Disco Divas and Heroic Knights: A critical multimodal analysis of gender roles in 'create the world' LEGO cards

Abstract

This chapter investigates how gendered identities are constructed in a promotional series of collectable cards, co-created by Sainsbury's (a UK Supermarket) and LEGO (an international toy manufacturer). It focuses on the representation of LEGO minifigures (small human-like toy figurines), presenting a systematic analysis of the way these figures are named, visually depicted and linguistically described on the cards. The findings reveal, first, that female is the marked form for a LEGO minifigure; female minifigures are represented half as often as male minifigures, and indices of gender are more compulsory for female characters. Second, the analysis shows that male and female minifigures are differentiated in distinct and restrictive ways, with women generally represented in terms of who and what they are (i.e. youthful and slim, with accentuated facial features) but men by what they do (i.e. being adventurous, taking risks, and having an occupation, especially one that relates to manual labour). These findings highlight some of the ways in which restrictive norms and sexist ideologies can be perpetuated through products targeting young children. They also demonstrate the importance of in-depth analysis and critique of such products, as a resource for resisting and challenging harmful and limiting gendered norms.

Full text

Chapter 4, in Innovations and Challenges in Language and Gender: Women and Sexism. Ed.
Carmen Caldas-Coulthard.
Disco Divas and Heroic Knights:
A critical multimodal analysis of gender roles in ‘create the world’ LEGO cards
Jai Mackenzie, University of Nottingham
Laura Coffey-Glover, Nottingham Trent University
Sophie Payne, University of Reading
Mark McGlashan, Birmingham City University
Abstract
This chapter investigates how gendered identities are constructed in a promotional series of
collectable cards, co-created by Sainsbury’s (a UK Supermarket) and LEGO (an international
toy manufacturer). It focuses on the representation of LEGO minifigures (small human-like
toy figurines), presenting a systematic analysis of the way these figures are named, visually
depicted and linguistically described on the cards. The findings reveal, first, that female is the
marked form for a LEGO minifigure; female minifigures are represented half as often as male
minifigures, and indices of gender are more compulsory for female characters. Second, the
analysis shows that male and female minifigures are differentiated in distinct and restrictive
ways, with women generally represented in terms of who and what they are (i.e. youthful and
slim, with accentuated facial features) but men by what they do (i.e. being adventurous, taking
risks, and having an occupation, especially one that relates to manual labour). These findings
highlight some of the ways in which restrictive norms and sexist ideologies can be perpetuated
through products targeting young children. They also demonstrate the importance of in-depth
analysis and critique of such products, as a resource for resisting and challenging harmful and
limiting gendered norms.
Introduction
In 2017, LEGO teamed up with major UK supermarket chain Sainsbury’s for the distribution
of a set of 140 ‘create the world’ collectible trading cards. Children were invited to join ‘Lily’
and ‘Sam’ on a round-the-world adventure, with each card featuring a minifigure (small
human-like toy figurine) or LEGO creation that they would encounter on their journey. The

cards captured a range of the minifigure characters offered by LEGO at the time, and many of
these minifigures displayed roles, occupations, hobbies and personality traits associated with
‘real’ life, such as Grandpa, Nurse, Skier and Clumsy Guy. The promotion was extremely
popular, and garnered attention from UK regional and national newspapers (Birmingham
Mail, June 13, 2017; Gazette Live, May 8, 2017; The Sun, June 28, 2017). In this chapter, we
investigate the extent to which the minifigures represented in the ‘create the world’ set are
gendered, and what particular roles, identities and actions are attributed to these figures. We
consider what children may learn about gender roles through playing with the cards, and
whether those lessons are likely to help or hinder young people in moving beyond the
limiting roles and commonly-held stereotypes that perpetuate sexism in the adult world.
We take the position that toys are significant artefacts for social semiotic analysis because
they facilitate the imaginative play through which children learn and become socially
aware (Caldas-Coulthard and van Leeuwen 2002; Kahlenberg and Hein 2010). Toys can
also be seen as ‘texts’, with designs that “communicat[e] symbolic and interactive
meanings” (Caldas-Coulthard and van Leeuwen 2003, 19). We therefore suggest that this
is an important area of research. Nevertheless, there has been little social semiotic work to
date that focuses on children’s toys and products, with the exception of Caldas-Coulthard
and van Leeuwen’s analyses of baby toys (2001), action figures (2002) and teddy bears
(2003). Through playing with toys such as the ‘create the world’ cards, children develop
their sense of what social roles and relationships are available to them, and what is
‘normal’, expected or desirable behaviour. For example, they may use the representations
on the cards as starting points as they engage in role play, building worlds, relationships
and qualities for the minifigures alongside their developing sense of their own worlds.
Given LEGO’s worldwide popularity and success – they reported their highest revenue in
85 years and introduced 355 new products in 2017 alone (The Telegraph, March 9, 2017)
– it is especially important to consider the socialising potential of their products. By
examining these cards, we hope to reveal the messages that children absorb when they
‘create the world’ through reading, collecting and playing with them, and to consider
whether these messages may perpetuate discriminatory norms and practices.
Gender and children’s toys
In recent years, toys have increasingly been developed and marketed for gender-
differentiated audiences, with distinctive sets of gendered semiotic resources coming to

