The Taiwanese people's sense of Taiwanese-ness and/or Chinese-ness has been a core subject of debate and discussion in various sociopolitical and cultural realms. The most salient and widely accepted origin of the consciousness of Taiwanese-ness dates back to the era of Japanese colonization when local intellectuals began a series of bottom-up initiatives of democracy and self-determination.Footnote 1 The sense of Chinese-ness, on the other hand, was introduced and inculcated in a top-down authoritarian manner by Chiang Kai-shek 蔣介石 and the Kuomingtang (KMT) following their retreat to Taiwan in 1949. Although the expression of Taiwanese consciousness was largely suppressed and considered politically incorrect, it has become uncensored, widely propagated and even a sign of pride for the people of Taiwan since the democratization process was initiated immediately after the lifting of martial law in 1987. The issue of people's recognition of their identity as Chinese and/or Taiwanese continues to be debated and researched, particularly because this self-identification is essential to the cross-Strait relational issue of claiming independence from or reunification with mainland China.
Existing studies on the national identity of Taiwanese citizens tend to utilize either macro or micro approaches to understanding the factors that influence a sense of Chinese-ness or Taiwanese-ness. The macro approach mainly involves the revision of the historical context, the trajectory of colonization and subjugation of the Taiwanese people, which shaped the Taiwanese consciousness.Footnote 2 The micro approach discusses the theory of socialization and the ways in which different socializing agents, such as parents and school, affect children's consciousness of their ethnic, cultural, national and political identity.Footnote 3 While the effect of parents on their children's political identity is significant, the pattern in terms of the effect of father or mother varies across different societies. This study explores the patterns of parental influence in the political socialization of children in Taiwan by looking at the association between parents’ identities and the extent of individuals’ Taiwanese-ness.
The study begins by discussing the oft-cited socio-demographic perspective to explain the formation and transformation of ethnic identity. This is followed by a literature review of the parental effect on children's political identity in other societies. In Asia, ethnic identity is often linked with political affiliation, a factor which tends to be overlooked in the North American literature and perspectives. Using Taiwan as a case study, we bridge the literature by leveraging the theory of compressed modernity – a path of modernization that is differentiated from advanced capitalist societies and socialist transitional societies. We argue that Taiwan exhibits a pattern of parental effect that diverges from other developed societies, such as the US and Japan, as well as transitional societies like China. Entrenched in Confucianism and a patriarchal system, Taiwan has experienced classic modernity (transitioning from an agrarian society to an industrial society) and second modernity (moving from an industrial society to a reflexive network or information society) in a compressed manner. Although the socioeconomic and political aspects of the society are well developed, the cultural aspect lags behind.Footnote 4 Consequently, fathers become more important than mothers in shaping the identity of offspring.
Our study makes several contributions to the literature on the effect of parental socialization on individuals’ national identification. First, by applying the theory of compressed modernity, we bring a macro perspective to ethnic identity formation and transformation, which are often perceived from a micro- or meso-level socio-demographic perspective in the literature. Second, our study contributes to the as yet inconclusive findings on the various patterns of parental influence in political socialization across various societies. Third, our findings also bridge theories from the macro- to micro-level to iterate the importance of contextual differences in determining the extent to which micro-level theories of political socialization are applicable. Fourth, we incorporate the political dimension, which, although neglected by many existing studies on ethnic identity, is relevant to the circumstances of most Asian societies, where ethnic identity usually is linked to a particular political affiliation. Finally, this study suggests that the consciousness of national and/or political identity among Taiwanese is dependent on the extent to which a patriarchal culture persists in the society.
Literature Overview
Debate on identity formation/change
There are three broad schools of thought on identity formation: primordialism, structuralism and constructivism. Primordialism claims that an individual's racial and ethnic identity is given and fixed and that identity formation or change is based on one's ineffable attachment to immediate and kin connections and the givenness from a particular social community.Footnote 5 Structuralism identifies the ways in which the structure of a society shapes the strength of an individual's ethnic identity.Footnote 6 Constructivism also looks at the ways in which identities are formed or transformed according to contextual characteristics, but it focuses more on the development of ethnic identity.Footnote 7 These three broad perspectives examine the ways in which group identities – rather than individual identities – are formed or changed. Consequently, several other perspectives that focus on structural and group dynamics, such as ethnic competition theory, boundary-making and pan-ethnicity, bridge the gap by discussing the manner in which social contacts affect an individual's sense of ethnic or group identity.
