie 112, 85-118.
11/4/03 2
LECTURE 3
Describing the problem of Gaël in the sociological and cultural terms of ‘didactic
contract’ rather than in psychological terms of intelligence quotient, or personality, was quite in
tune with the spirit of the times. In the second half of the 70s, there was, in the explanations of
the school failure, a shift from the macro-sociological theories which would put the blame on the
socio-cultural background of the students (e.g. the theories of Basil Bernstein regarding the
school failure of working class children), and on the workings of the educational system as a
whole (e.g. Bernstein’s theory of ‘transmission of culture’ stressing the conservative character of
educational systems), towards micro sociological theories which focused on classroom
interactions. This ‘interactionist trend’ in educational sociology had a well-developed theoretical
background in ‘symbolic interactionism’ and ‘ethnomethodology’, a research paradigm that
emerged in USA in 1930s (the so-called ‘Chicago school’ of sociology: W.J. Thomas, F.
Znaniecki, R. Park, H. Mead, G.H. Blumer). The papers and books of Ernst Goffmann that
appeared in mid-70s became very popular in intellectual circles and had a strong impact on the
way people started thinking about even their everyday interactions with other people and
institutions. Today, the ideas of Goffmann and others working in the interactionist paradigm of
sociology have become part of textbook knowledge and any Cegep student enrolled in the Social
Sciences program has to learn about them.
It is Goffmann who started identifying the various ‘contracts’ that bind our interactions
with other people in everyday and professional lives. He called them ‘frames’ (kind of scenarios)
that can be played in different ‘keys’, (e.g. as comedies or tragedies). Applied to the classroom
context, one could speak of several identifiable ‘frames’, such as ‘lecturing’, ‘questioning’,
‘reprimand’, ‘praise’, etc. The frame of ‘school questioning’ is very different from the frame of
‘asking a question’ in a non-didactic situation: in the latter, the person who asks normally does
not know the answer! A child who comes to school first time may be quite astonished that the
teacher is asking questions she certainly knows the answers for! When she accepts this strange
situation as normal, she has already understood the frame and became aware of the existence of a
definite didactic contract that binds her own and teacher’s behavior.
Paradoxes of the didactic contract
The student expects the teacher to teach, and for some students, this simply means to tell the
student how to solve the assigned problems and what answers to give. But if the teacher complies
with this expectation, the student will not learn anything, because she will not have had to make a
choice of one strategy among other possible strategies, and of one interpretation among other
possible interpretations (p. 41). This is analogous to the information value of a message: if the
11/4/03 3
LECTURE 3
probability of a message is 1, then the amount of information it carries is 0. Solving a problem

one knows how to solve adds nothing to our knowledge.
So the didactic contract puts the teacher in front of a paradox: everything that the teacher
undertakes to make the student produce expected behavior tends to deprive the student of the
necessary conditions for understanding and learning of the notion she aims at; if the teacher tells
the student what she wants, she can no longer obtain it’ (p. 41).
So how can a teacher convey new knowledge to the students? How can a student learn
new knowledge?
The Theory of Didactic Situations assumes that learning in a school situation is an
adaptation to a milieu. The task of the teacher is then to organize the milieu in such a way that
the adaptation will result in the student developing the target knowledge. This is, however, easier
said than done.
In the second part of the class we shall be analyzing the ‘division problems’ of last week,
from the point of view of their ability to produce, in the students, an adaptation resulting in an
understanding of the operation of division.
PART II - EXAMPLES OF THE IMPACT OF THE DIDACTIC CONTRACT ON THE MEANING OF DIVISION
IN TWO GRADE 6 TEXTBOOKS
Results of the activity of classification of ‘division problems’ in Week 2
Each student received a copy of two sets of ‘division problems’, one taken from a 1936 textbook
(label this set ‘I’), and the other - from a 1998 textbook (label this set ‘II’).
Students worked in four groups of three: G1, G2, G3, G4, for about 20 minutes. Their task was to
classify the two sets of ‘division problems’ according to some criteria of their choice. The last 15
minutes of the class were devoted to a presentation, by each group, of the criteria they chose.
A priori, many different criteria could be chosen. For the 1936 textbook, one criterion
could be ‘the seasons’: there were problems for Halloween, problems for Thanksgiving,
problems for Christmas, for Valentines, for spring rains, for summer camping, for summer jobs,
etc. Another criterion could be ‘types of magnitudes’: pure numbers, physical magnitudes
(measures and ratios of measures), money. Yet another could be ‘kind of numbers involved’:
whole numbers, fractions, decimals, or natural numbers, integers, rational numbers. One could
also classify the problems into one-step problems, two-step problems, etc.
In class, students chose different criteria:
Here are these criteria:
Group G1: classification according to an aspect of the operation of division
Division as the inverse operation with respect to multiplication (e.g. I.6)
11/4/03 4
LECTURE 3
Long (synthetic) division

… of whole numbers (e.g. I.11)
… division convertible to division by a whole number (e.g. I.15)
… division convertible to division of whole number by a whole number (e.g.
I.20)
Number of parts in a whole (e.g. I.1, I.2, I.4)
Repeated subtraction (e.g. I.4)
Groups G2 and G4
Procedural problems
Disguised procedural problems
Conceptual problems
Group G3
Enigma type problems (the problem can be solved backwards) (e.g. II.3,4)
Problems involving judgment (II.8)
Only group G1 used criteria specific to the notion of the operation of division. Other groups used
more general criteria that could be applied to problems related to any mathematical notion. These
are important criteria but they are not helpful in the study of the meanings of the operation of
division aimed by the textbooks.
Study of the meanings of division conveyed by two grade 6 textbooks
If our aim is to study how the meanings of division conveyed by the problems, it may be a good
idea to find out
- What mathematical notions and techniques are sufficient to solve each problem,
as compared to
- The mathematical notions and techniques expected to be used by students by virtue of
the implicit ‘didactic contract’, whose rules are conveyed by the context of the problem.
Analysis of the 1936 textbook problems
1. Alice, Ruth and Mary were the Pop-corn Committee for the Pearson School Halloween party. The girls
bought 3/4 of a quart of popcorn and divided it equally among themselves to pop. Each girl took what fraction
of a quart of corn to pop?
11/4/03 5
LECTURE 3
The problem can be solved by dividing one whole number (3) by another whole number (3): 3
quarters of a whole to be shared by 3 girls.
Here, division is an answer to the question: A quantity Q is divided into n parts: how much of the
quantity in one part?
The division does not have to be done numerically. It is enough to visualize 3 things and their
distribution among three people:
2. Tom and Jimmy were to make a box for a game to be played at the Halloween party. They needed 4 boards
each 3/4 ft long. The janitor gave the boys a board 3 ft long. How many boards each 3/4 ft long could they
have cut from the 3 ft board?
The problem is not a ‘division problem’; it is an ‘addition-in-disguise problem’. It can be solved
by adding fractions, or even whole numbers. Four boards each 3 quarters of a foot add up to 12
quarters of a foot board, which is a 3 feet board. So the janitor gave Tom and Jimmy enough
board to make their box: they could cut their 4 boards from the 3 feet board.
3. Henry brought 3/4 of a bushel of walnuts to the party. He divided the nuts into 50 equal shares. Each share
was what fraction of a bushel?
One could solve the problem by dividing 3/4 by 50 (a fraction by a whole number), but this is not
the only way. Actually, one can do with the notion of sharing a quantity Q (= one quarter of a
bushel of nuts) among n (=50) people, and continue with the operation of addition. To answer the
question in the problem one needs to understand fractions as ‘p parts out of a whole made of q
parts’. Here is how the reasoning could go:
Henry had 3 quarters of a bushel of walnuts. He divided each one of these quarters into 50 equal
parts. The whole bushel would thus be composed of 200 such parts (a bushel is composed of 4
quarters of a bushel, and 50 + 50 + 50 + 50 = 200). He took one such part from each quarter to
make a share (1+1+1): so he took 3 out of 200 parts, or 3/200 of a bushel.
4. Children had a peanut relay race. Each team ran 7/8 of a block, and each pupil on the team ran 1/8 of a
block. How many pupils were on each team?
This problem can be solved by dividing one whole number (7) by another whole number (1).
The problem is a division problem of the type: There are m things and these things are divided
into groups of n things: how many groups can be formed? The division can be performed by
repeated subtraction: 7 - 1 - 1 - 1 - 1 - 1 - 1 - 1 = 0, so 7 ones in 7. It can also be solved by
counting, after having visualized 7/8 as 7 segments of equal length. It is completely irrelevant that
eighths of a block are considered.
5. Each of the girls on the Refreshment Committee served 1/2 of a pumpkin pie at the party. The pies had been
cut so that each piece was 1/8 of a whole pie. Into how many pieces was each 1/2 pie cut?
11/4/03 6
LECTURE 3
The textbook probably expected the students to write 1/2 : 1/8 in answer to this problem, but the
problem can be solved by multiplying a number by 1/2: there are 8 parts in each pie, so half of
the pie has half the number of parts, i.e. 4. Here the fraction 1/2 is seen as an operator.
Again, visualizing and counting could be enough to find the answer, without writing any
arithmetic operations.
6. 9/10 ˆ 15/16
The text probably expects the students to use the procedure for the division of fractions:
‘When the divisor is a fraction, we must rewrite the example, making two changes:
1. Change the division sign to a multiplication sign
2. Invert the divisor (p. 69)’
introduced in the same section of the book.
This procedure uses the algebraic notion of division as multiplication by the inverse.
This notion is in sharp contrast with the arithmetic notions of sharing, partitioning or counting
that were sufficient for solving the introductory word problems, in which numbers did not appear
as pure numbers but as quantities of something. In fact, these ‘introductory problems’ did not
build the expected scaffolding for the notion of division now officially introduced, and no
conceptual link can conceivably be formed on the basis of just the problems such as the above.
Any simple arithmetic problem about concrete quantities can be solved with some arithmetic
notion of division. The notion of division as multiplication by the inverse is necessary in the
problem of creating a unified theory of number systems, and this is certainly not a task accessible
to 6 grade children. If the formal rule is given at this point, the students will never be given a
chance to make this link, because the work needed to make it is far too costly in terms of
conceptual effort in comparison with the mechanical application of the rule.
Let us try to understand what would be involved in solving the present problem without
knowing the ‘rule’. I claim that treating this problem as a ‘real problem’ and giving it to the
students prior to giving them the rule, in the form of: ‘What would it mean to divide 9/10 by
15/16?’ could actually create a situation in which the students would have the chance to finally
going beyond their whole number conceptions of division and reorganizing it into a new
conception, namely the ratio conception of division. This notion is still an arithmetic notion,
albeit a quite elaborate one. The link with the previous conceptions would thus be made, and the
leap into the abyss of empty formalism would be avoided.
Let us speculate how a student, who was not given the ‘rule’ before, could solve this
problem. Suppose the student understands this question as: "how many 15/16 in 9/10?". By
representing these two fractions as parts of a whole (e.g. a strip of 10 squares), the student may
11/4/03 7
LECTURE 3
realize that 15/16 is greater than 9/10 by a little and it does not fit into 9/10 a whole number of
times. He must, therefore, understand the question differently and thus change his notion of
division: not ‘how many parts worth 15/16 of the whole in a part worth 9/10 of the whole’,
because ‘how many’ presupposes a whole number as an answer, but ‘what is the ratio between a
part worth 9/10 of the whole to the part worth 15/16 of the whole’? He could then see the whole
as composed of, say, 160 little parts; 9/10 of the whole is composed of 9*16 = 144 such little
parts, and 15/16 of the whole is composed of 150 such little parts. So the ratio of the 9/10 part to
the 15/16 part is the same as 144 to 150, which is the same as 24 to 25.
We can see that in this reasoning, the hypothetical student went from viewing division as
partitioning or grouping (how many groups of so many elements) to viewing it as the operation of
estimating the ratio of two magnitudes.
Let us notice also that the notion of division as multiplication by the inverse is not
necessary for the solution of the problem. Indeed, a generalization of the solution given above
could lead to the following algorithm of the division of a fraction by a fraction: to divide a/b by
c/d , represent the whole as bd little parts. Then a/b is ad of these parts, and c/d is cb of
these parts. Then a/b divided by c/d represents the ratio of ad little parts to cb little parts.
Symbolically: a/c : c/d = ad : cb. Of course, the ratio can sometimes be simplified, like in
fractions.
7. 5/8 ˆ 15/16
The method elaborated in the previous problem allows us to treat this question as a simple
application of the method: we can see the whole as composed of 128 little parts, 5/8 as 80 of
these parts and 15/16 as 120 parts; the answer is the ratio of 80 to 120, which can be represented
by the reduced fraction 2/3.
8. The cookie recipe that Mrs. White planned to use called for 3/8 cup of chocolate. She had only 1/4 cup of
chocolate. What fraction of the full recipe could she have made with that amount of chocolate?
The problem can be solved using the notion of division as representing a ratio of quantities and a
notion of the equivalence of one quarter and two eighths: one quarter of a cup is the same as 2
eighths of the cup. The amount of chocolate that Mrs. White had was thus to the full amount
needed as 2 to 3. So she could only make two thirds of the recipe. This is a kind of
proportional reasoning that does not require the writing of any arithmetic operations, and
certainly not the operation implicitly expected by the authors of the textbook, namely 1/4 : 3/8 =
1/4 * 8/3 = 1/1 * 2/3 = 2/3.
9. Divide and put your answer in simplest form: 9/10 ˆ 3/5.
11/4/03 8
LECTURE 3
Division as ratio can be used here again; 3 fifths is the same as 6 tenths, so 9 tenths is to 3 fifths
as 9 tenths to 6 tenths, i.e. as 9 to 6 or as 3 to 2.
This appears to be quite simple, but this is not what the text expects the students to do:
the ‘simple form’ means the ‘mixed number’ form, i.e. 3/2 is expected to be represented as 1 1/2.
But this representation requires to understand division not arithmetically, as ratio, but
algebraically, as an operation on numbers, yielding a number, and not some new kind of entity.
Seeing ratios as numbers and extending the four arithmetic operations from whole numbers to
fractions is epistemologically quite difficult, because, from the former domain to the latter, these
operations change their meaning. One can no longer think of addition as ‘bringing together, or of
division as sharing, or partitioning a quantity. In the history of mathematics, a unified notion of
number has been, de facto, an achievement of the nineteenth century.
The solution of the problem, as it stands, without the implicit ‘didactic contract’
conveyed by the kind of worked out examples preceding this problem, does not require the above
sophisticated notion of division as an arithmetic operation on numbers. The notion of ratio of
whole numbers is sufficient. If a student solves the problem with the notion of division as ratio,
and is punished by the teacher because she did not transform 3/2 into 1 1/2, this is a sign for her
that the rules of the game are not what she taught they were. She does not know what is wrong
with her thinking and she is unable to yet understand the notion of division of fractions in an
algebraic way. Different scenarios are possible from this point on for the student, some ending in
frustration and failure, some - in the student suspending her ‘situational’ understanding of
fractions and happily engaging in the formal game on fractional expressions, changing division
into multiplication, inverting, simplifying, extracting the whole part, etc.
We can see now that, in the textbook, the ‘algebraic notion of division’ does not emerge
as a ‘conceptual necessity’ for solving problems, but is enforced by a ‘didactic contract’ in a
situation of institutionalization, as if it were a law and not a mathematical concept.
10. On Halloween the Pine Hill School had some Hard Luck races. The route for the races was in three laps.
The first lap was from the school to Five Corners: 1/4 mile. The second lap was from Five Corners to Orr’s
Sawmill: 7/8 mile. The third lap was from Orr’s Sawmill to the school: 3/4 mile. Helen said that the second lap
of the route was 3 1/2 times as long as the first. Jane said that it was 3 3/8 times as long. Which girl was correct?
The problem does not require division, but multiplication of fractions and comparison of
fractions. A possible line of reasoning could be: If Helen is right then the second lap would have
to be 3 1/2 times 1/4 mile. This is equal to three quarters of a mile plus a half of one quarter, i.e.
six eighths of a mile plus one eighth, i.e. seven eighths of a mile, so she is right. If Jane were also
right then the second lap would have to be 3 3/8 times 1/4 mile. Here the multiplication becomes
more complicated, so a mental calculation may be risky and it is better to write down the
11/4/03 9
LECTURE 3
operations:
3/4 mile + 3/8 * 1/4 mile = (3*8)/(4*8) mile + 3/(8*4) mile = (3*9)/(4*8) mile = 27/32 mile.
Now, 7/8 mile is the same as 28/32 mile, so Jane’s estimation is inaccurate by 1/32 mile.
If a student chose to solve the problem by dividing 7/8 by 1/4 and checking whether it is
equal to Helen’s proposal or Jane’s proposal, then a new conceptualization of division would be
required; ‘new’ with respect to the conceptualizations needed in the previous problems. To solve
the problem with division, the student must first understand the mutually inverse character of
division and multiplication: if y is a times x then a is y divided by x. In the problem, y and
x are magnitudes - distances, in miles; a is a factor, a ‘pure number’, so to say. Division could be
understood as a ratio, but the ratio must be treated as a number, otherwise the expression ‘a times
as long as’ wouldn’t make sense. The ‘must’ is a constraint of the problem and the chosen
strategy, not of a didactic contract, as in the previous problem.
Here is a possible line of reasoning: 1/4 of a mile is the same as 2/8 of a mile, so the
second lap is to the first as 7 eighths to 2 eighths. So the ratio is 7/2 or 3 1/2. Therefore the
second lap is three and a half times longer than the first, which is what Helen said, and not what
Jane said. So Helen is right.
11. Divide: (a) 912 / 76 (b) 35351 / 431
Question (a) could be solved by repeated subtraction and counting how many times 76 fit into
912 (12 times). But applying the same strategy to (b) would be a bit tedious: 35351 = 431*82 +
9. One could, of course, use some informed guessing: there are 80 four hundreds in 320
hundreds, so let’s try 431*80: this is 34480. Adding 431 twice to the result yields 35342 and
there is 9 left. In this reasoning, numbers are whole numbers and the division is the ‘Euclidean
division’, which identifies the integer number of times a number fits into another number and
computes the remainder. The textbook expects the students to use the long division algorithm
and represent the result in the ‘mixed number’ form: 82 9/431.
12. Woods family went to the State Fair. Father and Andy drove to the fair in the truck, taking some cattle to
be entered for prizes. Mother and Ruth drove the family car. On the way to and from the Fair, Father used a
total of 24 gallons of gasoline and 5 quarters of oil for the truck. The gasoline cost 18 cents per gallon, and the
oil cost 30 cents per quart. Father drove the truck 107 7/10 miles in going to the fair and 108 3/10 mile in
returning. Besides the cost of the gasoline and oil, the expenses for the truck were $1.00 for repairing a tire. To
the nearest cent, what was the cost per mile for the truck for the round trip?
This is ‘multiple step’ problem with several calculations to be made before the answer can be
given, but the ultimate operation is division.
(24 gallons of gas * .18 $/gallon + 5 quarters of oil * .30 $/quarter + 1.00 $) : (107 7/10 + 108 3/10)
= 3 cents per mile
11/4/03 10
LECTURE 3
Division has to be understood as a ratio here, but the ratio is not a pure number, it is a magnitude,
because magnitudes of different kinds are compared: cost per mile.
This is again something new with respect to the conceptions needed in the previous
exercises, where ratios of homogeneous magnitudes were considered only. In the history of
mathematics and physics, the notion of ratio of magnitudes of different kinds was very slow to
appear. In Euclid’s ‘Elements’, Book V devoted to the ‘theory of proportions’, starts with several
definitions regarding ‘magnitudes’ and ‘ratio of magnitudes’. Definition 3 states: ‘A ratio is a
sort of relation in respect of size between two magnitudes of the same kind’. For Euclid, area and
length, weight and volume were not magnitudes ‘of the same kind’. Also areas of circles and
areas of squares were not of the same kind, because figures bound by straight lines and figures
bound by curves were regarded as totally distinct entities. Thus, one of the theorems states:
Circles are to each other as the squares built on their diameters. In today’s terminology and way
of thinking about numbers we would say: the ratio of the area of a circle to the area of the square
built on its diameter is constant and equal to the number Pi. But Euclid refused to measure the
area of the circle with the area of a square and thus he could not obtain, in his system, the number
Pi. The requirement of comparing magnitudes of the same kind only has been an obstacle to the
development of mathematics for many centuries, but when it was finally overcome and
mathematicians started to allow themselves to use ratios of non-homogeneous magnitudes and
think of ratios as numbers which could equal to each other and not be just ‘proportional’, they
could invent the notions of velocity (distance per time), acceleration (increase or decrease of
velocity per time), and the differential calculus.
The point of all these historical remarks is to stress the non-trivial character of the new
conceptualization involved in solving the present problem using division.
13. Divide 3/4 by 5/9.

