Linguistics and Education 35 (2016) 88–99

Contents lists available at ScienceDirect

Linguistics and Education

journal homepage: www.elsevier.com/locate/linged

Modes and meaning in the classroom – The role of different semiotic

resources to convey meaning in science classrooms
Kristina Danielsson
Department of Language Education, Stockholm University, 106 91 Stockholm, Sweden

a r t i c l e i n f o a b s t r a c t

Article history: This study is framed within social semiotic perspectives on multimodality, and it has a twofold aim. The
Received 5 July 2014 primary aim is to analyze the ways in which teachers draw on different semiotic resources when intro-
Received in revised form 29 June 2016 ducing a new scientific concept in secondary school science classrooms, and to link the results to modal
Accepted 13 July 2016
affordance. A secondary aim is to try out parallel analyses of different modes in multimodal meaning
making using the ideational meta-function of the SFL framework. Analyses are based on instructional
Keywords:
episodes when chemistry teachers introduced the atom as a scientific phenomenon. The main focus of
Multimodality
the analyses is on processes used in different modes and how these depict the atom as either static or
Disciplinary literacy
Semiotic affordance
dynamic. The framework proved fruitful, and analyses revealed important patterns as to what aspects of
Systemic functional linguistics the atom were given through what mode(s), something which could partly be linked to modal affordance.
Science teaching The results are discussed in relation to its implications for research and education.
Classroom communication © 2016 Published by Elsevier Inc.

1. Introduction Lemke’s (1998) case study, focusing on one student in secondary

Over the last decades, the role of language in meaning making
has been emphasized by numerous scholars (Halliday & Martin,
1993; Lemke, 1990; Norris & Phillips, 2003; Schleppegrell, 2004;
Shanahan & Shanahan, 2008; Wellington & Osborne, 2001). How-
ever, multimodality is ubiquitous in all communication, and a
specialized use of modes for representation (e.g. writing, speech,
gesture, and image) is central in every content area (e.g. Danielsson
& Selander, 2014, 2016; Kress et al., 2005; Kress, Jewitt, Ogborn, &
Tsatsarelis, 2001).
In this article, mode is defined as a meaning-making resource
system which is shaped and developed over time in a commu-
nity. Examples of modes are speech, writing, gesture, and image.
Multimodality deals with representations or communicative events
involving more than one mode, for instance a drawing consisting of
both image and writing, or spoken words combined with gestures
(see, Jewitt, 2009a, for definitions and further discussions of these
concepts). Meaning making in science classrooms, which is the
focus of the present study, has repeatedly been described as highly
multimodal, involving action, speech, writing, subject specific sym-
bols, graphs, and diagrams, et cetera (e.g. Axelsson & Danielsson,
2012; Jakobson et al., forthcoming; Kress et al., 2001; Lemke, 1998;
Simpson & Walsh, 2010; Tang, Tan, & Yeo, 2011).
E-mail address: kristina.danielsson@isd.su.se
school during a single day in physics and chemistry classrooms
is elucidating as to the complex demands for students to “juggle”
between and integrate, for instance, verbal, chemical-symbolic, and
visual–graphical symbolic systems for different aspects of scientific
content. Yet, Lemke claims that the same information needs to be
accessible through as many types of media or channels as possi-
ble, not only since exactly the same content cannot be expressed
through any mode, but also since there needs to be a level of redun-
dancy in the learning situation. Another example is the study by
Tang and colleagues (Tang et al., 2011), where ninth-grade physics
students were followed during a teaching and learning cycle on the
concept of work-energy. Here students’ understanding of multiple
representations in different modes was focused. The students had
difficulties in constructing a scientific understanding of the content,
something which was discussed in regard to differences in the cat-
egorical, quantitative, and spatial meanings of the representations.
The findings in both these studies gave rise to reflections about the
importance of meta-discussions in classrooms as to how content is
expressed.
Since each representation of a concept (in the form of a gesture,
an image, words, etc.) has its specific potential for meaning making
(see affordance, below), certain aspects will be toned down while
others will be emphasized depending on the representation used.
Thus, to be able to “juggle” with representations, or to make mean-
ing from the teacher’s juggling, is a demanding task for the novice,
who is about to learn something new.

