Photographs and Cartoons: Differences in Interpretation of the Visual Semiotics ()
1. Introduction
The news media is now both inherently multi-modal and most often accessed on-line. This means that images (both still and moving) are now a key part to the interpretation of what is being said as much as, possibly now even more than, any actual text. This has implications both for the construction of each individual item within an image, how these are combined and, equally importantly, how they are interpreted by the viewer (and possibly a lack of clarity as to how they were meant to be interpreted by the original creator). Multimodality [1] [2] provides an analytic framework that works across these three different levels. At one level are the individual semiotic modes (text, image, background colour, sound and so on) and then how these are combined to build up the system of meaning embodied in an image or video [3] [4] [5] . This process of analysing how they are combined (intersemiosis) follows carefully structured approaches [6] [7] and is key to building meaning. Intersemiosis is thus critical in the creation of understanding as meaning should “not [be] analysed in terms of discrete building blocks or structures, but in terms of socially meaningful tensions and oppositions which could be instantiated in one or more ways” ( [5] , p. 31). In effect, we rarely will base interpretation on a single semiotic mode (whether textual or visual) and instead it is the interaction of two or more such modes that underpin meaning making.
The focus of this study is the portrayal of women in Saudi Arabia following from the recent changes to the male Guardianship laws [8] . As with many social issues in the Kingdom, these have been contentious and interpreted in very different ways. The stated goal is to reduce the extent that women need the permission of male family members to travel and interact outside the home. Part of this has been to lift the previous ban on women driving cars. Those who support the reforms note how much this changes practical life and removes long standing restrictions. On the other hand, critics note the extent that gender-based discrimination remains and the attitudes that restrict women’s practical freedoms remain in place.
To explore these themes this study considers the content of two groups of images. Some were produced to explain the recent social changes to their readership, some were generated by organisations arguing that, at best, the changes are superficial. Most of the first group are photographs and all of the second group are some forms of cartoon style images. This distinction is useful as the former may be less clear as to the wider interpretation (resemiosis) while the latter were designed for a very particular role. In effect, in one group the challenge is to consider how meaning making is managed within the image and the extent that interpretation by the observer is clear or not. The expectation in the second group is of less ambiguity of interpretation, even if different observers may agree or disagree with the intended message. By combining photographs (which may be more complex) with cartoons (which are usually designed to carry a very specific message) this study allows an exploration of whether the internal structure of the image alters the process of resemiosis.
2. Literature Review
2.1. Media Portrayals of Women in Saudi Arabia
The representation of the Middle East generally, Islam and the role of women in the Western Media has often been challenged due to the underlying orientation and assumptions. In the Saudi context, the focus has been on the substantial restrictions facing women in public life but with this analysis often takes little note of how women have always been able to run their own businesses [9] or access higher education. Issues such as a ban on driving, restrictions on travel and exclusion from many public spaces have all been regularly raised by critics of the Saudi regime. The legal changes since 2017 have sought to change this in particular in terms of male guardianship rules, the right to drive and more freedom to travel. In turn this has attracted attention both in the Western press [10] and across the Arabic speaking world [11] . More widely these changes are promoted by their supporters as a radical reset of social relations in Saudi Arabia and by others as far too limited and possibly cosmetic in nature [12] . In addition to the legal aspect of the changes, the Saudi state has started to produce public information material calling for a reduction in domestic violence and generally greater rights for women.
2.2. Visual Grammar
While much of this narrative has been created and sustained in the form of print, increasingly visual images are important as tools to transmit information and claim the attention of potential readers. In this respect, in the same way that grammar both structures the format and reading of text, visual grammar is an important tool for reading images [13] [14] . Chen and Gao (2014) note how this approach allows for consideration of the various visual modes in the image, the relationship of the image to surrounding material (text, other images, headlines), the intended interpretation by the creator and the possible range of interpretations by the readers [15] .
As noted above, the presentation of the role of women in Saudi society is contested (both before and after the recent reforms) and this combination of being able to study the process by which meaning is built up, and to consider issues of variable interpretation makes Kress and van Leeuwen’s approach valuable.
