P. L. Lijnse
Centre for Science and Mathematics Education, Utrecht University, The Netherlands


In the preface of his famous book "The Process of Education" (1960), the psychologist Bruner wrote about "a conviction that we were at the beginning of a period of new progress in, and concern for creating curricula and ways of teaching science" . He argued that a general appraisal of this progress and concern was in order, so as to better guide developments in the future.

At the time, this same optimism spread to other countries, leading to the well known curriculum wave of the sixties and seventies which flooded the world of science education. Now, 35 years later, it may be appropriate to look back for a while and ask ourselves what progress this curriculum development and related research has brought us.

To guide this reflection, it may be instructive to look somewhat further into what were considered to be the main problems and perspectives in 1960. Let me therefore briefly summarize some of Bruner's main conclusions.

Regarding 'the importance of structure' the following was said: ".. the curriculum of a subject should be determined by the most fundamental understanding that can be achieved of the underlying principles that give structure to that subject. Teaching specific topics or skills without making clear their context in the broader fundamental structure of a field of knowledge is uneconomical in several deep senses. In the first place, such teaching makes it exceedingly difficult for the student to generalize from what he has learned to what he will encounter later. In the second place, learning that has fallen short of a grasp of general principles has little reward in terms of intellectual excitement. (...) Third, knowledge one has acquired without sufficient structure to tie it together is knowledge that is likely to be forgotten."

As far as content choices are concerned, this idea of emphasizing the 'structure-of-the-discipline' seems to fit rather well with what academic physicists usually think to be of importance in teaching about their subject. Bruner, however, added a psychological rationale to this emphasis.

Regarding a second theme, 'readiness for learning', Bruner advanced his famous and much debated hypothesis that any subject can be taught effectively in some intellectually honest form to any child at any stage of development. This hypothesis was said to imply three aspects: the process of intellectual development in children, the act of learning (in particular the act of 'discovery') and the notion of a 'spiral curriculum'. Since then, these aspects have received considerable attention in curriculum development, as we shall see below.

A third theme of Bruner's relates to the fact that "the emphasis in much of school learning and student examining is upon explicit formulations, upon the ability of the student to reproduce verbal or numerical formulae. It is not clear, in the absence of research, whether this emphasis is inimical to the later development of good intuitive understanding indeed, it is even unclear what constitutes intuitive understanding".

"Usually", it is said, "intuitive thinking rests on familiarity with the domain of knowledge involved and with its structure". However, "the complementary nature of intuitive and analytic thinking should be recognized", particularly, as "the formalism of school learning has somehow devalued intuiton".

So, what to do about this? "Will the teaching of certain heuristic procedures facilitate intuitive thinking? For example, should students be taught explicitly that : "When you cannot see how to proceed with the problem, try to think of a simpler problem that is similar to it; then use the method for solving the simpler problem as a plan for solving the more complicated problem?"" In these statements, we see a foreshadowing of the cognitive swing which psychology has undergone since the sixties, a swing which has had much influence on research in physics education.

"In assessing what might be done to improve the state of the curricular art, we are inevitably drawn into discussion of the nature of motives for learning and the objectives one might expect to attain in educating youth", is said to introduce the fourth theme. This discussion is relevant to all levels involved, from the individual teacher and student in the physics classroom to the role of physics education in a society at large. Thus this theme will always demand serious attention.

Finally, regarding 'aids for teaching', it is concluded that "the teacher's task as communicator, model and identification figure can be supported by a wide use of a variety of devices that expand experience, clarify it, and give it personal significance". When we compare the personal computer with the 'teaching machines' of the sixties, we can see how this theme has acquired completely new significance in physics teaching since Bruner first wrote these words.

It is striking to realize how much of the above still applies today. In physics curriculum development and in research on physics teaching we are still struggling with the same problems. Nevertheless, in the past 35 years, a lot of work has been done. Much of the progress that has been made, if it may properly be called so, should be apparent from this volume.

In this chapter, I will restrict myself to the main experiences in physics curriculum development (as I see them). Many questions may be raised. For example, are we now (or still?) teaching the structure of the discipline, as Bruner advocated? Do we have physics curricula that are adapted to the intellectual development of children, and if so in what way? Is discovery learning still on the list of usual teaching strategies in physics? What different goals and objectives do we aim at now, and how do we deal now with motivation to learn ?

In dealing with some of these questions, I will structure my description along three main lines: aims and content, teaching and learning, and ways of curriculum development and implementation.

As we all know, physics education is not a constant but a variable. It changes in direct relation to the developments in the society of which it is a part, to the developments in that society's view on education and science and to developments in physics and technology themselves (Lijnse, 1983). Next to traditional schoolbook writing, professional curriculum development has come into existence as a means to adapt education to these continuing changes. And although in itself it is not usually considered to be part of proper research, it has stimulated the development of many research studies (Fensham, 1994)..


2.1. 'The structure-of-the-discipline'

The first major project, the PSSC physics course, primarily meant for "the academically superior collegebound students" (French, 1986), was very influential internationally (PSSC, 1960). As Matthews (1994) describes: "Its intention was to focus upon the conceptual structure of physics, and teach the subject as a discipline: applied material was almost totally absent from the text. Air pressure for instance is not mentioned in the index, it is discussed in the chapter on "The Nature of Gases", and the chapter proceeds entirely without mention of barometers or steam engines, the former making its first appearance in the notes to the chapter". The PSSC way of teaching included lots of experiments, reflecting an aim that the pupil should be 'a scientist for the day'. This latter characteristic seemed to apply even more to the equally influential English Nuffield-Physics projects (O-level, 11-16; A-level, 16-18). These projects all focused largely on teaching the basic disciplinary structure, although in a somewhat different way (Ogborn, 1978). As Rogers (1966), the man behind Nuffield O-level Physics, put it: "And for the things we do teach we should choose topics that have many uses. I do not mean practical applications, but rather linkages with other parts of physics. Science should appear to our pupils as a growing fabric of knowledge in which one piece that they learn reacts with other pieces to build fuller knowledge".

