706074

research-article2017
NMS0010.1177/1461444817706074new media & societyZhao and Zappavigna

Article

new media & society

Beyond the self:

2018, Vol. 20(5) 1735­–1754
© The Author(s) 2017
Reprints and permissions:
Intersubjectivity and the sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1461444817706074
https://doi.org/10.1177/1461444817706074
social semiotic interpretation journals.sagepub.com/home/nms

of the selfie

Sumin Zhao
University of Southern Denmark, Denmark

Michele Zappavigna
The University of New South Wales, Australia

Abstract
As an iconic image of our time, the selfie has attracted much attention in popular
media and scholarly writing. The focus so far has been on the representation of the
self or subjectivity. We propose a complementary perspective that foregrounds the
intersubjective function of the selfie. We argue that the presence of selfhood is often
an assumption. What distinguishes the selfie from other photographic genres is its
ability to enact intersubjectivity – the possibility for difference of perspectives to be
created and this difference to be shared between the image creator and the viewer.
Based on a social semiotic analysis of selfies on Instagram, we identify four subtypes of
selfie, each deploying a combination of visual resources to represent a distinct form of
intersubjectivity. Our analysis suggests that the potential for empowerment is inherent
in the visual structure of the selfie, and that, as a genre, it is open for recontextualisation
across contexts and social media platforms.

Keywords
Instagram, intersubjectivity, selfie, social semiotics, visual grammar

Corresponding author:
Sumin Zhao, Department of Language and Communication, University of Southern Denmark, Campusvej
55, 5230 Odense M, Denmark.
Email: [email protected]
1736 new media & society 20(5)

Introduction
A defining image of 2016 US presidential race is a photograph of a crowd in a campaign
event in Colorado – largely made up of young female supporters – turning their back on
Democratic president nominee Hillary Clinton in order to capture selfies (Figure 1).
When the photo went viral on social media and hit international news front-pages, a
wave of online critique ensured, with a dramatic headline from the Telegraphy decrying
‘the age of the selfie’ (Graham, 2016). Many critics view selfie taking as an inward look-
ing and self-centred act, inhibiting the so-called Millennial who attended the event from
engaging with the broader social political world. This view of selfies as the epitome of a
shallow, self-absorbed and narcissistic modern mentality has dominated popular debate
and public discourse on the issue (cf. Miltner and Baym, 2015). In research literature,
despite calls for more nuanced understandings of selfies as emerging social practices, the
echo of this populist reading of selfies can still be identified. For instance, various
attempts have been made by psychologists (e.g. Sorokowski et al., 2015; Weiser, 2015)
to identify the association between self-photography and personality traits such as nar-
cissism. ‘Selfie humanitarianism’ has been criticised as the ‘reframing of “helping oth-
ers” in terms of entrepreneurial and narcissistic self work’ (Koffman et al., 2015: 158).
One of the assumptions at the core of such criticism is that the naturalised reading of
a selfie is ‘look at me’. In the context of the Colorado campaign event, a young woman’s
selfie with Clinton could thus be interpreted along the lines of ‘look at me not Hillary’ or
‘look, it is me who is with Hillary’. It is a proposition difficult to refute, as the self/sub-
ject or subjectivity is the primary representational object in selfies as a visual media
(Pham, 2015). Nevertheless, in this article, we would like to propose a complementary
reading – a young woman’s selfie with Clinton can also be read as ‘look, it is my perspec-
tive on Hillary (and the presidential campaign)’ or ‘let’s look at the Clinton campaign

Figure 1.  Crowds taking selfies at Clinton campaign in Orlando, Florida. (Barbara Kinney/
Hillary for America).
Zhao and Zappavigna 1737

through my perspective’. The essence of our understanding of the selfie resides in this
subtle shift of focus from ‘me’ to ‘my perspective’, and underpinning this shift are two
fundamental assumptions about the selfie.
First, embracing the Hallidayian social semiotic theory of communication (Halliday,
1978; Kress and Van Leeuwen, 2006) – a theory that explores how language, image
and other modes of communication such as music function in various social contexts
– we view the selfie as a visual sign or a (multimodal) textual artefact that simultane-
ously fulfils the function of representing the self/subject, and the function of enacting
certain social relations between the maker and the potential audience of the selfie (for
details, see section ‘From subjectivity to intersubjectivity: a social semiotic approach’).
From a social semiotic perspective, a selfie contains certain ‘social propositions’ to be
negotiated between the image marker and the viewer, and the visual structures of a
selfie will provide the textual clues for the types of ‘social propositions’ to be negoti-
ated. Second, acknowledging the intersubjective nature (Gillespie, 2009) of the selfie
as a visual sign, we suggest that what is being negotiated in a selfie is not the relation-
ship between the visually represented self and the viewers (i.e. me vs the other) but the
different perspectives or points of view on visually represented phenomena (i.e. my
perspective on me vs other’s perspective on me). Simply put, the ‘social proposition’
to be negotiated in a selfie is the relations between various perspectives. These per-
spectives can belong to the selfie taker/poster, the visually represented self, the pri-
mary audience (families, friends and followers on social media), the ambience audience
(Zappavigna, 2011, 2012), media commentators and social media onlookers, and so
on. These two theses provide the basis for contesting the mainstream reading of the
Colorado selfie, since the ‘social propositions’ being put forward by these young
women are not necessarily just self-representations (i.e. look at me) but also their ‘per-
spective’ on the social political world around them (i.e. look, this is my point of view).
In other words, when the young women look inwardly into ‘the self’, they also create
a means for the self to engage outwardly with the world. The gendered criticism of
young people’s (in particular young women’s) selfie practices (Albury, 2015), which
focuses solely on the inward ‘narcissistic’ nature of the selfie, could be interpreted as
a rejection of their perspectives articulated through the emerging visual media of the
selfie. In proposing this alternative reading, we endeavour to engage in a more nuanced
debate on the selfie, and open up the possibility for understanding the selfie as a means
for critical social engagement.
In the remainder of this article, we shall elaborate on these two basic arguments
and substantiate this social semiotic interpretation of the selfie. We will first contex-
tualise our study by mapping out the key themes in existing selfie scholarship. We
will then outline the key theoretical concepts of a social semiotic approach to the
selfie, focusing on interpersonal functions in images and the notion of intersubjectiv-
ity. We will also discuss the methods we adopted for mapping out various forms of
intersubjective relations. Following on from this, we shall look in details at the types
of visual resources used in selfies for construing different forms of intersubjectivity,
illustrated by examples from a qualitative sample of images. Finally, we will discuss
the implications of our approach for understanding the selfie as a ubiquitous social
practice.
1738 new media & society 20(5)