Their model makes use of different analytic concepts to reflect how meaning is created within the image by the interactive participants using four main systems: contact; social distance; perspective; and, modality [13] . Contact depends on how the represented characters interact with the viewer, if they are looking directly at the viewer this can be seen as “demand” in that some form of direct response is expected (ranging from empathy to disgust). If this is lacking then the image is an “offer” with a degree of passivity [16] . In terms of demand, facial expressions are an important part of structuring the desired response. Social distancing relies both on the composition within the image (how close the represented participants are, technical issues such as camera view (close up or long range) and how much of the participants is shown [17] . Camera view also builds up the perspective in the sense of looking down at the participants, up at them or at eye level. As with other social interactions these positions are often held to reflect the relative power and importance between the participants.
While these three are, to some extent, matters of technical image composition, modality [18] captures the viewpoints of those who constructed the image, how it was used and also, to a lesser extent, the interpretations of the viewers. Initially [13] this focused on issues of truthfulness and credibility (which, of course, capture different criteria). Traditionally it was possible to make a-priori assumptions about truthfulness from the image composition. It was technically easier to fake an image that had little or no colour, limited background or other internal information. However, recent developments in terms of image manipulation make these rules of less value [19] suggesting both a need for greater digital literacy among the wider population and that credibility (and how this is built up) becomes a critical part of the interpretation process [20] [21] [22] .
These different modes are summarised in Figure 1 and used to analyse the chosen images. This captures the distinction above between technical composition and intended meaning. In effect the former should be clear, and thus open to clear description, the latter needs some degree of interpretation both from the semiotic resources used and the wider framing.
3. Research Methods
The way that Saudi women are represented in western media in the form of text has been analysed for some time both from the perspective of multimodality [10] and wider political, social and journalistic studies [23] . This study seeks to focus on non-verbal modes in part as this has been less studied and in part as, with the growing reliance on the internet for information, such modes become more important. This allows the study to both concentrate on the construction of meaning with the image using both interactive and compositional meaning. These capture ways in which meaning can be created within the provided image but there is also a need to consider how an image is interpreted by a viewer [15] . The latter captures both the original image maker (what meaning they might have wished to convey) and the various ways different viewers might respond. Functionally what is expected is the degree of clarity (in other words can the image be interpreted in many ways) and just how this meaning making is constructed.
The approach is qualitative [24] [25] , using a small number of images that in turn allows for in-depth study as to how meaning making is constructed. The sampling frame was derived from an advanced Google image search for direct semiotic reference to Saudi women and had been published on a western news website. All the images were from the English speaking media and span the period before and after the recent Saudi reforms.
4. Findings
The first batch of images (Figures 2-11 below) are analysed in terms of how visual contact and social power is used to create meaning.
![]()
Figure 3. Trapped with abuse—end male guardianship in Saudi Arabia.
Source, respectively from Human Rights Watch videos, No Freedom to Travel [26] and Trapped with Abuse – End Male Guardianship in Saudi Arabia [27] .
Source respectively from the CNN article Saudi Women are Tweeting for their Freedom [28] and the CNN report on the arrest of the person who made the Mecca Girl music video [29] .
Source (Figure 6 and Figure 7) respectively from the CNN article about a Saudi teenager given the right to stay in Thailand [30] and about al-Hathloul [31] .
Source (Figure 8 and Figure 9) respectively from the CNN article about Maha Taibah, an advisor to Saudi Arabia’s Labour Minister [32] , and an article on the BBC about the “we drove” music video [33] .
Source (Figure 10 and Figure 11) respectively from the CNN article “Saudi Women are tweeting for their freedom” [28] and a CNN report on the King Khalid Foundation addressing the problem of domestic abuse [34] .
Source, a New York Times article about al-Sharif [35] .
The first step in the analysis takes the images as presented without the contextual information provided by the wider publication or article.
In Figures 4-11 the participant looks directly at the viewer. This creates a demand, in particular that the viewer identify with them and, to some extent, have an understanding of their world. However, there are important differences within this batch. For Figure 7, in particular, there is a very direct interaction with the viewer, no other context is present apart from the woman’s head.
Of those without direct eye contact, in Figure 2 the woman is looking down without any contact which is usually a submissive pose asking for pity and empathy. In contrast in both 3 and 12, the woman is looking up but not at the viewer. Figure 12 in particular shows someone at rest and ease in contrast to 5 where there is a direct demand on the attention of a viewer.