The O-level project aimed to 'teach for understanding' and at 'physics for all'( "a course suitable for the general educated man and woman" ). Later, however, it was realised that these curricula were not really geared to 'physics for all', but were best suited for the more able scientifically oriented pupils. Therefore, to indicate their rationale, such curricula were better described by one of Roger's own titles 'Physics for the Inquiring Mind'.

One particular aspect of these particular curricula is that they also played an exemplary role in tackling the problem of updating physics teaching from a disciplinary point of view, particularly in relation to the problem of teaching 'modern physics'. French described some of PSSC's main choices as follows: "The most basic and universal features of the physicist's description of nature such matters as orders of magnitude and the effects of changes of scale would be stressed. There would be a unifying theme the atomic, particulate picture of the universe in the presentation and discussion of the subject matter. Also, in the interests of achieving depth of treatment, substantial areas of traditional material (such as sound) would be omitted". And, as the developers of Nuffield A-level said: "One of our basic decisions has been to sacrifice a wide acquaintance with many ideas for a deeper understanding of fewer". In fact, following this principle, they developed really innovative introductions to topics like quantum physics, statistical mechanics and electronics (at an 'advanced' level).

In spite of the immense international influence of these projects, and although similar innovations have been tried in many countries (GIREP, 1973; Aubrecht, 1987; Fischler, 1993), one cannot yet conclude, I think, that, at the secondary school level, the didactical problems of the why, the what, and  the how of including basic modern physics have been solved satisfactorily. The more so as, due to the rapid development of physics, we do not only have to deal with basic ideas of quantum physics and relativity. Further new topics are already knocking at the door of the curriculum, like chaos, condensed matter physics, computational physics, high energy physics and cosmology (GIREP,1995, 1993, 1991, etc.). In fact, this rapid development causes the physics curriculum to be under a continuous top down pressure, with the serious danger of becoming ever more overloaded. In this respect, the structure of physics can, because of its largely hierarchical nature, not only be regarded as a curriculum guide, but to a certain extent also as a hindrance, as it is often much clearer what has to be included than what can be left out.

So far, no consensus seems to exist about how to deal with this pressure on the curriculum. In view of the time needed to reach understanding, Arons (1990), for example, still chooses to accept something like the Bohr atom as a useful endpoint for an introductory physics course. "What seems to me to be feasible and highly desirable in an introductory course is to get to the insights gained in early twentieth century physics: electrons, photons, nuclei, atomic structure and (perhaps) the first qualitative aspects of relativity". And even for that, hard choices have to be made: "To achieve this, it is impossible to include all the conventional topics of introductory physics. One must leave gaps, however painful this may seem. How does one decide what to be left out? One powerful way, in my experience, is to define what I call a 'story line'. If one wishes, say to get to the Bohr-atom, one should identify the fundamental concepts and subject matter from mechanics, electricity, and magnetism that will make understandable the experiments and reasoning that defined the electron, the atomic nucleus, and the proton. The selected story line would develop the necessary underpinnings and would leave out those topics not essential to understanding the climax. For students continuing in physics, the gaps would have to be recognized, accepted, kept in mind by the faculty, and closed in subsequent courses".

To my opinion, the problem of how the physics curriculum as a whole can be constructed as a set of usefully intertwined gradually developing 'story lines', now needs renewed attention (Ogborn, 1978). Or, in other words, we have to ask once again how can the structure of physics (in a broad sense) be turned into a better teachable curriculum structure (De Vos et al., 1994))?

2.2. Process and Processes

In the projects mentioned above, much attention was also given to the 'process of physics' and to letting pupils experience the 'process of discovery (or inquiry)'. In the PSSC-texbook "physics is presented not as a mere body of facts but basically as a continuing process by which men seek to understand the nature of the physical world".

But pupils should not only learn about a process that others, in particular 'great' physicists, have gone through, they also should experience this process themselves. To quote Rogers (1966) again: "Practical work is essential not just for learning material content, but for pupils to make their own personal contact with scientific work, with its delight and sorrows. They need to meet their own difficulties like any professional scientist and enjoy their own successes, so that the relation of scientific knowledge to experiment is something they understand".

So, one could say that this emphasis on process was in the first place justified by internal reasons. It is part of understanding physics to know about how knowledge of physics is generated and how it develops. And as physics is an empirical science, it is considered an inherent part of physics to learn about nature by finding out, hypothesizing, testing and experimenting for yourself, i.e. students should be learning physics by doing physics. Since then, the use of "practical work" in physics education has increased enormously as it has become an integral part of many curricula and textbooks. This trend has developed to such an extent that the learning of experimental skills sometimes seems to have become an aim in itself, almost unrelated to the purpose of experimenting, i.e. developing new knowledge (Woolnough and Allsop, 1985; Woolnough, 1989; Wellington, 1989; Hegarty-Hazel, 1990; Hodson, 1993).

It also has become clear, from research on physics learning, that the original idea of discovery learning may have been somewhat too naive (Driver, 1983). On the other hand, learning physics by doing has nowadays almost acquired an extra dimension because of the possibility of modelling 'artificial worlds' that has come into reach by means of the microcomputer (Mellar et al., 1994).

Returning to history, in Harvard Project Physics (1970), another American project that earned much international applause, the attention to the internal process of physics was placed within a much broader intellectual perspective, in that external influences were also considered. In the famous words of Rabi: "I propose science be taught at whatever level from the lowest to the highest, in the humanistic way. It should be taught with a certain historical understanding, with a social understanding and a human understanding in the sense of the biography, the nature of the people who made this construction, the triumphs, the trials, the tribulations". Because of this particular emphasis, it was expected to attract a broader group of pupils (particularly girls). From a physicist's point of view, again, this project developed marvelous curriculum materials. Nevertheless, it not only did not really succeed in attracting significantly more students (French, 1986): for a long time its historical and philosophical approach seems to have been adopted by only a few teachers. Only recently this curriculum focus on history and philosophy, whilst always somewhere in the background, has acquired new impetus (Matthews, 1994). Attention to the 'nature of physics', to its historical, epistemological and methodological aspects, is now becoming a regular part of physics curricula (Aikenhead, 1991; Solomon, 1991). In England, it has even been included in the prescribed National Curriculum, while, e.g., in The Netherlands a new curriculum for "general science" is being developed in which this perspective is given much attention.