Ethnic competition theory discusses how friction between different groups increases with frequency of contact, especially in the competition for resources in the labour market. This competition triggers conflicts that push individuals and groups to mobilize along ethnic lines through collective cultural experiences and grievances, thus reinforcing individuals’ ethnic identity.Footnote 8 In other words, ethnic identity fluctuates with group dynamics in a society. Similarly, the boundary-making perspective emphasizes the clarity of ethnic or group boundaries. When a boundary is clear and rigid, it is difficult for individuals to cross it and adopt a different or new identity, thus rendering a stronger identity among those already within the boundary.Footnote 9 Pan-ethnicity, however, examines the formation or transformation of ethnic identity with the assumption that people are naturally rational and calculative. Thus, the ethnic identity of an individual shifts between different and/or even politically contradictory identities depending on which identity is conducive to the individual gaining resources.Footnote 10 Empirically, these approaches apply separately to individuals from different immigrant generations. They fail to discuss the retention or changes of ethnic identity between generations. Hence, the socialization approach – a micro perspective – tries to bridge this gap in the literature.
Parents’ role in political socialization – a micro perspective
The socialization perspective emphasizes the transmission of norms, values and beliefs from an agent with higher-order power to individuals with lower-order power in a unidirectional manner.Footnote 11 For instance, in the US, level of education, which is a socializing agent, has been positively associated with Asian immigrants’ sense of ethnic identity, because education transmits knowledge and awareness of race and ethnicity, which enhances Asian Americans’ ethnic identification.Footnote 12 Parents, as primary agents of socialization, also shape the political identity of their children significantly.Footnote 13 Such findings demonstrate that the structural dynamic perspectives of ethnic competition theory, boundary-making and pan-ethnicity are dependent on the micro-level interactional perspective. Robert Hess, Judith Torney and Jaan Valsiner targeted pre-teenage and pre-adult individuals to examine the ways in which parents influence their children's voting behaviour, political participation and party affiliation.Footnote 14 However, their study fails to discuss the effects of fathers and mothers separately or the significance of the role they each play in political socialization. Existing debate on the question tends to be an “either-or” proposition.
The father-dominant thesis is a functional perspective that emphasizes the father's role as the instrumental-adaptive guide to the extra-familial world, given his role as the major breadwinner within the family and the mother's expressive-nurturant role.Footnote 15 In this perspective, the social division of labour in families accentuates the role of the father as the chief agent of socialization in political matters, representing instrumental action in relation to the world beyond the family.Footnote 16 However, contradictory to the father-dominant thesis, many studies have demonstrated empirically that mothers exhibit greater influence than do fathers in political socialization, even with changes to the family structure whereby mothers are increasingly integrated into the labour force in modern society.Footnote 17 Interestingly, in circumstances where parents are discordant in their political identities, mothers still have a slight edge in transferring their political values to their children in the US.Footnote 18
Although culturally different from the US, China also displays a pattern of political socialization that undermines the father-dominant thesis to some extent. Confucianism explicitly inculcates and normalizes gender inequality and endorses traditions for a patriarchal society and respect for elders.Footnote 19 Even after a series of modern reforms, Confucianism remains the core culture that defines the roles of different members in a family. Thus, it is intuitive to conclude that the father-dominant thesis should be salient in the Chinese context. In one study, however, even though talking with fathers had a higher correlation coefficient than talking with mothers in influencing children's political interest, it was clear that mothers still played a significant role as agents of political socialization for families.Footnote 20
Japan, like China, is also heavily influenced by Confucian values. Therefore, it would be intuitive to deduce that Japanese mothers and fathers are significant agents of political socialization for their children, with fathers having a slight edge in the process. Counterintuitively, Japan follows a pattern similar to that found in the US, where children display a higher degree of agreement on party support with their mothers than with their fathers,Footnote 21 and the relationship between children's interaction with their fathers and their interest in politics is insignificant.Footnote 22 Such findings on the substantial influence of mothers on the political socialization of children are often explained by social learning theory and the different styles of verbal modelling adopted by fathers and mothers. It is suggested that mothers assert more influence on children's political affiliation because they are more likely than fathers to encourage open discussion rather than imposing their values.Footnote 23
Nevertheless, social learning theory cannot explain how the influence of fathers varies in different societies. Contextual differences from a cultural perspective also fail to resolve the conundrum of the different roles of fathers and mothers in political socialization. Therefore, it is not sufficient to hypothesize from the father-dominant thesis, social learning theory, or from the cultural perspective of Confucianism, to explain the patterns of parents’ effect on the political socialization of their children.