The problem can be reduced to finding a ratio of two whole numbers, 27 to 20: 3/4 is the same as
27/36 and 5/9 is the same as 20/36, so 3/4 to 5/9 is the same as 27 to 20.
14. Nancy earned her Christmas money making Christmas cards. She bought 2 sheets of cardboard at 5 cents
each, a bottle of drawing ink for 25 cents, and some watercolors for 25 cents. (A) How much did all these
things cost? (B) The cardboard sheets were 22 inches by 28 inches in size. She cut each sheet into strips 22
inches long and 5 1/2 inches wide. How many of the 5 1/2 inch strips did she cut from the 2 sheets? How many
pieces were too narrow for her to use?
Both questions (a) and (b) can be solved by addition.
A. 5 + 5 + 25 + 25 = 60 (cents)
B. Suppose Nancy cuts across the longer side. One strip of 5 1/2 and another of 5 1/2 make 11
inches; with another pair of 5 1/2 inch strips she has already disposed of 22 inches, so she could
11/4/03 11
LECTURE 3
cut yet another strip, and thus 5 strips come up to 27 and a half inches, and a strip of one half inch
wide and 22 inches long is left over from one sheet of cardboard. So, from the two cardboards
she could cut 10 5 1/2 inch strips and had two pieces two narrow for her to use.
15. Sally and Ruth decided to make some valentines which would be different from those they could buy in the
stores. They bought a sheet of red paper 22 inches by 28 inches. Each girl took 1/2 of it. How many hearts
could each girl have cut from her share, if each heart used up to 1 square inch of paper?
The only division that has to be done in this problem is 22:2 or 28:2 and the only interesting
question that the students may have about this problem is whether it matters how the paper is cut
in two: across the longer side of across the shorter side. Otherwise the problem can be solved by
multiplication understood as repeated division, or even by drawing and counting:
15. 155.8 / 2
16. 512 / 32
17. .828 / 6
The textbook regards the above three exercises as practice in applying the algorithm of the long
division of decimals. But problem # 15 could be done mentally, by representing it as sharing the
amount of 155 and 8 tenths of something between two people. Splitting 155 into 2 gives 77 and a
half. Splitting 8 tenths in half gives 4 tenths. So altogether each person gets 77+(1/2 + 4/10) =
77 + 9/10 = 77.9
Problems # 16 and 17 differ from # 15 in that the divisor is greater than the dividend, so
division has to be seen as a ratio. In # 16 the dividend is still a number greater than 1, in # 17 it is
less than 1, but this may not make a big difference.
How small is 5.12 compared to 32? Changing the unit one can say that 5.12 is to 32 as
512 is to 3200, which can be simplified to 128 to 800 (dividing by 4), and then to 16 to 100
(dividing by 8). The ratio 16 to 100 can be written in decimals as .16.
Problem #17 could be interpreted as an exercise in division of a whole number by a
whole number, in its conception as sharing or partitioning: changing the units one could think of
.828 as representing 828 grams of something (e.g. chocolate), and then partition or share this
amount into 6, obtaining 138 grams. This can then be represented as 0.138 of the kilo.
In each case, as we could see, the long division algorithm for decimals could be avoided
at almost no cost.
18. During 8 hours on Tuesday there was .96 inch of rainfall. This was an average of what decimal fraction of
an inch per hour?
Like in the problems above, long division could be avoided. We could think of .96 of an inch as
being 96 ‘centi-inches’, and partition these 96 units into 8 hours (by repeated subtraction, or
11/4/03 12
LECTURE 3
repeated addition or a combination of multiplication and addition). This gives 12 units per hour.
The 12 ‘centi-inches’ can be written as .12 of an inch.
19. Mr. Burns and his family drove their car and trailer to Arrow Head camp to spend a few days. They drove
297.5 miles in 8.5 hours in travelling to the camp. How many miles per hour did they average?
In this problem division has to be thought of as a ratio resulting in a new magnitude and not a
pure number. Average velocity in miles per hour has to be calculated: 297.5 / 8.5 [mi/h]. Long
division algorithm for decimals can be avoided by a process of rough estimation, multiplication
and addition. For example, let’s first look at how many times the 8 hours could fit into 297 miles.
If our guess is 30, then we calculate 8*30 = 240 and we add 30 halves, which is 15. So 8.5 * 30
= 255. This falls short of 297.5 miles by 42.5. Now 8*5 is 40 and 5 times a half is 2 and a half,
so 8.5 * 5 = 42.5 which is exactly the missing remainder of the road. Hence 8.5 * 35 = 297.5 and
so the average velocity was 35 mi/h.
20. 7581.6 / 7.8

This problem can be reduced to division of whole numbers, sharing or partitioning 75816 into 78.
The result is a whole number, 972.
21. Mr. Mills told Ned and Alice that they could sell vegetables during the summer and keep half of the profits.
Mr. Mills helped Ned build a stand. To make the boards below the shelf, they sawed up some 14-foot boards.
How many boards 3.5 ft long could they have sawed from each 14-foot board?
The problem could be solved by addition: adding 3.5 + 3.5 + .... until something close to 14 is
obtained. In this case, adding 3.5 four times gives exactly 14, so the answer is 4.
22. 3 / 1.25
This problem can be reduced to finding the ratio of 3 to 1.25. Making the unit hundred times
smaller, we can think of the problem as finding the ratio of 300 to 125. This ratio can be
simplified to 12/5. Representing this ratio in mixed number form we get 2 and 2/5. The ratio 2/5
can be represented as 4/10, so the result is 2.4 in decimal notation.
Analysis of the 1998 textbook problems
1. Look at the two series of operations. How do the divisors and results change? Can you find the missing
results?
8:8=1 3/16 : 8 = 3/128
8:4=2 3/16 : 4 = 3/64
8:2=4 3/16 : 2 = 3/32
8:1=8 3/16 : 2 = 3/16
8 : 1/2 = ? 3/16 : 1/2 = ?
8 : 1/4 = ? 3/16 : 1/4 = ?
8 : 1/8 = ? 3/16 : 1/8 = ?
11/4/03 13
LECTURE 3
(a) Add two more operations to each column

(b) By what number should 3/16 be multiplied in order to obtain 3/128?
(c) By what number should it be multiplied to obtain 3/64?
(d) What operations could replace each of these divisions? Do you see a rule?
(e) Write a similar series of operations and give the results.
The notion of division that appears to be aimed at by this series of exercises is the formal
mathematical one: to divide by a number means to multiply by its inverse.
Question (a) of this problem could be solved without even looking at the left hand sides
of the equalities, and the implicit didactic goal of the exercise could be missed. Questions (b) and
(c) appear to aim at re-focusing the students’ attention on the left hand sides of the equalities.
Then, perhaps, looking for a pattern in the first column the student might come to a
generalization: when you divide a whole number a by a fraction 1/c (c being a whole number),
the result is the whole number a*c. Looking for a pattern in the second column, the student
might come up with two ‘rules’: (1) in dividing a fraction a/b by a whole number c, the result
is a/(b*c), (2) in dividing a fraction a/b by a fraction 1/c , the result is a times the ratio of c
to b. To answer question (b), the first rule could be reformulated as, ‘a/b : c = a/b * 1/c’.
However, the rules obtained by the students might be a lot more specific than the target
general notion of division as multiplication by the inverse.
2. Mom said to Jacek: I bought 6 liters of honey. We’ll pour it into 1/2 liter jars. Bring the jars from the cellar.
(a) How many jars should Jacek bring? (b) How many jars of 1/4 l would he have to bring? And how many
jars of 3/4 liter would he have to bring?
The notion of division as multiplication by the inverse is not necessary to solve the problem. One
can solve it by drawing a diagram and counting:
(a) The diagram could be made of 6 pairs of squares, each square representing 1/2 l. So 12 jars
are needed for 6 liters of honey.
(b) The diagram could be made of 6 groups of 4 squares, each square representing 1/4 l. So 24
such jars are needed for 6 liters of honey.
(c) The diagramcould be made of 24 squares; partioning these into groups of 3 gives 8 (jars).
3. A quotient is equal to the divisor and it is 4 times larger than the dividend. What is the dividend?
To solve this problem it is not necessary to understand division as multiplication by the inverse.
Division as ratio, with ratio understood as number, is enough. But it is necessary to understand
the transitive property of equality (if A = B and A = C then B = C), and it is quite useful to be
able to represent unknowns by letters. This is more an exercise in logical and algebraic thinking
than in division of fractions. Here is how the reasoning might go: Let the dividend be a and the
divisor b. We are told that a : b = b and a : b = 4a. This implies that b = 4a (by transitivity of
equality). Then a : b = a : 4a = 1/4 (a whole to four times a whole is as 1 to four).
11/4/03 14
LECTURE 3
But a : b = b, so b = 1/4 (transitivity of equality). But b is 4 times as large as a, so a is a

quarter of b; but a quarter of a quarter is 1/16, so a = 1/16.
However, the notion of division as a ratio is not given a chance to develop in the textbook, so
the students can be left with the formal notion of division, and if they are not too good at formal
algebraic and logical thinking, they have no chance of solving the problem unless they treat it as a
riddle to be solved by guessing or experimenting on their calculators.
4. Find a number which is 4 times as large as the quotient of the numbers 3 1/2 and 2 4/5 increased by 1.
Division as ratio is sufficient to solve the problem: 3 and a half can be seen as 7 halves; 2 4/5 is
14 fifths; finding a common measure (one tenth) leads to seeing 7 halves as 35 tenths and 14
fifths as 28 tenths; so the ratio is 35 to 28 which is the same as 5 to 4. Now the unknown
number is 4 times 5/4 + 1, so it is equal to 5 + 4, i.e. the number is 9. The problem is mainly
an exercise in translating verbal representations of relations between numbers into arithmetic
operations. It is also a preparation for algebraic thinking.
5. 2 1/3 + 3/4 : 1/2

Students guided by this textbook will use the rules of arithmetic on fractions, and the algebraic
notion of division as multiplication by the inverse. One can, however, expect a lot of mistakes,
because the students have no means to verify, by themselves, if their solution is correct, and the
meaningless rules are easily forgotten. If the students could create for themselves a model of the
expression, then they could perhaps judge of the validity of their result.
6. -12,8 x (-0,2)
Early introduction of operations on rational numbers (and not just decimals) is another symptom
of the algebraic character of the notions of operations on numbers favored by the textbook.
7. 3 1/3 : (-5/6) : (-2)

The main problem here for the students is the order of operations, which is based on pure
convention, and not on any rational base.
8. Decide which product is less expensive

