This article was downloaded by: [Universitaets und Landesbibliothek]

On: 16 January 2014, At: 06:09

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number:
1072954 Registered office: Mortimer House, 37-41 Mortimer Street,
London W1T 3JH, UK

International Journal of
Science Education
Publication details, including instructions for
authors and subscription information:
http://www.tandfonline.com/loi/tsed20

History and philosophy of

science through models:
some challenges in the case
of 'the atom'
Rosaria Justi & John Gilbert
Published online: 20 Jul 2010.

To cite this article: Rosaria Justi & John Gilbert (2000) History and philosophy of
science through models: some challenges in the case of 'the atom', International
Journal of Science Education, 22:9, 993-1009, DOI: 10.1080/095006900416875

To link to this article: http://dx.doi.org/10.1080/095006900416875

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all
the information (the “Content”) contained in the publications on our
platform. However, Taylor & Francis, our agents, and our licensors
make no representations or warranties whatsoever as to the accuracy,
completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of
the authors, and are not the views of or endorsed by Taylor & Francis.
The accuracy of the Content should not be relied upon and should be
independently verified with primary sources of information. Taylor and
Francis shall not be liable for any losses, actions, claims, proceedings,
demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in
relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study
purposes. Any substantial or systematic reproduction, redistribution,
reselling, loan, sub-licensing, systematic supply, or distribution in any
 form to anyone is expressly forbidden. Terms & Conditions of access
and use can be found at http://www.tandfonline.com/page/terms-and-
conditions
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014
 INT. J. SCI. EDUC., 2000, VOL. 22, NO. 9, 993- 1009

History and philosophy of science through models:

some challenges in the case of ‘the atom’
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

 ria Justi, Departamento de Qu õ mica, Universidade Federal de Minas

Rosa
Gerais, Belo Horizonte, Brazil, and John Gilbert, School of Education,
The University of Reading, Reading, UK; e-mail: [email protected]

It is suggested that the contribution of history and philosophy of science (HPS) to science education can
be enhanced through a consideration of scientific models which are relevant to major sectors of the
curriculum. The possibilities for so doing are considered through the discussion of six assertions. A way
of characterizing such models, based on the work of Lakatos (1970, 1978), is outlined and applied to a
typically important sector, that of the nature of the atom. An analysis of the way that the curriculum for
14-16 year olds and typical textbooks in Brazil and the UK treat historical models of the atom is given.
The use of ‘hybrid’ models was identified in those treatments. Hybrid models, by their very nature as
composites drawn from several distinct historical models, do not allow the history and philosophy of
science to make a full contribution to science education. They do this by denying the role of distinct
models in the history of science and of the role of progression between these models in the philosophy of
science. The consequences for the teaching of science of an appropriate treatment of historical models
are outlined.

Introduction
In recent years the systematic inclusion of the history and philosophy of science
(HPS) in science education has been viewed as a way of making the latter more
closely related to the processes and conduct of science per se (Allchin 1995, Carson
1992, Hodson 1985, 1992, Matthews 1994, Monk and Osborne 1997). The inten-
tion has been that, if students’ understanding of how scientific knowledge is devel-
oped, and of how historical, philosophical and technological contexts influence its
development, then they will acquire a more comprehensive view of science and, as
a consequence, become more engaged by the learning of science.
How might the inclusion of HPS in the science curriculum be brought
about? This paper proposes that the field of ‘models and modelling’ can provide
one suitable basis. Some possibilities for this inclusion are discussed with the
aid of six assertions i.e. that: models are a suitable basis for HPS in science
education; historical models can be characterized; a finite number of models of
the atom exist; school curricula do not refer clearly to historical models; textbooks
do not make appropriate use of historical models; and hybrid models are used in
teaching.

International Journal of Science Education ISSN 0950- 0693 print/ISSN 1464-5289 online # 2000 Taylor & Francis Ltd
http://www.tandf.co.uk/journals
 994 R. JUSTI AND J. GILBERT

Assertion 1: models are a suitable basis for HPS in science

education
We suggest that ‘models and modelling’ is a suitable topic on which to base such
initiatives. The nature of the roles played by modelling and models in science is
much-debated (see, for example, Giere 1990). They do seem to play an intermedi-
ary role between scientific theory and the world-as-experienced, although the
precise nature of that role depends on how one interprets the meaning of ‘theory’,
‘model’ and ‘the world-as-experienced’ (Gilbert et al. in press). Moreover, models
are both important products of science in themselves and a means by which
research outcomes are disseminated (Giere 1990, Gilbert and Boulter 1995,
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

McMullin 1968, Nersessian 1992, Norman 1983).

The topic of ‘models and modelling’ needs a robust terminology. A model can
be taken to be a representation of an idea, object, event, process, or system (Gilbert
and Boulter 1995). The assumption that, as a result of individual enterprise,
whether alone or within a group, a mental model is produced is at the core of
the cognitive science perspective on learning (Johnson-Laird 1983, Gentner and
Stevens 1983). Mental models cannot be directly accessed, only inferred from the
major modes of human communication: gesture, speech, and writing. When men-
tal models are placed into the public domain through any mode of representation
they become expressed models (Gilbert and Boulter 1995). Those expressed models
which gain social acceptance following testing by the community of scientists
became scientific models. Those scientific models at the forefront of research at
any one time may be termed consensus models. A scientific model produced in a
specific context, but which has been superseded at the cutting edge of science, may
be called an historical model (Justi 1997). Within this latter definition, a context is
taken to mean a given system of philosophical, scientific, technological, and social
beliefs. A historical model is not therefore necessarily a model developed by an
individual or a small group of scientists, although this is apparently often the case,
nor is its development and use situated within a specific time period. The key issue
is that it achieved consensus status within a particular context. A curricular model
is a simplified version of any consensus or historical model, which is included in
the science curriculum at any level of the educational system.
The study of science at university level can hope to come close to the frontiers
of knowledge, where discoveries are being made, on many occasions. This will
make extensive use of consensus models. However, because school education is
commonly seen to provide a ‘foundation’ for science education, the emphasis there
will be on the past and consequently on historical models.