Selfie, the self and subjectivity

Historically, the self-portrait has regularly been at the forefront of cultural development,
exerting influence on how identity and selfhood are conceived within a culture (Hall,
2014). In contemporary contexts arising with the development of digital technologies,
the most prominent form of self-portrait is no doubt the selfie. The practices of selfie
making and sharing, as documented extensively in the research literature, exist in social
contexts as diverse as war and militarism (Brager, 2015; Kuntsman and Stein, 2015;
Silvestri, 2014), education (Charteris et al., 2014), pregnancy and breastfeeding (Boon
and Pentney, 2015; Tiidenberg, 2015b) and funerals (Gibbs et al., 2014; Meese et al.,
2015), and are being carried out by both private individuals, in particular young people
(Albury, 2015), and groups visible in news and entertainment media such as politicians
(Baishya, 2015; Coladonato, 2014) and celebrities (Collings, 2014; Nandy, 2015;
Shipley, 2015).
Despite its ubiquitous presence, to arrive at a definition of the selfie is not an easy
task. One of the commonly accepted criteria for defining a selfie is the representation
of the photographer in the image via the face or parts of the body, or through traces of
the body such as the shadow or reflection (Walker-Rettberg, 2005). Compositional
principles, or more precisely the deviation from the established professional princi-
ples, such as the Golden Ratio rules, have also been suggested as a distinctive feature
(Bruno et al., 2014). However, visual clues – be it representational or compositional
– alone are not sufficient for identifying a selfie. Recognising an image as a selfie
rather than a portrait can require viewers to ‘make inferences about the non-depictive
technocultural conditions in which the image was made’ and to have been socialised
into reading these types of image (Frosh, 2015: 1608). Many approaches to defining
selfie take into consideration the complex cultural, historical and technological factors
that give rise to modern selfies. Some view the selfie as a recent iteration of historical
genres such as self-portraiture in Western Art since the Renaissance (Fausing, 2014),
while others focus on its contemporariness, referring to it as a ‘platform-specific pho-
tographic genre’ (Meese et al., 2015: 1819), a ‘multimedia genre of autobiography or
memoir’ with ‘recognizable, though shifting, aesthetic parameters of posing and com-
position’ (Shipley, 2015: 403) and a ‘true vernacular genre’ (Walker-Rettberg, 2014:
9). For many scholars, however, the selfie is more than a genre or form of photography,
instead it is manifest as ‘social-technical assemblages’ that involve ‘entanglements of
subjectivities’ (Hess, 2015: 1629).
The complex nature of the selfie is perhaps best captured in Gunthert’s (2015)
description of a ‘bouquet of practices’ that are ‘reunited a posterior in a cultural con-
struction’ (p. 7). In the past few years, this complexity has given rise to multidiscipli-
nary study, spanning media studies, photographic theory, psychology and informatics.
This work has generated insights across a broad range of (emerging) social practices.
Some scholars have observed, for instance, the role of selfie in perpetuating consumer-
ism and facilitating the commodification different aspects of social lives (e.g. Kwon and
Kwon, 2015; Marwick, 2013). Others have commented on the way in which the selfie
politicises ‘discourses about how people ought to represent, document and share their
behaviour’ (Senft and Baym, 2015), and regulates the forming of gender, race and
Zhao and Zappavigna 1739

sexual identity (Burns, 2015; Williams and Marquez, 2015). While most research into
selfies tends to be qualitative, there is also some emerging work attempting to quantify
broader patterns of use such as the ‘selfiecity project’ (Tifentale and Manovich, 2015)
which ‘considers both the individual artistic intentions of a singular image and the over-
all patterns revealed by large amount of selfies made in a particular geographic location
during one week’ (Tifentale, 2014: 4).
One of the key themes in selfie scholarship concerns, unsurprisingly, the notion of
the self, and whether, depending on disciplinary orientation, it refers to notions of
personality, identity or subjectivity. Research psychologists, for example, have
attempted to determine whether selfies reflect particular personality traits, such as
agreeableness, conscientiousness, neuroticism and openness (Qiu et al., 2015).
Researchers adopting a (critical) socio-cultural perspective consider the selfie taker
‘not just a psychological type’ (Burns, 2015: 172). Rather, the process of selfie mak-
ing is seen as ‘reflexive processes of visual self-authoring’ (Lüders et al., 2010: 947),
and the end result – the selfie – as forms of personal quotidian narrative (Tiidenberg,
2015a) and personal confession (Hall, 2015). The authenticity of self-representation
in selfies is a contentious topic. It is often figured as having a temporal dimension
where selfies produced in ‘real-time’ engender forms of ‘liveness’ that bring an audi-
ence closer to the situation represented. For instance, Schleser (2014) considers the
selfie as a ‘micro-autobiography of the present moment’ (p. 155, italics added).
Hammelburg (2015) suggests that ‘stemfies’ (selfies taken by voters as they cast their
vote in an election) enact ‘the desire to be there “while it happens” and consequently
become part of a shared narrative, establishing the historicity that strengthens both
immediacy and affinity’ (p. 94). Similarly, the fleeting nature of Snapchat selfies has
been seen as an attempt at ‘a more authentic and raw representation of the self and the
moment’ (David, 2015: 91), as social streaming ‘brings the past, present, and future
of many users closer together as a simultaneous duration of multiple temporalities’
(Hochman, 2014: 3).
The notion of the selfie as an authentic form of self-representation, however, has been
questioned by many scholars, who points to its potential for generating/reinforcing dis-
courses and discursive norms of sexuality, and gender and race identities (e.g. Albury,
2015; Tiidenberg, 2015b; Williams and Marquez, 2015). This normative function is
highlighted when the selfie practices of different genders are compared. In her study of
young people’s use of selfies in sexting practices, for instance, Albury (2015) observes
that the self-representation of the body is a gendered practice. The discursive norms sur-
rounding the sharing are different for young women and young men. Young men’s selfies
are not subject to the same level of adult surveillance and peer condemnation that young
women’s self-representations (of the body) attract. Burns (2015) makes a similar obser-
vation in his comparison of female and male selfie practices:

By being discursively constructed as both problematic and feminine, the selfie enable the
targeted discipline of young women by perpetuating stereotypes, maintaining hierarchies, and
normalizing the punishment and correction of subjects perceived to be abnormal. Once the
subject is discursively brought into being as devalued, selfie taking then becomes an accepted
indication of a subject in need of regulation … (p. 1730)
1740 new media & society 20(5)

This emphasis on the discursive function of the selfie underpins our analysis in this
article. Of particular interest are three specific perspectives. The first is Gunthert’s (2015)
observation regarding the conversational use of the selfie – that it is

… specifically addressed to a receiver. The image here becomes a visual message, which
interpretation depends heavily on the triangle formed by its sender, the occasion represented
and the intended recipient. (pp. 8–9)

The second is Pham’s argument that the selfie has ‘moved us into a new visual
paradigm where the relations of power between the object of looking and the looking
subject are significantly more dynamic’. While the two remarks are not necessarily
comparable, both highlight the potential of selfie to discursively enact certain social
relations between the represented self, the maker of the selfie and the viewer. A third
perspective that has provided inspiration for us is Bellinger’s (2015) argument that
the selfie ‘invokes a particular stance towards representation’ (p. 1815). In his analy-
sis of a controversial selfie posted by former UK Prime Minister David Cameron,
Bellinger (2015) argues that the selfie is ‘constituted not only by a particular set of
visual markers but also by a specific relationship to photographic representation’ (p.
1809). We shall re-interpret ‘this relationship to photographic representation’ through
the theoretical lens of social semiotics, as well as the concept of intersubjectivity in
the following sections.

From subjectivity to intersubjectivity: a social semiotic

approach
The starting point of the discussion in this section is Gunthert’s (2015) observation that
the selfie is a visual message addressed to a certain audience in the context of social
media. From a social semiotic perspective, the selfie, as a visual message, simultane-
ously construes three kinds (or dimensions) of meaning:

•• Ideational: the representation of the self/subject;

•• Interpersonal: the enactment of social relations;
•• Textual: the organisation of the information in the image.

These three dimensions of meaning are referred to as metafunctions, a notion devel-

oped originally in Halliday’s (1978) description of language and later adopted by Kress
and Van Leeuwen (2006) in their work on visual ‘grammar’. While majority of the exist-
ing selfie literature can be seen as having addressed the ‘ideational’ or ‘representational’
aspect of the selfie, our focus in this article is on the interpersonal or interactional.
According to Kress and Van Leeuwen, the interactional structure of an image enacts
an imagined relationship between the visually represented participants and the viewer. In
a photo, there are several resources that can be used to construe this imagined relation-
ship, including gaze (if the visually represented participant is looking at ‘us’ – the viewer,
and if they are looking at us at an eye level or from above/below) and social distance (the
distance at which a person is being photographed).1 To use Dorothy Lange’s famous
Zhao and Zappavigna 1741

Figure 2.  Examples of interactional resources in images: Portrait versus Selfie – (a) Dorothy
Lange (1936) Migrant Mother and (b) Selfie: an example.

photo Migrant Mother as an example, Figure 2(a), we observe the mother at a close
social distance and at the eye level. She is looking away rather than gazing at us. As a
result, the image can be seen as creating an imagined social relationship in which we are
positioned as an ‘acquaintance’ of the mother and a sympathetic onlooker. When we
applied these descriptions to examining the selfie, we were unable to account for several
types of selfies, for example, those where body parts (e.g. feet or hands) are used to rep-
resent the self. In these types of selfies, Figure 2(b), since a human face is not present,
the description of gaze as a visual resource becomes redundant. Of course, it can be
argued that the relationship between the represented participant(s) – the ice cream and/or
the Ferris wheel – and the viewer is always one of observing in this type of image.
However, as both the ice cream cone and the Ferris wheel have been given saliency (in
terms of size and position in the frame) in the visual composition, ‘the object of seeing’
remains ambiguous, that is, if the image is to be read as a photo of an ice cream and a
Ferris wheel or an ice cream set against the background of a Ferris wheel.
Furthermore, what has not been accounted for in Kress and Van Leeuwen’s ‘grammar’
of images is the relation between the photographer and the viewer (rather than simply the
represented participant and the viewer). In Lange’s example, we are ultimately seeing the
mother as mediated by the photographer’s perspective. The photographer’s perspective
is implicit as there is no visual element indicating their presence. Nevertheless, we are
‘seeing with’ the photographer, despite this absence. In many traditional photographic
genres (cf. Caple, 2013), the perspective of the photographer is implicit, allowing the
‘personality’ of the represented participant to be foregrounded. In a selfie such as Figure
2(b), where the photographer is presumed to be the same as the visually represented
participant, the social relations being enacted include both the relationship construed
between the viewer and the visually represented participant, as well as the relationship of
1742 new media & society 20(5)

