Promoting filial piety through public service advertising: a multimodal discourse analysis approach

2.1 Filial piety in Chinese culture

The Chinese character for filial piety, xiao (孝), is formed by two vertically ordered radicals: 子 ‘the child’ and 老 ‘the elderly,’ with the former supporting the latter from underneath (Nichols 2013: 204). The ideograph is a representation of two important levels of filial piety under Chinese conceptualization – 养亲 ‘providing for one’s parent at all costs’ and 敬亲 ‘respecting one’s parent’ (Zeng 2002: 48–49). As a deep-rooted tenet, filial piety is positioned as the most fundamental virtue in the Confucius tradition, or in terms of a metaphor drawn from The Analects, filial piety is what constitutes the “root” for the “tree” of morality, providing water and nutrition so that all other branches may grow (Lau 2008: 3). As a result, filial piety has become a cornerstone of Chinese society, serving as the guideline for generations of Chinese children and the key to harmonious intergenerational relationships (Yeh et al. 2013: 278).

In early studies on filial piety, the concept was theorized as entailing an inherent hierarchical relationship which assigns parent–child roles along a vertical line (e.g. Ho and Lee 1974). Children’s obligations of filial piety are thus often understood as obedience and self-oppression to parental authority. Later studies, in contrast, manifested an increasing emphasis on emotion and reciprocal affection in parent–child relationships, in which the roles are in equal positions, and these studies thus understand filial behaviors as guided by spontaneous affection (Sung 1995; Yeh et al. 2013). Rather than being mutually exclusive, these two conceptual orientations represent the vertical and horizontal duality of filial piety. While reciprocal filial piety appears to be more potent in today’s context, the two dimensions remain coexistent in Chinese societies (Yeh et al. 2013).

2.2 Challenges for elderly care in contemporary China

In this subsection, we briefly discuss three social factors related to filial piety in contemporary China, mainly with regard to the challenges for elderly care in contemporary China. These include the aging population, insufficient social resources for elderly care, and weakening commitment to supporting the elderly among the younger generation. The first factor is China’s aging population, mainly owing to decades of implementation of the one-child policy. The policy was introduced in 1979 in the context of economic stagnation following the aftermath of the “Cultural Revolution” from 1966 to 1976 (Hesketh et al. 2005: 1171). Facing widespread poverty, the Chinese government acknowledged the new mandate of economic development, an important factor of which is its measurement per capita GDP. To increase this indicator, China deemed it essential to bring down the denominator, which is the country’s total population (Song et al. 1985), and therefore launched a nationwide campaign of population containment, relying on both legal stipulation and government rhetoric to compel and convince people to adhere to the one-child policy (Li and McKerrow 2019). In 2015, the one-child policy was replaced by a universal two-child policy, but thirty-nine years of the implementation of the one-child policy is estimated to have resulted in a net total of 458 million fewer people in the Chinese population (Tao and Yang 2011). Meanwhile, China’s average life expectancy extended from 68 years in 1981 to 77 years in 2018. Sliding fertility rate as a result of the one-child policy, coupled with extended life expectancy led to the aging of the Chinese population. With 7.08% of the population aged over 65 in 2002, China became an aging society according to international standards (Chen et al. 2019). Worse still, China’s aging shows no sign of abating – the population is predicted to become aged and super-aged in 2025 and 2035 respectively, as the elderly expand to 14 and 20% of the total population (Chen et al. 2019).

Second, with such a huge and rapidly expanding elderly population, China’s elderly care resources have become insufficient. Chinese elderly people live on government provision, children’s support, and self-support. Government provision mainly comes in the form of a pension, which has undergone reforms throughout the 21st century with its coverage expanded to a large extent (Liu and Sun 2016). However, due to the low average benefit level and huge regional differences under the current system, many disadvantaged groups still live on very little pension (Liu and Sun 2016). This financial difficulty is coupled with insufficient social infrastructure and support for the elderly. The number of institutions for the elderly (such as nursing homes) in China is still in shortage (Xu and Chow 2011), and only a limited number of care providers in these institutions are trained professionals, whereas quality care usually comes at a high price and is therefore unaffordable for many (Zhang and Yan 2015). Existing pension problems and underdeveloped infrastructure mean that government provision cannot be solely counted on. It is in this context that offspring support for the elderly has been emphasized more than ever before (Zhang and Yan 2015). To this end, China has revised the law to protect the rights of the elderly (Gov.cn n.d.), which not only stipulates financial resources to be provided by children but also specifically states that it is an obligation of children who reside away from their elderly parents to provide regular companionship. By 2020, over 16 Chinese provinces and cities had announced a new scheme under which only-child employees are entitled to “dependency leave” of up to 20 days when their elderly parents are hospitalized (The Economic Daily 2020).