(a) Margarine sold in cups of 250 g for 1,32 zl vs margarine sold in cups of 500 g for 2,49 zl.
(b) Yogurt sold in cups of 150 g for 0,93 zl vs yogurt sold in cups of 500 g for 2,60 zl.
Problem (a) can be solved by proportional thinking: since 500 is the double of 250, it is enough to
find the cost of two cups of the margarine sold in cups of 250 g for 1,32 zl each: that would be
2,64 zl, which is more than 2,49. So the margarine sold in cups of 500g is less expensive.
Also problem (b) can be solved avoiding division, and using only proportional reasoning:
3 cups of the first product is only 450 grams, but it would cost 3*0,93 zl = 2,79 zl, which is
11/4/03 15
LECTURE 3
already more than the 500 grams of the second product. So it is less expensive to buy yogurt in
the larger cups.
11/4/03 16
1999 MATH 645 CONCORDIA UNIVERSITY
LECTURE NOTES ON THE THEORY OF DIDACTIC SITUATIONS – ANNA SIERPINSKA
LECTURE 4
THE TOPAZE, THE JOURDAIN AND THE DIENES EFFECTS IN THE
PROCESS OF MATHEMATICS TEACHING
PART I - PROBLEMS OF TEACHING DIVISION - CONTINUED
In analyzing the two sets of division problems in Lecture 3, we found the following conceptions
of division:
Elementary arithmetic conceptions:
Sharing: A whole number N of objects is shared among a whole number of persons P (or is put in
boxes, in piles, etc.). N / P represents the number of objects that each person gets (there might be
a remainder).
Partitioning: A quantity Q is distributed in packages of q; Q / q represents the number of
packages. (Q and q can be any numbers)
A geometric conception:
Ratio as a comparison of two magnitudes of the same kind: if A and B are amounts of two
magnitudes of the same kind then one can sometimes express the relation between these amounts
as a proportion, ‘A is to B as p is to q’, where p and q are whole numbers, and write the ratio as a
fraction p/q, without necessarily understanding it as a number in the same sense as whole
numbers are numbers; in this conception, a ratio could still be understood as a pair of whole
numbers.
A physical conception:
Ratio as a compound magnitude: if A and B are amounts of two magnitudes, not necessarily of
the same kind, then the ratio of A to B expresses the amount of the first magnitude per a unit of
the second magnitude.
An advanced arithmetic conception:
Ratio as pure number: in the conception of ratio as a compound magnitude, the ratio of the
amounts A to B is abstracted from the magnitudes, and treated as a pure number.
An algebraic conception:
Multiplication by the inverse: a/b = a*b-1.
These conceptions form a hierarchical list in the sense that every next conception (except
perhaps the first two) can be construed on the basis of the previous ones, through generalization,
abstraction, and extension of the meaning of terms. But every next conception has to be
construed ‘against’ the previous ones because it requires a change of perspective, a certain
reorganization of knowledge and even a rejection of beliefs related to the previous ways of
understanding division.
In both textbooks, division was defined algebraically as the multiplication by the inverse
but the students were not given a chance to construct this conception on the basis of and against
the previous conceptions. The word problems given to the students could all be solved either
with the elementary arithmetic conceptions or were so formulated that the use of the division
operation was almost explicitly suggested. Having been given the procedure(s) for performing
division the students were able to solve the problems without having to develop the geometric
and physical notions of division. But without these conceptions, they had no grounds for
developing a meaning for the procedures, and therefore could only learn the procedures but not
the algebraic conception of division. The huge gap between the arithmetic conceptions of
division and the algebraic conception has not been filled by an appropriate choice of problems
and didactic contracts.
Everything would be fine if the authors of the textbooks said: our aim is only to teach the
students some algorithms of dividing ordinary and decimal fractions or rational numbers.
However, if asked, they might claim that they teach the students the notion of division, and
support their claims by arguments such as:
(1) Aren’t the students choosing the operation spontaneously in solving problems when
appropriate?
(2) Don’t the students actually correctly perform the operation of division?
Our answer would be: Yes, the students were choosing division in solving problems but
this operation was strongly suggested by the text or the context of these problems1; the hint was
1
Most problems in the 1936 textbook were included in sections whose titles explicitly mentioned division, like
‘Problems using division of fractions’ (p. 71), and many had verbal clues within the text of the problem, like ‘He
divided the nuts into 50 equal shares’, in Problem # 3, or ‘what fraction of the full recipe…’ in Problem #8 , or
‘what was the cost per mile…’ in Problem # 12.
so strong that the students did not have to choose among other possibilities - they knew, by virtue
of the implicit contract, that division was what they were expected to do. This phenomenon of
giving the answer in the question has been called by Brousseau: ‘the Topaze effect’ (p. 25).
Concerning argument (2), what can be called ‘performing the operation of division’ by a
mathematician, is not that at all from the point of the student, who is just, for example, changing
the sign of “/” to “*”, inverting the divisor, and using the multiplication table. This second
phenomenon, of giving a scientific name to a trivial activity has been called ‘the Jourdain effect’
(p. 25-26).
Exercise
Find a didactic situation through which the student would be led to construct one of the ratio
conceptions of division without being coerced into using this concept by an explicit hint from the
teacher.
Here is what I think might work with students (but I would need to experiment it to verify
my conjecture) with respect to the ‘physical conception of ratio’. I would give the students a
project titled: ‘How much of your own weight do you eat?’ Students could investigate this
problem concerning themselves, members of their family, their pets (situation of action). They
would come back with some data to class, and present the data to the class, perhaps in the form of
tables. The problem would then arise of comparing, who eats more than someone else, not in an
absolute way, but relative to one’s weight (situation of formulation). For example, children may
find out that a cat weighing 5 kg eats 200g of food daily and a dog weighing 20 kg eats 1 kg of
food daily. Ways of making such comparisons would then be proposed and defended by different
students or groups of students (situation of validation). For example, it could be proposed to
compare the ‘food weight per body weight’ quantity. In the case of comparing the absorption of
food by the cat and the dog, the question would be reduced to comparing the ratios 200 g to 5000
g and 1 kg to 20 kg. The main problem would then be: how can one compare such ratios?
Maybe some students would propose to represent the ratios by fractions: 200/5000 and 1/20, and
use the techniques of comparing fractions. If not, then the teacher could suggest this and
negotiate the sense of using this representation with the students (institutionalization). Once the
students are convinced about this representation, the teacher would pose a question in which the
weights themselves would be represented by fractions of some whole and then the ratio would be
a ratio of fractions, thus leading to considering the division of fractions as a ratio. For example,
the teacher could propose to compare the food weight/body weight ratio of a chickadee and a
sparrow, saying that a chickadee eats 9/10 of a small bird feeder daily and weighs 15/16 of the
weight of the feeder, while for the sparrow the ratio is 21/25. Who eats more of his own weight
and how much more? The students would then be left on their own to figure out the answer
(action), then formulate their strategies (formulation), and compare their strategies for validity
(validation). The teacher would then ask the students if it would be possible for them to
generalize their results and say how to compare such data as in the situation with the cat and the
chickadee. With some hints from the teacher (institutionalization) a formulation, verbal or using
letters, of some rule of dividing fractions in the sense of finding the ratio between two fractional
quantities could be written. The reason why not just ratios are considered in this sequence of
problems but comparison of ratios is that division is necessary only when a comparison has to be
made. Otherwise one would be satisfied with saying that, for example, for the chickadee, the
food weight / body weight ratio is as 9/10 to 15/16, and no further processing of the data
(representing the ratio in the form of one number) would be necessary.
PART II - THREE DIDACTIC PHENOMENA: THE ‘TOPAZE EFFECT’, THE ‘JOURDAIN EFFECT’, AND
THE ‘DIENES EFFECT’.
1. The ‘Topaze effect’
The name of the ‘Topaze effect’, which has been described above as ‘giving away the answer in
the question’ in teaching, comes from a play by Marcel Pagnol, written in 1928, in Paris. Pagnol,
born in 1895 in Aubagne, near Marseilles, had been a teacher and taught English in various lycées
in Southern France and then in Paris. He abandoned this profession in 1922 and devoted himself
entirely to writing for the theater and then for the cinema. The play ‘Topaze’ is set in a private
boarding school. Topaze is a teacher in that school. The first scene of the play shows Topaze
giving a dictation to a pupil during the recess. The boy, as described by Topaze in Scene III, ‘is a
conscientious worker but he had some trouble keeping up with the class because no one seems to
have taken an interest in him until now’, and Topaze decided to help him a little in his free time.
Here is how the first scene starts:
As the curtain rises, Mr. Topaze is giving a dictation to a pupil.… The pupil is a 12-year-old boy. He is
turning his back to the public. One can see his ears that stick out and his thin bird-like neck. Topaze is
dictating and from time to time bends over the shoulder of the little boy to see what he is writing.
TOPAZE: (He dictates while he walks up and down). “Some… lambs… Some lambs… were safe… in a
park; in a park. (He bends over the shoulder of the pupil and continues) Some lambs… lambz (The pupil
looks at him, bewildered). Now, child, make an effort. I am saying lambz. Were (he repeats very
distinctly) we-re. That shows that there was not only one lamb. There were several lambz”. (The pupil
looks at him, dazed).
(Marcel Pagnol, Topaze, Translation and Introduction by Renée Waldinger. Great Neck, N.Y.: Barron’s
Educational Series, Inc., 1958, p. 2).
Topaze wants the student to succeed; after all, part of the didactic contract is the obligation, for
the teacher, to do all he or she can to help the student succeed. But the way he is going about it,
is not leading to the student’s learning, but to the student’s producing a correct answer in spite of
not having learned anything.
I found myself caught into the trap of this phenomenon in my linear algebra class. After
having introduced the notion of elementary matrices and discussed and illustrated the fact that an
elementary row operation on a matrix A has the same effect as multiplication of A, on the left, by
an elementary matrix, I gave the students a problem to solve. In the first question of the problem,
a 2x2 matrix was given and the task was to find two elementary matrices E2 and E1 such that
E2 E1 A = U, where U is the 2x2 identity matrix. The problem, I said, is a typical final
examination problem; this immediately arose the students’ interest. The students were working
on the problem in their exercise books, and I was walking around, looking above the shoulders of
the students. A few of them were row reducing the matrix, but many were slowly just copying
the matrix, and obviously did not know what to do. After 7 or 8 minutes some 6 students out of
the 32 in the class had solved the problem. I let the other students work for another 2 minutes,
and then got to the board, and said: ‘You have to row reduce the matrix and keep track of the
elementary operations; then you just translate the operations into the elementary matrices’. So I
told the students how to solve the problem, but did nothing to help them make the conceptual link
between elementary matrices and row reduction, or, in other words, to help them understand the
theorem. I thought I had explained the theorem sufficiently before, and I simply ignored the fact
that, most of the time, explanation does not automatically cause understanding.
2. The ‘Jourdain effect’
The name of this phenomenon of teaching alludes to the play by Molière, ‘The Cit turned
Gentleman’ (Le bourgeois gentilhomme), which was first acted at Chambord in October 1670,
during the reign of Louis XIV. The play is about the sin of vanity in men who endeavor to appear
above of what they actually are. In the play, Mr. Jourdain, a good but simple and not too highly
educated citizen with no relations to aristocracy, aspires to becoming part of ‘la noblesse’, if not
by blood then at least by manners and education. So he hires a music master, a dancing master, a
fencing master, and a philosophy master. The name of ‘Jourdain effect’ refers to Jourdain’s
lesson with the philosophy master (Act I, Scene 6), where Mr. Jourdain learns that he had been
speaking prose all his life without knowing it:
Mr. Jourdain: … I must commit a secret to you. I’m in love with a person of great quality, and I should be
glad you would help me to write something to her in a short billet-doux, which I’ll drop at her feet. …
Philosophy-Master: Is it verse that you would write to her?
Mr. Jourdain: No, no, none of your verse.
Philosophy-Master: You would only have prose?
Mr. Jourdain: No, I would neither have verse or prose.
Philosophy-Master: It must one or t’other.
Mr. Jourdain: Why so?
Philosophy-Master: Because, sir, there’s nothing to express one’s self by, but prose or verse.
Mr. Jourdain: Is there nothing then but prose, or verse?
Philosophy-Master: No, sir, whatever is not prose, is verse, and whatever is not verse, is prose.
Mr. Jourdain: And when one talks, what may that be then?
Philosophy-Master: Prose.
Mr. Jourdain: How? When I say, Nicola, bring me my slippers, and give me my nightcap, is that prose?
Philosophy-Master: Yes, sir.
Mr. Jourdain: On my conscience, I have spoken prose above these forty years, without knowing anything
of the matter; and I have all the obligations in the world for informing me about this.
We have to do with the ‘Jourdain effect’ each time we describe the productions of our students in
mathematical terms, which presuppose an elaborate conceptual activity, while having no evidence
of such an activity. We say that a student has divided a fraction by another fraction, while it
would have been more appropriate to say that the student has changed the sign of division into a
sign of multiplication, put the second fraction upside down and used his memory of the
multiplication table to produce two numbers separated by a horizontal little line.
We also say that ‘a student has solved an equation’ to refer to an activity of transforming
an expression containing numbers and letters into another such expression, according to certain
rules. Most students have no notion of equation, just as Mr. Jourdain had no notion of ‘prose’. In
a recent research made by an M.T.M. student, Sonia Manago2, a pair of 16 years-old students
were given a set of algebraic expressions to classify according to criteria of their own choice, and
their classification proves quite clearly that ‘solving an equation’ does not mean, for these
students, finding those values of the variables for which the equality condition expressed by the
equation is satisfied. The girls came up with the following classification, and names for the
categories:
2
Manago, S. (1999): Two Students’ Thinking Process on the Notions of Equation and Solution. Research Intership
Project, presented in the Master in the Teaching of Mathematics programme, Concordia University, Montreal.
Equal equations: 2=2, x+1=x+1, 3=7-4, x2=9, x/5 = 0.2 x
Not exactly equal equations: 0.99 = 1, 8=9, 3/4 = 5/6
Formulas: A = bh/2, P = xy/2, h = 2A/b, y= 2p/x, Ax+B = Cx+D
Equations that equal to zero: 2X = 0, 2a - 5 = 0, 2x - 5 = 0, X = 0, X + 5 = X
Simple equations: y = 2x, y = ax+c, y = x+1, y= x2 + 1, x2 + 1
Complicated equation: 2x - x
Y or X equals a fraction: y = 5/2h, x = 5/2, y = 5/2 x
In the interview with Sonia, one of the students expressed her doubts with accepting x=0
as an equation; she said, ‘I don’t really think it’s anything. I think it’s an error!’. Asked, ‘what is
a solution to an equation’, one of the students said, ‘The answer to an equation’, and the other -
‘It’s the answer to the equation you are trying to figure out. So if you have a long equation, you
do all the steps to get to the answer… It’s the final answer’. For the latter student, the solution to
x+5 = x was ‘5=0’, because this was ‘the final answer’.
2.1 An extreme case of the Jourdain effect: the mathematical program of Zoltan
Dienes.
Extreme cases of the ‘Jourdain effect’ could be observed during the wave of the so-called New
Math reforms in 1960s and 70s. Children were given all sorts of ‘manipulatives’ to play with,
toys, dolls, and blocks. When children were sorting toys, their activities were called using the
terms of set theory such as ‘finding the intersection of two sets’. The case of claiming that some
kindergarten or first grade children manipulating cups of yogurt ‘constructed the group of Klein’
became legendary; it was then used as a joke to ridicule the New Math approaches and unjustified
ambitions (p. 139).
A side explanation: the group of Klein
Here is an explanation of what the group of Klein is and what it may have to do with the
manipulation of cups of yogurt, for those of you who never heard of it before.
A ‘group’ in mathematics refers to an algebraic structure of a special kind. An algebraic
structure is a set closed under some operations defined on the elements of that set, and satisfying
some conditions (‘axioms’). From this point of view, the set of all real numbers with the
operations of addition and multiplication, is an algebraic structure; because of the properties of
these operations (commutativity, associativity, distributivity, existence of a neutral element for
each operation, namely 0 and 1, existence of the additive inverses and multiplicative inverses)
this structure is called a ‘field’. Now, a ‘group’ is any non-empty set, say G, closed under one
operation, say *, which satisfies the following conditions:
(1) the operation is associative
(2) there exists, in G, a neutral element for that operation (i.e. an element e such that for element
x in G e*x = x*e = x)
(2) for every element x in G there exists an element y in G such that x*y = e (i.e. every
element in G has an inverse in G).
‘Group’ is, obviously, a theoretical model of such familiar structures as the set of all
integers with respect to addition, the set of all real numbers except zero with the operation of
multiplication, the set of all real numbers with respect to addition, the set of all rotations of the
plane around the same point, the set of all dilations of the plane with respect to the same point,
etc.
All these examples are examples of infinite groups, i.e. groups composed of an infinite
number of elements. But there exist groups with a finite number of elements, and these were
considered appropriate to be introduced to very young children at the time of the New Math
reforms. The reformers were supporting their projects by reference to Jean Piaget’s
psychological theories, and, in particular, to his thesis that the most primitive cognitive
‘operational structures’ in the child resemble the three most general mathematical structures,
which he called ‘mother structures’, because all other can be derived from them: the algebraic
structures, of which the structure of group of transformations is the prototype; the order structures
for which the structure of lattice of relations is the prototype; the topological structure of objects
invariant under continuous transformations (Piaget, 1969, p. 673). According to Piaget, child’s
mental development follows the logical development of the theory, from the most general ideas to
the most particular ideas, rather than the historical order of the development of mathematics, from
knowledge of individual cases, through abstraction, to the general cases. The child is cognitively
organizing her experience in transforming objects according to the possibility of reversing and
combining these transformations in her mind if not in reality. The child’s world in organized in
hierarchies of relations between things, events and people that can be modeled by the structure of
lattice. The child’s geometry is closer to topology than the Euclidean metric geometry which is
taught at school; in particular, any closed smooth curve is put in one class by children and called
‘a round’ - topologically, all such curves are indeed, continuous one-to-one images of a circle.
3
Piaget, J. (1969): Psychologie et pédagogie. Éditions Denoël.
From this ‘logical parallelism’ he concluded that teaching of mathematics should be focused, in
early grades, on set theory and isomorphisms of structures. The New Math curricular movement
fully endorsed this point of view. However, Piaget was warning the reformers against assuming
that the child’s cognitive structures are some kind of conscious knowledge, and was even evoking
Molière’s Mr. Jourdain to make his point. For him, the discovery of the ‘logical parallelism’ does
not solve the pedagogical or didactic problem of teaching mathematics; the problem remains in
the form of finding a way of helping the students pass from an unconscious use of the structures
to a reflection and theorization of these structures in a symbolic language.
… il se trouve que ces trois structures mères correspondent d’assez près aux structures opératoires
fondamentales de la pensée. Dès les «opérations concrètes»… on trouve des structures algébriques dans les
«groupements» logiques de classes, des structures d’ordre dans les «groupements» de relations et des
structures topologiques dans la géométrie spontanée de l’enfant (qui est topologique bien avant d’atteindre
les formes projectives ou la métrique euclidienne, conformément à l’ordre théorique et contrairement à
l’ordre historique de la constitution des notions)… S’inspirant des tendances bourbakistes, la
mathématique moderne met donc l’accent sur la théorie des ensembles et sur les isomorphismes structuraux
plus que sur les compartimentages traditionnels, et tout un movement s’est dessiné qui vise à introduire de
telles notions le plus tôt possible dans l’enseignement. Or, une telle tendance se justifie pleinement,
puisque précisement les opérations de réunions ou d’intersections d’ensembles, les mises en
correspondance sources des isomorphismes, etc., sont des opérations que l’intelligence construit et utilise
spontanément dès 7 ou 8 ans et bien plus encore dès 11-12 ans… Seulement l’intelligence élabore et
utilise ces structures sans en prendre conscience sous une forme réfléchie, non pas comme M. Jourdain
faisait de la prose sans le savoir, mais plus encore comme n’importe quel adulte non logicien manipule des
implications, des disjonctions, etc., sans avoir la moindre indée de la manière dont la logique symbolique
ou algébrique parvient à mettre ces opérations en formules abstraites et algébriques. Le problème
pédagogique subsiste donc entièrement, malgré le progrès de principe réalisé par le retour aux racines
naturelles des structures opératoires, de trouver des méthodes les plus adéquates pour passer de ces
structures naturelles mais non réfléchies à la réflexion sur de telles structures et à leur mise en théorie
(Piaget, ibid. P. 67-68).
But let us come back to our explanation of the notion of ‘group of Klein’. Felix Klein was a
German mathematician of the second half of the 19th century, who contributed to the theory of
groups, among others. The group of Klein is a four-element group, whose all elements are
inverses of themselves. More precisely, any set {e, a, b, c} with an operation * such that the
following ‘multiplication table’ holds,
* e a b c
e e a b c
a a e c b
b b c e a
c c b a e
One can construct many different models of this group. One is, for example, the set of all
plane symmetries that transform a figure made of two parallel segments of equal length, such as
the equal sign "=", into itself. There are four such symmetries: the identity transformation, a
vertical flip, a horizontal flip and a central symmetry. It is easily verified that each of these
symmetries, when repeated, brings the figure back to the initial position and that the combination
of any two non-trivial out of them gives the third.
Another example of a model of this group can be constructed using some transformations
of the position of a cup of yogurt, which has some inscription or decoration on its side. Suppose
at the beginning the cup is standing upright with its decoration in the front. Four transformations
of the cup’s position are considered: the cup is left untouched (e); the cup is put upside down so
that the decoration goes to the other side (a); the cup is put upside down so that the decoration
stays on the same side (b); the cup is given a half turn (c). It is easy to verify that these
transformations form a group of Klein. As children were playing with the cups of yogurt, they
were performing such transformations and their combinations and were, maybe, noticing, that
doing the same transformation twice would put the cup back in the original position, and that
combining any two of the transformations a, b and c would give the third one, and from this, the
‘structurally minded’ observers were concluding that the children ‘have constructed’ the group of
Klein. But what the children were actually doing had nothing to do with the identification of the
group structure in their manipulations. They were even more in the dark than Mr. Jourdain,
because they would not have been able to identify the part of their activity called ‘construction of
the group of Klein’, as Mr. Jourdain could identify when he was speaking prose and when he was
not; and they would not be able to produce, as Mr. Jourdain could, further examples of their
activity, now sanctified by a scientific term.
3. The ‘Dienes effect’
The Dienes effect has more to do with the work of researchers and innovators in mathematics
education than with the classroom practices of teachers, like the previous two phenomena. But it
is important for us to understand because we are all, in this class, at least part time researchers
and innovators in mathematics education. It is important also for our understanding of the theory
of didactic situations, because this theory was born, in part, as a result of Brousseau’s
understanding of what was wrong with Dienes’ theory of mathematical instruction.
We start with presenting, briefly, the person of Zoltan Paul Dienes and his theory.
3.1 Zoltan P. Dienes
During the period of the New Math reforms Zoltan P. Dienes, a Hungarian by origin, but fluent in
English, French, German and Italian, was a professor at the University of Sherbrooke, in Québec
(Servais & Varga, 1971, p. 394). He became well-known to teachers and parents of elementary
school children around the world for his blocks (ibid., p. 38, 108) designed for the teaching of
position systems of writing numbers in various bases, as well as blocks for the teaching of logic
(the set, available commercially in some countries, was often called ‘Dienes blocks’). For
mathematics educators he became known for his theory of the ‘psychodynamic process’ of
teaching and learning mathematics. When the New Math reforms started being criticized and
withdrawn from schools, Dienes left Sherbrooke, and now lives in Wolfville, Nova Scotia,
Canada. He still writes books. His latest is "I'll tell you algebra stories you have never heard
before" (2002; Upfront publishing, Leicestershire). [the last three sentences were written in
2003].
3.2 The concerns of Zoltan Dienes in mathematics education; the definition of the
‘Dienes effect’
Dienes’ main problem with mathematics education was no different than that of Brousseau: the
discrepancy between the stress, put in mathematics education, on teaching of techniques and the
attention awarded to the teaching of the understanding of mathematical ideas (Dienes, 1960, p.
45). But, unlike Brousseau, he did not seek to find the causes of the phenomenon and ways of
shifting the pedagogical scales more towards understanding of the actions of the teacher and her
interactions with the students and the mathematical content. He ‘rule[d] out bad teaching as a
regular contributory cause to the present state of affairs’ (ibid., p. 5), and set to devise an
instructional theory and exemplary teaching materials that would be, in a sense, ‘teacher-proof’:
if administered according to precise instructions of the teacher’s guide, they should work with
any teacher.
But they didn’t; Brousseau has tried them himself and observed them when used by other
teachers and discovered that they only ‘work’ with a committed teacher who, wanting to prove
that they do work, would make special interventions and modifications, in order to give meaning
to the students’ actions and help the students become aware of this meaning. The materials did
4
Servais, W. & Varga, T (eds). (1971): Teaching School Mathematics. A UNESCO Source Book. Middlesex,
England: Penguin Books.
5
Dienes, Z.P. (1960): Building Up Mathematics. London: Hutchinson International.
not induce understanding in students who would work with a teacher who would only just hand
out the worksheets, and encourage the students to continue. This belief in the existence of some
kind of infallible ‘artificial genesis of mathematical knowledge’ that would be independent of the
teacher’s personal investment in the learning process has been called the ‘Dienes effect’ (p. 37).
3.3 Dienes’ instructional theory
The central concept of both Dienes’ theory and Brousseau’s theory is ‘game’. But for Dienes, a
game is a special type of play: it is defined as a rule-bound play (Dienes, 1963, p. 24) 6, and it is
not assumed that, in a game, one must have winners and losers, and that there are some strategies
for winning the game, as in Brousseau’s theory. In the theory of didactic situations (TDS), a
game is more of a problem situation; the problem is to find a way or a strategy to win the game.
The discovery of the finding of this strategy is done through the construction of some new
knowledge or a new understanding of some old knowledge, and it is this knowledge that is the
true outcome of the game for the winner: she has learned something new.
In TDS knowledge is the outcome of the game but the game does not resemble this
knowledge, and the players are not playing with that knowledge; they are using or expected to use
that knowledge to win the game. But ‘playing with knowledge’ or ‘playing the target
mathematical knowledge’ is exactly what the games are all about in Dienes’ theory. In order to
teach the students a mathematical notion, Dienes proposed to give students sets of manipulatives
so structured as to faithfully represent this notion. The students would thus be given some
physical representations of the mathematical notion. They would first be allowed to freely play
with one representation (Brousseau, TDS, p. 139). Then the play would become more structured:
the students’ attention would be directed to some properties or actions that would be specific of
the target mathematical notion. The students would then be given several other physical
representations of the notion and would be directed to focus on the things that were common in
all those different ‘games’ they were playing (the stage of isomorphic games and abstraction).
Next, the students would be encouraged and helped to construct a schema of all these games, and
formulate it in words or represent it by spontaneously drawn diagrams (the stage of
schematization and formulation). The sequence would stop at that stage in earlier grades. In
secondary school, the sequence could continue with a symbolic representation of the schema in
some accepted mathematical notation, and the formalization of the properties of the target notion
in the form of an axiomatic theory (the stages of symbolization, formalization and
6
Dienes, Z.P. (1963): An Experimental Study of Mathematics-Learning. London: Hutchinson.
axiomatization) (ibid.). The ‘notions’ that Dienes had mostly in mind were those closely related
to Piaget’s ‘mother structures’ and logic: algebraic structures such as groups of transformations or
vector spaces, combinatorial problems of ordering, sets and operations on sets, laws of predicate
calculus.
3.4 Examples of Dienes’ ‘lessons’
Example 1: Vector spaces
During a sequence of activities supposedly leading to the identification, by the children, of the
structure of vector space, children are given objects (or pictures of objects) of several kinds: two
kinds, three kinds, four kinds. The objects are so chosen as to appear in two ‘opposite’ states.
For example, in one activity, they are given cups, gloves and boxes. The cups can be right side
up, or upside down. The gloves can be right side out and inside out. The boxes can be open or
closed. The children are asked to count how many items of each kind they have, assuming that
the rule of this game of counting is that opposite states cancel each other out, so that if a child
has, in her set, an open box and a closed box, she puts the pair aside and does not count it
(Dienes, 1963, p. 32-33). For example, if a child has 3 closed boxes, 5 open boxes, 2 gloves right
side out and 1 glove inside out, 6 cups right side up and 3 cups upside down, her ‘book-keeping
record’ would show:
Boxes Gloves Cups
2o 2r 3u
The children would be then encouraged to put their and a neighbor’s collections together
and count the items. A bookkeeping record of such a ‘merger’ of possessions could be:
Boxes Gloves Cups
Jane 2o 2r 3u
Mary 5c 3r 1d
Together 3c 5r 2u
Children would also be asked question such as, ‘What would be the state of your record if
you doubled/tripled, etc. your possessions?’, ‘What would happen if your neighbor had the same
number of items of each kind but in the opposite state?’, etc.
Playing this way with a variety of collections of objects, the children would be
encouraged to speak about the common features of all such situations and notice some common
features, which could eventually be schematized in the form of, say, a set of book-keeping
records regarding a number of kinds of things, each kind of thing appearing in two possible
states; the totals of the records can be made by addition and subtraction of numbers in each
rubric.
From this kind of plays to the concept of vector space, claimed, by Dienes, to be the
target mathematical notion, there is still a long way to go. First of all, in a vector space, one must
be able to multiply vectors by any number, not just a whole number. In plays with discrete items
such as boxes or gloves, it does not make sense to multiply even by fractions: what is 1/2 of a
glove?! But Dienes was not claiming that the target had to obtained in a short sequence of
sessions. The sessions could be scattered along one year, two years or more, and the ultimate
stage of axiomatization would not be attained till the end of secondary school.
Example 2: A lesson in logic
The description of this lesson, actually conducted by Dienes in a French-speaking classroom in