Assertion 2: historical models can be characterized

The second step in developing a models-based approach to HPS is to identify a
way of characterizing the historical models in any field. A framework developed
from Lakatos’ (1970) notion of ‘scientific research programme’ has proved both
valid and practicable (Justi 1997, Justi and Gilbert 1999a, b). In Lakatos’ view,
three elements constitute a scientific research programme: the ‘hard core’, which
comprises the major assumptions that guide all who work within a given research
programme; the ‘protective belt’, which comprises the auxiliary hypotheses that
protect the hard core from refutations; and the ‘positive heuristic’, that is a set of
 HISTORY AND PHILOSOPHY OF SCIENCE THROUGH MODELS 995

suggestions about how to modify the protective belt (Lakatos 1970). According to
Lakatos, when a hard core of a given research programme has to be changed, it is
overthrown. This is a result of a competition between the ‘progressive problem-
shifts’ of one research programme (in which novel empirical facts are both pre-
dicted and discovered) and the ‘degenerating problemshifts’ of another (in which
successful predictions can no longer be made) (Lakatos 1970).
This approach can be used to characterise the major historical models in
topics, which are important for science education (Justi 1997, Justi and Gilbert
1999a, b). As an exemplar, this is done below in outline for ‘the atom’. At each
stage in the historical development of understanding of the nature of ‘the atom’ we
have identified:
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

(1) The deficiencies in the explanatory capability of a given model;

(2) The features of that given model that were modified and incorporated
into the new model;
(3) The way by which the new model overcame the explanatory deficiencies
of its antecedents;
(4) The subsequently emergent explanatory deficiencies of the new model.
(Justi 1997)

Assertion 3: a finite number of models of the atom exist

We have identified six models of the atom, which are relevant to the compulsory
science curriculum in many countries. These are sketched below because their
natures are relevant to the discussion later in this paper.

Ancient Greek model

Early ideas about the existence of atoms are usually attributed to the philosophers
of Ancient Greece. The hard core of this model contains two ideas: matter is
composed of very small indivisible corpuscles (probably first called ‘atoms’ by
Leukippos); atoms are infinitely hard and differ in form, order, position, and
size. The assumptions of the positive heuristic of this model include those related
to atomic motion (they did not move) and to the ‘forces’ that governs the behaviour
of atoms (those of ‘love’ and ‘hate’).
The influence of the Greek atomic model was persistent. During the seven-
teenth and eighteenth centuries, many scientists e.g. Bacon, Descartes, Gassendi,
Boyle, and Newton discussed the constitution of matter. With the exception of
Descartes, who thought matter was continuous and infinitely divisible (Partington
1939), all of them accepted one or both of the core ideas of the Ancient Greek
model.

Dalton’s model
One major problem with the Ancient Greek model was that it provided no basis for
distinguishing between types of atom i.e. between the elements. Dalton’s model
addressed this directly, within the core notion of ‘atom’ as a indivisible entity
carried forward from the Greek model. The hard core of Dalton’s model was that:
 996 R. JUSTI AND J. GILBERT

The undecomposed chemical elements were specific elementary atoms of differing

and distinct physical weights.
(Nye 1976)
Dalton did not refer to forces between atoms, which were thus seen as being
stationary. However, his model did make both the calculation of masses and the
establishment of combinatorial relations between different types of atom (ele-
ments) possible. This meant that chemistry became a quantitative science,
which was a radical departure from the previous perspective. In order to do
such calculations, Dalton had to develop the protective belt of his model: some
assumptions about how atoms combined to form ‘compound atoms’. Despite this
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

success, Dalton’s model was only slowly accepted, being viewed as ‘audacious’
because of the degree of novelty that it contained relative to the centuries-old
Ancient Greek model (Calvet et al. 1992).
From the second half of the nineteenth century onwards, there was a great deal
experimental evidence amassed, for example from studies of specific heats, which
could not be explained by Dalton’s model.

Thomson’s ‘embedded mass’ model

In his exploration of the nature of cathode rays, J. J. Thomson was able to show
that they consisted of negatively charged particles of tiny mass - which he called
‘electrons’ - for which he calculated a charge-to-mass ratio. The discovery of
electrons required a new model of the atom. Although electrons had a negative
charge, atoms overall have zero charge. This implied that each atom must contain
an equal number of positive and negative charges. Therefore, Thomson proposed
that ‘the atom’ constituted of electrons embedded in a sphere of uniform positive
charge. Whilst building on the notion that the atoms of different elements were
unique, Thomson’s model superseded Dalton’s model in that it sparked a discus-
sion over the internal structure of atoms.
However, other experimental evidence soon showed its explanatory deficien-
cies. Geiger and Marsden’s investigation of the scattering of alpha particles when
they were fired at very thin sheets of metals produced an unexpected result. Many
of the alpha particles were deflected by a wide angle, some by more than 908. If
atoms were as predicted by Thomson’s model, all the alpha particles (which are
positively charged) would easily pass through the foil, with only an occasional
slight deflection in their paths. Their failure to do so lead to Rutherford’s model.