Compressed modernity and Taiwanese identification
Modernization is another contextual pattern of the contemporary era, but it remains underexplored in the political socialization of children by their parents. Classical or first modernity in a society is marked by the rapid transformation from an agrarian to an industrial order. During the process of transition, social and institutional arrangements undergo obvious changes.Footnote 24 Although many Western or other well-developed societies have shifted away from industrialization, they have developed areas of finance, commerce and technology. These metamorphoses also disrupt the old social orders and introduce new ones. Such societies are considered to be in the stage of “second modernity,” which features global risks, individualization and cosmopolitanism. Footnote 25
Although many societies have moved from first to second modernity, the trajectory of modernization differs across contexts, depending on the degree of compression and condensation across time and space.Footnote 26 There are three types of society that can be differentiated in terms of their relative compression and condensation. The first is the advanced capitalist society, represented by Western liberal, social democratic and developmental states, and Japan, which is autonomous and has little external pressure to achieve second modernity. The advanced capitalist society comprises radicalized reflexivity stemming from its own intent, which is the driving force of scientific, cultural, political and economic inputs and interests. Such a society experiences a low degree of time–space compression.
The second type of society is the late-developing or underdeveloped capitalist society, which is characterized by “dependency.” Its transition to first and then second modernity is subject to the risks of radicalized reflexivity because of its subordination to more advanced capitalist societies and global actors, and/or its seeking collaboration, assistance and learning opportunities from them. South Korea, Hong Kong and Taiwan are examples of such societies.
This dependency also is found in the third type of society, which is the formerly socialist society transitioning to a capitalist society – for example, Russia and China. This society, however, is differentiated from the second type by its initially planned socialist modernization, which relies on a centralized and closed sociopolitical and economic system. This then leads to structural, economic and social crises resulting from “the built-in failures of the endlessly self-reflexive command economic system.”Footnote 27 For these societies, the time–space compression is much greater. Moreover, the degree of compression is contingent on the “historical and contemporary entanglements between Western and non-Western societies in colonial and postcolonial contexts.”Footnote 28
Among East Asian societies, South Korea is an exemplar of a late-developing society that experienced a high degree of compressed modernity towards the end of the 20th century. While advanced capitalist societies took at least one and a half centuries to develop their first and then their second modernity, South Korea achieved the same within less than half a century. South Korea's rapid achievement of Western modernity has brought risk consequences of accidents and hazards that in many ways are similar to those experienced by Western societies earlier, but which were overcome through social development. This experience has made South Korea not only a risk society but also one that is highly complex and unique because of a “cultural lag” in the rapid process of industrialization and material expansion that hinders the development of scientific knowledge and expertise, improvements in the technology of risk assessment and enforcement of safety measures.Footnote 29 The cultural lag stems from the deeply rooted Confucian value of respect for the seniority of elders and superiors rather than for meritocracy or expertise. The serious consequences of this lag were demonstrated by a number of Korean airliner crashes towards the end of the 20th century. The deference paid to senior pilots meant that junior pilots felt unable to report problems or recommend solutions.Footnote 30