Sherbrooke, comes from (Servais, Varga, ibid. p. 38-40).
Children are given a set of the so-called ‘attribute blocks’ (also called Vygotsky’s logic
blocks, or Dienes’ logic blocks). Each of the blocks can have one of the 4 kinds of attributes:
Shape: round, triangular, square, rectangular (but not square)
Color: yellow, blue, red
Size: big, small
Thickness: thick, thin
The children work on various problems in groups. Not all groups work on the same problem. In
one of the groups, the teacher suggests that the children pick out of the whole set four blocks.
Suppose the children have picked the following four blocks:
one triangular, red, small, thick
one square, yellow, small, thin
one square, blue, big, thin
one square, blue, small, thick
The teacher asks the children to formulate sentences of the form ‘If… then… ‘ about this set of
blocks and then decide which are true and which are not.
Children formulate sentences and write them down in two columns, one for true sentences and
one for false sentences. Here is a sample noted by the observers of the lesson:
True False
If red then triangle If red then square
If triangle then red If triangle then blue
If blue then square If square then blue
If thick then small If small then thick
If large then blue If blue then large
Now the teacher changes the assignment. He writes the last true sentence on a slip of paper and
asks the children to add a fifth block, which would make the sentence false. A child proposes to
add a large yellow and thin round block. Then the children are invited to produce sentences that
are true about this new set. One of the sentences is: ‘If large then thin’. The teacher again
proposes to add a block that would make that sentence false. And so it continues. When the
children arrive at ten blocks it is not easy to find a simple sentence that is true. So they start the
game over game, now with one of the children proposing a true statement about a few blocks, and
another falsifying it by enlarging the set.
LECTURE 5
MORE PHENOMENA OF TEACHING:
THE METACOGNITIVE SHIFT, THE METAMATHEMATICAL SHIFT, THE IMPLICIT SUGGESTION OF
ANALOGY, THE PARADOX OF LEARNING BY ADAPTATION, THE PARADOX OF THE ACTOR
This week we continue discussing the phenomena of mathematics teaching as identified by