http://dx.doi.org/10.1016/j.linged.2016.07.005
0898-5898/© 2016 Published by Elsevier Inc.
 K. Danielsson / Linguistics and Education 35 (2016) 88–99
Regarding the role of multiple representations in knowledge
building in science, Tang and colleagues (Tang, Delgado, & Moje,
2014) have noted two main foci. Firstly, research on multiple rep-
resentations concerning how multiple representations of scientific
concepts in classroom practices affect students’ understanding;
secondly, research on multimodality dealing with students’ under-
standing of scientific concepts through the simultaneous use of
various modes within and across representations. The present
study can be seen as yet another alternative, through its focus on
the ways in which various semiotic resources used for describing a
specific scientific concept convey different meanings of the particu-
lar concept; in this case, the extent to which one curricular entity in
the science classroom, namely the atom, is represented as dynamic
or static in teacher introductions of the concept.
A key notion for the study is that of affordance (Gibson, 1977; van
Leeuwen, 2005; Kress, 2010), here defined as the ‘meaning poten-
tial’ (van Leeuwen, 2005) or ‘potentials and limitations’ of modes
(e.g. Kress, 2010:84). This notion relates to the fact that different
modes, as they are realized in social contexts, are governed by dif-
ferent logics. Examples are speech, which is governed by the logic
of time (speech sounds happen across time) and image, which is
governed by the logic of space, with still images also being gov-
erned by the logic of simultaneity, and moving images by the logic
of time. This is one reason why speech is regarded as especially
apt for describing temporal aspects, or issues of cause and effect.
Images, on the other hand, are considered to be more apt for spa-
tial properties, such as size and relative position between parts and
wholes.
Also, aspects such as valued modes (see Jewitt, 2009b or Kress,
2009 for discussions) need to be taken into consideration. In
schools, writing has long been valued, and the mode which students
normally are supposed to use when their learning is assessed, that
is, for their learning to be recognized (Kress, 2010; Kress & Selander,
2012). The present study uses data from a larger project (see Data,
below). One finding in that project was that teachers’ multimodal
board notes appeared to be highly valued in the classroom prac-
tices. What was written in the students’ notebooks was considered
decidedly important for passing the final written assessment of the
course, and with few exceptions these notes were direct trans-
formations of the teachers’ board-notes. Thus, information given
through other resources, such as speech or gestures, was generally
disregarded (Danielsson, 2011a).
The primary aim of the present study is to reveal patterns in
chemistry classrooms regarding what aspects of a specific scientific
phenomenon (in this case, the atom) is given through what mode,
or combinations of modes in multimodal ensembles (e.g. Jewitt,
2009b; Kress et al., 2001), and to link these results to the notion
of affordance. The main focus is whether the atom is depicted
as static or dynamic through the different modes. A secondary
aim is to contribute with methodological insights regarding multi-
modal analyses, by using the ideational meta-function in systemic
functional grammar (Halliday & Matthiessen, 2004) for parallel anal-
yses of different modes (see below). With this twofold perspective,
the present study can contribute with theoretical and analytical
insights as well as knowledge with relevance for classroom prac-
tices.
To accommodate for these aims, the study is theoretically
and methodologically framed within social semiotics (Halliday,
1978), and specifically on multimodality (van Leeuwen, 2005;
Jewitt, 2009a; Kress, 2010). A key notion in social semiotics is
that semiotic choice both reflects and shapes content. Further-
more, choice of semiotic resource or mode for representation is
seen as a result of social, cultural and situational factors in the
context in which the communication takes place, including par-
ticipants and available modes and resources. Hence, from such
perspectives, learning something new is intrinsically interwoven
89
with ways of expressing that knowledge through different semiotic
choices.
Studies within social semiotics are connected to Halliday’s sys-
temic functional grammar (SFG) (Halliday & Matthiessen, 2004), the
theoretical foundation of systemic functional linguistics (SFL); the
latter term is often used for applications of the SFG system on com-
munication in different modes, such as speech, writing, images, and
gestures.
Social-semiotic theory rests on three meta-functions: the
ideational (representing meanings about actions, states and events
in the world), the interpersonal (representing meanings about the
social relations of those engaged in the communication) and the
textual meta-function (representing the communicative channel
and textual organization). These meta-functions reflect different
aspects of meaning making (see Halliday & Matthiessen, 2004, for
a thorough presentation of the framework). The analyses in the
present study are done in relation to the ideational meta-function
(sometimes also referred to as experiential meaning), that is, the
field of communication, or what content is in focus in the commu-
nicative situation.
A foundation of the ideational meta-function is the transitivity
system in SFG (see Halliday & Matthiessen, 2004:302, for a model).
Transitivity analyses involve three major features in a clause,
namely the type of process (i.e. that something happens, is said, is
or is perceived: ‘the electrons move in a high speed’), what partici-
pants are involved in the process (i.e. who/what does/is/says/owns
what, etc.: ‘the electrons move in a high speed’), and the circum-
stances around the process (where, how, when, etc., the participant
did/thought/said: ‘the electrons move in a high speed’).
This article focuses on process types in the various resources
for meaning making that teachers use when introducing the atom
as a phenomenon (see Section 4). Previously, ideational analyses
connected to different disciplines have mainly been performed on
writing (e.g. Halliday & Martin, 1993; Schleppegrell, 2004). When
the SFL framework has been used for other modes, such as images
(Kress & van Leeuwen, 2006; Martin & Rose, 2008), other aspects
of the ideational meta-function have been highlighted. In their
ideational analysis of images in scientific texts, Martin and Rose
focus on aspects such as whether the scientific phenomenon is an
entity or an activity or if the categories are explicit or implicit. Even
though Martin and Rose (just like Unsworth, e.g. 2001) discuss
relations between image and writing, the type of detailed analy-
ses to compare modes which is performed in the present study,
are scarce. Here Martinec’s (2004) model for gesture analyses, is
interesting from a methodological point of view. In his data, Mar-
tinec analyzed both gestures and linguistic choice with respect
to experiential meaning (the ideational meta-function), and found
that process types were often the same in both modes. The fact
that Martinec analyzed gestures and verbal language as equivalent
is relevant for the present study. However, in the present study,
a number of modes (speech, writing, images, gestures, etc.), at
times appearing in multimodal ensembles, are analyzed through
the same framework, not only speech and gesture (cf. Martinec,
2004), or writing in connection to images (cf. Martin & Rose, 2008;
Unsworth, 2001).
1.1. The atomic model
Since the atomic model is in focus for they study, some com-
ments on the nature of the atom need to be given for the sake
of clarity for readers who might not be familiar with this subject
content.
At the beginning of the 20th century, Niels Bohr presented his
model of the atom, a model which is widely used even today, not
the least in schools. A typical image of the model is the third from
90 K. Danielsson / Linguistics and Education 35 (2016) 88–99
Fig. 1. Different models of the atom during the 20th century (Wikipedia 2014: “The evolution of the atomic model in the 20th century”). (The image is free to re-use: http://
en.wikipedia.org/wiki/File:Evolution of atomic models infographic.svg.)
the left in Fig. 1, with a positively charged nucleus1 surrounded by
negatively charged electrons, which, in Bohr’s model, are found in
orbits (or “shells”) around the nucleus. These orbits are supposed
to correspond to the approximate distance from the nucleus that
the electrons can be presumed to be, according to their level of
energy. Note that all these models are to some extent multimodal,
with symbols, such as plus and minus signs, integrated in the visual
depiction.
The orbits, in combination with the number of valence electrons
(the number of electrons in the outer orbit), are the basis for the
position of the chemical substance in the periodic table.
Today we also know that electrons appear in pairs (see
the rightmost model in Fig. 1, depicting the model of Heisen-
berg/Schrödinger), and that these pairs have different levels of
energy. Simple models of the atom often neglect these pairs, and
a common simplification is to describe the electrons as particles
spinning around a nucleus in a fashion similar to that of the planets’
orbits around the sun. However, it is more correct to view the elec-
tronic orbits as “electronic clouds”, where the density of the cloud
describes the probability to find an electron at a certain position.
A previous article (Danielsson, 2011b) describes how atoms
are depicted through different semiotic resources in the chemistry
classrooms focused on in the present study. That analysis revealed
that the content that students came across varied greatly as to the
qualities of the atom, sometimes with partly contradicting infor-
mation, such as viewing the atom as either static or a dynamic.
The present study aims at looking deeper into the role of different
modes for representation in relation to that aspect.
2. Data
The data used in the present study was collected within the
multidisciplinary project Chemistry text as tools for scientific learn-
ing (Eriksson, 2011), financed by the Swedish Research Council,
though the present study is independent of that project. The over-
all project followed all lessons during a content area dealing with
the atomic model, the periodic table, and chemical bonds. Between
20 and 40 lessons, normally around 40 min long, in altogether 7
Finland-Swedish (in Finland) and Swedish (in Sweden) mainstream
secondary school classrooms were studied, with students aged
between 14 and 15 years.
In the present study, data from 4 classrooms have been used,
with one teacher and around 20 students in each classroom; two
Finland-Swedish and two Swedish. In both countries, one class-
room was in a suburban school outside a main city, and one
was in a school in a mid-sized, rural town. The schools were
1
The positive charge of the nucleus is due to the positively charged proton(s) in
the nucleus. Apart from protons, the nucleus can also contain neutrons, which are
neutral.
linguistically and culturally relatively homogenous, with a major-
ity of the pupils speaking the language of instruction as their first
language (Finland-Swedish and Swedish, respectively), though the
teacher and some of the students in the suburban Finland-Swedish
classroom were bilingual in Finland-Swedish and Finnish. In all
classrooms, some students (2–4) spoke an additional language,
though none of them were newly arrived in the country. The teach-
ers were all qualified science teachers. For the sake of anonymity, all
participants have been assigned pseudonyms. The Finland-Swedish
teachers were given names starting with an F (Fia and Fred), while
the Swedish teachers were given names starting with an S (Sören
and Stig).
The data used in the present study consist of video record-
ings and digital photographs collected during whole class teaching
episodes when the teachers introduced the atom as a scientific phe-
nomenon. The video recordings were made with one video camera
placed at the back of the classroom, focusing on the teacher. Digital
photographs of the teachers’ notes on the whiteboard or black-
board were taken continuously by the author of the present study.
In all classrooms, these teaching episodes were a starting point for
the overall content area, and the introductions were given through
“chalk talk” (Artemeva & Fox, 2011), that is, a spoken exposition
combined with notes on the board. The introductions were typ-
ically around 8–12 min in length. Stig’s introduction was longer,
mainly because it contained more social talk besides the actual
content. Such talk was excluded from the analyses. Thus, only the
actual communication around the atom as a scientific phenomenon
is presented in the results section. All in all, around 40 min of video
recordings were used in the present study.
3. Methods of analysis
The video sequences were transcribed and analyzed using the
software Multimodal Analysis VideoTM , allowing for parallel anal-
yses of various meaning making resources (O’Halloran, Podlasov,
Chua, & K.L.E., 2012). In the teachers’ introductions, speech (here
defined as spoken words) was accompanied by gestures, and to
highlight certain aspects of the content, the teachers made board
notes, consisting of images (non-verbal visuals), writing (written
words), and written chemical or mathematical symbols.2 Also, one
model (a wooden ball normally used to represent a coal atom) was
used once in one of the classrooms. These modes (italicized) were
the basis for the analyses. However, since the teachers used writ-
ten mathematical or chemical symbols only to a small extent, these
modes were merged with writing. Also, the sole use of a model was
merged with image. Thus, four main categories are presented in
2
Chemical or mathematical symbols are considered separate modes from writing,
since they are governed by specific culturally and socially developed “grammars”.
 K. Danielsson / Linguistics and Education 35 (2016) 88–99 91