Historically, this broader perspective meant that the emphasis was (partly) shifted from teaching as inquiry to teaching about inquiry. An even more drastic shift, I think, was implied by what Shulman and Tamir (1973) called teaching of inquiry. This step was taken to its extreme in a third influential approach, developed by the U.S. project SAPA, which stands for Science A Process Approach. In the words of the psychologist Gagné, this project "rejects the 'content approach' idea of learning highly specific facts or principles of any particular science or set of sciences. It substitutes the notion of having children learn generalisable process skills which are behaviorally specific, but which carry the promise of broad transferability across many subject matters".

Scientific behavior was analysed in terms of its simpler constituent 'scientific process skills', such as: observing, classifying, measuring, communicating, and making inferences, that were thought to be learnable and teachable as such. Since then, the debate about whether one should emphasize scientific knowledge and/or scientific processes has been ongoing (Millar and Driver, 1987). Nowadays, it is even more topical than ever as many cognitive psychologists advocate the learning of even broader 'general skills' (thus not only scientific see below), not only as an aim in itself but also, as already proposed by Gagné, as the appropriate way to deal with the mentioned threat of 'elephantiasis' of the curriculum.

2.3. Broadening of aims

As already indicated above, the curricula that focused on physics-as-a-discipline appeared, both in its rationale and in its cognitive demands (see below), to be more geared to the gifted science interested pupils than to 'physics for all', thereby leaving a curriculum gap for the less able, less scientifically interested pupils. For them, (more) integrated science as well as technology projects were developed (see, e.g., Brown, 1977). These may be interpreted, in line with the spirit of that time, as a shift from discipline centred to more pupil-centred education. A main rationale behind integrated science was that a division in seperate disciplines does not coincide with the way in which pupils experience their world (however, as Black (1985) argued, pupils do not experience their world in an integrated-science way either). As a consequence, integrated science has been implemented in many countries, although some seem to have returned to coordinated science. Other countries, have resisted this trend and have not gone for integration at all.

Early technology projects were mainly developed as add-on activities to the physics curriculum, which reflects an application-of-physics view of technology (e.g. Schools Council, 1975). At present, this view of technology is no longer regarded as adequate, resulting in a gradual emancipation of technology to become a separate school subject (Layton, 1993).

In the seventies, another emphasis gradually developed towards what is now called STS (see, e.g, Solomon and Aikenhead, 1994), although that acronym still stands for a number of considerably different approaches. One of them deals with explicit reflection on the relation of science, technology and society (e.g. the English Science in Society project (SiS)), thus emphasizing social implications and issues. Another approach places more emphasis on relevancy of content for pupils, by teaching science in daily life and issue-related contexts (e.g., the Dutch PLON project for physics; Satis, 1992). Roughly, both approaches have also become known as 'science for the citizen' and 'science for action', or as contextualized science (or physics).

The SiS-project is an example of a project in which the social dimension is treated as an add-on to the regular curriculum. In the PLON project, however, attention for social scientific issues, 'consumer physics' and other 'pupil-relevant' contexts are integrated in the physics curriculum itself. If, however, the boundaries set on the curriculum are such that the physics curriculum should keep its identity as 'proper' physics, such a contextualized approach may result in considerable tension between the knowledge that seems to be relevant for the contexts chosen, and that which demands inclusion from the perspective of physics. Or, in other words, one has to try to find a balance between the 'structure-of-physics' and the structure-of-the-contexts (Lijnse, et al., 1990).

Both approaches, however, imply a broadening of traditional aims (Fensham, 1988), related again to the idea of 'science for all' alhough, in this context, this phrase is now to be interpreted differently from above. In connection to this broadening, in the eighties, new topics, such as environmental education and information technology had to find a curriculum place as well. A matter of discussion was and is whether these should be part of regular physics education, or are should be taught as seperate subjects.

New problems have also arisen from a societal point of view, for example the adaptation of physics education to the needs of girls (Bentley and Watts, 1986) and to the needs of a multicultural society (Reiss, 1993).  In fact, how broad can we make our aims and still remain within the borders of physics education? Or even might it be better to remove physics from the school time table?

All this also relates to another trend that is now attracting considerable attention, the new emphasis on scientific and technological literacy for all, including out-of-school ways of educating the general public.

However, next to this broadening tendency (the spirit of the late seventies and early eighties), we see another tendency showing up that focuses less on physics for 'citizenship', and more on the value of physics in the education of a highly qualified workforce (the spirit of the late eighties and nineties). Vocational qualifications are formulated and physics teaching is required to contribute to their attainment. Consequently, we note again a change in curriculum discussions, characterisable as a shift from pupil and relevance centred to 'client' and achievement centred, even with special attention for the gifted child. This trend can lead to pressure to restrict the content of physics curricula to its 'hard core' (preferably described in attainment targets, that can be regularly tested). However, the content of this 'hard core' is now not so much to be decided by 'pure' academic physicists or physics educators (as was done in the past), but by those who form the 'market' for which we educate our pupils (such as employers and institutes for higher education).