signal whether a toy is for boys or girls (Auster and Mansbach 2012; Kahlenberg and Hein
2010; Martinez et al. 2013). For example, toys marketed towards boys typically include
action figures, vehicles, building toys and weapons, thus promoting strength, outdoor play
and physicality, whilst ‘girls’ toys’ tend to include dolls, cosmetics and jewellery, promoting
the values of beauty, indoor play, domesticity and motherhood (Auster and Mansbach 2012;
Caldas-Coulthard and Van Leeuwen 2002; Kahlenberg and Hein 2010; Martinez et al. 2013).
In terms of colour, boys’ toys (and the boys who play with them) tend to be depicted in
colours such as red, blue, black or grey, whilst girls’ toys (and the girls who play with them)
are usually associated with pastels such as purple and pink (Auster and Mansbach 2012;
Caldas- Coulthard and Van Leeuwen 2002; Kahlenberg and Hein 2010). The pervasive
association of boys with blue and girls with pink has led some to suggest that there is now a
ubiquitous visual discourse of gender difference (Baker 2008, 96), and that the visual
(re)production of such difference serves to enforce and perpetuate limiting stereotypes
(Cunningham and MaCrae 2011, 608). Pink is shown to be the most gender-marked colour
for toys, being strongly associated with girls (Auster and Mansbach 2012; Cunningham and
MaCrae 2011; Wong and Hines 2015). This is consistent with Koller’s (2008) claim that
pink is also used pervasively in marketing that targets women.
LEGO and LEGO bodies
Historically, LEGO has been a toy that boys play with: in 2011, boys made up 90% of
LEGO’s customer base (The Guardian, June 4, 2017). However, in recent years, the
company has made a number of efforts to target girls. For example, they moved to increase
the representation of women in STEM professions, and present women in more intellectually
challenging roles, through their introduction of the Research Institute set in 2014, which
featured three female scientist figures. In 2017, they went a step further with the Women of
NASA set, which included four female scientists who made significant contributions to major
NASA missions. Despite their popularity and apparently progressive aims, however, the
LEGO group have often been the subject of controversy with regard to the way they represent
different groups in society. For example, the ‘LEGO Friends’ range, which was launched in
2012 and includes ‘mini-doll’ figures and sets with themes such as a bakery, riding camp and
amusement park (The Guardian, June 4, 2017; Johnson 2014), was criticised for the over-use
of pastel colours and promotion of stereotypical female roles, with some even condemning
the range as a regression to 1970s gender stereotypes (The Telegraph, June 1, 2017). This
range changed the shape of the minifigure for the first time, with its exclusively female

characters including slimmer bodies, shaped breasts and small waists. Similar markers are
now painted on to the standard body shape for female minifigures in the mainstream range,
although there have historically been far fewer of these explicitly marked female minifigures
than the standard (presumably male) form (Johnson 2014).
In this chapter, we are particularly interested in LEGO’s representation of human or human-
like figures in children’s products. Although there has been little social semiotic research that
explores children’s toy figurines, Caldas-Coulthard and van Leeuwen (2002) have analysed
the popular action figure franchises of Action Man and Barbie, focusing on the toys’ design
and movement, use of colour and the language used in advertising materials. They observe
that these semiotic resources are harnessed to demarcate clearly gendered roles for the toys:
Action Man appears in professional or action-oriented settings, whilst Barbie occupies settings
and roles associated with stereotypical femininity such as shopping, hairdressing, nursing and
motherhood. They show that naming strategies for Barbie are linked with idealised notions of
womanhood and romance (e.g. Blushing Orchid Barbie, Harpist Angel, Summer Dream),
whereas Action Man’s personae relate to power, danger and authority (e.g. Dr X, Bungee
Jumper, Ninja). The world of Action Man is also characterised by dark colours ‘evoking
mystery or danger’, whereas Barbie’s world is populated by pinks and purples which create ‘a
sense of romance’ (Caldas-Coulthard and van Leeuwen 2002, 102). LEGO minifigures are
quite different from these other action figure toys in size, feel and playability. Most notably,
and in keeping with LEGO’s primary remit as a building toy, the legs, bodies, heads and hair
can be taken apart and reconstructed, enabling different assemblages to be imagined and
constructed. The LEGO company themselves seem to encourage play with the bodies of these
figures. For example, the ‘build a minifigure’ feature in large LEGO stores invites customers
to create their own characters from a range of different body parts. In 2017, the Manchester
LEGO Discovery Centre had a ‘gay pride’ display in which traditional gender norms were
subverted through the configuration of minifigure parts. One minifigure in this display, for
example, had a bearded face, long blonde hair and wore a bikini.
The ‘create the world’ minifigure cards may be a small data set, but they have been extremely
popular with young children in the UK, are part of a large, successful franchise that has made
efforts to break gendered stereotypes, and represent many real-world characters and roles.
We therefore believe that they are an important site for influencing young people’s
developing concepts of their own, and others’, position in the world. In producing these

