Interactive Multimodal Metadiscourse in COVID 19 Campaign Posters by Ministry of Health, Kenya

The purpose of this study is to examine the interactive multimodal metadiscourse features of campaign posters developed exclusively by Kenya’s Ministry of Health (MoH) or collaboratively with religious groups and development partners. The objectives of the study were to establish the generic structure, categories, frequencies and functions of interactive textual metadiscourse and visual resources. Using a mixed research design employing qualitative and quantitative research approaches, the study analyzed a corpus of 25 posters on six thematic areas including social stigma, immunity, signs and symptoms, handwashing, immunity, mental health. The frame work of analysis is partly drawn from Hyland’s interpersonal model of metadiscourse (2005) and Kress (2010) and Kress and Van Leewun (2001, 2002, 2006) on visual interactive resources. The findings revealed presence of evidentials, code glosses and frame markers in the textual metadiscourse used by the poster designers. Evidentials were the most frequent used textual metadiscourse markers followed by code glosses and frame markers. On use of interactive visual resources, the research established use of information value, framing, connective devices and fonts to supplement the textual interactive metadiscourse devices to guide the readers. The results expand our knowledge on multimodality communication in respect to guiding readers when engaging in multimodal public health communication. Keywords— interactive multimodal metadiscourse, interactive metadiscourse, visual interactive resources,


INTRODUCTION
A poster is a temporary promotion of an idea, product or event put up in public space for mass consumption (Lippert, 2017). It is classified as multimodal because it contains both textual and graphic elements, although they may be wholly graphical or textual. Research on posters can be categorized into four areas specifically, entertainment, commercial, political and educational posters performing the key functions of informing, persuading and entertainment (Holloway, 2014). The poster is a condensed and specialized genre. This research focuses on educational posters and more specifically public health posters which seek to inform and educate the general public on the global pandemic, COVID-19. The emergence of the global pandemic, COVID-19 has created a specialized form of discourse manifest in the way government communicates to the public.
In Kenya, the Ministry of Health has used posters among other communication initiatives as a preventive strategy to control the pandemic and deal with the risk factors associated with the pandemic. In its attempts to effect behaviour change in the members of the public, the ministry in collaboration with religious groups and civil society has created campaign posters with several messages on pandemic. This study seeks to establish how the campaign poster designers guide the readers to COVID-19 information using interactive linguistic resources-metadiscourse and visual grammar-despite their physical absence on the Ministry's website. Metadiscourse is a term used in discourse analysis based on a view of writing or speaking as a social and communicative engagement and reveals the way writers project themselves into their discourse to signal their attitudes and commitments to matters in the text (Hyland, 2005). This linguistic concept of metadiscourse enables writers to create considerate texts because they take into accounts the reader's needs and expectations. A considerate text presupposes two levels of    With regard to the global pandemic, COVID-19 there has been attempts to analyze multimodal metadiscourse using Kress and Leewun visual grammar on the COVID-19 pandemic discourse. Yuigin and Rui (2020) analyzed the discourses of words and images of collected during the COVID-19 pandemic in China based on the visual grammar of Kress and Leewun. The findings reveal that the meanings of words and images of COVID-19 in China are constructed through the three aspects of interactive, representative and compositional meaning. The authors explain that the images and words work together to achieve the overall interpretation of the text. The current study borrows visual grammar approach in combination with underutilized Hyland's interpersonal model of metadiscourse. Some researchers have pursued multimodality on a purely interactive aspect with focus on social media. Baraza, Khasanda and Nyandoro (2020)performed a discursive analysis of the interactive meaning in COVID-19 containment discourses in social media using Critical Discourse Analysis (CDA) and Multimodal Discourse Analysis (MDA) within Systemic Functional Linguistics (SFL) as its theoretical bases. The purpose of the study was to analyze interactive meaning is facilitated in linguistic and visual mode in 28 texts sampled from the Internet and WhatsApp groups. The findings revealed several thematic family issues ranging from poverty, strained couple relationships, spouse cheating, spousal phone swooping, couple dominance and power struggles among other issues. However, the current study is on campaign posters. The potential of metadiscourse expansion to the visual realm to help readers navigate texts has been explored by various researchers. Kumpf (2000) explains how metadiscourse acan utilized by authors as a design criterion when considering the needs and expectations of readers of technical documents. Kumpf's work was applied in the reader orientation's work  Although D'Angelo's framework focus is academic posters, this study seeks to employ the framework to analyze public health posters and more specifically, COVID-19 campaign posters. The communication of public health information is crucial in the containment of contagious diseases. The framework will help understand how the poster designers guide the readers using interactive metadiscursive textual and visual resources in reading information on COVID-19.