Third, despite these legal measures, offspring support could be unreliable due to shrinking family size and weakened sense of responsibility in supporting the elderly (Wang 2016). Incidents of disrespecting, ignoring, or even abusing the elderly continue to emerge, particularly in China’s rural areas, where the mentality of “being old is being worthless” seems to have spread (B. Zhang 2018: 107). Mistreatment of the elderly is one manifestation of “poverty in spiritual life” (Ma and Liu 2017: 153) despite the increasingly affluent material life. Indeed, economic leaps brought by the reform and opening up since 1978 have come at the expense of morality (Peng 2012), which has downgraded rapidly in the last four decades as people scramble to accumulate wealth. In this context, China deemed it necessary to preserve and promote the sense of filial responsibilities among the public and has been using PSAs an important media strategy to achieve that.

3.1 Data

Our data is collected from China Central Television (CCTV), a state broadcaster under China Media Group. As an official organ, CCTV is regarded as an influential mouthpiece of the authority, and it enjoys large viewership in China (Zhu 2012). The website of CCTV Advertising Center offers a curated collection of PSAs previously broadcast on CCTV channels. Among all the PSAs on this website, eleven were found to promote filial piety as the main theme. In Table 1, the PSAs aired in 2014 are the 弘扬孝道 ‘Promoting Filial Piety’ series, and the other three in 2018 are the 时光逆行 ‘Flashback’ series. The two in 2015 and 2016 were broadcast during the special occasion of Chinese New Year. The duration of each PSA varies from 30 s to 1.5 min.

3.2 Analytical approach

We study filial piety PSAs under the broad conceptualization of discursive governance, which is a form of “implicit mechanisms of governance” that aims to “affect political and social representations within the public sphere in accordance with the wishes of political authorities” by relying on various discursive resources (Korkut et al. 2015: 2). Discursive governance is conceived as a mechanism widespread in all forms of politics, but it also exhibits culturally specific features when implemented in unique sociocultural contexts such as in China (Wang 2017). Notably, for example, China’s discursive governance is found to have integrated cultural resources from Chinese traditional culture (e.g. Perry 2013; Wang 2017).

We see filial piety discourse initiated by an authoritative source as a significant component of discursive governance. This discourse represents “a complex set of discursive strategies” (Fürsich 2009: 240) which integrates linguistic and visual resources in the PSAs. To investigate how these resources are strategically deployed for discursive governance, we take a multimodal discourse approach to studying filial piety PSAs. Specifically, we consider filial piety values as represented through characters’ emotions and attributes in the PSAs, which are realized by linguistic and visual resources on the one hand and are shaped by the broader sociocultural context on the other hand.

The emotions and attributes are modeled using the attitude system proposed by Martin and White (2005). The system includes three subcategories, namely, values of emotional responses (Affect), values according to which human behaviors are socially assessed (Judgement), and values which address the aesthetic qualities of objects and entities (Appreciation). Affect evaluates resources for construing emotional reactions and is divided into un/happiness, in/security, dis/satisfaction, and dis/inclination. Un/happiness refers to the extent of feeling happy or unhappy; in/security deals with anxious or assured feelings about our surroundings; dis/satisfaction refers to feelings of frustration or fulfillment; dis/inclination is about the desire for a condition of events. Judgement assesses human behavior according to social sanction and social esteem. While Judgement of social sanction concerns veracity (i.e. how truthful someone is) and propriety (i.e. how ethical someone is), Judgement of social esteem involves the subcategories of normality (i.e. how special someone is), capacity (i.e. how capable someone is), and tenacity (i.e. how resolute someone is). Appreciation is the evaluation of things, which can be divided into reaction, composition, and valuation. Reaction refers to the degree to which things catch our attention; composition concerns balance and complexity; valuation is to do with the value of things, such as originality and usefulness.