cards, LEGO and Sainsbury’s have the power to offer transformative positions; the creative
potential of the malleable minifigures and the inclusion of many fantasy characters means
that they have plenty of opportunities to transgress restrictive social norms and boundaries or
damaging stereotypes. This chapter will critically examine whether that is achieved.
Methodology
Data collection
The data for this study comprise the names, visual depictions and textual descriptions from a
set of 140 ‘create the world’ LEGO collectible cards. These cards were the product of a joint
promotion with the UK supermarket Sainsbury’s, and were distributed to customers who
spent over £10 between 3 May and 13 June 2017. They could also be bought from
Sainsbury’s for 50p, along with a £2 collector’s album to display the cards. We only included
cards that featured minifigures for our analysis, resulting in a data set of 104 cards. We
documented these data by first photographing the collection, and then transcribing the textual
data, which included the card number, title and full description, into a spreadsheet. This
textual dataset amounted to 2,265 words.
Each of the minifigure cards is composed in the same way (see Figure 1
i
). The visual
depiction of a minifigure occupies the centre position, where it is framed by a range of
marginal elements, including strongly demarcated frame lines and textual information such as
the name of the minifigure character, a description of that character, and the number of the
card in the series. For example, the Boxer is the central figure of card number 11, framed by
a red background with a faded plant motif. Below the image is the typed description (see
Figure 2). Because it is at the centre of the composition, it may be argued that the image of
the minifigure is attributed a greater degree of salience than the other semiotic elements, and
thus has high ‘information value’ (Kress and van Leeuwen 2006, 177). The prominence of
the visual is not surprising given the young audience for the cards, many of whom will be
emergent readers, and pay little attention to the linguistic elements
ii
. However, the
construction of information value in the cards can also be analysed from top to bottom, with
what is nearer the top taking the ‘ideal’ position - the ‘generalized essence of the
information’ (Kress and van Leeuwen 2006, 187), and the information at the bottom being
positioned as more ‘real’, detailed, specific information. In this respect, the linguistic
resources mobilised in the lower portion of the cards also take a prominent position, and are

likely to be read by older children, or adult carers, in prolonged engagement with the cards.
Because prominence is given to both visual and linguistic modes in the depiction of the
minifigures, our analysis takes account of the range of semiotic resources that are deployed
in relation to both of these modes.
Figure 1. Composition of a LEGO ‘create the world’ card
Figure 2. Card 011: BOXER
Before analysing the content of the cards, we first attempted to categorise each of the
minifigures represented therein as either male, female or gender-neutral. These identifications
were checked by all four authors of this chapter and it was agreed that all representations
could be identified as either female minifigures (henceforth FMs) or male minifigures
(henceforth MMs). We then created two sub-corpora for the text of the FM and MM cards.
Our coding of the minifigures by gender was checked by a group who represent the target
audience for these cards: ten children between the ages of 6 and 9
iii
. Each of these children

was separately shown each minifigure card in turn and asked to state whether the figure on
the card was male, female, or they couldn’t say, which would allow for interpretations of

gender-neutrality or non-binary gender. Overall, there was 99% agreement between the
authors and the children. The generally agreed lack of gender-neutral or non-binary
minifigures is indicative of the binary world depicted on the cards.
The Social Actor Framework
Our analysis of the ‘create the world’ minifigure cards focuses on the following research
questions:
1.
Are the ‘create the world’ minifigures differentiated by gender? If so, what gendered
identities and roles are constructed for them?
2.
What actions and functions do the minifigures perform?
3.
What wider social norms are indexed and (re)produced through these depictions?
We seek to answer these questions through a multimodal social actor network analysis (van
Leeuwen 2008). In doing so, we treat the minifigures as ‘social actors’, on to which an
identity and a place in society is projected, and the cards as recontextualisations of social
practices (van Leeuwen 2008). The social actor framework supports our exploration of the
kinds of socio-cultural norms and expectations at work in the construction of the minifigures
and, importantly, what they are teaching the children who play with them about their place in
the world.
In order to theorise the connections between particular visual and linguistic resources, we
draw on the concept of indexicality. An index is a type of sign that stands for something else.
It is distinguished from other signs because it stands for an object ‘by virtue of a real
connection with it’ (Peirce, 1998 [1895]: 14), or as Atkin (2005: 163) puts it, ‘through some
existential or physical fact’. Peirce (1998 [1895]) offers as an example ‘a low barometer with
a moist air’ as an index of rain; Atkin (2005) suggests that smoke is an index of fire. Both of
these indices have a causal connection with the phenomenon to which they point.
In this chapter, we do not employ the concept of indexicality in relation to physical objects or
environmental phenomena such as fire and rain, but in relation to social constructs, especially
gender. Ochs (1992) has developed the concept of indexicality in this respect, distinguishing
two types of index. ‘Direct’ indices, such as the pronouns ‘he’ and ‘she’, or the explicitly
gendered categories ‘man’ or ‘waitress’ (all of which are used in the minifigure cards),