In many ways, Taiwan is similar to South Korea in terms of its history of Japanese colonization, authoritarian rule, rapid industrialization in the latter half of the 20th century and the cultural influence of Confucianism. In addition, both Taiwan and South Korea have lived with the threat to their security by Communist regimes for 70 years, even as they transitioned from military rule to democracy. Moreover, both have been allies of the US and subject to its vast cultural, political and economic influence. The high degree of compressed modernity is undeniable. The “Taiwan economic miracle” – of rapid industrialization and economic growth, and quick transition and advancement as one of the world's leading manufacturers in the technology sector – has occurred within half a century. This phenomenon of highly compressed modernity, with the resulting cultural lag, has been used to explain Korea's unique and complex risk society. In 1949, Taiwan, depleted of resources, received a mass influx of immigrants from the mainland. Under Chiang Kai-shek and KMT rule, rapid economic development was prioritized, and classical modernity was introduced, transforming Taiwan from an agrarian to an industrial society for the first time in history. By 1960, owing to the limited capacity for consumption on the island, the domestic market for light consumer goods was saturated, and so, from 1960 to 1973, the KMT switched from import substitution to export substitution industrialization, pushing Taiwan's economy towards greater liberalization by establishing export processing zones (EPZs) and bonded factories.Footnote 31 By the late 1960s, manufactured goods had replaced agricultural goods as Taiwan's main export, and by the early 1970s, the manufacturing sector overtook the agricultural sector to become the chief sector in economic growth.Footnote 32
However, the infrastructure, which was conducive to the rapid expansion of Taiwan's economy through industrialization, was no longer adequate for economic growth. The “Ten major construction projects” were implemented to improve the island's transportation, communication and port facilities, paving the way for its second modernity, which transformed Taiwan into a knowledge- and information-based society.Footnote 33 By the late 1970s, as a result of increasing wages and land prices, taxation related to pollution and other protectionist legislation related to industry and manufacturing, Taiwan was losing its labour-intensive advantage and many domestic and foreign factories had begun moving to China or South-East Asia. To keep Taiwan's economy competitive globally, the government transformed the island's industrial structure with the introduction of a plan to increase the share of capital- and technology-intensive industries in the manufacturing sector.Footnote 34
In the early 1980s, Taiwan's economic growth was driven by the knowledge-based industry.Footnote 35 By the 1990s, rising wages pushed greater numbers of labour-intensive industries to move offshore. With a decreased job market for labourers, Taiwanese people were forced to become a high-quality human resource. In addition to comprehensive industrial clusters, there was considerable expansion of the information technology (IT) industry in Taiwan.
At the same time, demand for specialized service industries was increasing.Footnote 36 In the 1950s, the primary sector amounted to more than 50 per cent of the total labour employment; in 1999, it accounted for only 8.3 per cent.Footnote 37 In contrast, tertiary sector employment increased from less than 5 per cent in the 1950s to 54.5 per cent in 1999. Looking at the structural transformation, the percentage share of gross domestic product (GDP) for the agricultural sector decreased from 32.2 per cent in 1952 to 2.6 per cent in 1999, but there were significant increases in the manufacturing sector (from 12.9 per cent to 26.4 per cent) and the service sector (from 48.1 per cent to 64.3 per cent) over the same period.Footnote 38 Taiwan's economic achievement has been miraculous, transforming within half a century to first and then second modernity, a transformation that took most other advanced capitalist societies one or two centuries to accomplish.
Politically, Taiwan was under an authoritarian regime during the time of Chiang Kai-shek and the KMT. Towards the end of the 1980s, during and after the Chiang Ching-kuo 蔣經國 era, the political system began to liberalize. Similar to the process of compressed economic development, the modernization of Taiwan's political system, including the development of the rule of law, the protection of civil liberties and political rights, was achieved within a quarter of the time it had taken the advanced capitalist societies.Footnote 39 However, the existing literature fails to acknowledge the ways in which such a shift is potentially derived from macro social changes affecting the pattern of political socialization.
Apart from compressed modernity, another contextual factor that should not be overlooked is the change in the institutional arrangements for the household registration system. Between 1945 and 1992, household registration in Taiwan had been the responsibility of the police department. A newborn's household registration was determined according to the father's place or province of origin.Footnote 40 Consequently, a child's officially registered identity was defined by his/her father's identity. While it is possible that an individual's officially registered identity may be associated with his/her perceived national identity, the extent to which it affects his/her national identification remains a puzzle we seek to explore in this paper alongside compressed modernity.