the overviews in the Results section: speech, gestures, image, and Table 1
Overview of process type in different modes in teacher introductions of the atom.
writing.
Frequencies and percentages for each mode.
As a first step, spoken language was transcribed. In parallel with
this, tags indicating different types of action connected to the con- Relational Material Verbal Mental Total
tent were made in the transcripts. Examples of actions that were Gesture 33 (40%) 51 (60%) 0 0 84
tagged are gesturing or the use of artefacts, such as concrete mod- Image 87 (86%) 14 (14%) 0 0 101
els, or writing and drawing on the board. In this article, “drawing” Speech 261 (56%) 109 (23%) 13 (3%) 84 (18%) 467
Writing 1 0 0 0 1
refers to teachers’ notes on the board when these consisted of an
Total 382 174 13 84 653
image or a combination of image and writing and/or symbols. Thus,
in the following, a drawing can be a multimodal ensemble, while
an image is not. Table 2
Relational and material processes in different modes in teacher introductions of the
As a second step, the various meaning making resources in
atom. Observed and expected frequencies (EF).
the different modes were analyzed through the ideational meta-
function. As was mentioned above, in this article only the results Relational EF Material EF Total
as regards processes will be presented. Gesture 33 58 51 26 84
Halliday and Matthiessen (2004) present six processes: material Image 87 69 14 32 101
(e.g. ‘draw’, ‘take apart’), behavioural (e.g. ‘yawn’, ‘stretch’), mental Speech 261 254 109 116 370
Writing 1 1 0 0 1
(e.g. ‘think’, ‘believe’), verbal (e.g. ‘said’, ‘declared’), relational (e.g. 381 174 556
‘is’), and existential (e.g. ‘live’, ‘exist’). However, in the analyses of
p < 0.001.
the present data, behavioural processes were merged with material,
since the former can be seen as a subgroup of the latter (Andersen,
Petersen, & Smedegaard, 2001; Holmberg & Karlsson, 2006). For the Table 3
Relational and material processes in gestures and images in teacher introductions
same reason, existential processes were merged with relational pro-
of the atom. Observed and expected frequencies (EF).
cesses. Thus, four process types are used in the following: material,
mental, verbal, and relational processes. Here, material processes Relational EF Material EF Total

can be expected to depict the atom as dynamic, while mental, ver- Gesture 33 54 51 30 84
bal and relational processes can be expected to depict the atom as Image 87 66 14 35 101
120 65 185
a static phenomenon.
The unit of analysis for the spoken language was the clause, and p < 0.001.
processes in both independent and dependent clauses were ana-
lyzed: “before, the scientists thought (mental process) that the atom Table 4
was (relational process) the smallest existing particle”. Relational and material processes in gestures and speech in teacher introductions
Only gestures with a clear relation to the subject content were of the atom. Observed and expected frequencies (EF).