Rutherford’s ‘nuclear’ model

The hard core of Rutherford’s model consists of two ideas. First, the mass of an
atom is concentrated into a very small, positively charged central kernal (which he
called ‘the nucleus’), which would be around 10 000 times smaller than the atom
itself. Second, the nucleus is surrounded by a system of electrons kept together by
attractive forces from the nucleus. This model both elaborated the internal struc-
ture of the atom and explained Geiger and Marsden’s data. Whilst some alpha
particles would be deflected sharply by the positively charged nucleus, the major-
ity of them would pass through the extra-nuclear space (where the electrons were)
with very little or no deflection.
 HISTORY AND PHILOSOPHY OF SCIENCE THROUGH MODELS 997

However, the electronic structure of his model was described by the use of
classical mechanics. It soon became clear that classical mechanics fails when it is
applied to electrons in atoms. If electrons were stationary, they must be drawn into
the nucleus due to the attractive force between positively and negatively charged
entities. On the other hand, if they were in circular motion, calculations from
classical mechanics predicted that they would lose energy and fall into the nucleus.

Bohr’s ‘orbit’ model

Being convinced by his previous studies that classical mechanics could not explain
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

atomic structures, Bohr applied the idea of quantization previously developed by

Planck to the study of atomic spectra (Heilbron and Kuhn 1969). He was able to
propose a new research programme whose hard core was identified by Lakatos as:
1. Energy radiation (within the atom) is not emitted (or absorbed) in the continuous
way assumed in the ordinary electrodynamics, but only during the passing of the
systems between different ‘stationary’ states.
2. The dynamical equilibrium of the systems in the stationary states is governed by
the ordinary laws of mechanics, while these laws do not hold for the passing of the
system between the different states.
3. The radiation emitted during the transition of a system between the frequency ¸
and the total amount of energy emitted E is given by E ˆ h¸, where h is Planck’s
constant.
4. The different stationary states of a simple system consisting of an electron rotating
round a positive nucleus are determined by the condition that the ratio between the
total energy, emitted during the formation of the configuration, and the frequency of
revolution of the electron, is an entire multiple of 1/2 h. Assuming that the orbit of the
electron is circular, this assumption is equivalent with the assumption that the angular
momentum of the electron round the nucleus is equal to an entire multiple of h/2p.
5. The ‘permanent’ state of any atomic system, i.e. the state in which the energy
emitted is maximum, is determined by the condition that the angular momentum
of every electron round the centre of its orbit is equal to h/2p
(1970: 141)
This represented a major refinement of perspective on the internal structure of
electrons within atoms and enabled spectroscopic data to be both co-ordinated and
explained.

Quantum mechanics model

In the later pages of his 1970 paper, Lakatos presents a comprehensible discussion
of Bohr’s model, pointing out the anomalies and inconsistencies which gradually
led to its abandonment in scientific research, in particular its to explain the spec-
trum of the alkali metals or of molecules generally (Lakatos 1970).
During this process, the evolution of quantum mechanics, particularly of
SchroÈ ndinger’s equation (concerned with the dual wave and particle nature of
light) and Heisenberg’s uncertainty principle (concerned with the position and
movement of an electron), lead to the development of the quantum mechanical
atomic model. In this model, every electron in an atom is distinguishable. To use a
visualization, the position of a given electron can be hypothetically ‘marked’ at
regular time intervals, giving a pattern of probabilities for its position. An arbitrary
 998 R. JUSTI AND J. GILBERT

surface enclosing say 99% of the probability can be drawn: the electron is almost
certainly within this ‘orbital’ envelope at any given moment. This approach has
proved very useful in explaining the more complex occurrences of bonding.

Assertion 4: school curricula do not refer clearly to historical

models
The next step in considering how HPS may be introduced into the teaching of the
atom is to explore what type and degree of treatment are currently expected in
school curricula. As an opportunity sample, we took curricula for 14- 16 years olds
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

in Brazil and in the UK

In Brazil, science is part of the curriculum for all students both between the
ages of seven and 14 years (the ‘fundamental’ level) and between the ages of 15 and
17 years (the ‘medium’ level). Teaching about the atom occurs mainly in the eighth
year of the fundamental level and in the first year of the medium level, in which
‘disciplinary’ courses i.e. physics, chemistry, biology are included. There is no
national curriculum in Brazil. The subject matter of the courses at the medium
level is mainly defined by the requirements of universities’ entrance examinations.
Moreover, there is no homogeneity between the entrance examinations of univer-
sities in different states. In the state of Minas Gerais, for example, the list of
subjects for the main university’s entrance exams includes:
Atomic structure of matter - constitution of atoms.
The importance of this subject stems from its application to other subjects. It is
advisable to emphasise both the importance and limitation of atomic models as well
as their evolution. There will not be questions about atomic spectra, quantum num-
bers, the orbital model, and the distribution of electrons in sublevels of energy. The
idea of quantisation of energy should be dealt with through the discussion of the
diagrams concerning with energy of ionisation of atoms for which the atomic number
varies from 1 to 18.
1. Atomic model of Dalton: description and applications.
2. The existence of electrons. Mass and charge of an electron.
3. Atomic model of Rutherford and atomic nucleus. Masses and charges of protons
and neutrons.
4. Atomic numbers and mass numbers: concepts and related calculations.
5. Energies of ionisation as an empirical evidence for the levels of energy. Atomic
model of Bohr: qualitative aspects. Distribution of electrons in levels of energy.
(UFMG 1997: 27)
In the UK, all students have to study science between the ages of five and 16 years.
There is a National Curriculum for Science, compulsory for all students in state
(as opposed to private) schools. The section for 14- 16 year olds decrees that
students should be taught:
Atomic structure:
a. That solids, liquids and gases are composed of particles;
b. That atoms consist of nuclei and electrons;
c. The charges and relative masses of protons, neutrons and electrons
d. About mass number, atomic number and isotopes;
e. About a model of the way electrons are arranged in atoms;
 HISTORY AND PHILOSOPHY OF SCIENCE THROUGH MODELS 999

f. That the reactions of atoms depend on the arrangement of electrons in atoms.