We argue that attitudes do not exist in a vacuum, but always arise out of various social activities (Feng 2016). In social activities, attitudes are organized as a schema which involves three components in a causal relation: the events, behaviors, and objects which cause the attitudes (eliciting condition), the inner feelings (attitudes), and the verbal or nonverbal actions motivated by the attitudes (resultant action), as illustrated in Figure 1. The inner feelings are realized by attitudinal lexis (typically adjectives), such as “happy” (Affect), “kind” (Judgment), and “delicious” (Appreciation). Eliciting condition and resultant action can be realized through verbal or nonverbal resources, for example, visual depiction of an event (eliciting condition) or facial expression (resultant action). To use the attitude toward a person as an example, linguistically, we can say “she is very kind” (directly through attitudinal lexis), “she works as a volunteer to help elderly people” (eliciting condition), or “we all need to learn from her and help elderly people” (resultant verbal action motivated by the attitude). Kindness can also be represented visually by depicting her helping elderly people (eliciting condition) or other people’s respectful or approving facial expressions or actions (resultant action).

Figure 1:

The attitude schema (Feng 2016).

In the PSAs we analyzed, attitudes are constructed in different ways. Linguistically, the most direct way, as mentioned earlier, involves attitudinal lexis such as 爱 (‘love,’ +happiness)^[1] and 挺好 (‘good,’ +propriety) which the PSAs use to encode Affect in the parent–child relationship and Judgement toward filial behavior respectively. This mode of realization is termed as inscription (Martin and White 2005). Apart from inscription, these PSAs also use less direct ways to invoke attitudes (i.e. invocation), through verbal or visual representation of the abovementioned eliciting condition or resultant action. In terms of eliciting condition, for example, Text 1 appears synchronically with the visual depiction of a son playing chess with his aged father, in which case the verbal and visual resources co-construct the event for positive Judgement. In terms of resultant action, we found instances of both verbal actions and material actions. For example, 别爱得太迟 ‘Don’t express your love too late’ is a verbal action motivated by Judgement that it is wrong to postpone filial behaviors.

Text 1: 平时不管再忙，我也会在网上陪我老爸过过棋瘾。 [No matter how busy I am, I always make time for my father by playing chess with him online.] (from Playing chess)

In the coding of the dataset, each attitudinal lexical item in the text is counted as one case of inscription. As for invocation, an instance of attitude is counted each time when its eliciting condition or the resultant action of a character is represented in the text or visual image. In the case where the same attitude is co-constructed both by the text and its synchronous visual image (as in the case with Text 1 and the accompanying visual image), only one instance of attitude is recorded to reduce redundancy. There are also cases where a single utterance or visual image invokes more than one attitude. For example, a visual depiction in which a mother holds her son in her arms invokes both affection (resultant action) and kindness (eliciting condition), so both attitudes are recorded.

4.1 Overall distribution

The distribution of the attitudes in the texts and visual images of the eleven PSAs is presented in Table 2. According to Table 2, filial piety PSAs are attitudinally saturated with 270 instances of Affect, Judgement, and Appreciation. As filial piety mostly concerns people’s emotions and behaviors, the PSAs contain very little Appreciation (3.3%), which is an evaluation of things, for example, 绚烂的烟花 ‘splendid firework’ and 饭菜的香 ‘fragrance of the meal.’ The most represented attitudes are Affect (59.3%) and Judgement (37.4%), two categories on which the following analysis is focused. Overall, the PSAs are instilled with positive attitudes (75.9%) which promote filial piety through socially desirable emotions and attributes.

A wide range of activity types are chosen to construct different attitudes, as shown in Table 3. Cases of different types of activities are counted based on what the participants are doing in each shot of a PSA. Even when a PSA shows only one main activity type, for example a son practicing filial piety toward his elderly father, rather than treating the entire PSA as one single case of “son/daughter practicing filial piety,” each shot showing the son behaving filially, be it helping his father trim nails, preparing food, or chatting with the father, is counted as one single case of “son/daughter practicing filial piety.” Most PSAs contain more than one type of activities, for example by starting with a few shots showing “parent looking after son/daughter” activity but later dedicating more shots depicting “son/daughter practicing filial piety”; therefore in these cases, both activities of “parent looking after son/daughter” and “son/daughter practicing filial piety” are present, with the number of the latter exceeding the former. As can be seen in Table 3, activities participated in by the parent and the son/daughter (“son/daughter practicing filial piety” and “parent looking after son/daughter”) constitute the absolute majority (82 cases), where most positive attitudes occur. In contrast, negative attitudes occur mainly during “parent being alone” (36 cases). This choice of activities constructs a clear message that the well-being condition of the elderly is positive when they are with their children (and grandchildren) and negative when they are not (cf. Diener and Seligman 2002).