This is a very brief and very subjective overview of about forty years of curriculum discussions about aims and content. What may we conclude from this description? Apparently, physics education has had, and still faces now, a continuous stream of 'top-down' innovations. A first obvious conclusion, however, could be that, as far as aims and content is concerned, the same themes seem to show up regularly in a kind of wave motion, driven by changing views on education in changing societies. May we, nevertheless, conclude that physics education is spiralling upwards in some sense that may be called progress (as Bruner expected)? Or should we conclude that physics education is walking around in circles, almost like a snake regularly biting its own tail? Or does this question for progress only represent a 'category mistake', as the late Dutch mathematics educator Freudenthal (1991) argued: "Once, asked by an interviewer whether I thought that attempts at innovating have improved education, I hesitated for a short while, only to eventually stamp it as a wrong question. Pictures of education, taken at different moments in history are incomparable. Each society at a given period got the education it wanted, it needed, it could afford, it deserved and it was able to provide. Innovation cannot effect any more than adapting education to a changing society, or at the best can try to anticipate on the change. This alone is difficult enough". Before going somewhat further into this question, let us first have a closer look at curriculum considerations that have resulted from research on (physics) teaching and learning.


3.1. Behaviorism and 'Piagetianism'

In the above, I have not focused on ways of teaching and learning and on the influence of research on this aspect of curriculum development. Let us therefore paint another broad picture.

In the fifties and sixties, the dominant psychological viewpoint in education was that of behaviourism. It focused on the formulation of educational objectives and aims, distinguishing between knowledge and skills, and organised in learning hierarchies and taxonomies (Bloom, 1956). In fact, Gagné's approach, mentioned above, is an example of this view (SAPA, 1968). Programmed instruction and teaching machines developed into individually paced study systems and mastery learning (Bloom, 1971; White, 1979). Despite research reports about successful implementations, these approaches have largely faded away, although, in some sense, they showed up again more recently in much computer assisted teaching.

According to this position the teaching process should best be split up into smaller and smaller steps, leaving the sequencing of content, however, to continue to follow the 'logical' disciplinary structure. In that sense, in behaviourism, curriculum content is not a variable and it therefore had only a weak link to development of the 'didactics of physics'. Its lasting contribution to physics education has not been spectacular.

Another psychological position which has had much greater influence on physics education is 'Piagetianism'. Thus, Bruner's recommendation, quoted above, has been taken seriously. Piaget's description of concrete and formal operational thinking has been and still is a useful global guide in designing teaching. Apart from having influenced many curriculum projects, (some of which adopted explicitly a Piagetian perspective, such as ASEP, 1974) the Piagetian stage theory has, particularly in the U.S.A., given rise to a wealth of quantitative studies relating pupils' cognitive growth to many other quantitative variables. In the end, this type of research seems to have had little practical influence. More useful was the U.K.-based use of stages as a tool to identify the excessively high demands set by many (newly developed) curricula, as well as a means to match them to assumed age-dependent capabilities of pupils (Shayer and Adey, 1980; Adey and Shayer, 1994). At first, this research played an important role in making 'tangible' the extent to which, and the ways in which, the 'physicist-for-the-day' type of curricula mentioned above were inclined to overestimate the capabilities of 'all' pupils. Thus, 'Piagetian-ism' made the important shift from taking only the curriculum-to-be-taught as the sole starting point for curriculum development, to including also the cognitive development of pupils. That means that from the Piagetian point of view, curriculum content is seen as a 'structural' variable, to be sequenced according to 'developmental logic'.

Later, based on Piagetian reasoning patterns, curriculum materials have been developed, to be implemented as intervention lessons within the science curriculum, that aim not so much at the improvement of science learning in a narrow sense, but much more at the advancement of children's cognitive development itself (Adey, Shayer and Yates, 1989). Nevertheless, the real significance and potential of the Piagetian stage theory is still a matter of debate (Carey, 1985). This is much less the case for another aspect of 'Piagetianism', i.e., its 'constructivist' foundation (Bliss, 1995; Adey and Shayer, 1994): the idea that a learner essentially constructs his own knowledge by acting on his environment. When first formulated, this gave a kind of psychological foundation to the attractiveness of 'discovery learning' for science education, as worked out in several modes of 'learning cycles': exploration (messing around), invention, discovery (application). Alhough, as argued above, discovery learning in its naive sense has disappeared again, constructivism is still around.

It is difficult to say in what way Piagetianism has made a lasting contribution to science education. It is striking that the stage theory is hardly mentioned in current literature. Alhough much literature of the seventies was very optimistic about its value, I think that we may conclude that nowadays most research is only globally influenced by Piagetian stage theory. Or maybe we should say, nowadays research in physics education does not try any more to develop its potential (see, however Lawson ,1994, for a new interpretation).

3.2. Constructivism

This change must be linked with the spectacular rise, since the late seventies, of what I like to call 'didactical constructivism'. Using this term, I'm referring to what started as the 'alternative framework' movement, as it is sometimes loosely called. This movement may be regarded as also having its roots in the (early) work of Piaget. In fact, it did build on the way in which Piaget investigated the content of children's ideas about specific phenomena, but not on his analysis in terms of hypothetical underlying logico-mathematical structures that led to the stage theory mentioned above. At first, this focus on children's content specific reasoning led to numerous diagnostic and descriptive research reports about all kinds of pupils' concepts and ideas about situations (Driver, Guesne and Tiberghien, 1985). This has since been extended to pupils' ideas about experiments (Carey, et al. 1991), about learning and teaching, and about their epistemologies (Butler Songer and Linn, 1991). Subsequently, the same has been done done for teachers' ideas and opinions (Tobin et al., 1990). Also developments in time of pupils' and teachers' conceptions have been studied, be it during a number of lessons, or over many years (Driver et al, 1994).

Apart from the usual 'implications for teaching' that seem to be an almost obligatory endpoint of too many research studies, experimental classroom studies have been and are done to find concrete ways to improve the teaching of certain topics, or to find more general and better teaching strategies (CLIS, 1990). Such studies make clear that this research field has important implications for curriculum development, that are still to be developed to their full potential. It even implies a certain change of view on how we think about a curriculum. As Driver (1989) writes: "Curriculum is not that which is to be learned, but a programme of learning tasks, materials and resources which enable students to reconstruct their models of the world to be closer to those of school science". An important consequence of this view is that "the curriculum is not something that can be planned in an  a priori way but is necessarily the subject of empirical enquiry".