the viewer and the photographer. Accounting for this ‘missing’ perspective of the pho-
tographer thus becomes necessary.
In their interpretation of the meanings made in images in children’s picture books, Painter
et al. (2013) made an attempt to expand Kress and Van Leeuwen’s description to include the
notion of perspective or as they termed it, focalisation, that is, through whose eyes we ‘see’.
Interestingly, they drew their inspiration from narratology, specifically the notion of point of
view or ‘who sees’. Their aim was to explore how picture books help children learn how to
‘read’ the viewer-image relationship construed in narrative texts by visual choices that guide
the child into adopting different kinds of ‘viewing personae’. Narrative studies of point of
view typically distinguish between the voice ‘who tells’ us the story and the persona through
whose eyes we ‘see’ the unfolding narrative, a perspective that may shift as the narrative
unfolds. In terms of visual images embedded in narrative texts, focalisation deals with the
viewing position that is represented within this unfolding narrative: ‘are readers in eye con-
tact with characters, or observing them, and if observing are they observing directly or vicar-
iously (through the eyes of one of the characters as it were)’ (Martin, 2008: 13). Focalisation
has been used to analyse images in various contexts from children’s picture book narratives
(Martin, 2008; Painter et al., 2013), comics (Mikkonen, 2012), animated movies (Unsworth,
2013), to fictional films (Schlickers, 2009).
While the work of Painter et al. has opened up the possibility for describing ‘perspec-
tive’ in images, it cannot be applied directly to the analysis of the selfie. Apart from the
differences in medium technologies, the main limitation, as we see it, lies in the fact that
their description drew on narrative theory and the relation that they aimed to capture was
between the characters in the narrative (which may or may not be visually represented
participants) and the viewer. The recognition of perspective in the picture books does not
rely solely on structural clues in the illustrations but also verbal cues in the accompany-
ing texts. Furthermore, the perspective often shifts as the narrative unfolds. In the case of
the selfie, however, the perspective is ‘fixed’ and is presumed to be that of the maker/the
poster of the selfie, that is, the self that is represented. What becomes critical is thus not
whose perspective (who sees) is construed, but how perspectives are realised visually.
More precisely, what needs to be determined is whether and how perspectives can be
represented differently in the visual structure of selfies and whether these differences
matter in various social contexts.
To be able to account for different perspectives realised visually in selfies, we draw
on the theory of intersubjectivity as developed by British psychologist Gillespie, who
conceptualises intersubjectivity as

… the variety of relations between perspectives. Those perspectives can belong to individuals,
groups, or traditions and discourses … (Gillespie and Cornish, 2010: 19)

For us, there are two theoretical advantages to Gillespie’s inclusive definition of inter-
subjectivity. The first and foremost is his recognition of the social and interactional
nature of the human signs, which is deeply compatible with the theory of communication
as propagated by social semiotic theory. The second is that his approach can be opera-
tionalised at an analytical level. That is, it allows us to map our accounts of visual semi-
otic structures onto different types of perspective. In the next section, we will elaborate
Zhao and Zappavigna 1743

on the ways in which we have adapted Gillespie’s framework to account for perspectives
in the selfie and discuss the types of intersubjectivity and their corresponding visual
structures. Before this, we will briefly describe the dataset and methodological proce-
dures we have used for our study.

Dataset
The analysis undertaken in this article followed the principles of grounded theory
(Strauss and Corbin, 1994), using a mixed-methods data collection strategy. The aim was
to describe the visual resources drawn upon by Instagram users’ self-portraits. We first
randomly selected three individual Instagram users and tracked the range of different
visual choices they used for self-representation. We then began to develop our descrip-
tion of intersubjectivity by classifying these resources and iteratively sampling new
images. Sampling ceased when saturation of description was reached, that is, at the point
at which collecting additional data did not appear to modify the descriptive tool that we
were developing (i.e. the description of a feature was exhausted). At this point, we began
testing our description against a dataset of 5002 images containing the hashtag #coffee.
This ideational hashtag (Zappavigna, 2014a) was chosen because it was a commonly
used tag featuring an everyday object and allowed us to make sure that the differences in
visual structure we identified were interpersonal. In other words, it enabled us to control
for representational content by looking at a collection of selfies with same tag.
Another key consideration in the process of data collection and analysis is the issue of
Internet research ethics, which we view as a dynamic and reflective process rather than
a set of static rules (for a useful overview, see Östman and Turtiainen, 2016). The under-
lying principle of our ethical consideration is to strike a balance between our strive for
empirical rigour and our responsibility to protect the privacy of individual social media
user. Specifically, five dimensions were used to guide our ethical choices: intrusiveness,
public/private, vulnerability, potential harm and confidentiality (Convery and Cox,
2012). In the data collection process, we chose to work with data that are available in the
public domain. The three users chosen for the first stage of analysis are celebrity
Instagramers with a large number of public followings. We did not obtain individual
consent prior to data analysis as the main purpose of our research is to generate a descrip-
tive framework for the selfie as a textual artefact rather than to probe the social practices
of a specific community or an individual. To preserve the anonymity of the users by
reducing the possibility that the images may be found using an image search (such as
searching for images using Google image search), the images presented throughout this
article have been modified using the sketch filter in the image editing software, Gimp, to
produce an outline sketch. In addition, the eyes, nose and mouth area have been blurred
to make recognising the individuals difficult. Permissions were obtained for the full fron-
tal images presented in Table 2 and Figure 3.

Types of intersubjectivity in selfies

In his framework for analysing intersubjectivity, Gillespie builds on the work of Stryker
(1956) and identifies three levels of intersubjectivity between the self (S) and the other
1744 new media & society 20(5)

Table 1.  Coding frame for dialogical analysis of intersubjectivity based on Gillespie and
Cornish (2010: 36).

Level Explicit Implicit

Direct Statement that explicitly Statement that can be interpreted as
perspective communicates a direct revealing a direct perspective
(S → X) perspective ‘I love my books’ (i.e. I don’t want to
‘I love my books’ use a Kindle)
Meta-perspective Quotation or characterisation Quotation or characterisation that can
(S → O → X) that explicitly communicates be interpreted as revealing a meta-
a meta-perspective perspective
‘He said “I love my books”’ ‘He said “I love my books”’ (i.e. he does
not want a Kindle for Christmas)
Meta-meta- Quotation or characterisation Quotation or characterisation that can
perspective that explicitly communicates be interpreted as revealing a meta-
(S → O → S → X) a meta-meta-perspective meta-perspective
‘He keeps telling people that I ‘He keeps telling people that I said I love
said I love my books’ my books’ (i.e. he is trying to make me
sound old fashioned)