analyzed, for instance if a teacher made a circular arm movement Relational EF Material EF Total
when talking about the way electrons move in orbits. The process Gesture 33 54 51 30 84
in such a gesture was coded as material (cf. “the electrons move Speech 261 240 109 130 370
in circles”). For still gestures, as when a teacher shaped his or her 294 160 454
hands spherically when talking about an atomic nucleus, the pro- p < 0.001.
cess was coded as relational (cf. “the nucleus is shaped as a ball”).
These two types of gestures (material and relational) are partic-
Table 5
ularly important in regard to depictions of the atom as a static Relational and material processes in images and speech in teacher introductions of
or dynamic concept, and they have a direct relation to Martinec’s the atom. Observed and expected frequencies (EF).
(2004) analyses of gestures in static or action processes.
Relational EF Material EF Total
In analysing images, two-dimensional images showing particles
Image 87 75 14 26 101
of the atom in static positions (as in Fig. 5) were coded as depict-
Speech 261 273 109 97 370
ing relational processes. If containing an arrow, or if in other ways 348 123 471
indicating a movement, the image was coded as depicting a mate-
p < 0.002.
rial process. The images were analyzed as to their final design (see
e.g. Fig. 7), but repeated codings were made during the teachers’
expositions, when images were built up gradually through “chalk- significant, chi square analyses were used to compare observed
talk”. In these cases, different aspects of the image was commented frequencies with the expected statistical frequencies. The statisti-
on through speech, thus being part of multimodal ensembles (for cal significances are indicated by p-values for the results presented
example the different stages of the image in Fig. 6). Images were in each table. p-values below 0.05 denote statistically significant
usually present for some time after they were drawn, or in the case differences between occurrences of process types in the differ-
of the chart with the periodic table, pointed at. If the teacher related ent modes. For readers unfamiliar with basic statistics such as chi
to the image again, for instance by connecting to it explicitly in square analyses, a brief explanation is given in Bryman (2012).
spoken language (e.g. “as you can see in the drawing”), pointing, or
clearly looking at it, the coding was repeated. The rationale for this 4. Results
decision was that when the teacher brought an image to the fore
again, the image was part of a multimodal ensemble, complement- In this section, to start with, quantifications regarding pro-
ing the spoken words or gestures. cess types expressed through the different modes are presented in
As a next step, basic calculations regarding the process types Tables 1–5. Then the results for each mode are discussed regarding
in different modes were made to give an overview of the data, modal affordance, with examples given from the teaching episodes.
and to reveal patterns of the extent to which the various modes Finally, patterns in multimodal ensembles are presented.
were used in specialized ways regarding process types (Tables 2–5). A total of 653 processes occurred in the data, and these are the
To ensure that differences noted in this regard were statistically basis for the quantifications in Tables 1–5. Table 1 gives an overview
92 K. Danielsson / Linguistics and Education 35 (2016) 88–99
of the frequencies of processes given through the different modes
used during the teachers’ expositions: writing (including mathe-
matical and chemical symbols), speech, image (including one use of
a concrete model), and gesture. As can be seen in the overview, a
majority of the processes were expressed through speech (467 out
of 653 instances). Fewer were expressed through image (101) or
gesture (84). Even though writing includes mathematical and chem-
ical symbols in the overviews, only one process was given through
this mode. This is due to the fact that teachers mostly wrote sin-
gle words on the board, usually to label different particles in an
image depicting the atom. In some cases they also used symbols of
chemical substances, such as H (for ‘Hydrogen’) or Li (for ‘Lithium’).
The single written process in the data was the mathematical symbol
“=”, analyzed as a relational process (“equals”). Some other symbols
also imply relational processes (see Multimodal ensembles, below).
The concrete model (a wooden ball) gave a static depiction, with
the process coded as relational.
Table 1 also reveals that the vast majority of processes in the
data were relational. Around half as many were material; fewer
were mental, while verbal processes were even more scarce. Rela-
tional and material processes occurred in all modes, while mental
and verbal processes only occurred in speech. As was mentioned
in the Methods section, mainly material processes have the poten-
tial to depict the atom as a dynamic phenomenon. Therefore, the
overview in Table 1 implies that the atom is primarily depicted as
a static phenomenon in these classrooms.
Since verbal and mental processes only occurred in speech, and
because they were seldom directly connected to the atom as a
concept (see below), a separate comparison was made between
relational and material processes (556 processes in total) in the dif-
ferent modes (Table 2). Especially gestures and images appear to be
used in certain patterns, with gestures depicting material processes
more often than expected from their statistical relative frequencies
(EF in Tables 2–5), while the opposite holds for images. The same
pattern persists when comparing only these two modes (Table 3).
From the overview in Table 1, it is evident that speech and
gestures are important for expressing material processes in the
classrooms (160 of the 174 material processes were given through
these two modes). However, when comparing these two modes
with respect to relational and material processes, gestures were
used for material processes significantly more often than expected
from their relative frequencies, while the opposite was the case for
speech (see Table 4).
A comparison was also made between image and speech
(Table 5). Here, image is used significantly more often than
expected to express relational processes compared to speech, with
an opposite result regarding material processes.
To connect to the key question about the extent to which differ-
ent modes appear to depict the atom as static or dynamic, gestures
and images stand out as being used in opposing manners. Gestures
were often coded as material, thus giving a dynamic depiction,
while images were often coded as relational, thus giving a static
depiction. Processes in speech, finally, were more often coded as
relational than material.
Before discussing the use of the various modes in relation to
affordances, the relatively high frequency of mental processes in
speech (see Table 1) needs to be commented on. Many of these
were of the following types: “nowadays we know that atoms can
be divided into parts” or “before, the scientists thought that. . .”
or “now you think about planet orbits and things like that”. In
fact, most main participants (see Halliday & Matthiessen, 2004)
in clauses with mental processes were either “scientists” or “you”
(the students in the classroom), and often these clauses were a
starting point for saying something about the atom in a dependent
clause, such as “before, the scientists thought that the atom was the
smallest existing particle”. However, some participants in clauses
involving mental processes were scientific content, like in the fol-
lowing anthropomorphic metaphors (e.g. Taber & Watts, 1996) in
which particles are humanized, with desires: “negatively charged
particles do not want to be close to each other”, “negatively charged
particles and positively charged particles want to be drawn towards
each other” (the use of metaphors in these classrooms is analyzed
in more detail in Danielsson et al., forthcoming).
4.1. Different modes for different aspects of the atom – the role of
affordance
All images, gestures and instances of writing analyzed were
directly connected to the atom as such. As regards speech this was
not always the case. Some processes were of the kind exemplified
around mental processes, above, with the students or scientists as
the main participant. Others were related to what the teachers did,
or what they wanted the students to do, such as “Here I have drawn a
simplified model of the atom” (as mentioned in Section 3, social talk
was already excluded from the analyses). With this said, interest-
ing patterns in the data can be discerned, which can be connected
to affordance. Since only one process was expressed in writing (see
Table 1), only gestures, images, speech, and multimodal ensembles
will be discussed in the following.
4.1.1. Gestures
With regard to their materiality, gestures are governed both by
the logics of space and time. Therefore a gesture can be used as a
kind of visualization, and by using gestures involving hands and
arms, it is possible to mimic a movement or a shape. The gestures
used in relation to the atom as a scientific concept were all con-
sistent with the affordance of the mode in that respect. Examples
of gestures in the teacher expositions are shown in Figs. 2 and 3.
In the left photograph in Fig. 2, the teacher moves her arms in a
circular manner to depict the ways in which electrons swirl around
the nucleus. Another type of gesture is shown to the right in Fig. 2,
and in the two photographs in Fig. 3. Here the teachers hold their
hands still to depict relative sizes or positions of particles in the
atom. Both these types of gestures could be regarded as iconic in
that they to some extent are used to resemble what they represent:
moving electrons and relative sizes or positions of particles (see
e.g. Martinec, 2004 or Kress, 2010 for discussions around Peirce’
classical classification of gesture types into iconic, indexical and
symbolic).
Regarding the key issue concerning the atom as a static or
dynamic phenomenon, the quantifications above revealed that
even though processes expressed through gestures were coded as
both material and relational, there was an over-representation of
material processes. Thus gestures appear to be used for highlight-
ing the dynamic aspects of the atom. One such typical gesture is
shown to the left in Fig. 2, where the teacher talked about elec-
trons swirling around the nucleus. All four teachers used similar
gestures for this aspect of the atom. Gestures that were coded as
depicting relational processes were used to demonstrate relations
between particles in the atom, such as the gestures in Fig. 3 and in
the right photograph in Fig. 2. Fig. 3 shows how Sören first depicted
the atomic nucleus with his hands pulled together as a sphere,
and shortly afterwards held his hands apart, depicting an empty
space between the nucleus and the electrons. The gesture shown
to the right in Fig. 2 shows how Fia made a gesture when talking
about “Penny Proton” who was really big (compared to “Elvis the
Electron”). Here we can note that Sören and Fia used quite simi-
lar gestures for different aspects of the atom: empty space around
nucleus, and size of particle in relation to other particles, respec-
tively. Thus, the meaning of a gesture in itself might not be possible
to interpret without further enhancement from other resources,
 K. Danielsson / Linguistics and Education 35 (2016) 88–99 93

Fig. 2. Teacher gestures: electronic movements (left) and “big proton” (right), Fia’s classroom.