Theoretically, the dominant position in this "paradigm" is that of 'constructivism and conceptual change'. Much research is aimed at explaining processes of conceptual change in terms of individual or social processes, and at finding general strategies to let such change take place. Part of these strategies is their emphasis on "higher order thinking skills" and metacognition (Baird and Mitchell, 1986). This reflects a strong link with present-day cognitive psychology. Many meta-level discussions are taking place examining different opinions about constructivism and related ideas about knowledge and epistemology (Matthews, 1995). In itself this may be very interesting, but it does not (yet?) lead, I think, to much progression in the practice of physics education .

In my opinion, the main importance of this paradigm lies in the fact that now learning of physics content itself has become a major variable in much physics education research. Research results are no longer, in the first place, only to be interpreted within a far-away psychological perspective that is often seen by many practitioners as something that, whilst not irrelevant, is mostly unusable. In my experience, these content-specific research outcomes seem to have a much more direct appeal to teachers, didacticians and curriculum developers, as they question precisely their level of intuitive practice-built expertise.

This is again a very rough description of research on teaching and learning. Did this research influence practice so far, and if so, in what way ? The main theories have certainly influenced the above described development of curricula. Writing about the period up to the early eighties White and Tisher (1986) nevertheless concluded as follows: "The great amount of energy that went into research did not spill over into seeing the results affected practice." Is the situation for the period since then different, or is it too early to judge? As said, the misconceptions-wave got much attention from a wide audience of didacticians. It has also had some impact on the formulation of curriculum attainment targets, in the sense that concepts have to be developed now more gradually in steps. It is my impression that much effort has also been expended in trying to get the messages across to teachers (e.g., CLIS, 1990). However, what was the message? So far, teachers often get the impression that they are not doing well enough, that they do not succeed in making pupils understand sufficiently what they teach and that they should take more account of pupils' misconceptions. At first sight, that seems to be a rather negative message, which makes it understandable that many teachers are not very eager to listen. So, how could we do better? General strategies for conceptual change do not really function for physics teachers as long as they cannot be translated into concrete practice.Furthermore, researchers do not yet have much to offer at that level (e.g, Tobin et al., 1994), a failing of which they seem to become increasingly aware (Fensham, Gunstone and White, 1994). Fortunately, I would say, because otherwise, in my opinion, research in didactics of physics would, after an encouraging period, be in danger of stagnation again.


4.1. University based approaches

In the part two above, I described some main trends in physics curriculum development as far as content and aims are concerned. In part three, I did the same for research on teaching and learning physics that has had more or less strong implications for curriculum development. In doing so, implicitly I have also touched upon some major developments in ways of curriculum development, related to problems of curriculum implementation and of use of research results in practice. In this part, I will elaborate this theme more explicitly, as it is my conviction that Bruner's expected progress has very much to do with the way in which we will be able to solve these problems in the future.

A first thing to note, however, is the difference in time scales between large scale curriculum development and "fundamental" research on teaching and learning. Curriculum projects often have to produce, within a limited time, teaching materials that can and will be used in schools. By contrast, research on teaching and learning often aims at longer term development of understanding, to be framed in applicable theory. A second remark concerns the fact that curriculum implementation is, in the first place, also very much a matter of educational politics. If, for instance, the political situation in a country is such that the government decides to implement a new curriculum for all schools from a certain set date, quantitatively the implementation will be necessarily "successful", even although in terms of quality the situation may be quite different. The other extreme of the spectrum is when the political situation is such that schools, or even individual teachers, are very much free to choose whether or not they will adopt a new curriculum. Then, as past experience has shown, curriculum implementation is quite another matter.

Most of the first curricula were developed in project teams, in which university physicists, educational specialists and physics teachers cooperated (e.g., French, 1986; Raizen, 1991). This meant a fundamental change from the usual method of textbook writing by one or two authors, not usually practising physicists themselves but experienced teachers. At least in the US, a "fundamental axiom of the program was that the improvement of curricula needed to enlist outstanding research scientists" (Raizen, 1991). Or, as Matthews (1994) writes, in the first wave, the scientists were put "firmly in the saddle of curriculum reform, teachers were at best stable-hands, and education faculty rarely got as far as the stable door. The PSSC project epitomized "top-down" curriculum development : its maxim was: "Make physics teacher-proof." This description makes clear that in general most emphasis was laid on the up-dating of scientific content, that the translation of general theories of teaching and learning into curriculum materials and classroom practice mostly resulted in considerable 'slippage' (as Fensham describes it), and that the role of teachers was restricted to"trying out" and not so much to "participating in" As Welch (1979) wrote: "Scientists were usually hesitant to accept the criticism of their "science" from school teachers unless very convincing substantiating data were provided."

Nevertheless, such top down projects developed in general beautiful and very original and innovative curriculum materials, both for students and teachers, that have had a broad and considerable influence. For example, French (1986) describes the PSSC course as being characterized "by originality and freshness of approach", and the same characteristic applies to many other curricula developed in that period.

Another main characteristic of the first wave was that it was characterised by a mainly university-based development. Central project teams of specialists developed marvellous materials, to be tried out in a limited number of schools, to be implemented top-down and on a large scale afterwards. However, probably precisely because of their innovative character and high standards, this implementation appeared not to take place as expected. Quite often adoption of curricula did not necessarily mean adoption of their spirit, or of their recommended teaching methods. Indeed , the problem appeared to be one of dealing with curriculum-proof teachers, rather than of implementing teacher-proof curricula.

Fensham notes that in the 1970s "evidence accumulated that many or most of the hopes and good intentions of the reformers were not being achieved in schools" And, according to Matthews: "Now, in the 1990s, when school science reform is once more on the agenda, it is timely to know how much of this failure and confusion was due to the curriculum materials, how much to teacher inadequacies, how much to implementation and logistic failures, how much to general anti-intellectual or anti-scientific cultural factors and how much to a residue factor of faulty learning theory and inadequate views of the scientific method that the schemes incorporated."