directly point to the gender of the referent. With ‘indirect’ indices, which are of particular
interest here, the connection between the sign and the individual, group or meaning for which
it stands is made through shared cultural knowledge and assumptions. As an example, the
word ‘glamorous’ is associated with women in a western cultural context, whereas ‘stocky’
tends to be reserved for men. Visual signs can also carry indexical meanings. For example, as
shown above, the colour pink has come to be strongly associated with women and girls. The
connection between indirect indices and the social meanings to which they point are not
physical. Nor are they static or universal in their reference points - As Bucholtz (2009) and
Johnstone and Kiesling (2008) have shown, the indexical meanings of particular signs are
variable and context-dependent. The words dude and güey, for example, have both been
associated with male referents, but in certain contexts these words can index a ‘stance of cool
solidarity’ (Kiesling 2004: 282) and a ‘hip urban Latino identity’ (Bucholtz 2009: 158) that
points not only to the gender of the referent, but also their age, ethnicity, and the stance of the
speaker in relation to them. Our analysis is therefore sensitive to the context-specific social
meanings that are indicated through a range of semiotic resources in the ‘create the world’
cards, and to the potential relevance of other factors such as age and body shape.
The analysis that follows will focus on the categorisation of the minifigures, which relates to
how identities, roles and actions are represented (van Leeuwen 2008). We explore these
categorisations through close examination of the names given to the minifigures in the titles,
the visual images and colours (including the minifigure itself and the background) and the
language of the descriptions at the bottom of each card. Within van Leeuwen’s (2008)
framework, categorisation is divided into several sub-types. We focus here on those types
that are most prominent in the collection: functionalisation and identification. When social
actors are functionalised, they are categorised according to their functions – what they do –
such as a paying job, or some other societal role (van Leeuwen 2008, 42). Identification
categories, on the other hand, categorise actors according to what they ‘unavoidably are’ (van
Leeuwen 2008, 42). In relation to identification, we explore the sub-categories of
classification, whereby actors are divided according to the contingent socio-historical
categories that a given society or institution uses to group and separate individuals, and
physical identification, which categorises actors according to their physical appearance or
part of their physical body (van Leeuwen 2008, 44).

Although van Leeuwen’s (2008) initial description of the social actor framework focuses on
linguistic realisations, he also shows that it can be adapted to analyse visual
representations. Our visual social actor analysis focuses on how the minifigures are marked
out as male or female through semiotic resources such as the colours used on the card
backgrounds and minifigure bodies, and the markings on the faces and bodies of each figure.
We consider whether meaningfully placed lines (implying, for example, make-up, facial hair,
wrinkles, body fat or a slim waist), are indexical of gender, and other macro-social categories
that may intersect with gender, such as age. We also consider how the items that the
minifigures are depicted as wearing or holding contribute to an impression of the characters’
roles, activities they engage in, or who they essentially are.
Our analysis of the textual descriptions considers how the minifigures are linguistically
classified and identified through an exploration of adjectival descriptions. Using Wmatrix
(Rayson 2009), we compare adjectival descriptions that are distinctive to each gendered
group of cards, teasing out both qualitative and quantitative differences. We employ a
modified version of Moon’s (2014) model for classifying adjectival descriptions, grouping
them together according to the categories of ‘classification’, ‘physical characteristics’,
‘personal characteristics’ and ‘appraisement’. These also map onto the categorisation types in
van Leeuwen’s (2008) framework. Where adjectives serve multiple functions (e.g. describing
a personal characteristic in addition to appraisement), they are coded in every relevant
category.
Assigning the titles, visuals and descriptions of the minifigure cards to analytical categories
was not always a straightforward process. In order to achieve some consistency and
reliability, two authors of this chapter conducted each analysis separately, before coming
together to discuss any divergent categorisations or problematic cases. This process of inter-
rater coding helped us to specify and refine the boundaries of each analytical category.
Analysis
The minifigure cards can be grouped into two gender types, male and female, and are
recognisable as such by adults and children alike, as explained in the previous section. An
important point to note from the outset is that gender representation in the ‘create the world’
cards is not equal, with 67% (N70) of the 104 minifigures being identified as male, but only

33% (N34) as female; male minifigures (MMs) appear twice as often as female minifigures
(FMs). This gender imbalance suggests that male is the default gender for a minifigure in this
set, whilst the female form is marked (see Mills 2008 on the prevalence of this pattern in a
wider social context). In the analysis that follows, we consider how the gender of the ‘create
the world’ minifigure characters is indexed through the full range of semiotic resources
available in the cards, through their titles, the visual resources employed in the images, and the
linguistic choices made in the textual descriptions. As the section progresses, we draw out key
themes, showing what kinds of gendered identities and roles are constructed overall in the
‘create the world’ set.
Identification: classification and physical identification
Identification is the second most frequent type of linguistic categorisation in the naming of
the ‘create the world’ minifigures, appearing in 53% (N55) of the card titles
iv
. Classification
by social category is by far the most common sub-type of identification, with 51% (N53) of
the card titles featuring some form of classification. In this section, we begin by exploring
the ways in which the minifigures are classified in the card titles, identifying three main sub-
types of classification: gender, fantasy figure and personality. As shown in Table 1, we found
that gender classifications (e.g. Tiger Woman, Spooky Girl and Pizza Delivery Man) were the
most common type in the card titles, and that the FM titles were classified by gender more
than twice as often (proportionally) as the MM titles.
Classification Type
Proportion of
classifications
(total N53)
Proportion of
MMs (total N70)
Proportion of
FMs (total N34)
Gender
62% (N33)
20% (N14)
56% (N19)
Fantasy Figure
42% (N22)
21% (N15)
21% (N7)
Personality
17% (N9)
10% (N7)
6% (N2)
Miscellaneous (historical &
narrative figures, size, socio-
political group)
8% (N4)
3% (N2)
6% (N2)
Table 1. Minifigure card titles by classification