In terms of expression, as shown in Table 2, rather than being explicitly inscribed (13%), attitudes are predominantly invoked (87%) by eliciting conditions and resultant actions, with both verbal and visual resources. In particular, the visual mode can be used alone to construct the attitude. Most notably, positive Judgement is commonly invoked with visual representation of the event (eliciting condition), for example, a young adult taking care of the elderly parent by preparing meals and helping him/her taking medicine, whereas Affect is most often expressed through the character’s facial expressions (resultant action). In these cases, verbal account, especially inscriptions, has been purposefully avoided so that viewers are given the agency to appraise on their own rather than being told directly what to feel. Table 4 shows a further breakdown of the realization in terms of inscription and invocation, which will be explained further in the next two subsections.

4.2 Affect

Filial piety PSAs are emotionally charged, containing 160 cases of Affect in which the emotional status of the parent emerged as the central theme (Table 5). Remarkably, the dominance of positive values in all but the in/security category is a pattern consistent both for the parent and the son/daughter. Among the 120 instances of positive Affect, 86 fall into the happiness category, represented as the reciprocal love (+affection) between the parent and child and the delight (+cheer) from parent–child interactions (“son/daughter practicing filial piety” and “parent looking after son/daughter”). Apart from happiness, the remaining 34 cases of positive Affect are scattered almost evenly among the other three categories and mainly function to reinforce the intimacy of the parent–child relationship. They are represented as positive intentions for and feelings of missing each other (+inclination) or satisfaction and relief felt during filial practices (+satisfaction, +security). The use of negative Affect occupies a much smaller proportion (40 cases, 25%). Among them, the majority (25 cases) center on insecurity resulting from the parent’s old-age vulnerabilities. It is often represented as the parent’s loneliness and helplessness, as well as the son’s/daughter’s concern over the parent’s failing health. Meanwhile, 12 instances of negative Affect center on unhappiness, which occur mainly during “parent being alone,” and in some cases, suffering from geriatric diseases. Overall, abundant representation of Affect plays a key role in constructing filial values, as it is the genuine inner feelings of love and respect that are the driving force of filial piety; without such Affect, filial piety cannot last (Zeng 2002).

In terms of realization, firstly, among the 138 instances of invoked Affect (Table 4), the majority (113 cases) are represented by resultant actions, realized mainly through facial expressions and body contact resulting from the inner feelings of the parent and the son/daughter (e.g. characters smiling while holding each other’s arm or a character looking upset, sobbing, or crying). Secondly, the remaining 25 cases of invoked Affect are represented by eliciting conditions which explain why the character is experiencing a certain emotion. Notably, eliciting conditions for negative emotions are all visually represented, whereas those for positive ones are sometimes described linguistically. One case in point is Father’s lie. In Figure 2, the father walking in an empty alley elicits a sense of loneliness (−insecurity). Juxtaposed with this negativity is the positive Affect invoked in the “lies” the father tells his daughter (Text 2), that he is happy and having fun with his friends. The positive emotion the parent is trying to project and the negative emotion he is actually experiencing form a sharp contrast, reminding viewers that they have to be there in person and provide the care and help that the parent needs but rarely asks for. Finally, inscribed Affect is drawn on much less frequently. A total of 22 cases of affective lexis such as 爱 (‘love,’ +affection), 想 (‘want to,’ +inclination), and 安慰 (‘comfort,’ +security) are mainly used to represent the voluntary support and expression of love between the parent and the son/daughter. It is obvious that the linguistic resource is very often reserved for positive Affect only. This may be explained by the tendency toward emotional restraint in Chinese culture, where people see the need to suppress free expression of emotions, especially negative ones, as an act to preserve interpersonal harmony (Wei et al. 2013).