(DFEE 1995: 31)
Note that the specific model of the atom to be taught is not specified.

Assertion 5: textbooks do not make appropriate use of

historical models
The next step is to see how HPS is treated in typical textbooks written to relate to
the curricular requirements summarized in the paragraph above. Nine textbooks
from Brazil and three from the UK were analysed by applying the descriptions of
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

the historical models of the atom given above. The textbooks are identified in the
Appendix and are referred to below by the numbers allocated to them there. The
analysis produced data, which can be presented in three categories.

The notion of ‘model’ adopted

Only one of the three UK textbooks says anything about the nature of models, and
even then it is particularly unhelpful, i.e:
Models are an essential part of science. Models help us to think about the way things
work
(t. 12 : 84).
Four of the nine Brazilian textbooks analysed assert something about models.
However, only two of them define what a ‘model’ means. The author of one
textbook makes the reasonable statement:
A model is always a representation of the reality, not the reality itself, that is, it is not a
copy of a given system
(t. 5: 7, t. 6: 1).
This textbook also states that the aim of a model is to explain a given phenomenon
and that a model is provisional: it can be changed or substituted by another one
when new empirical observations cannot be explained by the original model.
These matters are also discussed in textbook 9, but it provides a weak definition
of ‘a model’ in stating that:
When a theory or set of ideas is utilised to explain or make predictions about the
structure and behaviour of a given system through comparisons, it is called a model
(t. 9: 134).
And that:
a scientific model is nothing more than an image of a system, developed from experi-
ments.
(t. 9: 143).
Of greater concern is that, in one of the UK textbooks, ‘model’, ‘theory’, ‘predic-
tion’, ‘experiment’ are conflated, rather than being distinguished between by defi-
nition:
A good model or theory is one which gives predictions which turn out to be correct.
The predictions are tested in the laboratory. If they do not pass the test, the theory or
model must be altered or thrown out.
(t. 10: viii)
 1000 R. JUSTI AND J. GILBERT

The relationship between model and experiments is perhaps over-emphasized in

the sole phrase that is written about models in textbook 3:
A good model should help to understand a number of properties of the system under
study; because of this a model has an experimental basis.
(t. 3: 91).
Moreover, the lack of distinction between ‘model’ and ‘theory’ - clearly expressed
in the definition given in textbook 9 - is also found in textbooks 5 and 7. The first
of these, after discussing some aspects of models in the early pages, uses the
expression ‘models (theories)’ (t. 5: 81), indicating that such words may be viewed
as being synonymous with each other. Textbook 7 uses the words ‘model’ and
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

‘theory’ in an alternating way, but without defining either of them.

A failure to discuss the relationship between ‘theory’ and ‘model’ was found in
the UK textbooks, probably as a consequence of the National Curriculum require-
ment that students should have opportunities to:
a. Develop their understanding of how scientific ideas are accepted and rejected on
the basis of empirical evidence, and how scientific controversies can arise from dif-
ferent ways of interpreting such evidence;
b. Consider ways in which scientific ideas may be affected by the social and historical
contexts in which they occur, and how these contexts may affect whether or not the
ideas are accepted.
(DFE 1995: 24)
Where no discussion of either ‘theory’ or ‘model’ is given. As would be expected in
these circumstances, the index to textbook 11 does not contain either word.

The basis for recognizing a ‘model’ of the atom

There was evidence in the Brazilian textbooks that only models that had an experi-
mental basis can be afforded the status of ‘a model’. The ideas of Ancient Greek
philosophers are dismissed by saying that they ‘only had philosophical thoughts as
their basis’ (t. 1: 14). At least one textbook takes a naive empiricist view:
The idea that matter consists of tiny particles - was first put forward by the Greek
thinker Democritos in 50BC. For centuries the theory met with little success. People
were not prepared to believe in particles which they could not see
(t. 10: 40)
Other textbooks (5, 6, and 9) are more incisive in their discrimination, saying that
Ancient Greek ideas cannot be classified as a scientific model simply because they
were not based on experiment. In like vein, textbook 12 states that:
The ancient Greeks first thought of the idea of atoms nearly 2500 years ago. They had
no experimental proof for atoms, but they liked the idea because it explained many
things about the world.
(t. 12: 69)
Such ideas suggest that the authors feel it appropriate to convey a simplistic view
of the development of scientific knowledge to students of this age.
The achievement of consensus status by each model of the atom is not dis-
cussed. Sometimes an assumption of the consensus status of a given model is
made, but this is done by the use of an incoherent set of criteria. For instance,
in textbook 4 it is said that:
 HISTORY AND PHILOSOPHY OF SCIENCE THROUGH MODELS 1001

As the Greeks did not have ways to check their ideas, they were not universally
accepted
(t. 4: 60).
The author does not specify what he means by ‘ways of check their ideas’, but from
the overall tone of the textbook this can be taken to mean ‘the conduct of experi-
ments that could be explained with the use of a given model’. However, if our
inference is correct, there would be an inconsistency in the use of such a criterion.
This is because, although the laws of proportionality, for instance, were explained
by the atomic model in the nineteenth century, atomism was not universally
accepted at that time (see above). This point is ignored in all the textbooks.
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

Aspects of behaviour that are explained by a given model are presented as if

they were the basis of that model. Textbook 4, for instance, describes the behav-
iour of electrons absorbing and liberating energy whilst moving between distinct
orbits as evidence that lead to the production of the Bohr model. Such behaviour
was the explanation of atomic spectra given by assuming Bohr’s model, as textbook
12 states:
Neils Bohr suggested that electrons move around the nucleus in orbits, rather like
planets around the Sun. - The electron orbits are called shells. - Bohr measured the
frequencies of the lines (in the atomic spectrum of hydrogen) and used them to work
out the energies of the electron shells.
(t. 12: 258- 259)
Even this is only partially correct, for the Lyman, Balmer, Paschen, Pfund series of
lines in the spectrum of hydrogen had been observed and measured prior to Bohr’s
model: the latter enables the former to be coordinated and justified.
The distinction between ‘basis for the production of a model’ and ‘explana-
tions given by a model’ is very important if we want students to understand how
models of a given phenomenon are changed.