Table 2 shows the linguistic forms used to classify the minifigures by gender. These can take
the form of ‘highly generalised gendered classifications’ (van Leeuwen 2008: 42), such as
‘man’ or ‘woman’, which can appear as pre- or post- modifiers (Lady Robot, Spider Lady) or
as compounds (Cavewoman). They can also take the form of derivations, created through
suffixes such as ‘ess’, ‘let’ and ‘maid’, which in these cases have a diminutive effect,
implying youthfulness, innocence or powerlessness. Others are more specific gendered
categories that are part of a gendered pair, such as King, Queen and Witch (the male
counterpart of the latter would be Wizard). ‘Girl’ (N7), a category that indicates youth, and
‘lady’ (N3), which can either indicate status, or be used as a euphemistic alternative to
‘woman’ (avoiding perceived connotations of sexual maturity), are the two most common
gendered classifications used for FMs. By contrast, the equivalent denotative gendered
counterparts for these categories - ‘boy’ and ‘gentleman’ are not used at all for the male
minifigures. Instead, ‘guy’ (N7), a category that is increasingly used in a gender-neutral way
in its plural form, yet at the time of writing still retains its gendered meaning in the singular,
is the most common gendered classification for the MMs, followed by ‘man’ (N6). These
findings reveal that the gender of the FMs is more often linguistically marked, and that
gendered labels for the FMs often imply youthfulness or innocence in a way that the MM
titles do not (see Mills 2008 for further discussion of inequalities in gendered categories).
Form of gender classification
Proportion of MM
gender classifications
(total N14)
Proportion of FM gender
classifications (total N19)
Generalised category: guy
50% (N7)
-
Generalised category: girl
-
37% (N7)
Generalised category: man
43% (N6)
-
Generalised category: lady
-
16% (N3)
Generalised category: woman
-
11% (N2)
Other gendered category
(king, queen, princess, witch)
7% (N1)
16% (N3)
Suffixes (-ess, -maid, -let)
-
21% (N4)

Table 2. Gender classification types in minifigure card titles
When it comes to the visual resources and textual descriptions that are used to represent the
minifigures, appearance takes primacy, and is closely related to the characters’ gender
classification. Our analysis now turns to the ways in which the minifigures are identified by
their appearance through a range of visual and linguistic resources.
Women as youthful with striking features, men as older with hairy faces
Visually, the physical attributes given to the minifigures sharply differentiate the FMs and
MMs. For example, facial hair (i.e. moustaches, beards and sideburns) is only found on those
minifigures linguistically identified as male. In addition to visual signs of facial hair,
linguistic references to hair in the textual descriptions are also restricted to the MMs. The
adjectival descriptions of MMs’ facial hair often carry positive evaluations, with the
adjectives ‘sensational’, ‘magnificent’ and ‘sweet’ premodifying ‘moustache’, ‘facial hair’
and ‘beard’ in descriptions of the Ringmaster, Evil Dwarf and Garden Gnome, respectively.
The one reference to ‘hair’ in the FM descriptions relates to head hair from the Medusa card,
referring to its functional attributes for this dangerous fictional character: “And don't offer to
cut her hair either - it's got some 'sssserious' bite!”
Just as the FMs are more often linguistically classified as female, they are also more often
visually identified as female through their physical appearance; all of the FMs have facial
markers that index femininity (i.e. implied make-up such as coloured lips and prominent
eyelashes), whereas only half of the MMs have facial markers that index masculinity
(moustaches, beard stubble, sideburns or bushy eyebrows
v
). However, since all the FMs’
gender is marked through implied make-up, the very absence of feminine indices can be said
to distinguish a minifigure as male. This lends further weight to the argument that male is the
default gender for a minifigure in this set, whilst the female form is marked.
Another more subtle distinction between representations of the minifigures’ faces is that the
MMs are more likely to be marked with signs of being older. This is indicated by lines
around the face in positions where we would typically expect to see wrinkles, namely across
the forehead and next to the eyes. Several of the MMs include such marks, whilst only one
of the FMs (Grandma) does. Indices of age therefore seem to intersect with gender here. The

implication that the MMs are older than the FMs is supported by the fact that many of the
MMs have white or grey hair whereas only two of the FMs do
vi
. Those playing with the
cards may therefore infer that, whilst it is normal and acceptable for men to age, women are
obliged to retain their youth.
Women as slim, curvaceous and well-dressed, men as larger and muscular
In terms of the minifigures’ bodies, only a small number of the MMs have visual indices of
gender, including chest hair and visible muscles. By contrast, the majority of the FMs have
visual indices of gender on their body, namely narrowed waists and/or breasts, making their
figures more curvaceous – less square – than the MMs. Of the 6 who do not have (visibly)
narrowed waists, in two cases the bodies are obscured so the waists cannot be seen, two are
wearing costumes and one is a robot. Additionally, several of the MMs have lines on their
bodies that imply they have fat around their waists, and/or their clothes are baggy, whilst
none of the FMs’ bodies include such marks
vii
. For example, the lines around the bottom of
Baseball Player’s body suggest he has a tucked-in baggy t-shirt, the lines above the waists
of Grandpa, Piggy Guy and Prospector imply a rounded stomach, and the lines around the
belts of Elf and Gnome imply body fat around a tightened area. The way the minifigures’
upper bodies are presented therefore suggests that it is normal and acceptable for men to
have muscles and body fat, whereas women are expected to have small waists and visible
breasts. Furthermore, the textual descriptions of the minifigures suggest that appearance in
general is more salient and important for women. Adjectives describing clothing, such as
“royal robes” (Queen) and “her robe is perfect” (Kimono Girl) are more frequent in the FM
descriptions compared with the MM descriptions
viii
. Other adjectival descriptors distinctive
to the FM set, such as ‘glitter-filled’ (Unicorn Girl) and ‘dressed-up’ (Bride), as well as
references to jewellery such as ‘sparkly stuff’ (Jewel Thief) also imply attentiveness to
looking special and striking through clothing and accessories. The inclusion of ‘glitter’ as a
premodifier is particularly notable in this regard, given its cultural associations as an
indirect index of femininity. Adjectives relating to physical appearance in the MM set are
more likely to focus on facial hair (see above).
Pink and blue as visual discourses of gender differentiation
Another feature of visual categorisation that is largely restricted to the FMs is the inclusion of
shades and hues of pink. Over half of the FM cards feature pinks and purples somewhere in
the visual composition of the cards (mostly in their clothing, but sometimes in the