(O) in relation to a phenomenon (X): (1) direct perspective, (2) meta-perspective and (3)
meta-meta-perspective, as shown in Table 1. These forms of intersubjectivity can find
their expressions in everyday language, explicitly or implicitly. When a person makes a
statement such as ‘I love my books’, they communicate explicitly a direct perspective – a
self’s (I) perspective on a phenomenon (books). Similarly, when a statement such as ‘He
said “I love my books”’ is uttered, a meta-perspective is being communicated – another’s
perspective (his) on a self’s perspective on ‘books’. Projection (i.e. quoting) here can be
seen as a linguistic means for realising meta-perspective.
Drawing on Gillespie’s approach, we see images as being able to communicate
perspectives in a similar way to the ways in which language can perform this func-
tion, albeit through visual structural elements. A traditional photograph, such as
Lange’s Migrant Mother, can be seen as communicating a direct perspective, that is,
the photographer’s perspective on the visually represented participant, the mother.
The photographer – the subject who is seeing – is however not present in the image,
unlike in English language where the first person pronoun has to be used (e.g. I love
my work.). Since the subject who is seeing is not explicitly represented, direct per-
spective in an image is always implicit, which allows the foregrounding of the imag-
ined relation between the viewer and the represented participant. Simply put, the
social relation being negotiated is the relationship construed between the viewer and
the object of seeing.
What distinguishes the selfie from traditional forms of photography is its foreground-
ing of the photographer’s perspective, and each different type of selfie realises a different
type of intersubjective relation in which the photographer’s perspective is open for nego-
tiation. We have identified four different subtypes of intersubjective relations in the
selfie, mapping these onto the four genres of selfies. Table 2 gives an overview of our
analysis.
Table 2.  Types of selfies and intersubjective relations.

Types of selfies Metafunctional resources Intersubjective relations

Representational Compositional
Zhao and Zappavigna

Presented + photographer’s image; (+ the photographer: Meta- Self’s perspective on technologically

other visual participants; + + centred perspective (camera phone) mediated
objects; + backgrounds) + foregrounded representation of the self (or group)
S → T → S

Mirrored + photographer’s mirror the photographer: Meta-meta- Self’s perspective on technologically

image + mirror; + the + centred perspective (camera phone) mediated
phone; (+ minor images + foregrounded representation of technologically
of objects; backgrounds; + mediated (mirror) representation of
other visual participants) the self
S → T → T → S
Inferred + body parts (e.g. feet, legs, + varied composition Self’s perspective
hand or arms); + objects, + + body parts foregrounded on technologically Explicit
other visual participants, + + other visual participants mediated   
scenery); salient representation of
self’s perspective
of certain
Implied + objects implying the + varied composition phenomenon (X)
photographer; + other + objects implying SP → T → SP → X Implicit
visual participant; + objects; the photographer
+ scenery foregrounded
1745
1746 new media & society 20(5)

Figure 3.  Examples of presented selfies and meta-perspective.

Meta-perspective: presented
Meta-perspective is the first type of intersubjectivity that can be realised in the selfie, specifi-
cally in a type of selfie which we term Presented (Row 1 in Table 2). A presented selfie
(Figure 3) is one in which the image of the photographer, usually the face and the upper body,
is visually presented. There could be other visual representational elements in the images,
including participants (as in the case of group selfies), objects or a background. In terms of
composition, the image of the photographer is typically placed at the centre of the frame and
is foregrounded. The presented selfie is the photographer’s perspective on a phenomenon,
that is, the self. However, we do not consider this a direct perspective but rather a perspective
mediated by technology – the camera phone, and hence a type of meta-perspective. Of course,
in a traditional photo, we see a phenomenon through the lens of a camera. Yet, the presence
of the camera, like the photographer’s perspective, is implicit. In a presented selfie, in con-
trast, visual cues such as the distortion and the reversed mirror view make explicit the tech-
nological mediation. In this way, the perspective of the technology user and the perspective
of the technology co-evolve in the selfie. We express this meta-perspective as S → T → S,
with S standing for the self and T for technology. A presented selfie thus communicates the
self’s perspective technologically mediated by the representation of the self (or group).

Meta-meta-perspective: mirrored
Apart from the represented type, all other types of the selfie realise meta-meta-perspec-
tive, varying in terms of the ‘object of seeing’ (X) and degree of explicitness. The first
type of selfie that construes a meta-meta-perspective is the Mirrored selfie (Row 2 in
Table 2). In a mirrored selfie (Figure 4), the mirror image of the self is captured.
Representationally, it often includes the image of the photographer (typically with face
partially or fully blocked by the phone), their phone and the mirror. Mirror images of
objects such as a designer bag, backgrounds and other visual participants could also be
Zhao and Zappavigna 1747

Figure 4.  Examples of mirrored selfies and meta-perspective.

present. The composition in a mirrored image tends to be similar to that presented with
the image of photographer centred and foregrounded, though this type of selfie tends to
be shot from a further distance. The mirrored selfie functions to communicate the self’s
perspective on technologically mediated (camera phone) representation of technologi-
cally mediated (mirror) representation of the self, expressed as S → T → T → S. Simply
put, a mirrored selfie construes a photographer’s perspective on the self mediated by two
layers of technologies. Unlike in the presented selfie, the technological mediation is
explicitly present as both the camera phone and the mirror are represented in the photo.

Meta-meta-perspective: inferred and implied

There are two other types of selfie that communicate meta-meta-perspective, Inferred
and Implied (Row 3 and 4 in Table 2). An inferred selfie is one in which body parts such
as hands or legs infer the presence of the photographer, while an implied selfie in one
where objects (e.g. a cup of coffee or a bottle of water) imply the presence of the photog-
rapher. In these two types of selfies, objects, other visual participants or scenery are
always present in the frame, and the composition structures are more varied. What is
being communicated in these two types of selfies is self’s perspective on technologically
mediated representation of self’s perspective of certain phenomenon, or S → T → S → X.
These two types of selfies are deeply distinctive from the presented and mirrored self-
ies, as the focus of ‘observation’ here is not the body parts of the photographer or the
objects used to imply them but the other objects, participants and background sceneries
included in the photo. For example, in Figure 5 (Left) the objects of seeing are the coffee
cup and the flower (vase) not the photographer. The hand does not simply function to
represent the photographer but their perspective. Similarly, in Figure 5 (Right), we are
seeing the customers in the café, with the coffee cup and the ginger bread man indicating
the perspective of the photographer. Simply put, we are seeing the café ‘as the photogra-
pher’ (Zappavigna, 2016).
1748 new media & society 20(5)

Figure 5.  Examples of inferred and implied selfies and meta-meta-perspective.