Change of models of the atom

The theoretical background on which each model is based is generally not clearly
discussed in the textbooks analysed. They all give names and dates and sometimes
describe experiments that resulted in discoveries that lead to the development of a
given model. However, on many occasions this is disconnected information since
the experiments are not related to the attributes of the previous model in the
historical sequence. For instance, textbooks 1, 2, 7 and 9 do not discuss the out-
comes of what they call ‘Rutherford’s Experiment’ (which should, of course, be
referred to as ‘Geiger and Marsden’s Experiment’) in the light of the Thomson
model. None of the textbooks presents a comprehensive discussion of the theor-
etical backgrounds of all the historical models or of the competition between the
research programmes associated with successive models. The only clear compar-
ison between the historical models is given in textbook 3, but even here the focus is
on the ‘empirical observation used in the adoption of the models’, not on the
differences between theoretical backgrounds and main attributes of successive
models.
Some textbooks assume that the introduction of a different positive heuristic
signals the introduction of a new model, rather than being either an improvement
to a given model or a step towards the refutation of that model. For instance, the
 1002 R. JUSTI AND J. GILBERT

proposal by Sommerfeld of the elliptical shape of electron orbits was the result of a
more detailed study of atomic spectra. This does not characterize a new model
since the hard core that based his studies was the one proposed by Bohr.
Nevertheless, among the five textbooks that make reference to Sommerfeld’s
work four (t. 2, 6, 8, and 9) say that he produced a new model. Even when the
textbook author recognizes that the two sets of ideas are very close to each other, it
is assumed that Bohr’s and Sommerfeld’s ideas reflect substantially different
models. This is the case of textbook 6, in which we read that:
Sommerfeld did not completely disregard the Bohr’s model. He proposed that the
circular orbit would be a special case of the elliptical one. Thus you see a model which
has changed from the previous one.
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

(t. 6: 21).

Assertion 6: hybrid models are used in teaching

In previous work we have showed that the model expressed by both a teacher and
by the textbook in use often do not correspond to any one of the historical models
previously identified in the field being studied (Justi 1997, Justi and Gilbert
1999a, b). This finding was confirmed in the present study. A hybrid model,
that which is commonly presented to students, is not a curricular model, that is
a simplified version of one distinct historical model. It is instead something in
which the characteristics of several distinct historical models have been merged.
The presence of hybrid models was identified in the texts examined in a range of
ways:

(a) There is no discussion when the attributes of a given model are

retained in the one that succeeds it
None of the textbooks recognize, for instance, the influence that Ancient Greek
ideas had even many centuries later: for example, that Dalton’s model retained
(rather than introduced) the notion of the atom as a solid entity. In textbook 3,
when models are explicitly compared, in a section entitled ‘main ideas about mod-
els’, only the new attributes of each model are presented. The existence of the
nucleus, for instance, is presented as an attribute of Rutherford’s model, but not of
Bohr’s model and of the Quantum Mechanics model, even though it is retained in
these latter. This would seem likely to contribute to students’ lack of a coherent
view about the processes of development of successive models. A fuller treatment,
which discusses all aspects, is needed in each case.

(b) An attribute is correctly associated with a given model when it is

presented in written form, but a (often physically close) pictorial
representation of that same model shows attributes of another model or a
mix of attributes of two distinct models.
We see this in the discussion of the Quantum Mechanics model. Textbook 8, after
saying that De Broglie proposed that electrons behave simultaneously as a particle
and as a wave, shows a pictorial representation of an atom in which the orbit has
 tron’ is
the shape generally used to represent a wave. See figure 1 (translations: ‘ele
Â
‘electron’ and ‘nu cleo’ is ‘nucleus’).
 HISTORY AND PHILOSOPHY OF SCIENCE THROUGH MODELS 1003
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

Figure 1.

Figure 2.

The same textbook asserts that the idea of an orbital is a basic characteristic of the
Quantum Mechanics model. It represents an atom in two ways. First, as a nucleus
surrounded by a circle called ‘an orbit’ in which there is a small ball (presumably
the electron). Second, as a nucleus surrounded by a region containing a lot of
points called ‘an orbital’ (presumably electron probabilities). See figure 2 (transla-
 rbita’ is ‘orbit’, whilst ‘orbital’ is ‘orbital’)
tion: ‘o
By so doing, the difference between ‘orbit’ and ‘orbital’ is visually shown to
students. However, after this, the textbook states that the identity of the electron is
given by its quantum numbers. It then gives a visual representation in which a
nucleus has semicircles cut into one of its sides. In figure 3, each semicircle or shell
(camadas) is identified by a letter (K, L, M,. . .) and by its main quantum number
(nu mero quaà ntico principal) (1, 2, 3, . . .) respectively.
 1004 R. JUSTI AND J. GILBERT
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

Figure 3.