background or number colour), compared with only a handful of the MMs, one of which is
Piggy Guy, whose outfit is the colour in which pigs are typically represented. Interestingly,
two of the children who acted as second raters for this study identified Piggy Guy as female,
presumably because of the strong association of pink with girls and women. The importance
of pink as a resource for the ideational representation of femininity is particularly notable
when considering the representation of non-human characters. As well as being linguistically
marked for gender, the Alien Villainess, Lady Robot, and Lady Cyclops cards are also
visually marked through the use of pink and feminising facial features, including lipstick
(which is always pink) and exaggerated make-up, such as Lady Robot’s pink cheeks.
Moreover, pink is integrated into the dress of Alien Villainess and Lady Robot. The most
frequently used colour on the MM cards, on the other hand, is blue (closely followed by
green); less than half of the FM cards feature these colours. Overall, the use of pink (or
purple) for the FMs and blue (or green) for the MMs reinforces and reproduces a visual
discourse of gender difference (Baker 2008)
ix
. In addition, the higher frequency of both
linguistic and visual indices of gender for the FMs, together with the fact that there are
double the amount of MMs in total, combine to create the very powerful impression that
female is the marked form for a LEGO minifigure.
Functionalisation
Functionalisation is the most frequent type of linguistic categorisation in the naming of the
‘create the world’ minifigures, with 62% (N64) of the card titles referring to the characters’
functionalised role (such as a paying job or other social activity). Slightly more MMs than
FMs are given a title that refers to a functional role, such as Janitor, Fisherman, Fortune
Teller and Disco Diva. These findings are echoed in the visual icons the minifigures are
shown to be holding, where again slightly more MMs than FMs hold an item that draws
attention to what the character does in their role, such as the Janitor and his mop, the
Rockstar and his guitar, the Cavewoman and her club, and the Nurse and her syringe
x
. These
findings suggest a slight disparity between the kinds of roles the minifigures are given, with
the MMs more likely to be categorised in terms of what they do in the world, and the FMs
more likely to be categorised in terms of who they are and what they wear. The difference
between the representation of the FMs and MMs becomes more marked when we consider
the more specific types of functionalised roles and occupations they are given. Table 3
presents a summary of these types in the minifigure card titles, organised by theme.

Theme
Proportion of MM
functionalisations
(total N42)
Proportion of FM
functionalisations
(total N17)
Entertainment
29% (N12)
18% (N3)
Manual labour
21% (N9)
-
Physical activity
19% (N8)
53% (N9)
War/Conflict
10% (N4)
6% (N1)
Adventure/travelling
14% (N6)
-
Crime
5% (N2)
12% (N2)
Service
5% (N2)
6% (N1)
Science
2% (N1)
-
Business
2% (N1)
-
Health
-
6% (N1)
Magic
-
6% (N1)
Table 3. Functionalisation in card titles, organised by theme
There are a number of functionalisation themes that apply significantly more often to MMs,
namely manual labour (e.g. Plumber and Welder), war/conflict (e.g. Knight and Trooper) and
adventure/travelling (e.g. Explorer and Sea Captain). The fact that multiple MMs, but almost
no FMs, have functionalised roles that relate to these themes implies that physical work,
conflict, adventure and travel are generally available to men, but not to women. Additionally,
a higher proportion of the MMs have a functionalised role that relates to an occupation
xi
.
This disparity is particularly apparent in the male-only manual labour category.