III. METHODOLOGY
This study adopts a mixed methods research design, which combines both qualitative and quantitative elements of research. The rationale for the choice of the design is based on the research objectives-frequencies, categories and functions of metadiscourse. Shorten and Smith (2017) explain that a mixed method design is appropriate for answering research questions that neither qualitative or quantitative methods can answer alone. The corpus for this research consists of 25 campaign posters developed with MoH exclusively and collaboratively with religious institutions and development partners. The disciplinary scope is the following thematic areas on COVID-19 messages are as follows: signs and symptoms, social stigma, mental health, how to maintain a strong immunity, wearing of masks, protecting the family, social distancing, handwashing recommendations for children, messages to parents, caregivers and guardians.  Scott (1997) was used for corpus compilation, development and analysis. The Concord was applied to the corpus to establish the distribution and frequency of metadiscourse. A qualitative functional interpretation of the quantitative patterns was done. The interactive visual resources were identified through a qualitative analysis and each poster considered having binary features for example +F-F-frame lines, no frame lines to indicate presence or absence of interactive visual resources.

IV.
RESULTS AND DISCUSSION Guided by the research objectives and analysis of MoH Corpus both manually and using concordance software, Oxford Wordsmith Tool 4, this section discusses the results of interactional metadiscourse and visual semiotic and grammar resources analyses with focus on types, frequencies and types within the e-posters.

Interactive Metadiscourse
The micro searches conducted on the MoH posters corpus revealed the presence of three traditional metadiscursive features (frame markers, code glosses and evidentials). Interactive metadiscourse is because of the writer's assessment of the assumed reader's comprehension capacities, understandings of related texts and used for interpretive guidance (Hyland, 2005). The distribution of interactive metadiscursive features is as follows: The most frequent interactive metadiscursive feature as captured in the above table are evidentials, followed by code glosses. Frame markers are the least used metadiscursive features. The absence of transition markers and endophoric markers can be attributed to the multimodal nature of posters. These features have been replaced by other visual interactive features.

Evidentials
These metadiscursive devices are representations of an idea from another source, which guide the reader's interpretation and establish an authorial command of the subject (Hyland, 2005). Table 4 indicates 3.8 evidentials per a thousand words, which is higher than Hyland's study of academic textbooks, which had 1.7 evidentials per a thousand words. The high frequency of evidentials in comparison with other metadiscursive devices can be attributed to efforts by the Ministry to convince the public of inclusivity of their representatives-the religious groups. Thus, attributions to sources with positive evaluations-religious groups and development partners-are used as endorse the Ministry's message to bridge the trust deficit of the general public with the government. In the posters, the logos and names of institutions involved implicitly establish an authorial command of the COVID-19 message.
Evidentials are used through attribution to stakeholder approach to the COVID-19 pandemic through logos placed at header and footer of the poster. For example, in Figure 1

Fig.1: Poster on signs and symptoms
The logos from the Ministry of Health, religious groups and development partners perform a metadiscursive role by constantly reminding and assuring the readers the source of information. Michael (2008) explains that evidentials are the elements that indicate source of information upon which an utterance is based. In this case, the COVID signs and symptoms the message source is what appears in the header and footer.

Code glosses
These metadiscursive features supply additional information in discourse by rephrasing, explaining or elaborating what has been said, to ensure that the reader is able to recover the writer's intended meaning (Hyland, 2005). Table 4 reveals a frequency of 1.75 code glosses per thousand words slightly higher than to Gonzalez (2005) corpus of virtual bank's web sites at 1.5 code glosses per a thousand words. Generally, the code glosses are used in explaining and rephrasing scientific terminology and jargon which all the readers may not be conversant with. In Figure 2, the poster designers are aware that they are addressing the general public who might not be conversant with the scientific terminology such as "processed foods" hence use of parenthesis (added sugar, fat, salt) to ensure readers comprehend the message in terms of what to avoid in building their immunity in the context of COVID-19. Similarly, in Figure3 above, the use of code glosses has been used to aid reader understand Vitamin B supplements by placing its appearance in brackets (blue capsules) and also expounding on meaning of five food groups daily in parenthesis (3 meals and 2 snacks a day).