In terms of social distance and frame size, 2, 3, 4, 5, 8, 9 and 12 are all medium shots showing the individual from head to waist but at some distance to the subject. In 6, 7, 10 and 11, these are much closer images, showing just the head and shoulders, creating a closer intimacy between subject and viewer.
The angles of the images also vary substantially. In 2, 3, 4, 6, 8, 10 and 11, the subject is at the viewers eye level (suggesting equity) while in 5, 9 and 12 the image looks up at the subject (suggest superiority). The outlier is thus 7 where the represented participant is viewed from a high level and the viewer holds the power in this particular relationship.
As discussed, considerations of modality encompass consideration of plausibility of the image. Some of this comes from issues in the composition and layout of the image. Colour saturation is important and 2, 3, 10 and 11 are in black and white (which often indicates low modality) which also applies to 7 due to the full colour saturation. The balance uses natural colours suggesting higher modality. Table 1 below summarises all the images in terms of their modality markers.
In effect, almost every image has a mix of high and low modality but the balance varies substantially. On one end of the scale 9 (showing a woman wearing a traditional Aqaba) scores as low modality for all the features and in contrast 5 and 6 only have one low modality score. Both of these show women fairly directly demanding the attention of the viewer, either by direction of gaze and composition (6) or actions (5).
The second batch of images are also used to explore the construction of salience. Note that these are all cartoon style drawings so in one sense are very much produced for a particular purpose.
Sources (Figure 13 and Figure 14), both [36] [37] from a New York Times article about Saudi Women driving [38] .
Source (Figure 15 and Figure 16) from the Human Rights Watch videos “No freedom to travel” [26] and “trapped with abuse – the Male Guardianship system” [27] respectively.
Source (Figure 17 and Figure 18) from the Human Rights Watch videos “No freedom to travel” [26] and “trapped with abuse – the Male Guardianship system” [27] respectively.
Sourced (Figure 19 and Figure 20) from the video “trapped with abuse – the Male Guardianship system” [27] and a CNN image entitled “veiled Saudi woman football fan” [39] .
Given that these are all constructed images, as opposed to photographs being used to supplement a wider argument, the focus is on the visual elements used to attract a viewer’s attention.
Figure 13 uses two narrative approaches, the obvious being of a Saudi man holding a Saudi woman (with a car) in his hands. The less obvious is whether this indicates care giving and safeguarding or an attempt to control female access to the new (and limited) right to drive on their own. The relative size of the participants (and the direction of gaze) very much emphasises the relative power relations between the man and the woman. The yellow background is designed to attract and hold attention, in part it may reflect the concept of desert widely associated with Saudi Arabia but it gives the image high visual appeal. Figure 14 uses some similar concepts, in particular that of a small female figure underneath a much larger hand with a car very much obscured as a background shadow. Again, this is ambiguous in that it may suggest that access to driving is emerging as a possibility or, instead, remains very much under patriarchal control.
The rest of the images use colours to form salience. Figures 15-17 show a Saudi woman wearing an abaya but the Arabic text (that draws attention due to its colour) says “together to end the male guardianship system”. In particular Figure 16 is taken from a video about domestic violence and the red text can be seen as stressing either fear of violence or empowerment to end violence. By contrast 15 is from a video about the changes to travel restrictions using green text to emphasise hope. The blue text for Figure 17 is chosen to emphasise a need (and some hope) for greater freedom and liberty.
Figures 18-20 all make use of red, suggesting the violence inherent in the old guardianship system. Figure 20 has a woman wearing a niqab and holding a Saudi flag, again suggesting the risks in the old system and need to change. This sequence of images was designed to identify existing problems and support the planned changes to the guardianship rules.
5. Discussion
The discussion above reflects the relative difficulties of using intersemiosis to explore how an image is constructed compared to resemiosis for how it is interpreted (and how the original producer intended for it to be interpreted). As noted, the images selected above can be split into two broad groups. While the first set includes both photographs and cartoon style images the second set are all structured as cartoon images. The result is some difference in the type and volume of semiotic modes and their interaction.
The first set could be seen as being richer in their semiotic construction (Table 1) even if their overall modality varies substantially. Many of the first group are framed with the subject looking directly at the viewer, in effect a demand for a response. The second set are usually simpler reflecting their design as seeking to explain the need for the recent social reforms.