This is not the place to discuss all these factors extensively. The important thing that I want to stress here, is that it now seems that a centralized-expert-project-team format of curriculum development, although seeming very reasonable at the time, is bound to very much underestimate the intricacies of curriculum implementation, and in particular of the teacher's role in it. As French (1986) noted: "the crucial ingredient for the success of any educational innovation is the classroom teacher."

4.2. School-based approaches

It is therefore understandable that a quite different school-based approach to curriculum development emerged. It seems probable that this arose in part as a reaction to the problems described above, and in part because, in tune with the spirit of the seventies and eighties, teachers became much more emancipated in general and more concerned about physics, about education and about physics education. As Eggleston (1980) writes in the preface of a book about the situation in Britain: "School-based curriculum development has, in the early 1980s, become the dominant form of the curriculum development movement. After a decade in which the main effort has been focused on the national project, we have come to realise that if change in the schools is the objective, then the initiative must also come from the schools. The result has been a gradual resurgence of curriculum development that arises directly from the needs and enthusiasms of the schools, of their pupils and of their teachers".

This "bottom-up" kind of curriculum development generally results in rather different types of materials, with different aims and pretensions. These emphasise use of teaching methods that are manageable by teachers, give less emphasis to the scientific content of physics and more to its possible relevance for students, are less glossy and more down to earth, and in some sense are less innovative and original but more usable and locally adaptable.

In terms of research, this change in the model of curriculum development more or less coincided with an advocated change in educational research attitude, away from academic research focusing on the development and subsequent application of general educational theories, and towards action research that was meant in the first place to support and help teachers in the direct achievement of their goals, thus leading to exemplary practices to be taken over by others.

Both of the "idealized" models of curriculum development described above have complementary roles to play. University-based projects, be it with scientists or teachers, may develop very innovative curricula that may not be directly implementable on a large scale. Nevertheless, their influence in the long run may be considerable and indispensable. In school-based, teacher-centred ways of curriculum development, attention is often given to more direct concerns of teachers, so that the development becomes an important mechanism for getting teachers involved in the direct improvement of their own teaching situation, leading to the availability of flexible and in principle rather easily implementable curriculum materials and experiences. It has often turned out that part of this improvement lies in a locally manageble adaptation of the products of large scale more innovative projects, which means that this knife may cut both ways.

4.3. Developmental Research

In my opinion, however, another third model also needs consideration, not to replace the two models described so far, but to fulfil another essential role, for which the first two models do not provide. The need for this model has to do, in my view, with the explicit linking of research on teaching and learning to curriculum development, and in that sense with bridging the gap between educational theory and curriculum practice. In the main projects of the past, as already described, general educational theory often only had its influence somewhere in the background, or in the curriculum rhetoric. In fact, in my opinion, that is not an unlucky coincidence, but has to do with the very nature of such theory. In reality, the actual development of such curricula was much more based on the intuitive content-specific didactical knowledge, views and experiences of the developers. The same applies, in fact, to school-based curriculum development. In fact, action research often results much more in action, than in development of empirically supported didactical theory. So, both models have, in my opinion, resulted in many important differences in and improvements of educational practice, but not in a systematic research-based way of making curricular progress.

At the same time, however, the described growth of research on the learning and teaching of physics seems to promise that such progress is within reach, provided that we succeed in making research on learning and curriculum development shake hands in a joint long term approach.

This can best be done, I think, in a rather pragmatic empirical process of closely inter-connected small scale research and development, that I like to call 'developmental research' (Lijnse, 1995), in which researchers (physicists, didacticians of physics) and physics teachers closely cooperate on a basis of equality. I envisage a cyclical process of theoretical reflection, conceptual analysis, small scale curriculum development (including teacher training and test development), and classroom research into the interaction of teaching-learning processes. The final, empirically based, description and justification of these interrelated processes and activities constitutes what we may call "possible 'didactical structures' " for a particular topic under consideration. A detailed description and justification of such structures may be given in terms of learning tasks, of their interrelations, and of the actions that students and teachers are supposed and expected to perform. In fact, such descriptions can be considered as empirically tested domain specific didactical theories (Klaassen, 1995), that are based on an explicit view of physics and of physics teaching. Reflection on such theories for various topics may lead to 'higher level' didactical theories. In the long run, as the disciplinary structure of physics is not the most suitable starting point for instructional design, developmental research should also lead to empirically supported didactical structures for teaching the whole of physics. As Freudenthal (1991) argues, the term 'implementation of results' may not be an adequate description in the case of developmental research. It asks much more for a gradual and continuous process of dissemination, use, reflection and further development of ideas, in order to establish change at all levels.

This third, additional model of developmental research is not a theoretical fata morgana, but a way of both pragmatic and reflective working that, in various ways, already takes place at quite a lot of places. In fact, it means that curriculum development and didactical research are merged. The CLISP-approach (Driver and Oldham, 1987) is a well known example that comes close to what I have described. The PEEL project in Melbourne has taken a similar route, though not focusing on the teaching of particular subject matter, but on the development of metacognition. In recent European summerschools for PhD's in science education, it turned out that many activities were dealing with the teaching of X, where X stands for a particular topic (Lijnse, 1994, 1996; see also Psillos and Meheut, this volume). In the US, some physics educators (e.g. McDermott and Shaffer, 1993 ) seem to be working along similar lines.

At the same time, however, this list reveals a particular weakness of the advocated approach, i.e. the absence of models and/or examples of ways of cooperating and building on one another's concrete experiences. This requires that detailed descriptions of research and curriculum materials be made available, descriptions which will have to be much more detailed than is common in the usual research literature. Could modern facilities, like the Internet, take that role in the future? `


Let me finish by briefly summarizing the above in terms of what I think to be the main conclusion. Starting from Bruner's description dating from the late fifties, regarding the expected progress in curriculum development, I have tried to describe the main trends in physics curriculum development. Much work has been done in trying to keep our physics curricula both conceptually and educationally up to date a task that will never be finished.

At the same time, and largely resulting from the first main curriculum effort, research on physics teaching and learning has shown that the difficulty of designing understandable curricula and teaching has been strongly under-estimated in the past. This points to a second long-term task that also needs unending attention in the future.