Frame markers
These metadiscourse features signal text boundaries or elements of schematic text structure (Hyland, 2005). The devices function to sequence, label, predict and shift arguments making the discourse clear to the public reading COVID-19 campaign posters. As indicated in Table 4, the frequency of frame markers is 1 per a thousand words which is less than Hyland's metadiscourse studies on textbooks (1998,2000) which had 3.8 per a thousand words and Khadohi (2009) on corporate websites with 3 per a thousand words.
In Figure 4, the frame marker has been used to announce goals of public health communication through the demonstrative pronoun these actions. The following campaign poster developed by the Ministry of Health uses numbers (01, 02, 03, 04, 05) to indicate to the reader the order in which the text should be read in terms of the vulnerable groups. The members with underlying medical conditions are considered most vulnerable and elderly placed at the end of the list. Apart from numbering, Figure 5 uses of bulletins and dots are used as metadiscursive features to draw to the attention of readers the preventive issues meant to instill behavioural change. These elements indicate to campaign readers the order in which the text should be read.

Interactive visual metadiscourse
The visual interactive resources used to complement the textual interactive metadiscourse are framing, fonts, connective elements and information value as illustrated in Table 5. Graphs which are extensively applied in academic posters to guide readers are sparingly used in campaign posters. The presence and absence of visual interactive resources is represented by the binary features as captured by information value left-right + and Tryptich-, the former implies presence while the latter absence of the resource.

2.1 Information Value
Information value is created through placement of elements. The posters arrangement whether done unilaterally or collaboratively determines placement of elements. For example, all the posters developed by Ministry of Health as represented by Figure 6 follow the top-bottom order. The posters developed collaboratively as represented by Figure 7 between the MoH and religious institutions follow the left to right order with two vertical columns. The designers of the posters have divided it into two vertical columns suggesting the reader move his gaze from left to right. Symbolically, the arrangement captures shared space with each party using its allotted space to communicate its message. On the left side of the poster, the Ministry communicates visually on social distancing while the religious groups represented by Muslims qualify the message with a religious interpretation on personal space.

Framing and connective devices
Framing plays an interactive metadiscursive role through use of frame lines, colour contrast and empty spaces between texts to distinguish sections of a text. Figure 8 communicates COVID-19 health messages for health care providers. The poster designers use colour contrast-blue and yellow-to guide readers to the propositional content-various coping strategies. Frames are captured in circular and rectangular shapes, the former bearing textual advisory while the latter visual images illustrating the advisories.   The empty spaces play a metadiscursive role in Figure 9 by separating the three messages and focusing our attention to each: first WHO general message on protection of rights within the context of COVID-19; second, the right to equality and dignity; third emphasis on protection of rights and access to information, health care, education for the vulnerable groups during the COVID-19 period.
The poster designers have also used connectives in the campaign posters to guide the reader to show the connection between various sections of textual or visual texts (Kress and Van Leewun, 2006). Figure 9 uses the hand as a symbol of the various rights captured in different colours-freedom, equality, dignity, peace and rule of law. The colour red representing danger is used for COVID-19 while green the target for the pandemic-vulnerable groups. The connective devices in Figures 8 are achieved through repetition of colour and shapes, the blue and yellow colours meant to help readers focus afresh on each coping strategy.
Other connective devices used are vectors achieved through geometrical symbolism and arrows. Figure 10 which communicates to the general public on social stigma presents the message in four boxes with different colours; the first box introduces us to the topic of social stigma through a vector relation by employing straight arrows arranged in form of a taxonomy to direct the reader to message-paying attention to language; the third and fourth vector presents variants of circular arrows reminding readers on personal responsibility in communicating COVID-19 messages to avoid social stigma.