Fig. 3. Teacher gestures: nucleus (left) and “empty space” (right), Sören’s classroom.

such as a spoken exposition, hence being a part of a multimodal
ensemble (see Section 4.1.4).
4.1.2. Images
During these teaching episodes, only still images were used.
Since the materiality of still images is governed by the logics of
space and simultaneity, they are especially apt for spatial aspects
of the atom, for instance relative size and position of particles. Thus,
according to the modal affordance of still images, we could expect
them to be used mainly for relational processes (cf. ‘the atom con-
sists of. . .’). This was also the case, with an over-representation of
relational processes in the images in comparison to other modes.
However, here it is important to note that processes in images
were analyzed each time they were brought to the fore in the
classroom. Therefore the same image appears more than once
in the analysis. Yet, since an image depicting the atom as either
static or dynamic was brought to the fore a number of times, this
meant that this way of viewing the atom was also recurrent in the
classroom.
In all classrooms, the teachers made drawings of the atomic
model in line with Bohr’s model (see Fig. 1, Section 1.1), with a
nucleus surrounded by electrons in clearly defined orbits. Typical
examples of such drawings are shown in Figs. 4–8 (these drawings
are also commented on below concerning their multimodality).
All images in these drawings were regarded as depicting rela-
tional processes, and they focus on the particles that build up the
atom. They also highlight spatial relations, such as where in the
atom the particles are found, as well as their relative sizes (though
in a very crude manner, which is sometimes pointed out by the
teachers).
The instances when images in drawings were coded as depict-
ing material processes were all connected to the drawing in Fig. 9,
which was built up in two steps. It was made when Stig told the
students about a university lecturer who used drawings and ges-
tures to explain that electrons in fact do not move in clearly defined
orbits. The first stage (left in Fig. 9) depicts the electronic movement
in the inner orbit, and the second step (right in Fig. 9) depicts the
electronic movement in the second orbit.

Fig. 4. Teacher’s drawing of the atomic model, first stages (Fia).

94 K. Danielsson / Linguistics and Education 35 (2016) 88–99

Fig. 5. Final drawing of atomic model (Fia).

Fig. 6. Image of the atomic model built up in stages (Sören).

However, it is important to mention that the extent to which an
image is actually depicting the atom as static or dynamic could be
discussed. Images containing circles illustrating orbits imply that
something is moving along an imagined line. In that sense, images
such as the one in Fig. 4 could to some extent be regarded as depic-
tions of a movement, even though the dynamic aspect is toned
down in comparison to images like the one in Fig. 9.
4.1.3. Speech
Due to its sequentiality, the materiality of speech is gov-
erned by the logic of time. In regard to affordance, it is often
claimed that speech (and writing) is especially apt for describ-
ing temporality or for logic reasoning (see Section 1) rather than,
for instance, spatiality. However, even though information given
through an image or a gesture can be taken in by the viewer
more or less as a whole, in many cases a similar content could
be expressed in words, even though quite a lengthy explanation
might be required. Regarding temporality and logic reasoning (such
as cause and effect), where speech (and writing) is expected to
work especially well, there were very few instances of such rea-
soning in the teachers’ introductions to the atom. Also, when
that happened, such reasoning was never expressed through pro-
cesses.
The affordance of speech might not affect whether the atom
was depicted as a static or dynamic concept. As was shown in
the overviews above, processes in speech depicted the atom as
both dynamic and static. However, speech was over-represented
in regard to relational processes when compared with gestures
(Table 4), while speech was over-represented in regard to material
processes when compared with images (Table 5), which indicates
that gesture and image were used in more specialized ways than
speech.
4.1.4. Multimodal ensembles
Multimodal ensembles were in many ways the “state of the art”
in the teachers’ expositions, where typical combinations of modes
were the following:
• Multimodal drawings (involving image, writing and/or mathe-
matical or chemical symbols)
• Speech and gestures
• Speech and images
 K. Danielsson / Linguistics and Education 35 (2016) 88–99 95

Fig. 7. Final drawing of the atomic model (Sören).

Fig. 8. Image of lithium in two stages (Stig).

Fig. 9. Image of swirling electrons, drawing in two stages (Stig).

• Speech and multimodal drawings
• Speech, gestures and images or multimodal drawings.
Most drawings could be regarded as multimodal ensembles,
combining images with writing and/or mathematical or chem-
ical symbols (the only exception being the drawing in Fig. 9,
which consists of an image only). Figs. 5 and 7 illustrate how
writing typically is used in the board notes, with headings, and
then single words labelling particles. Fig. 5 shows the sole pro-
cess expressed in writing: the mathematical symbol “equals”.
In that drawing, plus and minus signs in direct relation to the
name of the various particles are also used. Even though no pro-
cess is explicitly expressed, the plus and minus signs in these
drawings imply relational processes (cf. “the proton is posi-
tively charged”), further strengthening a static depiction of the
atom.
Altogether, almost all drawings (including image, writing and
symbols) depict the atom as static. The only image clearly depicting
the atom as a dynamic concept is the one shown in Fig. 9. We can
also note that this image is the only one dissimilar to the classic
models in Fig. 1. The image, which in itself contains no writing or
symbols, functioned as part of a multimodal ensemble where Stig
used his whole body: while making the drawing he gave spoken
comments on what his lecturer had done, and imitated him, moving
his arms round and round in circles, almost jumping while drawing
(see also the excerpt in Table 7).
96 K. Danielsson / Linguistics and Education 35 (2016) 88–99

Table 6
Excerpt from multimodal exposition, Fia’s classroom. Processes in italics; process type indicated for processes related to the atom in teacher’s communication.