Men as daring and dangerous, women as fun, frivolous and decorative
There are a number of visual and linguistic markers that suggest the MMs are more
dangerous and aggressive than the FMs. For example, more MMs than FMs hold an item that
can be used to attack, such as Evil Dwarf’s axe, Heroic Knight’s sword and Alien Avenger’s
gun. Several MMs, but no FMs, hold some kind of tool that is used to cut or fix, such as
Mechanic’s spanner, Carpenter’s saw and Butcher’s knife. The visual implication that MMs
do dangerous things becomes even more apparent when looking at what they wear, with
many MMs wearing some kind of equipment that protects them from serious injury (usually
helmets), whereas only one FM wears such equipment (Snowboarder, who wears a helmet)
xii
.
The association of MMs with danger is also reflected in the textual descriptions of the
minifigures. For example, in the MM descriptions there are a small number of adjectives that
imply danger: Gangster is described metonymically as ‘dangerous’ (“there’s something
dangerous in that case and it’s not an out-of-tune violin!”); the Pirate Captain is characterised
as a “merciless, rotten old sea dog”, and Ocean King is described as a “hot-headed monarch”,
which implies he is easily angered. There are also some FMs whose descriptions imply they
are dangerous, but these characters are more likely to be mythical, suggesting that being
dangerous is not an everyday trait for human women. For example, Wacky Witch is
evaluated as “wicked”, and Medusa is described metonymically as dangerous through
reference to her hair: “don’t offer to cut her hair either, it’s got some ssserious bite!” Surfer
Girl, a human-like FM, is described as “hot-blooded”, but (by contrast with “hot-headed”)
this phrase has a semantic prosody of sexual, rather than dangerous, behaviour. This
implication of sexual behaviour is reinforced with the description “She certainly knows how
to make waves at the beach!”, which points not only to the literal waves that she rides, but
also the metaphorical waves of admiration that she inspires.
The functionalisation sub-category of ‘physical activity’ includes the largest number of FM
cards, with more than half of the female minifigures’ functionalised titles relating to physical
activity
xiii
. A third of the FMs with physical roles are dancers: performers who depend on
viewers’ evaluation for success. These functionalised roles are represented both
linguistically, through the form ‘dance + -er’ (Flamenco Dancer, Hula Dancer), and through
the minifigures’ clothing: a traditional red and black dress and fan for Flamenco Dancer, and
flower garland and grass skirt for Hula Dancer. The items held by the FMs engaging in
physical activity tend to be symbolic, decorative, or serve to enhance their performance in

some way. For example, Flamenco Dancer carries a fan, Hula Dancer holds maracas and
Disco Diva a microphone. In addition, Disco Diva is described in the textual descriptions as
“the belle of the glitterball”, which echoes the idiomatic English phrase ‘the belle of the ball’,
serving as an intertextual reference to the fairy-tale genre, where women traditionally occupy
more passive, decorative roles. Even though she is visually active, linguistically Disco Diva
is therefore represented as a passive object to be looked at. There are no visual or linguistic
references to dancing for the MMs who engage in physical activity. These minifigures
instead perform sports and leisure activities such as climbing (Mountain Climber), boxing
(Boxer) and weightlifting (Weightlifter).
Discussion and Conclusion
In the LEGO ‘create the world’ minifigure cards, a range of semiotic resources combine to
create a powerful impression of distinct and binary gender roles, with ‘femaleness’ being the
marked form. All of the minifigure cards in the 2017 collection are linguistically and visually
gendered in some way, there are twice as many MMs as there are FMs, and the gender of the
FMs is foregrounded twice as often, through a range of linguistic and visual indices. Linguistic
and visual patterns in the cards suggest that different roles, identities and activities are
available to men and women: FMs are represented as younger and slimmer, with more
emphasised facial features (such as eyelashes and lips), whereas MMs are depicted as
physically stronger, larger and more mature. MMs are also more likely to be functionalised in
more dangerous or physical occupations, whilst FMs take up less adventurous, and more
frivolous, roles.
It is worth pointing out that there are some exceptions to these general patterns. For example,
the MM cards Small Clown, Mime and Thespian break out of their typical gendered moulds
because they wear make-up or have slightly narrowed waists. There are also several FMs
who wield weapons (such as Alien Villainess, Lady Cyclops and Cavewoman), or who
engage in adventurous and/or dangerous activities (such as Surfer Girl and Snowboarder).
However, there are limited types of minifigures that are able to transcend the gender
boundaries that pervade the ‘create the world’ set. For example, most MMs possessing visual
characteristics more typical of FMs are performers. FMs who wield weapons are all either
fantasy figures (Alien Villainess, Lady Cyclops and potentially Tiger Woman, who has a
whip) or an exaggerated caricature of an historical figure (Cavewoman). This creates the
impression that gender norms can only be transgressed within the confines of role play and

fantasy. Even when depicted in violent fantasy roles, however, the femininity of the women
in the FM cards is emphasised. Indeed, indices of femininity are particularly prevalent in the
textual and visual representations of fantasy figures that are not from the human (or even
animal) world, such as Lady Robot, Alien Villainess, and Lady Cyclops. This excessive
indexing of femininity on fantasy figures restricts the potential of children’s imaginative and
fantasy play to go beyond the limits of their immediate social worlds.
Our analysis has systematically evidenced the (re)production of gender norms and
stereotypes in the ‘create the world’ cards, showing that this is one of many sites in which
children learn that their place in the world is sharply determined and restricted along
gendered lines. Whilst some of the minifigures do overcome dominant gender norms to an
extent, the cards overall are overwhelmingly reliant on restrictive, hegemonic gender norms.
The limiting and unbalanced nature of these constructions is not always immediately
apparent, but as we have shown, it can be revealed through the kind of critical multimodal
analysis deployed here. Given the important role toys play in children’s development, it is
imperative that we use methods like this to challenge discriminatory messages in children’s
toys and products, if we are to continue the work of transforming sexist ideologies in wider
society.
References
Atkin, A. (2005) ‘Peirce on the index and indexical reference’, Transactions of the Charles
S. Peirce Society 41(1): 161-188.
Auster, C. J., and Mansbach, C.S. (2012) ‘The gender marketing of toys: An analysis of
color and type of toy on the Disney store website’, Sex Roles 67: 375-388.
Baker, P. (2008) Sexed texts: Language, gender and sexuality. London: Equinox.
Bucholtz, M. (2009) ‘From stance to style: Gender, interaction, and indexicality in Mexican
immigrant youth slang’, in Jaffe, A. (ed.) Stance: Sociolinguistic Perspectives, 146-170. New
York: Oxford University Press.