Since many selfies with meta-meta-perspective include both body parts and objects,
for example, the hand holding the ice cream cone in Figure 2(b), the difference between
the inferred and implied selfies (Figure 5) can be seen as topological, i.e. differences in
degree, rather than typological, i.e. difference in category (as indicated in Table 2, bot-
tom-right). What distinguishes the two is the degree of explicitness in which they com-
municate the meta-meta-perspective, in other words, the easiness in which the viewer
can recognise the perspective of the photographer. Simply put, it is easier to recognise
the perspective of photographer in an image that includes a human hand than in a photo
that shows an empty coffee cup. We nevertheless set up the inferred selfie as a separate
subcategory, as we believe it epitomises the dilemma of modernity created by new tech-
nologies, where ‘the expression of subjectivities intensifies’ (Carvalheiro and Telleríra,
2015: 211) albeit as an extension of the object. The items used to signify the presence of
a perspective are often ‘fetishized’ objects – a cup of coffee, a designer bag or a rifle gun,
engendering forms of individual and community identities (e.g. Zappavigna, 2014b).
So far, we have discussed four types of selfies – presented, mirrored, inferred and
implied, and illustrated the different ways in which the photographer’s perspective can
be represented in these different types of selfies and how it is related to the object of see-
ing. Our aim to show that the function of selfies is not solely to represent the self or
subjectivity rather to enact intersubjectivity, that is, to generate various possibilities of
relations between perspectives. Although our description is by no means exhaustive, it
has allowed us to address and problematise some fundamental assumptions of the selfie,
a point that we will elaborate on in the final section of the article.

Questioning the self: intersubjectivity and its implications

for studying selfie
When we talk about the selfie, we assume a conflation between the photographer and the
represented participant. In effect, whether they are the same person is not important and
Zhao and Zappavigna 1749

often cannot be taken for granted, in particular in inferred and implied selfies. This is
perhaps the most critical insight we have gained from our social semiotic analysis of the
selfie. What ultimately distinguishes the selfie from other types of portraiture is not ‘the
self’ or the representation of subjectivity but the possibility for difference of perspectives
to be created and this difference to be shared between the image creator and the viewer,
that is, the enactment of intersubjectivity. Simply put, the selfie is a distinctive visual
genre as it can articulate and generate perspectives in different ways. There are two main
implications that can be drawn from our analysis of intersubjectivity in the selfie.
While several scholars have argued for the empowering nature of the selfie (e.g. Albury,
2015; Nemer and Freeman, 2015), our analysis has helped to illustrate that this potential for
empowerment is inherent in the visual structure of the selfie. To illustrate this point, we
shall return to the starting point of this article, the Clinton selfies. As we have discussed
earlier, if we consider the sole function of the selfie to be representing the subject, it would
be difficult to provide a counter argument to the popular view that young women taking
selfies at a campaign event is a sign of narcissism and political disengagement. Recognising
the function of the selfie for enacting intersubjectivity and for construing perspectives,
however, opens up new meaning potential. In the Clinton example, the selfies that would
have been taken do not simply provide a medium for the young women to represent them-
selves as ‘objects of seeing’, but also for articulating their perspectives on the political
events, that is to be ‘the seeing subjects’ at the same time. Selfies allow their perspectives
(and other traditionally marginalised voices) to be put forward and negotiated in public
discourse in a way other visual media (e.g. press photos) cannot. The act of selfie taking
thus can be seen as an act of political engagement rather than disengagement. It seems to
us that the moral panic and popular critique surrounding the selfie often arise from failure
to understand this unique intersubjective ‘grammar’ of the selfie, as well as the difference
between a selfie and a self-portrait in representing ‘points of views’, or to borrow Bellinger’s
(2015) term, ‘stances of representation’. As a result, when looking at a selfie, in particular
those of young women, many can only see the ‘objects of seeing’ but fail to recognise the
‘seeing subjects’. For us, to engage the public in a more nuanced debate about the selfie
ultimately requires critical visual literacy, or selfie literacy.
Our analysis of intersubjectivity also highlights the importance of distinguishing sub-
types of selfie and their distinctive discursive functions in the ‘context-collapsed’ (Wesch,
2009) social media environment. While all genres of selfie concern the representation of
the subject, they differ greatly in terms of intersubjectivity. Each sub-type of selfie realises
a different type of intersubjective relation, including the relation between various perspec-
tives, and the phenomenon receiving attention. For instance, a represented selfie concerns
the self’s perspective on technologically mediated representation of the self, while an
Inferred selfie realises the self’s perspective on technologically mediated representation
of self’s perspective on certain phenomenon. For us, the inferred and implied selfies – the
two types of selfie that convey meta-meta-perspective – are particularly fascinating. In
both types, the presence of the subject is most illusive yet the presence of perspective is
unambiguous. On the one hand, the conflation of the photographer and the represented
visual participant cannot be taken for granted. That is, we cannot be certain that the hands,
feet or ice cream cones belong to the photographer, and our assumption that the image is
a selfie is based on the anchoring elements present in the social media environment, such
as the existence of the personal profile or accompanying post or captions. On the other
1750 new media & society 20(5)

Figure 6.  Public announcement #binyourbutts.

hand, the perspective of the other is clear, as we do not simply see a Ferris Wheel, see
Figure 2(b), as in the case of traditional photography; instead, we see the wheel through
the photographer who stands in front of it holding an ice cream cone (though we do not
see the actual person). This intersubjective functionality is essential as it allows an image
to be transformed from one context to another context; even when the idea of a specific
‘self’ is no longer important, the relations between perspectives remain. Interestingly, in
the process of collecting and analysing selfies, we started to observe the spread of the
selfie-like structures or ‘quasi’ selfies in media environments that are not traditionally
associated with the selfie such as women’s magazines and billboards (e.g. Figure 6).
Our analysis of the intersubjective function of the selfie is by no means exhaustive.
For instance, our dataset is scraped exclusively from Instagram, where selfies are often
edited using filters to create certain stylised look. We nevertheless hope our micro ana-
lytical approach will provide a useful complementarity to macro social political readings
of the selfie. We envisage several lines of inquiry to be followed from our study. One is
the ways in which different social groups (e.g. mummy bloggers) articulate their per-
spectives and shape the public debate around issues through the visual medium of the
selfie. Another is the ways in which different genres of selfie have been deployed across
different social media platforms as well as in traditional media. We hope our micro-
structural analysis may also be useful for annotating and categorising selfies in quantita-
tive and (semi) automated investigations.

Acknowledgements
We thank Professor Anne Cranny-Francis (University of Technology Sydney) for her valuable
comments on subjectivity and identity in critical theory in early stage of this project.