Speech Action (blackboard notes and gestures)

001 Teacher: in the nucleus there are (RP) two of those particles/. . ./do you
remember charges?
002 Student: the one in the middle is neutral
003 T: and what is the neutral particle called? (RP) starts drawing a small circle on the blackboard beside the model (I:
RP), equivalent to one of the particles in the nucleus (Fig. 5)
004 S: nucleus
005 T: no. but is it a neutron proton or electron? (RP)
006 S: neutron
007 T: writes on blackboard: “neutron (neutral)” (W: RP [implied])
008 T: it’s easy to remember since it starts neutron. neutral. points at the blackboard at the beginning of the words `neutron' and
`neutral'
009 T: it has no charge (RP) points at the blackboard at the small circle in the model, depicting a
neutron (I: RP)
010 T: Do you remember the copy we distributed?
011 S: “Elvis the electron” (in English)
012 T: it was (RP) Penny Proton who was (RP) so big and round and really gestures with full arms forming round sphere (G: RP) (right, Fig. 2)
positive charge
013 T:. and the proton was. it is positive (RP) draws a plus sign in one of the small circles depicting the proton in
the model; writes a plus sign and a circle around to the right of the
model, depicting proton (I: RP; W: RP [implied]) (Fig. 5)
014 T: And it’s easy to remember since it starts with a P like in positive writes on blackboard, beside the small circle: “proton (positive)”
(Fig. 5) (W: RP [implied])
015 T: And neutrons and protons are (RP) the same size. approximately the points at the blackboard at the circle depicting a nucleus (I: RP; W: RP
same size. they are (RP) in the nucleus [implied])
writes on blackboard: “nucleus”
016 T: And the little tiny particles. that were called (RP) points at the blackboard at the circle depicting an electron (I: RP)
017 S: electrons
018 T: yes. that was (RP) “Elvis the electron” (in English) writes a minus sign in the circle depicting an electron (W: RP
[implied])
019 T: yes which swirls (MP) around there all the time. and just rolls (MP) gestures with full arms making moving circles (G: MP) (left Fig. 2)
020 writes on blackboard: circle with minus sign and “electron” (Fig. 5)

I: image; G: gesture; W: writing; RP: relational process; MP: material process.

Table 7
Excerpt 1 from multimodal exposition, Stig’s classroom. Processes in italics; process type indicated for processes related to the atom in teacher’s communication.

Speech Action (whiteboard notes and gestures)

001 Teacher: then I write two writes “2” on the electronic orbit (see Fig. 8, left image)
002 T: there are (RP) two electrons there. I will show you later how they really move (MP)
003 Student: are they at the same place then?
004 T: no holding (MP) handsa .. no.. give me two minutes and I will explain how it really is (RP)
005 T: you know. they are (RP) there in approximately the same orbit Makes circular gesture (G: MP)

G: gesture; RP: relational process; MP: material process.

a
The teacher has a humorous style and uses uncommon metaphors. In this case, “no holding hands” is a way of expressing the fact that the electrons are never at the same
place.