In our view, the semicircles strengthen the idea of an orbit as a path in which
electrons move. This is exactly contrary to the distinction made two pages earlier
(t. 8 : 91). Such ambiguities result from an attempt to create a pictorial representa-
tion of a model that was essentially the result of a mathematical-statistical treat-
ment. This seems to be evidence that the author does not understand what the
phrase ‘quantum mechanics as a theoretical background of a model’ means.
Textbook 12 introduces the confusion, perhaps as the result of an attempt to
simplify matters:
The electrons move around the nucleus. They move at random and in a chaotic way,
but to make the picture simpler we’ve shown them as if they travel in ‘orbits’ around
the nucleus.
(t. 12: 248)
Authors of other textbooks are more coherent whilst discussing the quantum
mechanics model by asserting that:
This model is recognised as more satisfactory although it is a little bit abstract.
(t. 3: 102)
or by explaining that:
The movement of electrons around the nucleus was described by SchroÈ dinger
through a mathematical equation that relates the corpuscular nature (particle), the
energy, the charge, and the mass of an electron. The numerical solutions of such an
equation, called quantum numbers, make the characterisation of each electron by its
energy possible.
(t. 9: 173).
 HISTORY AND PHILOSOPHY OF SCIENCE THROUGH MODELS 1005

(c) Attributes of a given model are presented as if they were part of

another model.
This was found in three examples: i) Textbook 4 asserts that, according to
Thomson:
Matter would be composed of atoms made up of an equal number of (two) main
particles: protons and electrons. As the mass of electrons is very small in comparison
with the mass of protons, the mass of an atom is that of the protons. The electrons,
homogeneously distributes among the protons, would keep the electric equilibrium.
(t. 4: 91).
When Thomson proposed his model, protons had not yet been characterized as a
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

class of particles. Therefore, the relationship between the masses of protons and
electrons was unknown at that time. ii) Textbooks 4, 7, 8, and 9 show pictorial
representations for Rutherford’s model in which there are circular, defined, orbits.
The movement of electrons in such orbits is part of the hard core of the later Bohr
model. iii) Textbook 1 associates attributes of the Quantum Mechanics model (the
existence of sublevels called s, p, d, and f, and the maximum number of electrons
in each of them) with the Bohr model. These ideas are introduced after the pre-
sentation of Bohr’s model in the text, but without saying that they constitute
another atomic model. These ideas are introduced to show how to distribute
electrons in sublevels. It seems that the author of textbook 1 assumes that ‘atomic
models’ and the ‘distribution of electrons’ are completely distinct subjects, that he
does not realise that they are closely related, and that the ‘distribution of electrons’
is an attribute of several atomic models. As the author states:
In each level of energy, electrons are distributed in sublevels of energy, represented
by letters s, p, d, and f in an ascending order of energy.
(t. 1: 31)
This suggests to students that the existence of sublevels is an attribute of the model
in which energy levels were defined (Bohr’s).

Discussion
Students, when taught in the conventional way, where little if any attention is paid
to the history and philosophy of science, inevitably develop a poor understanding
of the processes of change in science and of the epistemology of science. There is
ample evidence for this outcome, whatever its cause e.g. Lederman (1992). For
example, Blanco and Niaz (1998) investigated students’ understanding of HPS in
the context of the ‘atom’. Their methodology involved the application of a ques-
tionnaire about a month after the teaching of atomic structure. Students made
explicit what they knew about important experiments from the historical devel-
opment of the structure of the atom, as well as how they interpreted some of the
findings of such experiments. It was found that students think that: there is an
absolute and unchanging conception of the atom; the production of experimental
data, and not the hypotheses on which they are based or the interpretations derived
from such data, are the most important part of a scientist’s work, and; if a given
model is changed by another scientist, the scientist who had proposed that model
had made experimental mistakes.
 1006 R. JUSTI AND J. GILBERT

Given a choice, students prefer models of the atom which are easily visualiz-
able. For example, Harrison and Treagust (1996) elucidated students’ models of
atoms and molecules in order to discuss their influence on further instruction in
chemistry. The study used semi-structured interviews. In the initial question,
students were asked to talk about their ideas of atoms and to draw their models.
Next, they were presented with a set of six diagrams of atomic models often used in
teaching the atom. They were asked to choose which diagram had a best fit with
their model. Although the understanding of each of the presented atomic models
were not probed in detail, some of the outcomes of this study are relevant here. For
instance, it was demonstrated that the majority of the students ‘prefer models of
atoms and molecules that depict these entities as discrete, concrete structures’
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

(Harrison and Treagust 1996: 532). The orbits model was preferred by 46% of
the students. 55% of the students thought atoms are balls or spheres, whilst 32%
drew simple diagrams containing a nucleus and electrons. Around 25% of the
students believed that atoms are protected by a substantial outer ‘shell’. As the
authors themselves emphasized:
In this situation students have shown that, in the main, they see atoms as concrete
particles with discrete parts.
(Harrison and Treagust 1996: 524)
Such outcomes could have come about as a result of (1) a gap of meaning between
teachers’ and textbooks’ discussions of different pictorial representations of the
atom, and (2) the lack of connection made between the presentation of a given
model in its written and pictorial representations - which is typical when hybrid
models are invoked.
The students may not have realized that their mental model was a hybrid of
several historical models. For example Harrison and Treagust also state that:
When the ‘‘clouds in the sky’’ ideas was pursued, several students added that the
electron cloud was like a cloud in the sky and the electrons were like the droplets of
water in cloud. Further discussion indicated that these students saw the cloud as a
separate entity, containing the electrons.
(Harrison and Treagust 1996: 525)
The students’ view is based on the hard core of Thomson’s model. However, the
electron cloud is presented as an attempt to create a pictorial representation of the
Quantum Mechanics model. If students were mixing such different representa-
tions and reducing a very abstract model to a more simple and concrete one, this
may be a result of the use of hybrid models in teaching.
Such views may be a consequence of an inadequate understanding of both the
historical development of models of the atom and of the epistemological status of
models in the development of scientific knowledge by teachers and authors of
textbooks. Both of them are aspects that contribute to the existence of hybrid
models in teaching. When hybrid models are used, the gaps of validity between
attributes of a given model cannot readily be addressed, no questions requiring
different ways of thinking about a phenomenon can be raised, and no different
approaches to the interpretation of a phenomenon are possible.
These problems may be compounded by students’ lack of comprehension of
what a model is. In the most cited study on this topic, Grosslight et al. (1991)
interviewed school students and ‘experts’ about their notions of modelling and
models. They identified three ‘levels’ of understanding: a naÈõ ve realist view in
 HISTORY AND PHILOSOPHY OF SCIENCE THROUGH MODELS 1007

which a model is a copy of some aspect of reality; the meanings currently accepted
by science, that see modelling as an imaginative attempt to represent some aspect
of a phenomenon so that predictions can be made and tested; and a rather unclear
intermediary level which has some elements of both the other two.