Funding
The author(s) received no financial support for the research, authorship and/or publication of this
article.
Zhao and Zappavigna 1751

Notes
1. For detailed explanation of interactional resources in images, please refer to Kress and Van
Leeuwen (2006: 119–158).
2. This is the maximum number of images which Instagram’s Terms of Service allow an indi-
vidual scrape and group into a collection.

References
Albury K (2015) Selfies, sexts, and sneaky hats: young people’s understandings of gendered prac-
tices of self-representation. International Journal of Communication 9: 1734–1745.
Baishya A (2015) NaMo: the political work of the selfie in the 2014 Indian general elections.
International Journal of Communication 9: 1686–1700.
Bellinger M (2015) Bae caught me tweetin’: on the representational stance of the selfie.
International Journal of Communication 9: 1806–1817.
Boon S and Pentney B (2015) Selfies – virtual lactivism: breastfeeding selfies and the performance
of motherhood. International Journal of Communication 9: 1759–1772.
Brager J (2015) The selfie and the other: consuming viral tragedy and social media (after)lives.
International Journal of Communication 9: 1660–1671.
Bruno N, Gabriele V, Tasso T, et al. (2014) ‘Selfies’ reveal systematic deviations from known
principles of photographic composition. Art & Perception 2: 45–58.
Burns A (2015) Self(ie)-discipline: social regulation as enacted through the discussion of photo-
graphic practice. International Journal of Communication 9: 1716–1733.
Caple H (2013) Photojournalism: A Social Semiotic Approach. Basingstoke: Springer.
Carvalheiro JR and Tellería AS (2015) Mobile and Digital Communication: Approaches to Public
and Private. Livros LBCOM BOOK. Covilhã, UBI: Portugal.
Charteris J, Gregory S and Masters Y (2014) Snapchat ‘selfies’: the case of disappearing data. In:
Hegarty B, McDonald J and Loke SK (eds) Rhetoric and Reality: Critical Perspectives on
Educational Technology, Proceedings ascilite Dunedin. pp. 389–393.
Coladonato V (2014) Power, gender, and the selfie. The cases of Hillary Clinton, Barack Obama,
Pope Francis. Comunicazioni Sociali 36: 394–405.
Collings B (2014) #selfiecontrol: @CAZWELLnyc and the role of the ironic selfie in transmedia
celebrity self-promotion. Celebrity Studies 5: 511–513.
Convery I and Cox D (2012) A review of research ethics in internet-based research. Practitioner
Research in Higher Education 6: 50–57.
David G (2015) All what we send is selfie: images in the age of immediate reproduction. In:
Carvalheiro JR and Tellería AS (eds) Mobile and Digital Communication. Covilhã: Livros
LabCom, pp. 79–100.
Fausing B (2014) Selfies, healthies, usies, felfies. Available at: http://www.kommunikationsforum.
dk/artikler/selfies-og-jagten-paa-anerkendelse
Frosh P (2015) Selfies – the gestural image: the selfie, photography theory, and kinesthetic socia-
bility. International Journal of Communication 9: 1607–1628.
Gibbs M, Nansen B, Carter M, et al. (2014) Selfies at funerals: remediating rituals of mourning. In:
Selected papers of internet research 15: the 15th annual meeting of the association of internet
researchers, Daegu, Korea, 22–24 October, no. 4.
Gillespie A (2009) The intersubjective nature of symbols. In: Wagoner B (ed.) Symbolic
Transformation: The Mind in Movement through Culture and Society. London; New York:
Routledge, pp. 23–37.
Gillespie A and Cornish F (2010) Intersubjectivity: towards a dialogical analysis. Journal for the
Theory of Social Behaviour 40: 19–46.
1752 new media & society 20(5)

Graham C (2016) Crowd turns its back on Hillary Clinton as photo captures the age of the selfie.
The Telegraph. Available at: http://www.telegraph.co.uk/news/2016/09/25/crowd-turns-its-
back-on-hillary-clinton-as-photo-captures-the-ag/
Gunthert A (2015) The consecration of the selfie: a cultural history [La consécration du selfie. Une
histoire culturelle]. Études Photographiques 32: 1–21.
Hall J (2014) The Self-Portrait. London: Thames & Hudson.
Hall K (2015) Selfies and self-writing: cue card confessions as social media technologies of the
self. Television & New Media 17: 228–242.
Halliday MAK (1978) Language as a Social Semiotic: The Social Interpretation of Language and
Meaning. Baltimore, MD: University Park Press.
Hammelburg E (2015) #stemfie: reconceptualising liveness in the era of social media. Tijdschrift
voor Mediageschiedenis 18: 85–100.
Hess A (2015) The selfie assemblage. International Journal of Communication 9: 1629–1646.
Hochman N (2014) The social media image. Big Data & Society 1(2).
Koffman O, Orgad S and Gill R (2015) Girl power and ‘selfie humanitarianism’. Continuum 29:
157–168.
Kress G and Van Leeuwen T (2006) Reading Images: The Grammar of Visual Design. London;
New York: Routledge.
Kuntsman A and Stein R (2015) Digital Militarism: Israel’s Occupation in the Social Media Age.
Stanford, CA: Stanford University Press.
Kwon YJ and Kwon K-N (2015) Consuming the objectified self: the quest for authentic self. Asian
Social Science 11: 301–312.
Lüders M, Prøitz L and Rasmussen T (2010) Emerging personal media genres. New Media &
Society 12: 947–963.
Martin J (2008) Intermodal reconciliation: mates in arms. In: Unsworth L (ed.) New Literacies and
the English Curriculum. London: Continuum, pp. 112–148.
Marwick AE (2013) Status Update: Celebrity, Publicity, and Branding in the Social Media Age.
New Haven, CT: Yale University Press.
Meese J, Gibbs M, Carter M, et al. (2015) Selfies at funerals: mourning and presencing on social
media platforms. International Journal of Communication 9: 1818–1831.
Mikkonen K (2012) Focalisation in comics. From the specificities of the medium to conceptual
reformulation. Scandinavian Journal of Comic Art 1: 69–95.
Miltner KM and Baym NK (2015) The selfie of the year of the selfie: reflections on a media scan-
dal. International Journal of Communication 9: 1701–1715.
Nandy S (2015) Persona, celebrity, and selfies in social justice: authenticity in celebrity activ-
ism. In: Carvalheiro JR and Serrano Tellería A (eds) Mobile and Digital Communication:
Approaches to Public and Private. Covilhã: Livros LabCom, pp. 101–122.
Nemer D and Freeman G (2015) Empowering the marginalized: rethinking selfies in the slums of
Brazil. International Journal of Communication 9: 1832–1847.
Östman S and Turtiainen R (2016) From research ethics to researching ethics in an online specific
context. Media and Communication 4: 66–74.
Painter C, Martin JR and Unsworth L (2013) Reading Visual Narratives: Image Analysis of
Children’s Picture Books. London: Equinox.
Pham MHT (2015) ‘I click and post and breathe, waiting for others to see what I see’: on #femi-
nistselfies, outfit photos, and networked vanity. Fashion Theory 19: 221–241.
Qiu L, Lu J, Yang S, et al. (2015) What does your selfie say about you? Computers in Human
Behavior 52: 443–449.
Schleser M (2014) Connecting through mobile autobiographies: self-reflexive mobile filmmaking,
self-representation, and selfies. In: Schleser M and Berry M (eds) Mobile Media Making in
an Age of Smartphones. Basingstoke: Palgrave Macmillan, pp. 148–158.
Zhao and Zappavigna 1753