Brousseau.
1. THE METACOGNITIVE SHIFT
The phenomenon of the ‘metacognitive shift’ occurs when a teaching aid becomes an object of
teaching itself (p. 26-7).
Teachers use all kinds of teaching aids to convey the meaning of the abstract
mathematical concepts. They can be material objects such as counters, sticks, blocks, or
graphical representations, or orally communicated metaphors. However, the interpretation of
actions on and with these objects as representations of the particular mathematical concepts that
the teacher has in mind requires that the students focus their attention on certain features of the
objects and not on others, and manipulate these objects in some appropriate ways for this
particular goal. Otherwise, they may miss the concept completely. If, in order to avoid this, the
teacher starts teaching the students the rules of interpreting and using things that were supposed
only to help the students grasp the meaning of a mathematical concept, we have to do with the
‘metacognitive shift’ in teaching.
Example 1: The Venn diagrams
When, during the New Math reform period in the 1960s, elementary school children were taught
operations on sets, they could not be expected to reason about them using the language of formal
logic. Thus a less formal language was needed, and it was proposed to use the so-called Venn
diagrams. A Venn diagram is composed of a number of circles, each representing a set, and the
mutual position of the circles is intended to represent the relation between the sets. Thus, for
example, the set A» (B« C) could be represented by the diagram:
11/4/03 1
The set (A» B) « (A» C) can be represented by the diagram
Comparing the two diagrams the students would be expected to ‘discover’ the distributivity law:
A» (B« C) = (A» B) « (A» C) .
The representation seems to be quite simple and straightforward, as long as it serves not
as a formal language but a heuristic tool; something one quickly draws such a diagram on scrap
paper to convince oneself about the truth or falsity of a relation between sets, and then writes a
proof in ordinary mathematical language (based on the definitions of set operations and the
tautologies of the predicate calculus). But, in the teaching of set theory at an elementary level the
‘ordinary mathematical language of set theory’ was not available. So there was no alternative
language to communicate and test the validity of statements: there was only the ‘language of
11/4/03 2
strings’1 or ‘Venn diagrams’. To smooth out communication between the teacher and the
students, and avoid misunderstandings, the ad hoc drawings for the personal use of
mathematicians had to be developed into a language. And this is indeed what we can observe in
the abundant literature for teachers produced at the time of the reforms. For example - in the
‘Unified Modern Mathematics’ series (1972) produced for the use of Teachers College, Columbia
University by an impressive international board of mathematics educators called ‘Secondary
School Mathematics Curriculum Improvement Study’ (including, beside the American teacher
educators, such ‘big names’ as Gustave Choquet from France, Lennart Råde from Sweden, and
Hans-Georg Steiner from Germany). The Volume 2.1 of this series contains 6 pages of all kinds
of rules concerning the interpretation of Venn diagrams. In particular, it is proposed to write the
symbol ø in a region of a Venn diagram to mark that it represents an empty set and the symbol
‘x’ to mark that the region is not empty (ibid., p. 14)2. In this approach, the students were given
exercises just for the practice of the conventions of the representation (ibid., p. 19). At the time
of the New Math reforms, Venn diagrams were used everywhere, but in some places this
representation received more attention than in others. In Europe the most famous advocate of the
New Math reforms who contributed a lot to the spreading of the use of the ‘language’ of Venn
diagrams and arrow diagrams (for relations and functions) was certainly the Belgian
mathematician Papy. He added color coding to his Venn diagrams and represented empty sets
not by the symbol ø but by hatching the regions of the diagram3.
When presenting the phenomenon of the ‘metacognitive shift’, Brousseau gives the
example of the abuse of the Venn diagrams used in teaching set algebra and arrow diagrams in
teaching about relations and functions, but can we think of other examples, not necessarily
coming from the history of the New Math reforms but from the present day mathematics
teaching? I have thought of two examples; can you come up with more?
1
In some American textbooks for teachers, the expressions ‘the language of strings’ rather thah ‘Venn
diagrams’ was used, e.g. Comprehensive School Mathematics Program 1978, CSMP Mathematics for
Intermediate Grades. Part III: The Languages of Strings and Arrows. CEMREL, Inc. (USA).
Experimental Version 5 -25401.
2
See Appendix 1.
3
Papy (with the collaboration of Frédérique Papy), 1968: Modern Mathematics, Volume 1. London:
Collier-Macmillan Ltd. (Papy signed his books with his last name only).
11/4/03 3
Example 2: The metaphor of scales in teaching equations
The metaphor of ‘scales’ and ‘balancing of scales’ has been used extensively for the purpose of
giving meaning to operations on equations such as, e.g. adding or subtracting the same number to
or from both sides of the equation4. For the metaphor to work, one must think not of the modern
scales, with just one plate, which automatically displays the weight (and sometimes the price)
when something is put on the plate, but of the old-fashioned two-plate scales with the product put
on one plate and weights on the other. Such scales are, presently, museum objects, and most high
school children have never seen them. The drawings of scales in textbooks thus become
representations of objects just as abstract as the concept of equation. The understanding of the
metaphor requires therefore the teaching of the rules of interpretation of the sequences of
drawings of the scales with different objects on both sides and the focusing of the students’
attention on whether the plates are drawn on the same or different level - they have no notion of
‘balancing the scales’. A teacher might, of course, just ignore the metaphor and teach the
operations on equations directly, but some teachers do work with the metaphor for a long time
with the students, and produce a whole ‘language of scales’ to reduce the ambiguity of
communication, just as their predecessors were producing the ‘language of Venn diagrams’. The
risk of ambiguity and unintended interpretations of the scales metaphor is real; there are accounts
of this happening in published research papers56..
Example 3: The use of technology in mathematics teaching: can it favor the ‘metacognitive shift’
phenomenon?
An important part of the controversy about using or not using calculators and mathematical
computer software is concerned with exactly the risk of teaching the teaching aid rather than the
mathematics it is supposed to help understanding. In my own practice of teaching linear algebra
with Maple, I remember that, at the beginning, I used to spend a lot of time teaching my students
the commands of Maple and the quite awkward syntax of the software. Students were spending a
lot of time trying to figure out why a command didn’t work they way they expected (just to find
out, for example, that they had forgotten to put the semi-colon at the end of the command line).
4
See Appendix 3: two pages from a Polish tetxbook for Grade 7 students:
Zawadowski, W. et al., 1996: Matematyka 2001. Podrecznik dla klasy 7. Warszawa: Wydawnictwa
Szkolne i Pedagogiczne. English translation of the text is provided.
5
Pirie, S.E.B. (1998): Crossing the Gulf between Thought and Symbol: Language as (Slippery) Stepping-
Stones. In H. Steinbring, M.G. Bartolini-Bussi, A. Sierpinska (eds.), Language and Communication in the
Mathematics Classroom. Reston, VA: National Council of Teachers of Mathematics, pp. 7-29.
6
MacGregor, M. (1998): How Students Interpret Equations: Intuition versus Taught Procedures. In H.
Steinbring, M.G. Bartolini-Bussi, A. Sierpinska (eds.), Language and Communication in the Mathematics
Classroom. Reston, VA: National Council of Teachers of Mathematics, pp. 262-270.
11/4/03 4
At the end, they were not sure what they are learning in the course: linear algebra or the Maple
language. Today, I no longer demand that the students do their homework assignments using
Maple; I allow them to do so, if they want. Each student in the class has the right to one-hour
tutorial on the use of Maple, and I provide help during my office hours and via e-mail. I am using
Maple for teaching: my lectures are written in a Maple worksheet. The worksheet is active
during the class; it is projected on a large screen and student generated examples and conjectures
are tested using the software, thus avoiding tedious calculations by hand. But, in my lectures, I
do not discuss the syntax of the language nor do I make any suggestions of the type, ‘Look, here
is a useful command’. However, in printing the class notes for the students I do not delete the
commands; I was explicitly asked by some students to leave them on, so that they can use them
when working with Maple on their own.
It is not clear, however, if technology can be classified as a ‘teaching aid’, aimed at
overcoming the difficulties in the weak student and enhance understanding in the stronger
student. In the mathematician’s hands technology is an instrument, useful or sometimes even
indispensable in the execution of certain tasks. And then, the teaching of how to use technology
in doing mathematics is not a symptom of meta-cognitive shift. It is the teaching of the craft of
using a tool in an intelligent way. Some research which has already been done on the teaching
and learning of mathematics with technology7, shows that for many students the calculator is the
ultimate reference and not a ‘mathematical instrument’ in solving problems. This research also
points to the need of preparing the students to correctly interpret and use computer’s outputs.
For example, if I ask Maple to solve the equation x2 - (1+√2)x + √2 = 0, I obtain 1, and
√2. But when I ask Maple to verify if √2 satisfies the equation, the response is ‘false’ (see the
bordered figure below). One has to understand that for Maple, √2 does not exist; ‘√2’ is only a
symbol for a decimal number, an approximation of √2. But no approximation of √2 satisfies the
above equation.
7
see, for example, Guin, D. & Trouche, L. (1999): The Complex Process of Converting Tools into
Mathematical Instruments: the Case of Calculators. International Journal of Computers in Mathematical
Learning 3, 195-227.
11/4/03 5
> f:=x->x^2-(1+sqrt(2))*x+sqrt(2);
2
f := x -> x - (1 + sqrt(2)) x + sqrt(2)
> solve(f(x)=0,x);
1, √2
> evalb(f(1)=0);
true
> evalb(f(sqrt(2))=0);
false
If I ask the calculator TI-92 to solve the same equation in the exact mode, the response I
obtain is ‘x = -(√(3-2√2) - √2 - 1)/2 or x = (√(3-2√2) + √2 + 1)/2’. Of course,
3 - 2√2 = (1 - √2)2 and the same numbers 1 and √2 are obtained, but the calculator will not
spontaneously simplify an expression. Besides, one has to know that the calculator is simplifying
an expression upon the command ‘expand’ and it’s no use looking for a ‘simplify’ command,
because it does not exist.
2. THE META-MATHEMATICAL SHIFT
This phenomenon, only briefly mentioned by Brousseau (p. 39), consists in the ‘replacing [by the
teacher] of a mathematical problem by a discussion of the logic of its solution and attributing all
sources of error [in its solution] to [a misunderstanding of this logic]’. An example of this could
be the situation where the teacher, in the aim of improving the students’ performance in solving
equations and understanding what they are doing, teaches the students a theory of equations:
gives a definition of an equation, and before that, a definition of a variable, an algebraic
expression, the logical axioms of equality (reflexivity, symmetry and transitivity) and the
algebraic properties of equality in number systems (e.g. if a = b and c = d then a + c = b + d),
etc. Such knowledge belongs to the so-called ‘meta-mathematics’, i.e. a theory of the language of
mathematics, which makes abstraction from the intuitive meanings of particular mathematical
statements or objects and occupies itself only with their general form.
The New Math reforms were fraught with proposals amounting to no more and no less
than teaching meta-mathematics to school-children; after all, logic and theory of relations can be
regarded as parts of meta-mathematics. Today, the shift is less felt, but some elements of it still
11/4/03 6
exist, I believe, in the introduction of equations in secondary schools via the notions of ‘open’ and
‘closed sentences’.
3. THE IMPLICIT SUGGESTION OF ANALOGY
Most of the phenomena of teaching mathematics we have reviewed so far represent the different
ways in which teachers try to
(a) maintain the fiction that learning does, indeed, take place, and that, therefore, they are doing
what is expected of them as teachers (Topaze, Jourdain), or,
(b) genuinely help the students learn better, but the method chosen does not and cannot bring
about the expected results (Dienes, the meta-cognitive and the meta-mathematical shifts).
The phenomenon discussed in this section (p. 27) belongs to the category (a) of these
phenomena. The teacher gives the students problems formulated so as to highlight their analogy
with problems previously solved by the teacher on the board or discussed in class. The teacher
does not want to explicitly say, ‘solve this problem just like we solved problem number so and
so’ but she gives a hint, sometimes just by asking the question in exactly the same form. The
students are expected, in this game, to get the hint. The teacher feels miserable when they don’t
because this forces him or her to explicitly point to the analogy and the falsehood of her game is
revealed.
4. THE PARADOX OF LEARNING BY ADAPTATION
The remarks that Brousseau is making in the section ‘Paradoxes of learning by adaptation’ (pp.
44-47) could be understood as a criticism of the constructivist epistemology and psychology of
learning and a promotion of the interactionist stance in these matters.
Constructivism as a psychology and epistemology made its way into mathematics
education in the late 70s and, in the USA, it generated a lot of basic research into children’s
processes of acquiring a notion of natural number, fractions, arithmetic operations, and also more
mature students’ processes of learning such notions as the exponential function8. It developed
into a certain ‘ideal’ of teaching mathematics, where there would be no lecturing, no drill
exercises, but individual children ‘constructing their own knowledge’ (a well known
constructivist slogan) by solving problems, with the teacher’s role reduced to that of an
interviewer (‘Tell me how you solved this problem’). By adapting to a stimulating environment,
with interesting problems, children’s ‘cognitive structures’ would grow and evolve in a natural
8
See, for example, the volume:
Steffe, L.P. & Gale, J. (Eds.), 1995: Constructivism in Education. Hillsdale, New Jersey: Lawrence
Erlbaum Associates, Publishers.
11/4/03 7
way. The metaphor here is that of a biological organism changing through adaptation to its
environment.
From the constructivist perspective, as Piaget stressed, knowing is an adaptive activity. This means that
one should think of knowledge as a kind of compendium of concepts and actions that one has found to be
successful, given the purposes one had in mind. This notion is analogous to the notion of adaptation in
evolutionary biology, expanded to include, beyond the goal of survival, the goal of a coherent conceptual
organization of the world as we experience it. (Von Glasersfeld, 1995, p. 7)9.
Constructivism has developed in opposition to Behaviorism, a theory of learning quite prominent
in the USA, based on the fundamental assumption that a rewarded response is the action that will
be repeated, ‘reinforced’. Behaviorism led to a theory of instruction based on drill and practice
and a system of rewards and punishments.
Constructivism attracted mathematics educators interested not only in early childhood
education, but also those working with university students. For example, the so-called APOS10
theory of learning mathematics advanced by Ed Dubinsky and his collaborators, is firmly based
in the constructivist epistemology. Here is how Dubinsky defines mathematical knowledge:
Mathematical knowledge is an individual’s tendency to respond, in a social context, to a perceived problem

situation by constructing, re-constructing and organizing, in her or his mind, mathematical actions,
processes, objects and schemas with which to deal with the situation (Dubinsky, 1997, p. 95)11.
Thus, for a constructivist, knowledge is a psychological entity: an individual’s network of

cognitive structures, schemas, constructed through the individual’s experience in solving all kinds
of problems, practical and theoretical. From this perspective, objectivist notions such as ‘truth’
and ‘validity’ of knowledge which refer to a ‘correct representation of reality’, do not make
sense; they are replaced, in constructivism, by the notion of ‘viability’. The very concepts of
‘correct’ and ‘reality’ are questioned by constructivism.
To the biologist, a living organism is viable as long as it manages to survive in its environment. To the
constructivist, concepts, models, theories, and so on are viable if they prove adequate in the contexts in
which they were created. Viability - quite unlike truth - is relative to a context of goals and purposes. But
these goals and purposes are not limited to the concrete or material. In science, for instance, there is,
beyond the goal of solving specific problems, the goal of constructing as coherent a model as possible of
the experiential world (Von Glasersfeld, ibid.).
9
von Glasersfeld, E. (1995): A Constructivist Approach to Teaching. In L.P. Steffe & J. Gale (eds.),
Constructivism in Education. Hillsdale, New Jersey: Lawrence Erlbaum Associates, Publishers, pp. 3-15.
10
‘APOS’ stands for Action - Process - Object - Schema.
11
Dubinsky, E. (1997): Some Thoughts on a First Course in Linear Algebra at the College Level. In D.
Carlson, C.R. Johnson, D.C. Lay, A. Duane Porter, A. Watkins, W. Watkins (eds.), Resources for Teaching
Linear Algebra. The Mathematical Association of America, MAA Notes, Volume 42, pp. 85-105.
11/4/03 8
Brousseau holds a very different view of learning and knowledge. He claims that the
notion of learning by adaptation is inconsistent (p. 44-45). ‘Adaptation - he claims - contradicts
the idea of new knowledge’. If a person solves a problem different from all the problems she had
solved so far, with some adaptation of the knowledge she already had, why would she think she
has invented some new knowledge? In a similar situation she could go, if necessary, through the
same process, from scratch, and therefore there is no need to identify the process as a new
‘method’ or ‘new knowledge’. But if this person shows her way of solving the problem to some
other persons who were also trying to solve it but were no able to, then their interest in it,
eagerness to understand it, and their appraisal of its more general and not just local value, will
indicate to her that some new knowledge has actually been invented. This is what appears to be
meant by the statement: ‘knowledge is almost the cultural recognition that direct knowing is
impotent to solve some situations naturally (by adaptation)’ (p. 45).
This view of knowledge is strongly reminiscent of the position taken by interactionism in
social psychology. Interactionism in social psychology has its roots in the pragmatic positions of
Peirce, James, Dewey (USA, end of the 19th and beginning of 20th century) and the sociological
research of the so-called Chicago school of sociology12. Its epistemology stresses the roles of
experience, the common sense, and the existence of multiple interpretations. All one knows is
experience; but this experience is not a sequence of isolated sensations but a culturally shared
world which we take for granted in the everyday life. Experience is thus the common sense,
where the ‘common’ means the ‘shared’. And ‘shared’ means ‘objective’. We do not want to
have ‘private knowledge’ or ‘subjective knowledge’ - we do not value it. When we notice that
our knowledge is different from the shared knowledge, because, for example, we are unable to
achieve goals that others achieve, we treat this knowledge as something subjective and we
reconstruct it until it allows us to achieve the goals. Then we treat it as objective again
(Hammersley, 1989, p. 56). The basic unit of the interactionist psychology is a goal directed
action; it is assumed that our actions are social, even if we perform them alone, because we are
able to view ourselves as objects. In undertaking an action we imagine the effects that it will
produce on others and how they could react (ibid., p. 59):
It is this socially generated ability to view oneself as an object, and to interpret the world in alternative
ways, that allows people to modify their interpretations and to choose different courses of action. As a
result, human action… is constructed through a reflexive process which takes the form of a person making
12
See an interesting account of interactionism in
Hammersley, M. 1989: The Dilemma of Quantitative Method. Herbert Blumer and the Chicago Tradition.
London: Routledge.
11/4/03 9
indications to himself, that is to say, noting things and determining their significance for his line of action.
(Blumer, cited in Hammersley, ibid., p. 130).
Applying this theory to learning, one might perhaps say that, from the interactionist perspective,
new knowledge coincides with choosing different courses of action, and different interpretations
of a class of situations, and this can happen only when the learner sees himself or herself as an
object - a member of a society and judges his or her actions from that external point of view. In a
sense, one can learn new knowledge if one is, at once, a learner, and one’s own teacher. In a
sense, also, one knows only if one ‘acts’ one who knows - ‘if one acts as a knowledgeable
person’, according to what the society takes for granted as the behavior of a knowledgeable
person. Only ‘shared’ or ‘public’, or culturally identified as such knowledge is considered to be
knowledge.
5. THE PARADOX OF THE TEACHER AS LEARNER
For constructivists, the ideal teaching situation is one where the teacher does not know the
solution of the problem given in a didactic situation. The problem could have been invented by
the student, as a result of being in some more general problem situation, and the teacher and the
student work together on it as partners-in-mathematics.
Brousseau compares this situation to one in a theater, where the actor would not just act a
feeling (e.g. joy or anger) but actually experience this feeling. Referring to Diderot’s analysis of
acting, Brousseau claims that this would result in quite poor performance: such acting might not
be very convincing: ‘the more the actor feels emotions he wants to play, the less he is able to
allow the audience to share this feeling’ (p. 46). Being on stage, visible for the audience, the
actor could become ashamed of his private feelings and try to conceal their perceptible symptoms
rather than amplify them, which is what he has to do if he wants the audience to understand what
is going on.
The point Brousseau is trying to make, I presume, is that if a teacher finds it useful to
act as if she did not know how to solve the problem, this should only be good acting and not the
actual state of the teacher’s mind. Not only should the teacher know the knowledge that she
intends to teach but she should use all the means in her repertoire of ‘didactic tricks’ to put on
stage and devolve (p. 31) to the students a problem situation they would consider it their own
responsibility to solve and which would lead them to develop or use the knowledge in question.
11/4/03 10
LECTURE 6
DIDACTIC ENGINEERING 1. THE NOTION OF FUNDAMENTAL SITUATION.
1. DEFINITION OF THE METHODOLOGY OF DIDACTIC ENGINEERING
Suppose you want to conceive and try a new way of teaching a piece of mathematical knowledge.
Then you have, roughly speaking, two ways of going about that.
Choice 1: The Comparative Study of Experimental and Control Groups
You write down a scenario of classroom activities, with a precise description of the role and
actions of the teacher and the expected reponses of the students. The scenario contains advice for
the teacher in case the students commit errors and mistakes of various types. The decisions made
in writing the scenario pertaining to the choice of the mathematical activities and problems could
be justified by curriculum prescriptions, some theory of learning, some principles of teaching,
knowledge of mathematics and personal teaching experience. But the evaluation of the teaching
project will not be evaluated of the basis of this justification. In fact, this justification may not be
even written down or otherwise made explicit in the final report of the project. The project will
be evaluated by testing the scenario on a group of students. A control group will also be chosen.
The control group will be taught the same mathematical content with traditional methods, and
both groups will be administered identical pre-tests and post-tests. In the case of similar results
on the pre-test, if the experimental group performs better on the post-test, then the teaching
project will be evaluated as ‘effective’.
Choice 2: The methodology of instructional development
You reject the ‘comparative study’ methodology because you do not believe that it is possible to
teach ‘the same mathematical content’ with two different sets of mathematical activities and
diferent pedagogical approaches. You also don’t believe that one can assess what the students
have learned by counting their scores on a standardized test. You start by writing a detailed
scenario, just like someone who picked Choice 1, but you do make explicit the rationale behind
all your decisions, based on your theory of learning, your instructional theory and your theory of
what it means to know the particular mathematical content that you plan to teach the students,
because this is going to be the ground with respect to which your project will be evaluated. You
make predictions concerning the knowledge that the students should construct as a result of
participating in the planned activities. Then you try out your scenario in a class with someone
else, not yourself, as a teacher. You are sitting in the classroom as an observer. You collect all
possible documentation concerning the students’ mathematical work. You audio- and videotape
the classes, and you collect all the students’ written work. Then you analyze this material with
the question: Has the anticipated knowledge developed in the students? If not, then what
knowledge has developed? What are the reasons behind the discrepancies between the
anticipation and the actual outcome? Can these be explained in terms of the theoretical
frameworks assumed a priori? Is there a need to search for an alternative theory, or for an
amendment of the theory that has been used? How can the scenario be improved to decrease the
difference between the anticipated and the actual knowledge produced by the scenario? On the
basis of this analysis, you re-design your scenario and try it again in the same way. Etc.
If you made Choice 2 and your theoretical framework is based on the Theory of Didactic
Situations, then you can say that you are using the methodology of Didactic Engineering in
developing a teaching project in mathematics.
The term ‘engineering’ in the name of the methodology comes from an analogy of this
kind of conception, design, and implementation work in mathematics education with the work of
a civil engineer. An engineer does not validate his or her design of, say, a bridge, on the basis of
a comparison with already constructed bridges (which have not fallen down), but on the basis of
(a) predictions of its properties (stability, capacity, etc.) which can only be deduced from the
mathematical and physical theory, and (b) the realization of the project, checking if the
predictions were correct1.
2. A TOOL FOR DIDACTIC ENGINEERING: THE NOTION OF FUNDAMENTAL SITUATION
If you plan to teach students some piece of mathematical knowledge so that they learn it, then the
theory of situations suggests that you have to organize the didactic milieu and the game of the
students with this milieu in such as way that this particular mathematical knowledge will appear
as the best means available for the understanding of the rules of the game and elaborating a
winning strategy (see Week 1 Class notes). You know from the theory, that there are many ways
1
An interesting analysis of the methodology of didactic engineering can be found in the article
Artigue, M. & Perrin-Glorian, M.-J., 1991: Didactic Engineering, Research and Development Tool: some
Theoretical Problems linked to this Duality. For the Learning of Mathematics 11.1, 13-18.
in which you may fail to reach this goal (Jourdain, Topaze, Dienes effects, and other phenomena),
so you have to organize the milieu so as to avoid falling into the trap of wanting to preserve the
fiction of ‘doing your job’ at all costs.
This assumption is based on a view of mathematics, advanced by the theory of situations,
according to which ‘mathematical knowledge cannot be apprehended otherwise than through the
activities that they allow us to realize and therefore the problems that it makes it possible to
solve. Mathematics is not simply a logically consistent conceptual system for the production of
rigorous proofs; it is, first of all, an activity which is realized in a situation and against a milieu.
The activity is a structured one, with distinguishable phases of action, formulation and
validation, devolution and institutionalization’2.
Theory of situations and this particular conception of mathematical knowledge can be
used both to identify what mathematical knowledge is being constructed by the teacher and his or
her students in an actual lesson and to ‘engineer’ situations aimed at the construction of a certain
knowledge by the teacher and the students.
In this class we are interested in the latter activity, and our questions is: How do we go
about finding out what kind of situation would generate a given mathematical knowledge?
We have to start by analyzing the knowledge K that we aim to teach. We will want to
define this knowledge by the general characteristics of a problem-situation to which it could be
considered as an optimal solution strategy. A problem-situation is more than just a problem (and
it is not a school exercise); it is characterized by what is at stake in solving or not solving it, the
possible states of the system it which it has appeared, rules of action, aims of solving the problem.
By listing those factors (variables) of the problem-situation which are pertinent from the point of
view of the knowledge K associated with it, one obtains a model called The Fundamental
Situation associated with K.
Symbolically, FS(K)=[C1, … , Cn].
Normally, a piece of knowledge does not appear in its full generality, but in a variety of
meanings or concepts depending on the domain of its use or application. We can assume that
each such ‘conception’ of K is associated with certain ranges of values of the variables Ci.
Thus, the FS(K) can be thought of as a generator of specific situations (SS) associated with the
different conceptions of K. We have to decide which conception of K we want the students to
construct.
2
Translated from the article
Bosch, M. & Chevallard, Y., La sensibilité de l’activité mathématique aux ostensifs. Objet d’étude et
problématique. Recherches en Didactique des Mathématiques 19.1, 77-123, p. 81.
Example of an analysis of a piece of knowledge in terms of variables of situations in which it