A way forward
The textbooks analysed do not reflect a reasonable interpretation of the nature of
models and the role of models and modelling in science and in science education. It
does seem essential that curriculum designers, teachers, and the authors of text-
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

books, should have a general overview of the historical development of specific

scientific themes. They will then not only be able to present isolated historical facts
but also to teach the content and processes of science. This would avoid comments
such as was found in one of the analysed textbooks:
Historical aspects, that are important for the consolidation and amplification of con-
cepts but that are unnecessary in an initial approach, are presented in appendices.
(t. 4: 3, our emphasis)

Several specific ways of introducing HPS into science education have been pro-
posed. Kauffman (1989), for instance, quotes reports of the use of a unit about the
history of a given subject after the completion of conventional teaching. Using this
approach, one could teach the development of the historical models after the
consensus model had been explored. Other researchers propose the introduction
of HPS by integrated it more with contemporary scientific content itself. Some
examples can be found in Ellis (1989) and Kipnis (1995), who propose the repro-
duction of several key historical experiments in classrooms: this could be done for
‘the atom’. Arons (1988) proposes the discussion of the historical contexts within
which the development of a specific theme took place. These contexts are readily
available, for much of the history of the atom, through websites such as http://
maple.lemoyne.edu/~giunta. Solomon (1989) advocates the dramatization of
historical debates: the discussions in Rutherford’s laboratory must have been
fascinating. It is not too much to hope that an entire course could be based on
the models and their development: Project Physics Course did this for the history
of physics in general (Brush 1978).
We believe that a comprehensive understanding of historical models of key
scientific themes is essential for the elimination of hybrid models. The intro-
duction of comprehensive discussions concerning historical models into science
teacher education courses does seem called for. By understanding not only isolated
episodes in the history of science, but also aspects concerning the development of
scientific knowledge and the characterization of distinct models, teachers will be
able to introduce HPS into science education in a more authentic and coherent
way. By understanding how the ‘research programme’ associated with a given
model was succeeded by that of the next, teachers will be able to introduce the
dynamism of science into science teaching. Science teaching should improve as a
consequence because students would be given a historically defensible view of the
achievements of science - a view in which the understanding of the ‘movement’ of
distinct models play an important part. An improved understanding of the nature
of science would seem to be an inevitable corollary.
 1008 R. JUSTI AND J. GILBERT

References
ALLCHIN, D. (1995) How NOT to teach history in science. In F. Finley, D. Allchin,
D. Rnees, and S. Fifield (eds), Proceedings of the Third International History,
Philosophy, and Science Teaching Conference, Vol. 1 (Minneapolis: University of
Minnesota), 13-22.
ARONS, A. B. (1988) Historical and philosophical perspectives attainable in introductory
physics courses. Educational Philosophy and Theory, 20(2), 13-23.
BLANCO, R. and NIAZ, M. (1998) Baroque tower on a gothic base: a Lakatosian reconstruc-
tion of students’ and teachers’ understanding of structure of the atom. Science and
Education, 7(4), 327- 60.
BRUSH , S. G. (1978) Why chemistry needs history - and how it can get some. Journal of
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

College Science Teaching, 7, 288-91.

CALVET, M., ANDREU, I. B. S., RIVERA, L., VERDAGUER , I. and IZQUIERDO, M. (1992)
Sequence of models of ‘‘gas’’ and ‘‘atom’’ adequate for producing explanations:
three examples inspired on the history of chemistry. In S. Hills (ed.), The History
and Philosophy of Science in Science Education, Vol. 1 (Kingston, Ontario: Queen’s
University), 135- 47.
CARSON, R. N. (1992) If science is presented as historical narrative: a sample chapter. In
S. Hills (ed.), The History and Philosophy of Science in Science Education, Vol. 1
(Kingston, Ontario: Queen’s University), 149-55.
DFE (DEPARTMENT FOR EDUCATION) (1995) Science in the National Curriculum (London:
DFE and The Welsh Office).
ELLIS, P. (1989) Practical chemistry in a historical context In M. Shortland and A. Warwick
(eds), Teaching the History of Science (Oxford: Basil Blackwell), 156-67.
GENTNER, D. and STEVENS, A. (1983) Mental Models (Hillsdale, N. J.: Erlbaum)
GIERE, R. N. (1990) Explaining Science (Chicago: University of Chicago Press).
GILBERT, J. K. and BOULTER, C. J. (1995) Stretching models too far. Paper presented at
the annual meeting of the American Educational Research Association (San
Francisco).
GILBERT, J. K., PIETROCOLA, M., ZYLBERSZTAJN, A. and FRANCO, C. (in press). Science and
education: notions of reality, theory, and model. In: J. K. Gilbert and C. Boulter (eds),
Developing Models in Science Education (Dordrecht: Kluwer).
GROSSLIGHT, L., UNGER, C., JAY, E. and SMITH, C. (1991) Understanding models and their
use in science: conceptions of middle and high school teachers and experts. Journal of
Research in Science Teaching, 28(9), 799- 822.
HARRISON, A. G. and TREAGUST, D. F. (1996) Secondary students’ mental models of atoms
and molecules: implications for teaching chemistry. Science Education, 80(5), 509- 34.
HEILBRON, J. L. and KUHN, T. S. (1969) The genesis of the Bohr atom. Historical Studies in
the Physical and Biological Sciences, 1, 211-90.
HODSON, D. (1985) Philosophy of Science, science and science education. Studies in Science
Education, 12, 25-57.
HODSON, D. (1992) In search of a meaningful relationship: an exploration of some issues
relating to integration in science and science education. International Journal of
Science Education, 14(5), 541- 62.
JOHNSON -LAIRD, P. N. (1983) Mental Models: Towards a Cognitive Science of Language,
Inference and Consciousness (Cambridge, MA: Harvard University Press).
JUSTI, R. S. (1997) Models in the Teaching of Chemical Kinetics. Unpublished Ph.D.
Thesis, (Reading: The University of Reading).
JUSTI, R. and GILBERT, J. (1999a) History and philosophy of science through models: the
case of chemical kinetics. Science and Education, 8(3), 287- 307.
JUSTI, R. and GILBERT, J. (1999b) A cause of a historical science teaching: the use of hybrid
models. Science Education, 83(2), 163- 77.
KAUFFMAN, G. B. (1989) History in the chemistry curriculum. Interchange, 20(2), 81-94.
KIPNIS, N. (1995) Blending Physics with History. In F. Finley, D. Allchin, D. Rnees, and
S. Fifield (eds), Proceedings of the Third International History, Philosophy, and Science
Teaching Conference, Vol. 1 (Minneapolis: University of Minnesota), 612- 23.
 HISTORY AND PHILOSOPHY OF SCIENCE THROUGH MODELS 1009