Schlickers S (2009) Focalization, ocularization and auricularization in film and literature. In:
Jannidis F, Martinez M and Schmid JPW (eds) Point of View, Perspective, and Focalization:
Modeling Mediation in Narrative (Narratologia). Berlin; New York: Walter de Gruyter, pp.
243–258.
Senft TM and Baym NK (2015) What does the selfie say? Investigating a global phenomenon.
International Journal of Communication 9: 1588–1606.
Shipley JW (2015) Selfie love: public lives in an era of celebrity pleasure, violence, and social
media. American Anthropologist 117: 403–413.
Silvestri L (2014) Shiny happy people holding guns: 21st-century images of war. Visual
Communication Quarterly 21: 106–118.
Sorokowski P, Sorokowska A, Oleszkiewicz A, et al. (2015) Selfie posting behaviors are associ-
ated with narcissism among men. Personality and Individual Differences 85: 123–127.
Strauss A and Corbin J (1994) Grounded theory methodology. Handbook of Qualitative Research
17: 273–285.
Stryker S (1956) Relationships of married offspring and parent: a test of Mead’s theory. American
Journal of Sociology 62: 308–319.
Tifentale A (2014) The Selfie: making sense of the ‘masturbation of self-image’ and the ‘vir-
tual mini-me’. Selfiecity.net, pp. 1–24. Available at: https://d25rsf93iwlmgu.cloudfront.net/
downloads/Tifentale_Alise_Selfiecity.pdf
Tifentale A and Manovich L (2015) Selfiecity: exploring photography and self-fashioning in
social media. In: Berry DM and Dieter M (eds) Postdigital Aesthetics. London: Palgrave
Macmillan, pp. 109–122.
Tiidenberg K (2015a) Great faith in surfaces – a visual narrative analysis of selfies. In: Allaste
AA and Tiidenberg K (eds) ‘In Search of …’ New Methodological Approaches to Youth
Research. Newcastle upon Tyne: Cambridge Scholars Publishing, pp. 233–256.
Tiidenberg K (2015b) Odes to heteronormativity: presentations of femininity in Russian-
speaking pregnant women’s Instagram accounts. International Journal of Communication
9: 1746–1758.
Unsworth L (2013) Point of view in picture books and animated movie adaptations. Scan: The
Journal for Educators 32: 28–37.
Walker-Rettberg J (2005) Mirrors and shadows: the digital aestheticisation of oneself. In:
Proceedings of digital arts and culture, IT University of Copenhagen, Copenhagen, 1–3
December, pp. 184–190.
Walker-Rettberg J (2014) Written, visual and quantitative self-representations. In: Walker-Rettberg
J (ed.) Seeing Ourselves through Technology: How We Use Selfies, Blogs and Wearable
Devices to See and Shape Ourselves. Basingstoke; New York: Palgrave Macmillan, pp. 1–19.
Weiser EB (2015) #Me: narcissism and its facets as predictors of selfie-posting frequency.
Personality and Individual Differences 86: 477–481.
Wesch M (2009) YouTube and you: experiences of self-awareness in the context collapse of the
recording webcam. Explorations in Media Ecology 8: 19–34.
Williams AA and Marquez BA (2015) Selfies – the lonely selfie king: selfies and the conspicu-
ous prosumption of gender and race. International Journal of Communication 9: 1775–1787.
Zappavigna M (2011) Ambient affiliation: a linguistic perspective on Twitter. New Media &
Society 13: 788–806.
Zappavigna M (2012) Discourse of Twitter and Social Media. London: Continuum.
Zappavigna M (2014a) Coffeetweets: bonding around the bean on Twitter. In: Seargeant P and
Tagg C (eds) The Language of Social Media: Communication and Community on the Internet.
Basingstoke: Palgrave Macmillan, pp. 139–160.
1754 new media & society 20(5)

Zappavigna M (2014b) Enacting identity in microblogging through ambient affiliation. Discourse

& Communication 8: 209–228.
Zappavigna M (2016) Social media photography: construing subjectivity in Instagram images.
Visual Communication 15: 271–292.

Author biographies
Sumin Zhao is a Carlsberg Distinguished postdoctoral fellow at University of Southern Denmark
and the book review editor for Discourse and Communication (SAGE). She publishes in the areas
of critical multimodal discourse analysis, mobile technologies and digital literacy. Her recent books
include Critical Multimodal Studies of Popular Discourse (with Emilia Djonov, Routlege, 2016
paperback), and Advancing Critical Multimodal Studies (co-edited, Routledge, forthcoming,)
Michele Zappavigna is a senior lecturer in the School of Arts and Media at the University of New
South Wales. Her major research interest is the discourse of social media. Recent books in this area
include: Discourse of Twitter and Social Media (Bloomsbury, 2012), Researching the Language
of Social Media (Routledge, 2014) (with Ruth Page, Johann Unger and David Barton) and
Searchable Talk: Hashtags and social media metadiscourse (bloomsbury, forthcoming).