appears as an optimal solution - Part 1: FS and SS
Knowledge K = Number
FS(K) := [size of set (small, big; finite, infinite), order type of set (continuous/discrete), context
of use (comparison of size of sets (which can be close to each other or far away), counting, measuring,
coding elements in a set, marking rythms (chanting)), representation (numerals (in various systems:
decimal, binary, roman, etc.; verbal, written), sets of physical objects, abaci, ...)]
Conception K1 of K = everyday names of natural numbers in their counting function

SS-K1 := [size of set (not very small, say 20-30; finite), order type of set (discrete), context of
use (comparison of size of sets (which are far away), counting), representation (oral numerals)
The next step is to conceive of a didactic situation in which the teacher would devolve to
the students a problem-situation whose constraints would satisfy the values of the variables of the
chosen Specific Situation. This didactic situation would be defined by some values of its
pedagogical variables that would ensure the devolution of the problem situation and would not
contradict the problem-situation.
Example - Part 2: variables of a didactic situation
DS(K1) := [motivation (K1 must be a means to solve the students’ problem, not a school problem);
prerequisite knowledge (the students know the names of numbers in proper order: they can chant, ‘one,
two, three,…’ till at least 20); kind of problem (the comparison of two biggish sets of objects placed far
apart is involved in the problem to be devolved to the students)]
Only after this preliminary analysis of the knowledge to be taught can one start thinking
of a possible realization of the didactic situation in the classroom. There are, a priori, many ways
in which such a situation could be materialized, theoretically conserving the property of leading
to the construction of the same knowledge by the teacher and the students3.
Example - Part 3: a classroom realization of a SS for the counting function of number
The teacher in a kindergarten class of 5 years old children has arranged the following:
In opposite corners of a large room there are two tables. On one table there are 23 pots of paint,
on the other, there are about 30 brushes. The teacher says: ‘Anybody who brings, from that table
over there, as many brushes as there are pots of paint on this table - I mean one brush for each pot
of paint - wins a prize’. Children are motivated: they want to win a prize. They are five year olds
3
Think of the various classroom situations proposed by the class in Week 5 with respect to the ratio
conception of division.
who know their number sequence till at least 30, so the prerequisites are there. There are too
many pots of paint for the children to grasp their number visually - so they will have to use some
coding. They will not be allowed to use marks on paper or beads or anything other than their
voice to code the number. The most economical solution will be to count the pots of paint,
remember the last numeral and then count the brushes till that numeral and take those counted.
Exercise 1
(a) Define the FS for the knowledge: operation of division (FS(Div))

(b) Define the SS for the physical ratio conception of division (SS-Div/r)
(c) Define a didactic situation for SS-Div/r
Hints for a solution
(a)
The Fundamental Situation for the operation of division could be defined by the following
variables:
• level of generality (an operation in abstract algebraic structures understood as the
multiplication by the inverse, an operation in a concrete number structure or measure space)
• kind of entities involved (numbers only, a number and another entity (e.g. a physical
magnitude, a transformation, a matrix, etc.), non-numbers only)
• kind of numbers involved, if at all (natural numbers, integers, rational numbers, real numbers,
complex numbers, etc.)
• kind of representation of numbers used, if numbers used (exact, decimal approximations,
decimal system, another position system)
• size of numbers, if numbers used (big, small, less/greater than 1)
• relative size of the dividend and the divisor, if numbers used (in a/b, a < b or a > b)
• context of application (sharing, partitioning, geometric ratio, physical ratio,
multiplying/combining by/with the inverse of an element in an algebraic structure)
(b) A Specific Situation related to the Physical Ratio conception of division could be defined by
the following values of the above variables:
SS-Div/r is concerned with an operation in a concrete measure space, involving two physical
magnitudes, expressed by rational but not integer numbers, represented in either a fractional or
decimal form, and involved in a problem of comparison of ratios (otherwise the ratio would not
have to be understood as a number).
Exercise 2
(a) Read Brousseau’s description and analysis of a classroom situation: ‘The thickness of a sheet
of paper’ (pp. 195-212)
(b) Determine the mathematical knowledge K and its specific conception K1 that this situation
is aiming at and describe K and K1 in terms of their defining variables
(c) Describe the didactic variables of the classroom situation corresponding to K1.
(d) In what way the classroom situation realizes SS-K1?
(e) By what kind of behavior does the teacher avoid (or not) the trap of the phenomena of Topaze,
Jourdain, the metacognitive shift, the metamathematical shift, the implicit suggestion of analogy?
For a better ‘feel’ of the above situation, I propose the following small activity in the class:
I’ll put five books on the table: 1. ‘Thinking mathematically’, 2. ‘Elementary Linear Algebra’, 3.
Sheakespeare, 4. Matematyka-7, 5. Robert Dictionary. I’ll ask you to evaluate the thickness of
the paper used in each of the books and order the books from the one with the least thick paper to
the one with the thickest paper. All you will have is a ruler. How will you go about it?
When I tried to evaluate the thicknesses myself, I was measuring the thickness of all
inside pages of the book and counted the number of pages. Here are the results I got:
width # sheets ratio
1 12 mm 120 0.1
2 20 mm 271 0.074
3 13 mm 120 0.108
4 21 mm 231 0.909
5 57 mm 525 0.108
When I then calculated the ratios width/# sheets, it turned out that the book (5) has
thicker paper than (1) which was obviously not true; I could feel it. Then I decided to measure the
width of the same number of pages in each book; I chose 240 pages or 120 sheets. I then
obtained the following table:
Book width # of sheets
1 12 mm 120
2 8 mm 120
3 13 mm 120
4 11 mm 120
5 7 mm 120
Now I did not have to calculate the ratios, I knew that the order of the books from the
book with the most thin pages to that with the thickest pages is: 5, 2, 4, 1, 3.
It has become clear for me in this exercise that had I chosen to take the same number of
pages from each book to start with, I would not have to calculate the ratios. So if the aim of the
situation has something to do with ratios, division or rational numbers, then assigning an exercise
like this one for each individual child to do on his or her own, would not help achieving it. Now I
understood why the situation as described in Brousseau’s book was a lot more complicated, with
children not just measuring and calculating individually but in groups and having to communicate
and compare their results. It is rather unlikely that all groups would have chosen the same
number of sheets to measure.
LECTURE 7
DIDACTIC ENGINEERING 2:
CAUSES OF DIFFICULTIES OF REALIZATION OF DIDACTIC ENGINEERING PROJECTS.
It is often the case that a teaching project which, a priori, appears to create the appropriate
conditions for the construction, by the students, of a given mathematical knowledge, is
disappointing when it is experimented. Students either do not learn anything new or learn
something different from the target knowledge.
What could be the objective1 causes of this failure? We shall discuss three types of
causes, labeled ‘didactic transposition’, ‘teacher’s epistemology’, and ‘obstacles’ of various
origins. It may also happen that a teaching experiment ‘works’ once, but the success is not
repeated when it is run for a second time. Why? This is the problem of reproducibility of the
products of didactic engineering.
1. DIDACTIC TRANSPOSITION
The didactic engineering methodology presupposes that, at least in the phases of action,
formulation and validation, the students will act not as students whose sole aim is to satisfy the
teacher and pass the course, but as learners whose aim is to gain some new mathematical
knowledge by solving a problem. It is implicitly expected that, at some point, the students will
behave as mathematicians.
This expectation can be justified only if the target knowledge is, indeed, of the same
nature as mathematicians’ knowledge. But this need not be the case.
In general, school mathematics knowledge differs quite substantially from
mathematicians’ knowledge, not only quantitatively: school mathematics knowledge is not a
small subset of the mathematician’s knowledge. The school institution has developed
mathematics as a teaching subject which may have parts not included in the mathematicians’
knowledge.
One could say that, in general, mathematical knowledge (or any kind of knowledge for
that matter) is a function of the institution in which it ‘lives’. The mathematics developed and
1
‘Objective’ here means: independent from the personal characteristics of the teacher and the students
(e.g. poor class management of the former and the incurable laziness of the latter).
used in an Arts and Science Department is different from the mathematics developed and used in
the Engineering Department, which is yet different from the mathematics as it serves the medical
and biology research centers. There is Actuarial Mathematics and Financial Mathematics. There
is not one ‘La Mathématique’, as it was held in the 1960s, during the New Math reforms. Of
course, all these various institutions are not hermetically closed and there is a continual flow of
ideas between them. When an institution picks up some body of knowledge from another one
and adapts it, changes it to fit her own goals and tasks, then we speak of an ‘institutional
transposition of knowledge’2. When the transposition goes from an institution which produces
knowledge to an institution which teaches it to students, then we speak of ‘didactic
transposition’3.
The process of didactic transposition is inevitable when it comes to teaching mathematics
at school; however teachers and curriculum developers must beware of producing knowledge that
would have only some kind of ‘internal’ value for the functioning of the school as an institution,
but not outside of it. We must constantly remind ourselves that the aim of the school is to prepare
the children for life and professions outside of the school. The risk is real; we have seen
monstruous examples of such unnecessary ‘didactic creativity’ in mathematics teaching, such as
the ‘language of strings and arrows’, in the 1960s New Math reforms. There are less monstruous
examples such as the ‘Big cosine’, Cos and the ‘Small cosine’, cos.
Exercise 1
Can you give other examples of school mathematical concepts which do not exist in the academic
mathematics?
Aside from creating new ‘objects of knowledge’, school mathematics differs from
‘research mathematics’ also in other ways.
A part of some general knowledge can be isolated and given an important status of a
‘topic’ with a special name and place in the curriculum. Example: ‘les identités remarquables’ in
the French curriculum. This refers to algebraic properties of operations on real numbers which
are, in the theory, simple consequences of the properties of the real number system
2
Chevallard, Y. 1992: Fundamental concepts in didactics: perspectives provided by an anthropological
approach. In R. Douady & A. Mercier (eds.), Research in Didactique of Mathematics. Selected Papers.
Grenoble: La Pensée Sauvage éditions, pp. 131-167, p. 165.
3
Actually, in participating in the present course, you are witness of a didactic transposition of a body of
knowledge that has first evolved as a university research domain, the ‘Theory of Didactic Situations’. I
shall not dwell on the differences between this academic body of knowledge and the knowledge that we
are constructing together in the course, but I am fully aware of the existence of such differences.
(commutativity and associativity of addition and multiplication, distributivity of addition with

respect to multiplication, etc.), e.g. (a-b)(a+b)=a2 - b2.4 The curriculum devotes several class
periods for this topic and students are given many exercises for their application; a special test is
usually designed for the assessment of the students’ mastery of these identities.
A technique that mathematicians are using as an instrument and are not studying per se
becomes an object of teaching: this is the case of ‘paramathematical notions’ such as equation,
parameter or proof5. A mathematician would not say: ‘today, I’ve been doing proofs’, or ‘I’ve
been solving equations with parameters’ because he or she is doing proofs every day or every
time he or she is doing mathematics, and in all his problems there are constants, variables and
parameters. But a teacher can be ‘doing proofs’, or teaching the ‘notion of equation’, or ‘the
notion of parameter’ to his or her students today and consider the topic finished and done with
tomorrow6.
Perhaps the most important difference between research mathematics and school
mathematics is the aim of the mathematical activity. The mathematician wants to know and in
the aim of knowing he uses his or her mathematicial competences; the student has to demonstrate
that he or she knows what he or she is expected to know and possesses the competences aimed at
by the curriculum. For example, a mathematician will transform an expression like 4x2 - 36x
into 4x(x-9) or (2x-6√x)(2x+6√x) depending on the assumptions and purposes of the problem in
which this expression appeared. But the student is quite likely to receive this expression isolated
from any mathematical problem, in an exercise like ‘Factor 4x2 - 36x’ and he or she will know to
be expected to answer with 4x(x-9) or (2x-6√x)(2x+6√x) depending on the didactic, not
mathematical context of this exercise, i.e. on what the teacher had been teaching before giving
this exercise. So, in choosing what to write the student is not so much solving a mathematical
problem as deciphering the rules of the didactic contract7.
A thorough analysis of the distinction between school and research mathematics in
general is available in Chevallard (1991) (see footnotes). An in-depth study of the didactic
4
Chevallard, Y. 1991: La transposition didactique du savoir savant au savoir enseigné. Grenoble: La
Pensée Sauvage éditions, p. 43.
5
Proviso: the ‘notion of proof’ is an object of study for logicians working in the so-called ‘foundations of
mathematics’. But even for them mathematical proof is an everyday tool; they justify their assertions by
means of proofs.
6
Chevallard, 1991, p. 49-51.
7
ibid., p. 52.
transposition of a particular mathematical concept, the concept of distance, can be found in