In both tasks, different participants, physicists, physics teachers and researchers of physics education, have different, but equally important, roles to play. As I have argued, in the past, these different roles have more or less led to three different models of physics curriculum development, that in some sense are equally important although aiming at different functions. For the future, the long-awaited realisation of Bruner's predicted curricular progress will, in my opinion, very much depend on the extent to which we will succeed in steering work in these different perspectives so that they can contribute, in a co-ordinated and cooperative way, to the development of new physics curricula and new ways of teaching.


Adey, P., M. Shayer and C. Yates (1989). Thinking Science. London: Macmillan.

Adey, P. and M. Shayer (1994). Really Raising Standards. London: Routledge.

Aikenhead, G.S. (1991). Logical Reasoning in Science & Technology (Student Text and Teachers Guide) Toronto: John Wiley.

Arons, A.B.(1990). A Guide to introductory physics teaching. New York: Wiley.

Aubrecht, G. (1987). Quarks, Quasars and Quandaries. Maryland: AAPT.

Baird, J.R. & I.J. Mitchell (1986). Improving the quality of teaching and learning: an Australian case study the PEEL project. Melbourne: Monash University.

Bentley, D. & D.M. Watts (1986). Courting the positive virtues: a case for feminist science. Eur. J. Sci. Educ.,8, 121 -134.

Black, P. (1985). Could physics be re-admitted to the curriculum? Phys. Ed., 20, 267 -271.

Bliss, J. (1995). Piaget and after: the Case of Learning Science. Stud. Sci. Educ. 25, 139 -172.

Bloom, B.S. (1956). Taxonomy of Educational Objectives, Handbook I: Cognitive Domain. New York: Longman.

Bloom, B.S. (1971). Mastery Learning and its implications for Curriculum Development. In E.W.Eisner: Confronting Curriculum Reform. Boston: Little, Brown and Co, 17 49.

Brown, S.A. (1977). A review of the meanings of, and argumentation for, integrated science. Stud. Sci. Educ, 4. 31 62.

Brown, S. & D. McIntyre (1981). An Action-Research Approach to Innovation in Centralized Educational Systems. Eur. J. Sci. Educ., 3, 243 258.

Bruner, J.S. (1960). The Process of Education, New York: Random House.

Butler Songer, N. & M.C. Linn (1991) How Do Students' Views of Science Influence Knowledge Integration? J.Res.Sci.Teaching, 28, 761 784.

Carey, S. (1985). Conceptual change in Childhood, Cambridge: MIT Press.

Carey, S., R.Evans, M.Honda, E. Jay & C. Unger (1991) 'An experiment is when you try it and see if it works`: a study of grade 7 students` understanding of the construction of scientific knowledge, Int. J. Sci.Educ.

Clement, J. (1993). Using bridging analogies and anchoring intuitions to deal with students' preconceptions in physics. Journal of research in science teaching, 30, 1241-1257.

CLIS (1987). CLIS in the classroom: approaches to teaching. Leeds: CSSME.

CLIS: Interactive Teaching in Science, Workshops for Training Courses (1990). ASE.

Driver, R. (1983). The pupil as Scientist? Milton Keynes: O.U.P.

Driver, R., Guesne, E. & Tiberghien, A. (eds.) (1985). Children's ideas in Science, Milton Keynes: O.U.P.

Driver, R. and V. Oldham (1986). A constructivist Approach to Curriculum Development in Science. Studies in Science Education, 13, 105 122.

Driver, R. (1988) Changing Conceptions, Tijds. Did. ß-Wetenschappen, 6, 161 198.

Driver, R., J. Leach, P. Scott and C.Wood-Robinson (1994). Young people's understanding of science concepts: implications of cross-age studies for curriculum planning. Stud. Sci Educ., 24, 75 -100.

Duit, R., F. Goldberg and H. Niedderer (Eds.) (1992). Research in Physics Learning: Theoretical Issues and Empirical Studies. Kiel: IPN.

Eggleston, J. (ed.)(1980). School-based curriculum development in Britain. London: Routledege.

Eijkelhof, H.M.C.& J.Kortland (1988). Broadening the aims of physics education experiences in the PLON-project. In: P.J.Fensham (ed.) Development and Dilemmas in Science Education, London: Falmer Press, 282 305.

Fensham, P.J.(1988)(ed.). Development and Dilemmas in Science Education, London: Falmer Press.

Fensham, P.J. (1992). "Science and Technology". In P.W.Jackson (ed.), Handbook of Research on Curriculum. New York: Macmillan, 789-829.

Fensham, P., R. Gunstone and R. White (Eds.) (1994). The Content of Science. London: Falmer Press.

Fischler, H. (1989) Quantenphysik in der Schule I: Tendenzen der didaktischen Diskussion und Aufgaben der Fachdidaktik, Physica Didactica 16, 21 -33.

Fischler H. & Lichtfeldt M. (1992) Modern physics and students` conceptions, Int.J.Sci.Educ. 14, 181 -190.

French, A.P. (1986). Setting new directions in physics teaching: PSSC 30 years later. Physics Today, Sept, 30 35.

Freudenthal, H. (1991). Revisiting Mathematics Education. Dordrecht: Kluwer.

Gabel, D. (1994)(Ed.) Handbook of Research on Science Teaching and Learning. New York: Macmillan.

GIREP (1973). A.Loria and P.Thomson (eds) Seminar on the Teaching of Physics in Schools: Electricity , Magnetism and Quantum Physics. Copenhagen: Gyldendal.

GIREP (1986). Cosmos: an Educational Challenge. Noordwijk: ESTEC.

GIREP (1991). H.Kühnelt, M.Berndt, M.Staszel and J.Turlo (Eds.) Teaching about Reference Frames: from Copernicus to Einstein. Torun: NCUP.

GIREP (1993). L.Chainho Pereira, J.Alves Ferreira and H.A.Lopes (Eds.) Light and Information. Braga: Universidade do Minho.