Given that Taiwan's experience of compressed modernity is theoretically different from the experience of advanced capitalist societies like the US and Japan or (formerly socialist) transitional societies like China, we suggest the following hypotheses.
Hypothesis 1: the pattern of influence of parents on their children's sense of Taiwanese-ness after experiencing compressed modernity should be different from patterns in the US, Japan and China.
Given that Taiwan is similar to South Korea in many ways in terms of historical contingencies and, most importantly, the experience of a high degree of compressed modernity, we assume that there is a cultural lag in Taiwan. The traditional Confucian cultural value of patriarchy penetrates many institutions in Taiwan, including families, and continues to be influential today despite modernization.Footnote 41 Therefore, we propose:
Hypothesis 2: after the experience of compressed modernity, the father is the sole agent of influence on a child's sense of Taiwanese-ness.
To corroborate the theory of compressed modernity and the experience of cultural lag, we look at differences in the extent to which respondents from different parts of Taiwan experience compressed modernity. Since compressed modernity refers to rapid industrialization, urbanization and (in Taiwan) advancement in technology, it is most likely to occur in urban districts, which therefore will experience a higher degree of cultural lag. Rural districts in Taiwan have significantly fewer demands for such rapid transformation, so the cultural lag is much less. Therefore, we propose:
Hypothesis 3: only individuals living in urban areas that have experienced compressed modernity will identify their fathers as the sole parental influence on their sense of Taiwanese-ness.
Research Method
Data
This data used in this study derive from the 2013 Taiwan Social Change Survey (round 6, year 4): National Identity. The survey and data collection were devised and executed by the Institute of Sociology, Academia Sinica, and include questions on “national identification” from the 2013 International Social Survey Programme (ISSP) focusing on three aspects. The first aspect concerns views on cross-Strait relations, policies regarding cross-Strait relations, and identification issues regarding Taiwan or the Republic of China (ROC). The objective is to understand whether people in Taiwan possess irreducibly essential values or whether there can be changes in the collective consciousness contingent upon changes in the actual situation. The second aspect focuses on the significance of the measurement of individual variables in connection to the extent of global transformation, as well as the density, breadth and significance of contact with mainland China through cross-border exchange. The third aspect is on understanding the importance of changes at various stages of an individual's life, so the survey also includes questions concerning respondents’ views on politics across different cohorts (which are discussed in the later part of this section). The finalized and cleaned dataset is available for public access, enabling us to use it for this research.
The sampling target of the full dataset includes residents of Taiwan who were 18 years of age at the time of data collection.Footnote 42 When the survey was administered, the investigators deployed stratified three-stage probability proportional to size (PPS) sampling, and 2,000 cases were completed for the group on national identification.
The first sampling unit was township, followed by village and then the individual. The sampling was originally stratified into seven strata based on the population and urbanization indices of 358 townships and cities.Footnote 43 However, owing to the extremely small population in the seventh stratum, it was combined with the sixth stratum for convenience of research administration.
Measurement and variable adjustments
(1) Sense of Taiwanese-ness
The concept of Taiwanese-ness is the dependent variable in this study. Indicators include the extent to which respondents identify themselves as Taiwanese on a scale of 1 to 10, where 1 represents not Taiwanese at all and 10 represents absolutely Taiwanese.
For this study, we use Walker Connor's concept of national identity.Footnote 44 According to Connor, the psychological and spiritual bond between co-nationals is rooted in the common belief that they are ethnically related, and this belief renders a dichotomous view of the social world.Footnote 45 In other words, for our study, the respondents’ perception of their own identity is either Taiwanese or otherwise. Moreover, we have run the regression for combining 8–10 versus 0–7 and 9–10 versus 0–8. The results are more or less the same as the results for 10 versus 0–9. Consequently, choosing “10” on the scale – absolutely Taiwanese (wanquan shi Taiwan ren 完全是台湾人) – suggests a very strong sense of Taiwanese identity, and respondents who do so form a group with a common conviction; those who choose values other than 10 do not always share the same conviction, and therefore are “other.” We readjusted the variable by changing “10,” which represents “absolutely Taiwanese,” to “1,” while combining all the cases that chose “9” or below into another group, “0,” which is redefined as “otherwise.”