Caldas-Coulthard, C. R. and van Leeuwen, T. (2001) ‘Baby’s first toys and the discursive
construction of childhood’, in Wodak, R. (ed.) Critical Discourse Analysis in Post Modern
Societies, in Folia Linguistica XXXV/1-2, 91-110. Austria: Vienna.
Caldas-Coulthard, C. R. and van Leeuwen, T. (2002) ‘Stunning, shimmering, iridescent: Toys
as the representation of social actors’, in Litosseliti, L. and Sunderland, J. (eds). Gender
identity and Discourse Analysis, 91-108. Amsterdam: John Benjamins.
Caldas-Coulthard, C. R. and van Leeuwen, T. (2003) ‘Teddy bear stories’, Social Semiotics
13(1): 5-27.
Cunningham, S. J., and MacRae, C.N. (2011) ‘The colour of gender stereotyping’, British
journal of psychology 102: 598-614.
Johnson, D. (2014) ‘Figuring identity: Media licensing and the racialization of LEGO
bodies’, International journal of cultural studies 17(4): 307-325.
Johnstone, B and Kiesling, S. (2008) ‘Indexicality and Experience: Exploring the
meanings of /aw/-monophthongization’, Journal of Sociolinguistics 12(1): 5-33.
Kahlenberg, S.G. and Hein, M.M. (2010) ‘Progression on Nickelodeon? Gender-role
stereotypes in toy commercials’, Sex Roles 62: 830-847.
Kiesling, S. F. (2004) ‘Dude’. American speech 79(3): 281-305.
Kress, G. and van Leeuwen, T. (2006) Reading images: The grammar of visual design. 2nd
ed. London and New York: Routledge.
Koller, V. (2008) ‘“Not just a colour”: Pink as a gender and sexuality marker in visual
communication’, Visual communication 7(4): 395-423.
Martinez, E., Nicolas, M.A. and Salas, A. (2013) ‘Gender representation in advertising of
toys in the Christmas period (2009-12)’, Comunicar 41(XXI): 187-194.

Mills, S. (2008) Language and sexism. Cambridge and New York: Cambridge
University Press.
Ochs, E. (1992) ‘Indexing gender’, in Duranti, A. and Goodwin, C. (eds) Rethinking
context: Language as an interactive phenomenon, 335–358. Cambridge: Cambridge
University Press.
Peirce, C. S. (1998) [1895] ‘Of reasoning in general’, in The Essential Peirce: Selected
Philosophical Writings. Volume 2 (1893 - 1913), edited by the Peirce Edition Project, 11–26.
Bloomington and Indianapolis: Indiana University Press.
Rayson, P. (2009) ‘Wmatrix: a web-based corpus processing environment’. Computing
department, Lancaster University. http://ucrel.lancs.ac.uk/wmatrix/.
van Leeuwen, T. (2008) Discourse and practice: New tools for Critical Discourse Analysis.
Oxford: Oxford University Press.
Wong, W.I. and Hines, M. (2015) ‘Effects of gender colour-coding on toddlers’ gender-
typical toy play’, Archives of Sexual Behavior 44, 1233-1242

i
Due to copyright restrictions, examples of the LEGO cards are included as illustrations. Neither
LEGO nor Sainsbury’s keep an accessible record of the 2017 cards, although a full list and selection of
images can be found at https://thecollector.io/features/2017/05/lego-create-the-world-complete-list/
ii
This claim is supported by the way a group of ten young second-raters engaged with the cards to
identify their gender: they all looked first at the image, then at the title, and finally, if necessary, at the
written description. Only three of the children read the written descriptions at all.
iii
This group consisted of five girls (one aged 6; three aged 7; one aged 9) and five boys (one aged 7; four
aged 9)
iv
Throughout this chapter, percentages are rounded to the nearest whole number.
v
This statistic includes four ‘fantasy figure’ MMs who have hairy faces.
vi
14% (N10) of the MMs have marks that imply wrinkles, compared with 3% (N1) of the FMs. 19% (N13) of
the MMs have white or grey hair, compared with 6% (N2) of the FMs.
vii
10% (N7) of the MMs and 82% (N28) of the FMs have visual indices of gender on their body. 11% (N8)
have lines on their body that imply body fat or baggy clothing.
viii
21% (N4) of the FM cards, and 1% (N1) of the MM cards, include adjectives describing clothing.
ix
59% (N20) of the FM cards and 6% (N4) of the MM cards feature pinks and/or purples; 69% (N48) of the
MM cards and 44% (N15) of the FM cards feature blues and/or greens.
x
60% (N42) of the MMs, and 50% (N17) of the FMs, have a title that refers to a functional role; 61% (N43)
of the MMs, and 50% (N17) of the FMs, hold functional items.
xi
53% (N37) of the MMs and 35% (N12) of the FMs are given occupational roles.
xii
14% (N10) of the MMs, and 9% of the FMs, hold an item which can be used to attack. 9% (N6) of the
MMs, but no FMs, hold a tool that is used to cut or fix. 17% (N12) of the MMs, but only 3% (N1) of the
FMs wear equipment that protects them from serious injury.
xiii
53% (N9) of the female minifigures’ functionalised titles relate to physical activity, compared with
19.04% (N8) of the MMs’ functionalised titles.