Hagerty-Hazel, E. (1990) (Ed.) The Student Laboratory and the Science Curriculum. London: Routledge.

Harvard Project Physics Course (1970). New York: Holt.

Hodson, D. (1993). Re-thinking old ways: towards a more critical approach to practical work in school science. Stud. Sci. Educ., 22, 85 -142.

Johnson, S. (1984). The underachievement of girls in physics: Towards explanations. Eur.J.Sci.Educ.,6, 399-409.

Klaassen, C.W.J.M., (1995) A problem posing approach to the teaching of radioactivity. Utrecht: CD-ß Press.

Layton, D. (1992). Technology's challenge to science education. Buckingham: Open University Press.

Lawson, A.E. (1994). Research on the Acquisition of Science Knowledge: Epistemological Foundations of Cognition. In D. Gabel (Ed.) Handbook of Research on Science Teaching and Learning. New York: Macmillan, 131 177.

Lijnse, P.L.(1983). Physikunterricht in einer sich wandelnden Gesellschaft. Physica Didactica, 10, 43 60.

Lijnse, P.L., P. Licht, W. de Vos and A.J. Waarlo (Eds.) (1990a) Relating Macroscopic Phenomena to Microscopic particles. Utrecht: CD-ß Press.

Lijnse, P.L., K.Kortland, H.M.C.Eijkelhof, D.van Genderen and H.P.Hooymayers (1990b). A Thematic Physics Curriculum: a Balance Between Contradictory Curriculum Forces, Science Education, 74, 95 103.

Lijnse, P.L. (1995). 'Developmental research' as a way to an empirically based 'didactical structure' of science. Science Education, 79, 189-199.

Matthews, M. (1994). Science Teaching: the role of history and philosophy of science. London: Routledge.

McDermott, L.C. & P.S. Shaffer (1993). Research as a guide for curriculum development: An example from introductory electricity. Am J. Phys., 60, 994 1003.

Mellar, H., J.Bliss, R.Boohan, J.Ogborn and C.Tompsett (1994). Learning with Artificial Worlds: Computer Based Modelling in the Curriculum. London: Falmer Press.

Millar, R. & R. Driver (1987). Beyond Processes. Studies in Science Education, 14, 33 62.

Nuffield O-level Physics. London: Longman.

Nuffield Advanced Science: Physics (1971; revised version: 1986). Harlow: Longman.

Nussbaum, J. and S. Novick (1982) Alternative frameworks, conceptual conflict and accommodation: toward a principled teaching strategy. Instructional Science 11, 183 200.

Physical Science Study Committee (1960). Physics. Boston: Heath & Co.

Ogborn, J. (1978). Decisions in curriculum development a personal view. Phys. Educ., 13, 11 18.

PLON (1986). Curriculummaterials, Utrecht University/Zeist NIB.

Posner, G.J., Strike, K.A., Hewson, P.W., Gertzog, W.A. (1982). Accommodation of a scientific conception: Toward a theory of conceptual change. Science Education, 66, 2, 211-227.

Raizen, S.A. (1991). The Reform of Science Education in the U.S.A. Déjà Vu or De Novo? Studies in Science Education, 19, 1 -41.

Redish, E.F. (1994). Implications of cognitive studies for teaching physics. Am.J.Phys. 62, 796 803.

Reif, F. and J.H. Larkin (1991). Cognition in Scientific and Everyday Domains: Comparison and Learning Implications. Science Teach., 28, 733 760.

Reiss, M.J. (1993). Science Education for a Pluralist Society. Buckingham: OUP.

Royal Society, (1985). The Public Understanding of Science.

Satis 8 14 (1992).Hatfield: ASE.

Science, A Proces Approach (1968). AAAS.

Schools Council Project Technology (1975). London: Heinemann.

Scott, P., H. Asoko, R. Driver and J. Emberton (1994). Working from children's ideas: An Analysis of constructivist teaching in the context of a Chemistry topic. In P. Fensham, R. Gunstone and R. White (Eds.) The Content of Science. London: Falmer Press.

Shayer, M. & P. Adey (1981). Towards a science of science teaching. London: Heinemann.

Shulman, L.S. and P. Tamir (1973). Research on Teaching in the Natural Sciences. In R.M.W Travers (ED.) Second Handbook of Research on Teaching. Chicago: Rand Mcnally, 1072 1097.

Solomon, J. (1988). Science Technology and Society courses: tools for thinking about social issues. Int.J. Sci.Educ., 10, 379 -387.

Solomon, J. And G. Aikenhead (Eds.) (1994) STS Education: International perspectives on reform. New York: Teachers College Press.

Tobin, K., J. Butler Kahle and B.J. Fraser (1990). Windows into Science Classrooms. London: Falmer Press.

Tobin, K., D.J. Tippins and A.J. Gallard (1994). Research on Instructional Strategies for Teaching Science. In D. Gabel (Ed.) Handbook of Research on Science Teaching and Learning. New York: Macmillan.

Vos, W. De, B. Van Berkel and A.H. Verdonk (1994). A Coherent Conceptual Structure of the Chemistry Curriculum. J. Chem.Educ., 71, 743 746.

Wellington, J. (Ed.) Skills and processes in science education, London: Routledge.

White, R.T. (1979). Achievement, Mastery, Proficiency, Competence. Stud. Sci. Educ., 6, 1 22.

Woolnough, B. And T. Allsop (1985). Practical work in science. Cambridge: CUP.

Woolnough, B.E. (1989). Towards a holistic view of processes in science education. In J.Wellington (Ed.) Skills and processes in science education, London: Routledge.

Wright, E.L. (1993). The Irrelevancy of Science Education Research: Perception or Reality? NARST News, 35, 1 -2.


Section E1, Curriculum Development in Physics Education  from: Connecting Research in Physics Education
with Teacher Education
An I.C.P.E. Book © International Commission on Physics Education 1997,1998
All rights reserved under International and Pan-American Copyright Conventions
Return to the Table of Contents