The drawings, which were the main board notes made during
the teachers’ expositions, were typically made in combination with
speech, and, at times, with gestures. While the drawings in the
teachers’ notes usually depicted the atom as static (e.g. Figs. 5–8),
both similar and dissimilar patterns were found in the accompa-
nying speech and gestures. Examples are given in the excerpts in
Tables 6 and 7, with transcriptions of speech in the left column
and comments on actions like drawing and gesturing in the right
column.
After a short introduction, in which all spoken processes related
to the atom were relational, Fia started making the drawing shown
in Figs. 4 and 5. When the communication shown in Table 6 took
place, she had finished the first stages of drawing the model (see
Fig. 4) and she went on to comment on the charges of the different
particles.
Similar to the introductory spoken text, most of the processes
directly linked to the atom as a concept were relational, mainly in
the form of are or has. This was also the case when Fia talked about
“Penny Proton” who was big and positive, making the static (rela-
tional) gesture shown to the right in Fig. 2. The two spoken material
processes, swirls and rolls (line 019), were combined with the
dynamic (material) gesture shown to the left in Fig. 2. As was noted
above, Fia and Sören used similar gestures for different aspects of
the atom (the relative size of the proton, and empty space between
nucleus and electrons). Here the information needed for the inter-
pretation of the gesture was given through speech (line 012).
In the multimodal ensembles during Fia’s exposition (combining
speech, gestures and multimodal drawings), the blackboard draw-
ing (image and writing) depicted the atom as a static concept, while
speech and gestures either enhanced this static depiction, or added
a dynamic component to it. Similar to Martinec’s (2004) study,
when speech was combined with gestures, both modes boosted
each other to give the same notion of the atom in Fia’s exposition.
This pattern was common in all classrooms. However, an excerpt
from another classroom (Table 7), reveals a somewhat different
pattern.
Just before the short dialogue shown in Table 7 took place, Stig
had started drawing an image of Lithium (Fig. 8, left).
In this excerpt, the gesture is the only resource offering the
dynamic aspect of the atom. We can also note that the teacher com-
bines this material gesture with a relational process in speech: “[the
electrons] are in approximately the same orbit”. Thus, instead of
enhancing each other, gesture and speech give opposing, or com-
plementary, representations. After a discussion about where and
 K. Danielsson / Linguistics and Education 35 (2016) 88–99
Table 8
Excerpt 2 from multimodal exposition, Stig’s classroom. Processes in italics; process type indicated for processes related to the atom in teacher’s communication.
Speech
001 T: he was going to illustrate how they really move (MP)
002 T: I asked the same question as you.. how do they really move (MP)?
003 T: how. where are (RP) they?
004 T: the first little orbit, I grasped that
005 T: because it goes (MP) like . . . this
006 T: but the electron. or electrons. could just as well be (RP) here.. or here
007 T: they can be (RP) out here
008 T: but usually the are (RP) here
009 T: so. this orbit. is where it usually is (RP)
I: image; G: gesture; RP: relational process; MP: material process.
how to draw electrons in orbits, the drawing was finished (Fig. 8,
right).
Later during the same teaching episode, Stig came back to how
the electrons “really move”, now telling the anecdote from his own
university years, mentioned above. He described how his univer-
sity teacher had explained how electrons move, drawing the image
shown in Fig. 9 accompanied by speech and action (Table 8).
Here, most spoken processes associated with the atom were
relational (except goes, line 005), while the processes given in the
accompanying image or gestures were material. Once (line 009) a
dynamic gesture was combined with relational processes in draw-
ing and speech. In the excerpt presented in Table 8, the same types
of processes were usually given through image and gestures. How-
ever, as opposed to the pattern in Fia’s classroom, processes given
through gestures quite often depicted the atom differently than
those given through speech, again giving opposing, or complemen-
tary, representations, allowing the two modes of communication
to convey separate meanings.
5. Conclusions and discussion
The primary aim, to analyze to what extent the atom was
depicted as dynamic or static through different modes, is depen-
dent on the secondary aim. Therefore, the secondary aim is
discussed initially.
The secondary aim concerned the potentials of parallel analyses
of processes in different modes. A challenge here is the fact that the
SFL framework was developed in relation to writing (and speech).
Therefore, any attempt to use the framework for other modes, such
as gesture or image, must involve some kind of “translation”.
In the present study, gestures were only analyzed when they
had a clear connection to the atom as a phenomenon, and here
the codings of the data as relational or material processes were
straightforward. Most gestures were used to depict electronic
movement (coded as material), while some depicted aspects such
as relative size of particles, or “empty space” (coded as relational).
These codings were parallel to Martinec’s (2004) analyses of ges-
tures in static or action processes.
The coding of processes in images was less clear-cut. As was
mentioned in Section 4.1.2, images with circles representing orbits
could perhaps have been analyzed as depicting a movement. Even
so, such images were coded as relational, since a circle in itself does
not imply a movement, unless a clear indication of movement, like
an arrow, is added.3 Another analytical choice regarding images
was to repeat analyses of images when they were brought to the
fore again. Such repeated analyses could be seen as a parallel to
wordings being repeated during the expositions (a word “disap-
pears” when it has been uttered, and needs to be repeated, while
3
Arrows were not added in the images of the atom during these introductions,
though such use was noted in one of classrooms further into the content area.
97
Action (whiteboard notes and gestures)
gesturing in circular manner with right arm (G: MP)
drawing/gesturing in circular manner (Fig. 9, left image) (I: MP) (G: MP)
drawing/gesturing in circular manner (Fig. 9, left image) (I: MP) (G: MP)
drawing/gesturing in circular manner (Fig. 9, left image) (I: MP) (G: MP)
drawing/gesturing in circular manner (Fig. 9, left image) (I: MP) (G: MP)
pointing/gesturing (Fig. 8, right image) (I: RP) (G: MP)
an image is still there, and can be recalled again by pointing or
clearly looking at it, or by referring to it in speech). Without such
repeated analyses, images would have been left out from analyses
of multimodal ensembles when the teachers commented on them
again, through speech (and gesture), for instance in the examples
given in Section 4.1.4.
Bearing the abovementioned downsides in mind, to make par-
allel process analyses proved to be fruitful, enabling systematic
comparisons between modes, which was the basis for the primary
aim to reveal patterns as to the use of different modes to depict
the atom as static or dynamic. These analyses revealed that differ-
ent modes appeared to be used in specialized ways, which could
be related to modal affordance. This was particularly true for ges-
tures and images. In most cases, these modes were used to depict
the atom in opposing manners. Gestures mainly highlighted the
dynamic dimension of the atom, with swirling electrons (process
type coded as material). Images, on the other hand, usually depicted
the atom as static, highlighting the various particles of the atom,
and their relative positions (process type coded as relational). How-
ever, even though gestures and image in particular appeared to be
used in specialized ways, all modes were to some extent used to
give both dynamic and static depictions of the atom.
When analysing processes in gesture and speech in multimodal
ensembles, these often boosted each other, parallel to Martinec’s
(2004) finding that the “same” content was given in both modes.
However, this was not always the case in the present study. In some
cases, such as in the excerpts from Stig’s classroom in Section 4.1.4,
a relational process was expressed through speech while the ges-
ture was dynamic, and vice versa. This might be unproblematic for
students who do not struggle with content or language. Yet, when
two resources used in parallel give opposing information, this could
be a potential obstacle.
An important result is that it was through the teachers’ use of
multiple modes that the students were given a relatively compre-
hensive illustration of the atom, including the various particles
and the dynamic aspect (though commonly, the descriptions of
the atom were in line with Bohrs’ simplified model, see Fig. 1,
Section 1.1). However, as mentioned in the introduction, results
from the main project revealed that the students in these class-
rooms transformed the teachers’ board notes into their notebooks
without adding information given orally or through gestures
(Danielsson, 2011a). Since the content given through gestures
or speech was seldom integrated into the students’ notes, the
dynamic aspects of the atom were toned-down (see also Kress’
discussion about ‘ontological weight’ as regards different modes,
2010:86).
Regarding the use of multiple resources used for meaning mak-
ing, the results are in line with previous studies from science
classrooms. The teachers in the present study used a variety of
modes to explain different aspects of the atom: writing, speech,
images, gestures, and combinations of these, in multimodal ensem-
bles. Even though the present study differs from Lemke’s (1998), in
98 K. Danielsson / Linguistics and Education 35 (2016) 88–99
which one student was focused on, the classroom practices were
similar, with information given through a variety of modes more
or less simultaneously. There was indeed redundancy during the