LAKATOS, I. (1970) Falsification and the methodology of scientific research programmes. In

I. Lakatos and A. Musgrave (eds), Criticism and the Growth of Knowledge (London:
Cambridge University Press), 91-196.
LAKATOS, I. (1978) The Methodology of Scientific Research Programmes (Cambridge:
Cambridge University Press).
LEDERMAN, N. (1992) Students’ and teachers’ conceptions of the nature of science: a review
of the research. Journal of Research in Science Teaching, 29(4), 331-59.
MATTHEWS, M. R. (1994) Science Teaching - The Role of History and Philosophy of Science
(New York and London: Routledge).
MCMULLIN, E. (1968) What do physical models tell us? In B. van Rootselaar (ed.), Logic,
Methodology and Philosophy of Science III (Amsterdam: North-Holland), 385- 96.
MONK, M. and OSBORNE, J. (1997) Placing the history and philosophy of science on the
curriculum: a model for the development of pedagogy. Science Education, 81(4), 405-
Downloaded by [Universitaets und Landesbibliothek] at 06:09 16 January 2014

24.
NERSESSIAN, N. J. (1992) How do scientists think? capturing the dynamics of conceptual
change in science. In R. N. Giere (ed.), Cognitive Models of Science (Minneapolis:
University of Minnesota Press), 3- 44.
NORMAN, D. A. (1983) Some observations on mental models. In D. Gentner and A. L.
Stevens (eds), Mental Models (Hillsdale, New Jersey: Erlbaum), 7- 14.
NYE, M. J. (1976) The nineteenth-century atomic debates and the dilemma of an ‘indiffer-
ent hypothesis’. Studies in History and Philosophy of Science, 7(3), 245- 68.
PARTINGTON, J. R. (1939) The origins of the atomic theory. Annals of Science, 4, 245-82.
SOLOMON, J. (1989) Teaching the history of science: is nothing sacred? In M. Shortland and
A. Warwick (eds), Teaching the History of Science (Oxford: Basil Blackwell), 42- 53.
UFMG (UNIVERSIDADE FEDERAL DE MINAS GERAIS) (1997) Manual do Candidato: Vestibular
1998 (Belo Horizonte: UFMG).

Appendix. The textbooks which were analysed

Textbook 1. CARVALHO , G. C. (1995) Qu õ mica Moderna, Vol. 1 (SaÄ o Paulo: Scipione).
Textbook 2. CARVALHO , G. C. (1995) Qu õ mica Moderna, Vol. 3 (SaÄ o Paulo: Scipione).
Textbook 3. COSTA, M. C., SANTOS, G. O. (1995) Qu õ mica: a visaÄ o do presente, Vol. 1 (Belo
Horizonte: Leà ).
Textbook 4. NETO, G. C. (1995) Qu õ mica: da teoria Áa realidade, Vol.1 (SaÄ o Paulo: Scipione).
Textbook 5. NOVAIS, V. (1993) Qu õ mica, Vol. 1 (SaÄ o Paulo: Atual).
Textbook 6. NOVAIS, V. (1993) Qu õ mica, Vol. 3 (SaÄ o Paulo: Atual).
Textbook 7. PERUZZO, T. M. and CANTO, E. L. (1996) Qu õ mica na Abordagem do Cotidiano,
Vol.1 (SaÄ o Paulo: Moderna).
Textbook 8. SARDELLA , A. (1997) Curso de Qu õ mica - Qu õ mica Geral (23a.edicË aÄ o). (SaÄ o
Paulo: Saraiva).
Textbook 9. USBERCO, J. and SALVADOR, E. (1995) Qu õ mica - Qu õ mica Geral (SaÄ o Paulo:
Saraiva).
Textbook 10. RAMSDEN, E. (1994) Key Science (Cheltenham: Stanley Thorne).
Textbook 11. HILL, G. (1995) Chemistry Counts, (2nd. Ed.) (London: Hodder and
Stoughton).
Textbook 12. HOLMAN, J. (1995) Chemistry (Victoria: Australia: Thomas Nelson).