Chevallard & Johsua (1991)8.
2. TEACHER’S EPISTEMOLOGY
Knowledge functions in the school institution in a very different way than in a research
institution. It may therefore not be very realistic to expect that, even through the best - a priori -
didactic engineering, students will develop mathematical ways of thinking and concepts similar to
those of research mathematicians.
The way mathematical knowledge functions at school is influenced by the teacher’s
knowledge relative to the acquisition of mathematical knowledge. It is not necessary that the
teacher be aware of that knowledge, that he be able to verbalize it and discuss it against other
possible views on the acquisition of mathematical knowledge. The existence of this knowledge
manifests itself in the teacher’s practices. Brousseau calls this knowledge ‘teacher’s
epistemology’ (p. 35)9.
Here are some examples of such manifestations.
1. A teacher will classify some students’ errors as ‘une étourderie’ (a result of absent-
mindedness), especially if they come as an answer to an ‘easy’ question that most students
already are familiar with. They will classify some other errors as ‘conceptual’ or simply
‘serious’. The former will not entail any didactic action on their part; the latter will - the teacher
will engage in ‘revisions’, ‘remedial activities’, etc., with respect to students whom they suspect
of having ‘conceptual problems’. This is a symptom of some kind of ‘spontaneous psychology’
that the teacher seems to profess, which has little to do with research in cognitive psychology,
where responses to cognitive problems are not classified into ‘serious’ and ‘not serious’.
2. The teacher expects the students to produce their answers according to a schema which
they consider as ‘correct’; they expect the students to produce their answers in an ‘intelligent’
way, acceptable within the mathematical culture they believe the school should represent. Some
teachers thus confuse the laws of the production of knowledge with the systematization and
organization of knowledge. But even those who know that, in the domain of real mathematical
research, solutions to problems are not found this way or that, in general, human cognition does
not work in such a systematic, orderly and logical fashion, will still hold such expectations and
8
Chevallard, Y. & Johsua, M.-A. 1991: Un exemple de transposition didactique. Grenoble: La Pensée
Sauvage.
9
The explanations of the phenomenon of ‘teachers’ epistemology’ that you will find in this section are a
free translation and adaptation of a part of Brousseau’s clarifications which I obtained from him in private
e-mail correspondence.
requirements with respect to the students’ answers: this is a necessary implication of the didactic
contract. The teacher cannot transform his or her class in a psychology lab, and he or she cannot
straighten up all the individual or even collective trajectories of learning because this would be
too time- and labor consuming and perhaps technically impossible.
Thus, in his or her practice the teacher uses certain ‘concepts’ or ‘laws’ related to
epistemological questions such as ‘what does it mean to investigate?’, ‘to learn?’, ‘to
understand?’ which define the students’ field for action and justify the teacher’s decisions. This
spontaneous epistemology is accompanied by a whole mythology of metaphors and symbols thus
forming a system - a praxis of the teacher - which makes his work appear possible and legitimate.
Here are some more examples of teachers’ decisions that reflect this spontaneous
epistemology and show how distinct it can be from an epistemology actually governing the work
of mathematicians.
In mathematics it does not matter very much how a result has been established: whether
by way of a witty reasoning or by way of a laborious calculation and verification of all possible
cases. What matters is to show that the solution solves the problem. But if a student solves an
equation by trying some numbers at random or even systematically, his solution will not be
considered correct by the teacher. It is worthless from the point of view of the teacher’s
epistemology, because it has no ‘positive didactic value’: the student has not demonstrated he or
she had learned the method. The aim, at school, is not to solve the problem but to demonstrate
one has learned what one has been taught10.
Teachers’ spontaneous epistemology of mathematics is often at odds with the foundations
of the theory of situations. Since it is this theory that underlies teaching projects conducted
according to the methodology of didactic engineering, this discrepancy may explain the
difficulties of the classroom realization of these projects.
Exercise 2.
In an article published in English in the journal For the Learning of Mathematics in 1991,
Michèle Artigue and Marie-Jeanne Perrin-Glorian11 report about a didactic engineering
experience in a middle school (10-12 years old children) with students who had difficulties in
mathematics (and other subjects). It turned out to be practically impossible to realize the didactic
engineering projects, and the researchers started to study the reasons for this failure. They have
studied, in particular, the teachers’ conceptions related to the teaching and learning of
mathematics.
10
This is the end of Brousseau’s explanations.
11
Artigue, M. & Perrin-Glorian, M.-J. 1991: Didactic Engineering, Research and Development Tool: Some
Theoretical Problems Linked to this Duality. For the Learning of Mathematics 11.1, 13-18.
Based on your reading of this part of the article, explain in what way could these
conceptions contribute to the failure of the didactic engineering? In what way are these
conceptions incompatible with the theory of situations?
3. REPRODUCIBILITY OF DIDACTIC SITUATIONS
It has proved extremely hard for teachers to reproduce a didactic situation so that the
mathematical meaning of the students actions is conserved. This phenomenon is called, by
Brousseau, the ‘obsolescence of didactic situations’ (p. 193).
By obsolescence we mean the following phenomenon: from one year to another, teachers have more and
more trouble reproducing the conditions likely to lead their students to create, perhaps through different
reactions, the same understanding of the notion taught. Instead of reproducing conditions which, while
producing the same result leave the trajectories free, they reproduce, on the contrary, a “history”, a
development similar to that of previous years, by means of interventions that, even if discrete, completely
change the nature of the didactic conditions guaranteeing a correct meaning for the students’ reactions; the
obtained behavior is apparently the same but the conditions under which it was obtained modify the
meaning (p. 193).
The failure of even the first realization of a didactic engineering project is related to this
difficulty in reproducing the mathematical meanings. In writing up a scenario for a didactic
situation, it is rather easy to describe the physical environment, the tasks, the verbal interventions
and the actions of the teacher, the expected reactions of the students. It is much more difficult to
enumerate and highlight those features of the milieu which are indeed responsible for the
emergence of the target mathematical meaning. Sometimes it can even be hard to tell what these
features are, and it may well be that the dynamics of a didactic situation represent a chaotic rather
than a stable system, meaning that a tiny alteration of its conditions (e.g. an apparently
unimportant remark or a hint from the teacher, an expression of approval or disapproval on his or
her face) may cause huge changes in the knowledge that it produces12.
In order to increase the chances of reproducibility, the didactic engineering methodology
is very demanding with regard to the a priori analysis of the target knowledge and the work of
identifying the critical variables of the didactic situations that are supposed to lead to it.
Example.
A group of researchers13 have studied the problems of the reproducibility of the products of
didactic engineering in the following setting: The same, a priori, didactic situation has been
implemented by two different teachers in two different classes of 13-14 years old students. The
12
ibid., p. 14.
13
Arsac, G., Balacheff, N., Mante, M. 1992: Teacher’s Role and Reproducibility of Didactic Situations.
Educational Studies in Mathematics 23, 5-29.
mathematical content of the situation was the activity of proving. The situation was composed of
two main phases:
(a) the research phase, in which the students are given a problem and write their solutions on a
poster; the problem is so chosen that there are many solutions possible, and it is likely that the
students will disagree on many points;
(b) the debate phase: the students’ solutions are written on large sheets of paper and are then
displayed as posters on the walls of the classroom. Students are divided into teams which analyze
the solutions. Each team then delegates one spokesperson who presents the result of the analysis
to the whole class. A whole class debate is then engaged.
The situation was planned to last over two one and a half hour periods.
In Class I, the teacher was a member of the research team. In Class II, the teacher was
not a researcher but the scenario was thoroughly presented and discussed with her.
In Class I the problem given to the students was the following:
‘Write for other students a message allowing them to come to know the perimeter of any triangle
a piece of which is missing. To do so, your colleagues will have at their disposal only the paper
on which is drawn a triangle and the same instruments as you have (rulers, etc.).’
In Class II the problem given to the students was the following:
M
‘The two lines intersect outside a page. Write a method allowing anybody to draw the line
through M and the point of intersection without going out of the page.’
The teacher was supposed not to interfere at the mathematical level during the whole
activity. The students had to feel that it is entirely their reponsibility to decide which solutions
are valid and which are not. Otherwise, the researchers claimed, the real activity of mathematical
proving will not emerge, and the students will only want to produce statements acceptable by the
teacher and the teacher’s authority will be the ultimate criterion of validity of solutions.
The teacher was allowed to intervene only at the level of the presentation of the problem
and the organization and chairmanship of the debate.
However, in the implementation in Class I, the teacher intervened in ways that did affect
the students’ attitude towards the mathematical activity and, in actual fact, relieved them from the
responsibility for the mathematical validity of their solutions.
Here are some examples of this kind of interventions, as reported by the researchers:
- in order to guarantee that the research phase be not too long, the teacher invited the students to
propose a solution as soon as she thought that what they obtained was sufficiently developed, and
not when she was sure that the students thought so;
- the teacher was drawing the students’ attention to some crucial words in the formulation of the
problem (in particular, the word ‘any’ in Class I);
- the teacher in Class I kept continuous contact with the students, making about one intervention
every minute over an 80-minute period (from ‘Are you okay?’, to ‘Have you read the problem
carefully?’);
- when she considered that a discussion is irrelevant from the mathematical point of view and
leads nowhere, she would urge the students to dismiss it and go further; she would also focus the
students’ attention on solutions and ideas that she considered relevant, and sum them up for the
students.
As a result of these interventions the students ‘got confused and were no longer
committed to any real discovery of the solution’.
In a repetition of the experiment in another class, the teacher was asked explicitly not to
intervene at all, and stay at her desk during the whole phase of discovery. She did. She only told
the students what is the problem that they have to solve and told them that they have so much
time for it. But this did not make the students engage in the activity of proving, either. The
students compared the solutions with respect to their simplicity, clarity, usability, but not their
validity. When, after one solution was accepted by the whole class, the teacher asked if the
authors of this solution are sure of it. They responded: ‘Yes, because we have done it in a lot of
cases’. So it was not even enough for the teacher to directly ask for a proof, to make the students
produce a proof, although, as it was found out, they were perfectly able to prove that their
solution was correct, in a mathematical way.
In Class II, the teacher intervened in a stronger way even than the teacher in Class I: she
asked leading questions (‘Do you all agree? It’s not proved’), she did not transcribe some
solutions on the big sheets of paper, she intervened directly on the content of the debate, she
reinforced some students’ ways of proving (‘Now you make a drawing in each group’. ‘It works,
but only if M is on the bisector’, ‘Look at your drawing and come to an agreement’).
The researchers concluded by saying that as soon as the teacher (a) tells the students that
they have a limited time to solve a problem, and (b) shows the students that she endorses the
‘epistemological responsibility’ for the mathematics produced in class (e.g. when she refuses to
write a false statement on the board), the students have no reason for entering a genuine activity
of proving in the mathematical sense.
It turns out that it is very difficult to implement a teaching design in a way which
conserves the intended mathematical meaning of the activities. The meaning of the activity may
change from one implementation case to another, due to the teacher’s interventions. The
teacher’s interventions are caused by her beliefs about her role and her duties as a teacher, and
when she acts on the basis of these beliefs, she may not support the development of the students’
development as autonomous thinkers.
Exercise 3.
Find a few critical features of the didactic situation ‘Race to 20’ (pp. 3-18) that, if altered, would
lead to its failure in attaining such objectives as: discovery and proof, by the children, of a
sequence of theorems related to the notion of Euclidean division in integers (‘division with a
remainder’).
4. COGNITIVE, DIDACTIC AND EPISTEMOLOGICAL OBSTACLES
When students start upon learning a new mathematical concept, their minds are not blank slates,
they are filled with all kinds of knowledge, beliefs, experience. New knowledge is not simply
added on, it must be merged with the old knowledge. But new knowledge can sometimes
contradict the old knowledge, and then the old knowledge functions as an obstacle to learning the
new one.
The contradictions are inevitable because whatever knowledge we have, is an answer to a
domain of questions and problems. This is usually a limited domain, but as long as we have not
transgressed its limitations, we believe that there are no limits and our knowledge is universal. So
when we find ourselves in front of new questions and problems we want to solve them with the
knowledge we have and this knowledge may not be capable of solving these new problems. We
need to reject parts of our old knowledge, or re-organize it, or generalize it and recognize that it
had a limited domain of application. This process feels as if there was something blocking our
mind - an obstacle.
It is not always possible to know in what ways the students’ old knowledge can function
as an obstacle in learning the knowledge aimed at by a didactic engineering project. But this old
knowledge may bias the students’ interpretation of the tasks given to them and they may thus
completely miss the mathematical point of the activities.
According to Brousseau, obstacles that appear in the teaching of mathematics can be of
various origins: ontogenic, didactic, epistemological, cultural (p. 86).
Ontogenic obstacles are ways of knowing whose limitations are due to the stage of the
mental development of the child. A 6 year old child cannot be expected to understand the
principles of an axiomatic theory.
Didactic obstacles are ways of knowing whose limitations stem from a certain way of
teaching. For example, in elementary courses in physics at high school, vector magnitudes such
as force or velocity are taught in a context of problems that minimizes the complexity of
mathematical computations. If the students have not studied the cosine theorem in mathematics
but did study the Pythagorean theorem, composition of magnitudes problems will deal with
orthogonal vectors only. The students risk to develop a conception that vector addition applies to
orthogonal vectors only, which will function as an obstacle in their study of vector geometry and
physics later on. Moreover, school problems related to vector magnitudes in physics tend to
concentrate on the measures of the magnitudes; the direction of action of a force or motion is
given either implicitly in a diagram accompanying the problem, or in non-mathematical terms
such as ‘North-West’ in the text. Direction is not something that is calculated and it is normally
not the object of the question in the problem. So it becomes unimportant. This may explain why
in vector geometry, many students believe that a vector can be completely defined by its length14.
Epistemological obstacles are those whose limitations are related to the very meaning of
the mathematical concepts. A mathematical concept has many levels of generality and
abstraction, and many aspects, developed during its history and depending on the context of its
use. Each level and aspect has its limitations, and if one thinks of a concept in a meaning that is
not appropriate for the given context or problem, then this way of thinking functions as an
14
Knight, R.D. 1995: The Vector Knowledge of Beginning Physics Students. The Physics Teacher 33, pp.
74-78.
obstacle and one makes mistakes or cannot solve the problem. For example, one can think of real
numbers as measuring numbers, or as elements of an algebraic structure called a well ordered
commutative field. The latter understanding is useless in the context of problems related to the
measurement of lengths, areas and volumes of geometric figures. The former is useless when one
deals with questions such as: could we put all real numbers in a sequence?15
15
More about epistemological obstacles can be found in:
Sierpinska, A. 1994: Understanding in Mathematics. London: Palmer Press, pp. 112-137.
Sierpinska, A. 1992: On Understanding the Notion of Function. In E. Dubinsky and G. Harel, The Concept
of Function, Aspects of Epistemology and Pedagogy. MAA Notes, Vol. 25, pp. 23-58.
Sierpinska, A. 1991: Some Remarks on Understanding in Mathematics. For the Learning of Mathematics
10(3), 24-26.
LECTURE 8
DISCOVERY TEACHING/LEARNING AND THE THEORY OF DIDACTIC SITUATIONS.
1. HOW ‘GOOD’ OR ‘BAD’ IS THE THEORY OF SITUATIONS?
IS THIS A GOOD QUESTION TO ASK ABOUT THE THEORY?
Lessons or lesson plans cannot be said to ‘conform’ or ‘not conform’ with the theory of situations
or didactic engineering. The theory of situations provides us with a perspective from which to
better understand what is going on in the classroom in terms of the students’ learning, the
mathematical meanings that the students construct, and the role, in this process, of the teacher.
When we design a lesson with the theory of situations as a guide then we are not trying to
conform to some set of rules or norms but we are trying to foresee the effects of our decisions and
choices with the tools of the theory. Thus, we may come to the conclusion, that in organizing the
classroom activities in such and such way, all we’ll obtain is an expected behavior in the students
(for example, they will produce an expected equation) but not the expected understanding. Then
we may try to change the lesson plan so that the expected understanding does develop in the
students. However, no matter how hard we try, we may still be wrong in our expectations, and
when experimented, the lesson will not produce the understanding and knowledge we aimed at.
If this happens, it is not the theory that is to blame but our design in which we may have failed to
take into account certain variables.
Here is what Brousseau wrote in respect to the status of the theory of didactic situations
in an e-mail message sent to me in the last week of November 1999.
‘The theory of situations is aimed to serve both the study and the creation of all kinds of learning and
teaching situations, whether they are “spontaneous”, or the product of an experience or of a special
engineering project, and whether they are efficient or not. It is not a method of teaching. The theory can
provide some methods of teaching, it can justify some methods and disqualify some other methods, as the
case may be. The theory contains models that may support certain plans of action aiming at making the
students (re)discover some mathematics. This way the theory can make suggestions for engineering. The
didactic situations of “rediscovery” can thus be, in general, linked to these models’ (my translation).
The practical significance of a didactic theory can be defined, partly, by its relationship
with a methodology of instructional design. In the case of the theory of didactic situations, the
corresponding methodology is that of didactic engineering. This is how Brousseau explains the
relationship between the two:
11/4/03 1
‘Theory of situations has been elaborated to provide a framework for the study and means of description of
any situation in which there is an intention of teaching someone some precise knowledge, whether it
succeeds in it or not. It can take into account all the forms of learning identified in all kinds of research.
The theory does not pretend to present all aspects of teaching situations and replace all the approaches:
psychological, psychoanalytical, linguistic, statistical, etc.. But it tends to put the contribution of these
approaches in the perspective of their function and their generality in the description of “didactic
phenomena”. One could use here an analogy with economics and its relation with commerce. Economics
does not deal with the psychology of the vendor or of the buyer but it can take into account their impact on
the macroeconomical phenomena.
Theory of situations is neither an ideology nor particular didactic method. In this sense, it has no
technical alternative. It does not directly recommend this or that particular didactic procedure. Its
theoretical concepts only allow one, for reasons of consistency, to predict the role of certain factors in some
circumstances. This way it puts limitations on what it is possible to do or change in teaching, just as
thermodynamics discards the possibility of building a perpetuum mobile but does not give precise
guidelines for the construction of an ideal engine.
On the other hand the theory manages well the contingency or the possibility that some of its
theoretical predictions be rejected as false. For example, at one time I claimed that “a situation of
formulation related to a precise piece of knowledge cannot function if a previous situation of action has not
allowed the students to develop an implicit model of this knowledge”. But, later, I found counter-examples
and I had to restrict the generality of the claim that, previously, made a lot of sense to me. Another
example: I claimed that “An implicit model of action that is not, rather quickly, supported by a formulation,
is lost as fast as it is learned”. I had to soften this declaration by provisions such as “in most cases”, “if the
situation of action is evolving”, etc..
In this sense, the relevance of the theory lies in its ability to raise questions and classify and order
the answers. (One of the difficulties of mathematics education is the multiplicity and diversity of the
research work, classified according to criteria which appear to have nothing to do with each other. The
important thing is thus to be able to judge of the relevance of the questions and the validity of answers).
Yet, if I have developed the theory of situations it is because I needed it for didactic engineering.
It is a product of my efforts to classify questions that are raised when one wants to organize, by whatever
means, the learning, by anyone, of some knowledge. This explains why one finds, in the description of the
types of situations, arrangements corresponding to the possibility of the construction of knwledge by the
students in a non didactic situation. This is what resembles most, it seems, to the settings in which one can
detect the “discoveries” of the students and imagine their role in the learning processes. There remains the
question of interpretation, in the theory of situations, of the conclusions and claims of the observers of
“discovery learning”.
This way, the theory of situations restricts the meaning that one can attach to “didactic
engineering”. The point is to distinguish between didactic “innovations” or “inventions” made for the
educational market, and those products of the didactic work that are based on proved techniques, described
by explicit and sensible technologies, which are, in turn, founded on falsifiable theories and verifiable
experiences. The distinction is not easy because, in any work of didactic engineering, inventions play a
very important part, as do empirical considerations which can be extremely complex and possibly
judicious. Moreover the success of these products at differents levels (students, teachers, society) is not
decisively determined by their belonging to this or that category.
I am presently trying to fill the gap that I have left between the theory which presents the wide
basic concepts and general didactic phenomena, and the different products of the didactics proper (the
teaching of natural numbers, decimal numbers, rationnal numbers, measure, statistics and probability,
reasoning and logic, space and geometry, elementary arithmetic and algebra) whose practice I have
organized and observed. I am dreaming of a big treatise on didactic engineering which I consider as the
didactics proper.’ (My translation)
11/4/03 2
DISCOVERY LEARNING APPROACH
I have judged one of the lessons presented in the class by students (related to Pythagoras
theorem) as an example of the ‘discovery learning’ approach. I did not mean to say that it is not
‘conform with didactic engineering’. I only meant to say that, if one analyzed this lesson from
the theory of situations perspective prior to actually conducting the lesson then one would either
change the plan of the lesson or interpret the outcome of the lesson differently: the evaluation of
what the students have actually learned in that lesson would be different.
What I understand by ‘discovery learning approach’ is an organization of the students’
activity in a sequence of exercises so that, at the end, some representation of the target knowledge
is produced by the students (e.g. a formula is written, a term is said, a drawing is produced), so
that, in the eyes of one knowing the target knowledge, the manifestation of this knowledge in the
lesson is obvious. In the approach presented by one of you, in the activity phase, the students
were already manipulating representations of the target knowledge; the formulation phase was
strongly guided by the teacher (Topaze effect), and the validation phase occurred after the
‘discovery’ (Jourdain effect): the students were then asked to prove a theorem which they already
knew to be true.
I asked Brousseau, by e-mail, for an explanation of the relations between the discovery
learning approach and the theory of situations and he kindly responded (last week of November
1999, same message as above). I partly translate and partly summarize, below, his answer.
‘The theory of situations makes it possible to identify the circumstances of the discovery by the students (or
anybody, for that matter) of some knowledge but, within the theory, the notion of “learning through
discovery” appears contradictory. On the other hand, the theory can recognise the notion of “teaching
through discovery” as referring to the organization of non didactic situations, and to the identification, the
interpretation and the institutionalization of the knowledge produced by the students in these situations.
The “teaching through discovery” approach creates situations based on “an epistemological and didactic
fiction” which favors an auto-didactic reading of these situations by the students.
Briefly speaking, the “discovery learning” appears to be a myth, a psychological extrapolation
invented to justify and conceal a “discovery teaching”: a method of rediscovery. The problem solving
method is based on heuristic interpretations of the mathematical activity, as opposed to the classical
axiomatic and purely deductive interpretations. It leads to epistemological commentaries and “original”
terminology and methods. Heuristics start to be taught as ordinary objects of teaching. These heuristics are
conjectures about the learning of mathematics that are false in all their generality but useful or valid in an
important number of cases. But they are indeterminate, otherwise they would become theorems. There is
no reason why these heuristics would work better than the “real theorems”; in that case one could look for
second order heuristics to produce and guide the first order heuristics. There is no doubt that heuristics,
like, for example, analogies, are useful for the researcher who engages them under his or her own
responsibility. But when they are proposed to the students as methods, in the frame of a didactic contract,
they are nothing but some of the invisible means of the teacher to conceal his or her didactic strategies and
produce the Topaze or Jourdain effects.
This being said, there is no reason why heuristics should not be used when they work. The theory
of situations does not make a mystery out of the theatrical component of any didactic situation. To play at
11/4/03 3
discovery is not as realistic as to play at a competition of theorems but it can be as exciting and maybe
more flattering, more gratifying for the ego of the student as well as of the teacher; briefly, it can be,
locally, just as efficient for the student’s learning.
From a strictly scientific point of view, it is not necessary to invent false psychological and
cultural mechanisms to justify this didactic approach. But in pre-service teacher training, if didactic
situations are presented for what they are, then perhaps the future teachers will not fall for it. It might be
that they need to “believe in the reality of movie characters”. To be successful in teaching one needs such a
will or conviction that perhaps the easiest thing to do would still be to justify this professional “faith” by
appropriate beliefs?’ (my translation)
It seems to me that the debate should focus on three points:

- The relations between the theory of situations and the didactic engineering;
- The relations between the ‘discovery learning’ approach and the theory of situations;
- The possible difference between the meaning of the word “theory” in Educational Sciences,
especially anglo-saxon, and its interpretation in natural sciences, and the difference of perspective
on the nature of didactic phenomena where the make believe and the masquerade are the standard
practice.
11/4/03 4
LECTURE 9
DIDACTIC VARIABLES. WHAT TO OBSERVE IN A CLASSROOM? HOW TO PRESENT A TEACHING
PROJECT?
1. DIDACTIC VARIABLES
Let us ask the question:

What is a teaching situation a function of?
To answer this question we must first decide what we shall understand by a teaching situation. In
the theory of situations, the teaching situation is a game between the teacher and the system
composed of the student and the milieu:
r
STUDENT MILIEU
TEACHER
Thus, the teaching situation depends on

- the personal characteristics of the TEACHER; in particular his or her personality, beliefs about
what it means to teach and to learn at school (i.e. his or her ‘epistemology’), his or her
mathematical knowledge, cognitive style,…
- the personal characteristics of the STUDENTS; in particular the stage of their cognitive
development, their cognitive style, their personalities (affective, motivational),…
- the composition and arrangement of the MILIEU; in particular, the task proposed to the students
(exercise, open-ended problem, project, …), the resources and tools put at their disposal such as
texts, game boards, geometer’s kits, calculators, computers, pens and pencils, crayons, scissors,
paper, exercise books, desks and chairs, the arrangement of the desks and chairs in the
classroom,…)
- the RELATIONS r between the students and the milieu; in particular
- the students’ relation to the task (school problem vs taken-as-own problem)
- the students’ contact with the task (mental vs physical)
11/4/03 1
- the students’ relations with the resources and tools (in particular, the students can be
using a calculator as a computational device only, or as a heuristic tool)
- the RELATIONS R between the teacher and the student - milieu system; in particular
- proportion of teacher’s interventions with respect to students’ time on task
- the subject of teacher’s interventions (organizational vs mathematical)
- the kind of teacher’s interventions (presentation of task, hints, genuine questions (e.g.
tell me how you solved this problem), rhetorical questions (i.e. questions to which the teacher
knows the answer), questions meant to maintain the contact and students’ attention (e.g. ‘Are you
with me here?’ and the teacher continues his or her lecture without waiting for a reply),
exposition of theory, solving exemplary test problems, discussion with students, etc.
I have thus listed a large number of variables on which a teaching situation appears to
depend. But not all of them will be called ‘didactic variables’; only those that can be controlled
by the teacher, not as a person, but as an element of the didactic system. From this point of view,
the didactic variables will be those related to R, r and MILIEU. Variables related to the Teacher
and the Student as persons are excluded, not because they are not important in the process of
teaching but because they would qualify rather as psychological and sociological variables and
not as didactic variables. The assignment of values to didactic variables in the design of a lesson
must take into account the sociological and psychological variables but it has no control over
them.
2. WHAT TO OBSERVE IN A CLASSROOM? HOW TO ANALYZE AN OBSERVED LESSON?
When you are sitting in a classroom observing, take note of those facts that are pertinent from the
point of view of the didactic variables. Make sure you are able to answer questions like:
- What were the tasks proposed to the students?
- What were the resources and tools put at the students’ disposal?
- How was the classroom furniture arranged and how were the students and the teacher
positioned?
- What were the students’ relations to (a) the task (b) the tools and resources ?
- What was the ratio of time of teacher talking to students talking or working on tasks?
- What kinds of interventions did the teacher use?
- What did the students do?
When analyzing the data obtained through observation, ask yourself questions related to
the mathematical content of the lesson:
- What mathematics did the teacher appear to want to teach the students?
11/4/03 2
- Were the didactic means he or she used likely to help him or her to meet this goal? (Why yes or
why no?)
- What were the students learning?
- How would you change the lesson if you yourself were to teach the same mathematical topic?
3. HOW TO PRESENT A TEACHING PROJECT?
In presenting your teaching project, first describe the mathematical knowledge that you expect
your students to develop through participation in the proposed activities. Then describe the
organization of the MILIEU and explain why do you think that this organization can help the
students to develop the target knowledge.
To describe the organization of the milieu means to assign values to the didactic
variables, and thus this description should contain answers to questions related to the variables M,
R and r:
- What are the tasks proposed to the students?
- What are the resources and tools put at the students’ disposal?
- How is the classroom furniture arranged and how are the students and the teacher positioned?
- What are the expected students’ relations to (a) the task (b) the tools and resources ?
- What is the teacher’s role in the situation and what kinds of interventions is he or she expected
to make?
- What are the students expected to do?
The justification of your choices in regard to these questions should be guided by your
aim of teaching the students the assumed knowledge.
11/4/03 3

Lecture Notes On The Theory of Didactic

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Lecture Notes On The Theory of Didactic

Încărcat de

Drepturi de autor:

Formate disponibile

Lecture notes

on the Theory of Didactic Situations in mathematics

Concordia University, Montreal, Canada

given in year 2003

in the frame of the program

Master in the Teaching of Mathematics

© 2003 Anna Sierpinska

1. WHAT IS A DIDACTIC SITUATION?

The context of the didactic situation

2. SITUATIONS OF ACTION, FORMULATION, VALIDATION AND INSTITUTIONALIZATION

3. EXAMPLE OF A SEQUENCE OF CLASSROOM SITUATIONS: THE RACE TO 20

Phase 2: The students play the game in pairs (4 rounds, 10 mn)

Phase 4: Game of discovery: Formulation of propositions

4. ANALYSIS OF THE EXAMPLE

Let us reflect on the following questions:

2. What does the term ‘dialectic’ refer to?

2.1 Meanings of ‘dialectic’ in the history of philosophy

PART II: AN EXERCISE IN THE STUDY OF MEANING OF A MATHEMATICAL CONCEPT: THE

1. Meaning of the operation of division for a research mathematician

2.1 Information about the textbooks

2.1.1 Categorization of division problems used by

Titles of chapters and sections related to division

Chapter 4 - Multiplying and dividing with mixed numbers

Chapter 8 - Multiplying and dividing with decimals

2.1.2 Categorization of division problems used by

Zawadowski, W. et al. (1998): Matematyka 2001. Podrecznik do klasy 6 szkoly

Titles of sections related to division

8:8=1 3/16 : 8 = 3/128

• Add two further operations to each column.

8. Decide which product is less expensive

2.2 Brousseau’s criteria of classification of division problems

Group 2: type of magnitudes

Group 3: type of didactic situation

Group 4: previously taught techniques of calculation

reduction to operations in natural numbers

(2) Conceptions of division

1999 MATH 645 CONCORDIA UNIVERSITY

LECTURE NOTES ON THE THEORY OF DIDACTIC SITUATIONS – ANNA SIERPINSKA

The historico-epistemological context of the emergence of the notion of didactic contract

Paradoxes of the didactic contract

probability of a message is 1, then the amount of information it carries is 0. Solving a problem

Results of the activity of classification of ‘division problems’ in Week 2

Long (synthetic) division

Study of the meanings of division conveyed by two grade 6 textbooks

Analysis of the 1936 textbook problems

9. Divide and put your answer in simplest form: 9/10 ˆ 3/5.

13. Divide 3/4 by 5/9.

20. 7581.6 / 7.8

Analysis of the 1998 textbook problems

(a) Add two more operations to each column

But a : b = b, so b = 1/4 (transitivity of equality). But b is 4 times as large as a, so a is a

5. 2 1/3 + 3/4 : 1/2

7. 3 1/3 : (-5/6) : (-2)

8. Decide which product is less expensive

LECTURE NOTES ON THE THEORY OF DIDACTIC SITUATIONS – ANNA SIERPINSKA

Elementary arithmetic conceptions:

An advanced arithmetic conception:

Multiplication by the inverse: a/b = a*b-1.

1. The ‘Topaze effect’

2. The ‘Jourdain effect’

A side explanation: the group of Klein

3. The ‘Dienes effect’

3.3 Dienes’ instructional theory