In terms of intersemiosis, both batches are relatively straightforward. The various images (and text where it is used) are being used to support a central message. What is less clear is what this message is (the process of resemiosis). Here there is a relatively clear break between the two groups. Those in the second batch can mostly be seen as non-ambiguous in this regard. Even Figure 14 which might be construed as the old guardianship model as one of caring rather than social restrictions, is relatively easy to interpret. The final group (18 - 20) are all designed to indicate the problems of the guardianship model and the need for significant change.
![]()
Table 1. Modality markers [13] , p. 160.
In this respect, that the second batch are clearly constructed around a position simplifies part of the problem with resemiosis. The intent of the designer is clear and each has a single message. For the viewer, the remaining ambiguity stems from their degree of understanding of the much wider context, and, perhaps more pertinent, whether they see the images as basically positive (a direction of change away from the old system) or indicative of just how substantial the degree of change is that is really needed.
For the first batch, there is not just the issue of the viewer/interpreter but also of what is being represented. Consequently, Figure 10 and Figure 11 could be seen as relatively traditional representations of Saudi women, but notably both score very low in terms of modality in every category. However, this points to a fundamental issue with moving from intersemiosis as the focus of a study to resemiosis. To many potential observers, those would be readily seen as very plausible images of Saudi women, perhaps even typical. Of course, other observers may see them as somewhat outdated and stereotypical representations and argue that Figures 5-9 are more credible representations of Saudi women after the recent reforms (and those images score more highly in terms of high modality).
If the process of interpretation was simply a consequence of analysis of the image the low modality of Figure 10 and Figure 11 should be critical in terms of their perceived plausibility. Accepting there is a disconnect starts to address some of the wider process of resemiosis. The core value of intersemiosis is that it combines all the presented semiotic modes into a coherent whole accepting that this or that mode may be contradictory.
Resemiosis may be about how we interpret a single image but the logic of that interpretation is not just bounded by the presented image. Even laying aside, individual biases and preconceived ideas, to interpret the validity of an image may well involve combining a wider set of information. In this case, the surrounding text may be important as may the headline for the article. Equally how the image was accessed maybe important. If the image was found running an internet search routine, then the framing of that enquiry will influence what is presented, as well as the likely interpretation framework of the observer. In that respect simply knowing the headline of the wider report is informative. That Figure 8 is of a government adviser and Figure 12 is of an activist in many ways is more important for wider meaning making than the specific framing of the image. In this case they are broadly similar in that respect but clearly reflect very different personal situations.
6. Conclusions
In summary this study points to the value of Kress & van Leeuwen’s (2006) interpretative approach to combining the semiotic modes in an image. However, it also points to the problem of how the observer will interpret this information, especially in an area that is contested with sometimes strongly held opinions and a lack of contextual knowledge. The strength of intersemiosis is that it moves from studying semiotic modes in relative isolation to bringing them together in a coherent whole. The weakness of current approaches to resemiosis is that this might be how the observer then places their understanding on the presented image (and some of this, inevitably, will draw on understanding and information outside the current frame of reference), however, in other cases it will involve the observer drawing on different components of the wider article to inform their interpretation. In effect, combining the various modes in images results from more than one instance of intersemiosis.
The difference between the photographs and the cartoons is marked especially in terms of how Interactive Meaning is constructed. As is to be expected the cartoons use less semiotic modes and have no secondary aspects such as background imagery that does not support the main focus. In turn the cartoons are more obvious in terms of both Representational and Compositional Meaning again reflecting their relative clarity of purpose. Having noted this, the wider framing (the title of the video or article) is still critical to any understanding.
In this case this is partly linguistic. Unless the reader can understand Arabic script, the full intent of the cartoon images will be lost. In effect, knowing the nature of the publication and being provided with a translation both compensate for what might otherwise be a missing semiotic resource.
The challenge for researchers is to accept this broadening of frame of reference and to consider how this more holistic view may be at variance to the implications of a study of an individual image. Thus, intersemiosis allows for the idea that not all the semiotic modes are complementary and that some may carry different messages or even competing messages (such as words and the presented image). In turn resemiosis becomes a process not just of building up meaning from mutually reinforcing semiotic structures but also of balancing potentially contradictory or even competing meaning systems.