teaching episodes analyzed in the present study, and thus students
were thus given the “same” information through different channels
(i.e. modes). However, due to their different potentials for meaning
making, different modes were used for slightly different aspects
of the content. Consequently, an important conclusion is that these
students needed to combine the information given through the var-
ious semiotic resources in order to pick up what particles the atom
consist of, what the they are called, their electric charge, and their
relative positions in relation to each other, at the same time as they
need to recognize that the atom is dynamic.
As both Lemke (1998) and Tang and colleagues (Tang et al.,
2011), argue, to “juggle” between modes and representations in
science, and to make meaning from them is a demanding activ-
ity. Similar to their studies, no discussions were noted in these
classrooms as to the affordance of modes, or how modes need
to be combined to give a comprehensive representation of the
content. When such discussions are connected to content, they
have the potential to enhance both content knowledge and a
more general awareness of the ways in which different modes
are often used in specialized ways. As Lemke argues in relation
to the need to teach students about multi-literacies and hybrid
genres, this is not a matter of adding new content to the cur-
riculum but instead about “foregrounding and thematising what
is already in the curriculum, getting teachers and students to
pay attention to what was formerly taken for granted” (Lemke,
1998:269).
To sum up, the parallel analyses of processes proved to be a fruit-
ful way of revealing how different modes were used for different
aspects of the atom. These results could be related to modal affor-
dance. Also, the analyses revealed that the teachers in this study
gave a relatively comprehensive picture of the atom, assuming that
they consider Bohr’s simplified model of the atom adequate enough
for students of this age. However, since there were no classroom
discussions about the affordance of modes, or what aspects of the
atom were highlighted through different representations, it was
left to the students to make meaning from the combined resources.
Also, since board notes appear to be highly valued in these class-
rooms, students might miss important aspects of the content given
through resources like speech and gestures. As regards implications
for education, the results imply that teachers need to become aware
of the ways they themselves, as well as other learning resources
used, employ different modes or combination of modes in teach-
ing. Also, they need to make the students aware of this. If focusing
both on content and on the ways in which the content is expressed
through different semiotic resources, students will get the oppor-
tunity to enhance both their content knowledge and a more general
awareness of how modes are used in specialized ways.
Acknowledgements
The author wishes to express her appreciation to the partici-
pants in the interdisciplinary project Chemistry texts as tools for
scientific learning, funded by the Swedish Research Council, from
which the data was collected. The author also wishes to thank the
anonymous reviewers for valuable comments on earlier versions
of the article.
References
Andersen, T., Petersen, U. H., & Smedegaard, F. (2001). (The language as a resource.
Danish systemic functional linguistics in theory and practice) Sproget som resource.
Dansk systemisk funktionel lingvistik i teori og praksis. Odense: Odense Univer-
sitetsforlag.
Artemeva, N., & Fox, J. (2011). The writing on the board: The global and the local in
teaching undergraduate mathematics through chalk talk. Written Communica-
tion, 28(4), 345–379.
Axelsson, M., & Danielsson, K. (2012). Multimodality in the science classroom. In
A. Pitkänen-Huhta, & L. Holm (Eds.), Literacy practices in transition: Perspectives
from the Nordic countries (pp. 142–166). Bristol: Multilingual Matters.
Bryman, A. (2012). Social research methods. Oxford: Oxford University Press.
Danielsson, K. (2011a). Att bli del av en naturvetenskaplig diskurs. In I. Eriksson (Ed.),
Kemiundervisning, text och textbruk i finlandssvenska och svenska skolor – En kom-
parativ tvärvetenskaplig studie (pp. 161–237). Stockholm: Stockholm University
Press.
Danielsson, K. (2011b). Då blir de fulla och glada. Multimodala representationer av
atommodellen i kemiklassrum. In J. Smidt, E. S. Tønnessen, & B. Aamotsbakken
(Eds.), Tekst og tegn. Lesing, skriving og multimodalitet i skole og samfunn (pp.
121–143). Trondheim, Norway: Tapir Akademisk Forlag.
Danielsson, K., Löfgren, R., & Pettersson, J. A. (forthcoming). Gains and losses.
Metaphors in chemistry classrooms. In K.-S. Tang, & K. Danielsson (Eds.),
Global developments in literacy research for science education. Dordrecht, NL:
Springer.
Danielsson, K., & Selander, S. (2014). (Multimodal texts in education) Se Texten! Mul-
timodala texter i ämnesdidaktiskt arbete. Malmö: Gleerups.
Danielsson, K., & Selander, S. (2016). Reading multimodal texts for learning. A model
for cultivating multimodal literacy. Designs for Learning, 8
Eriksson, I. (Ed.). (2011). Kemiundervisning, text och textbruk i finlandssvenska och
svenska skolor – En komparativ tvärvetenskaplig studie (Chemistry teaching, text
and text practices in Finland-Swedish and Swedish schools – A comparative
multidisciplinary study). Stockholm: Stockholm University Press.
Gibson, J. (1977). The theory of affordances. In R. Shaw, & J. Bransford (Eds.), Perceiv-
ing, acting and knowing. Toward an ecological psychology (pp. 67–82). Hillsdale,
NJ: Lawrence Erlbaum.
Halliday, M. A. K. (1978). Language as social semiotic. London: Edward Arnold Ltd.
Halliday, M. A. K., & Martin, J. R. (1993). Writing science. Literacy and discursive power.
London: University of Pittsburgh Press.
Halliday, M. A. K., & Matthiessen, C. M. I. M. (2004). An introduction to functional
grammar (3rd edition). London: Arnold.
Holmberg, P., & Karlsson, A.-M. (2006). (Grammar with meaning) Grammatik med
betydelse. Uppsala: Hallgren & Fallgren.
Jakobson, B., Danielsson, K., Axelsson, M., & Uddling, J. (forthcoming). Measuring
time. Multi-lingual elementary school students’ meaning-making in physics. In
K.-S. Tang, & K. Danielsson (Eds.), Global developments in literacy research for
science education. Dordrecht, NL: Springer.
Jewitt, C. (Ed.). (2009). The Routledge handbook of multimodal analysis. London: Rout-
ledge.
Jewitt, C. (2009). An introduction to multimodality. In C. Jewitt (Ed.), The Routledge
handbook of multimodal analysis (pp. 14–28). London: Routledge.
Kress, G. (2009). What is a mode? In C. Jewitt (Ed.), The Routledge handbook of
multimodal analysis (pp. 54–67). London: Routledge.
Kress, G. (2010). Multimodality. A social semiotic approach to contemporary commu-
nication. London: Routledge.
Kress, G., Jewitt, C., Bourne, J., Franks, A., Hardcastle, J., Jones, K., & Reid, E. (2005).
English in urban classrooms: A multimodal perspective on teaching and learning.
New York, NY: RoutledgeFalmer.
Kress, G., Jewitt, C., Ogborn, J., & Tsatsarelis, C. (2001). Multimodal teaching and
learning. The rhetorics of the science classroom. London: Continuum.
Kress, G., & van Leeuwen, T. (2006). Reading images. The grammar of visual design
(2nd edition). London: Routledge.
Kress, G., & Selander, S. (2012). Multimodal design, learning and cultures of recog-
nition. The Internet and Higher Education, 15(4), 265–268.
Lemke, J. L. (1990). Talking science. Language, learning and values. Norwood: Ablex
Publishing.
Lemke, J. L. (1998). Multimedia literacy demands of the scientific curriculum. Lin-
guistics and Education, 10(3), 247–271.
Martin, J. R., & Rose, D. (2008). Genre relations. Mapping culture. London:
Equinox.
Martinec, R. (2004). Gestures that co-occur with speech as a systematic resource. The
realization of experiential meanings in indexes. Social Semiotics, 14(2), 193–213.
http://dx.doi.org/10.1080/1035033042000238259
Norris, S., & Phillips, L. M. (2003). How literacy in its fundamental sense is central to
scientific literacy. Science Education, 87(2), 224–240.
O’Halloran, K., Podlasov, A., Chua, A., & K.L.E., M. (2012). Interactive software for
multimodal analysis. Visual Communication, 11(3), 363–381.
Schleppegrell, M. (2004). The language of schooling: A functional linguistics perspective.
Mahwah, NJ: Lawrence Erlbaum.
Shanahan, T., & Shanahan, C. (2008). Teaching disciplinary literacy to adoles-
cents. Rethinking content-area literacy. Harvard Educational Review, 78(1),
40–59.
Simpson, A., & Walsh, M. (2010). Multiple literacies: Implications for a changed
pedagogy. In F. Christie, & A. Simpson (Eds.), Literacy and social responsibility.
Multiple perspectives (pp. 24–39). London: Equinox.
Taber, K. S., & Watts, M. (1996). The secret life of the chemical bond. Students’ anthro-
pomorphic and animistic reference to bonding. International Journal of Science
Education, 18(5), 557–568. http://dx.doi.org/10.1080/0950069960180505
Tang, K.-S., Delgado, C., & Moje, E. B. (2014). An integrative framework for the
analysis of multiple and multimodal representations for meaning-making in
science education. Science Education, 98(2), 305–326. http://dx.doi.org/10.1002/
sce.21099
 K. Danielsson / Linguistics and Education 35 (2016) 88–99 99

Tang, K.-S., Tan, S. C., & Yeo, J. (2011). Students’ multimodal constructs of the work-
energy concept. International Journal of Science Education, 33(13), 1775–1804.
http://dx.doi.org/10.1080/09500693.2010.508899
Unsworth, L. (2001). Teaching multiliteracies across the curriculum: Changing con-
texts of text and image in classroom practice. Buckingham: Open University
Press.
van Leeuwen, T. (2005). Introducing social semiotics. New York: Routledge.
Wellington, J., & Osborne, J. (2001). Language and literacy in science education. Buck-
ingham: Open University Press.
Kristina Danielsson, PhD, has a position as professor in reading and writing devel-
opment at Stockholm University and in Swedish at the Linnaeus University. Her
main research area is disciplinary literacy with a special focus on multimodality
and learning in the science disciplines.