Figure 2:

Example of invoked affect through eliciting condition (screenshot from Father’s lie).

4.3 Judgement

Judgement also plays an important role in the construction of filial values. As introduced above, Judgement targets the attributes and behaviors of the son/daughter and the parent, as shown in Table 6. We look at positive and negative Judgements toward the son/daughter before discussing those targeting the parents. In the category of the son/daughter, one salient feature is that among the total 46 instances of Judgement toward them, all but three fall into the im/propriety category. Judgement of propriety (43 cases) targets the behaviors of the son/daughter, which are mostly kind, caring, and respectful toward the parent (“son/daughter practicing filial piety”) and are judged to be desirable practices. Although small in proportion (3 cases), positive social esteem (+capacity), which represents the capacity of the grownup child, functions to put him/her in a position of physical and financial strength – a precondition for practicing filial piety. In contrast, in the parent category, loss of social esteem is typical for the elderly parent as he/she ages (18 cases of −capacity, −tenacity, and −normality), exemplified in failing health (−capacity), stubbornness (−tenacity), and eccentric behaviors (−normality), for example, grabbing dumplings and pocketing them in a restaurant (Text 3). Nevertheless, this negativity is outnumbered by positive Judgement targeting the parent (37 cases of +propriety and +tenacity), which are used to highlight the selflessness (+propriety) and tirelessness (+tenacity) of the parent in his/her caring for the child (“parent looking after son/daughter”).

In terms of realization, Table 4 shows that Judgement is predominantly invoked, with eliciting conditions bearing much of the attitudinal load (84 cases), different from the pattern observed for Affect. In these cases, eliciting conditions provide the actual behaviors of the character for evaluations, more specifically, by describing and/or depicting the practice of the filial son/daughter and caring parent as well as symptoms of senility experienced by the parent. For example, Text 3 elicits negative Judgement of the parent’s capacity and normality by describing the loss of memory and odd behaviors. Text 4 elicits Judgement of propriety targeting the son and parent by describing how they look out for each other (“son/daughter practicing filial piety” and “parent looking after son/daughter”).

In most of these cases, verbal descriptions are accompanied by visual representation of such behaviors, lending support to the credibility of the descriptions. For example, the son’s narration of his own filial effort (e.g. Text 1) occurs with images of him going on a business trip and playing chess with his father via computer tablet from a hotel room. Notably, visual images are also often used alone to construct eliciting conditions, for example, in Figure 3 the son covering the parent with a blanket elicits positive Judgement of propriety. In contrast, resultant actions (12 cases), which occur much less frequently in the construction of Judgement, are represented entirely with the verbal resource. Although small in number, they take up a unique position in constructing Judgement of im/propriety in the PSAs. In nine out of the eleven PSAs, resultant actions in the form of directives occur and do so only at the very end, recommending what should and should not be done, as an encapsulation of the key message the PSA intends to send. For example, Text 5 ends with the “do’s” and “don‘ts” which appear only after being instantiated with representation of the absence of filial behaviors and preceding emotional context.

Text 3: 不知从什么时候开始，我爸的记性越来越差。冰箱在哪，厕所在哪，他刚刚做过的事，他都忘了 …… 有一天中午，我带他到餐厅吃饭。我爸发现那个盘子里啊还有两个饺子，他竟然用手直接拿饺子装进了口袋。[I’m not sure when it started, but my father’s memory is getting worse and worse. He couldn’t remember where the fridge and washroom are. He forgets things he just did … I took him out for lunch at a restaurant one afternoon. When my father saw there were two dumplings left on the plate, he grabbed them with his bare hands and put them directly in his pocket.] (from Packing food)

Text 4: 现在有车了，想带他们（父母）出去玩。他们却总说不用，说坐地铁方便，说我工作累，该多在家歇歇。总是为孩子着想，就是他们爱的表达方式。[I want to take my parents out now that I have a car. But they always decline, saying that riding the subway is convenient, and that I should rest more at home after having worked so hard. Always prioritizing the child’s needs is how parents express their love.] (from Let’s go together)

Text 5: 别爱得太迟，多回家看看。[Don’t express your love too late. Go home more often.] (from Mother’s wait)

Figure 3:

Example of invoked judgement through eliciting condition (screenshot from Packaging food).