We realize that identity is dynamic and mutable. However, it is not the purpose of our study to investigate the factors of identity transformation or to attempt to foresee how national identity varies over time. Rather, we stress the fundamental feature characterizing nationalism, which is that “belonging to a nation does not only imply knowing who ‘we’ are but also recognizing who are the ‘others’.”Footnote 46
(2) Parents’ identity
One of the independent variables in this study is the identity of respondents’ parents. As we intend to compare the influence of father versus mother on the respondents’ sense of national identity, the indicators are based on respondents’ reports of their parents’ places of origin.Footnote 47 The choices originally included “Fukienese of Taiwan,” “Hakka of Taiwan,” “Aborigine,” “Mainlander,” “Second-generation mainlander,” “Kinmen or Matsu,” “South-East Asian” and “Other.” “Mainlander” and “Second-generation mainlander” represented those whose household registration was for Chinese provinces other than Taiwan.Footnote 48 Because the focus of the study is on the ways in which parents’ identification as Taiwanese or (mainland) Chinese affects their children's self-identification, we created a new variable for the identities of father and mother by combining the first three categories into “Taiwanese native,” combining “Mainlander” and “Second-generation mainlander” into “Mainlander,” and omitting the last three categories.
(3) Experience of compressed modernity
Compressed modernity is a macro-level social phenomenon that began with modernization. Although there is no indicator in the survey asking respondents about their experience of compressed modernity, we still are able to create a variable based on year of birth. As the development of contemporary Taiwan has been dependent on the way in which the state has been governed, we divided respondents into three groups: those born before 1949, who did not grow up in any form of modernity; those born during the period when Chiang Kai-shek was in power (1949–1977), who experienced only first modernity while growing up; and those born when Chiang Ching-kuo was in power or later (1978 to present), who grew up in compressed modernity.
(4) Rural/urban resident
To further test the validity of the explanation of compressed modernity with the experience of cultural lag, we decided to compare residents in rural areas with those in urban areas. The indicator used in this study is the respondents’ own report of whether they live in a city or village. The questionnaire provided them with five options: a big city, the suburbs or outskirts of a big city, a small city or town, a country village, or a farm or home in the country. We categorized the first three options as “urban” and the last two options as “rural.”
(5) Control variables
Previously, Kuan Chen Lee and Wan-Ying Yang used the same dataset and concluded that there was a generational difference in the way that the 228 Incident affected respondents’ sense of Taiwanese-ness.Footnote 49 The 228 Incident was an anti-government uprising triggered by the police brutality of a 40-year-old widow, Lin Chiang-mai 林江邁, who demanded the return of her confiscated contraband cigarettes. Prior to the event, local inhabitants were growing increasingly dissatisfied and resentful of the frequently corrupt activities of the KMT authorities, such as the arbitrary seizure of private property, mismanagement of the economy and exclusion of local residents from political participation. Therefore, we accounted for this possibility in our models by controlling for the extent to which respondents consider the 228 Incident a significant historical event that the next generation should remember. The first control variable is place of residence in Taiwan. Because different industries and industrial sectors are concentrated in different parts of Taiwan, residential location could demonstrate how compressed modernity differs by region. According to the administrative division of Taiwan, the northern regionFootnote 50 has a high concentration of electronic information, semiconductor and photoelectric industries; the core industry of the middle and eastern regions is agriculture,Footnote 51 although the middle region also has some industries related to machinery and mechanical tools, precision machinery and photovoltaic panels.Footnote 52 The southern region is renowned for solar panels and medical devices.Footnote 53 Although the survey does not ask about place of residence in terms of region, we created this variable by combining locations that are in the same region.
Second, we controlled the level of education for both mothers and fathers. By doing so, we could identify whether the influence of each is based solely on the cultural beliefs of the society rather than the power or status obtained by the parents through education. Although there are 22 options in the survey for level of education, we simply re-categorized them into none/illiterate, self-study, elementary school, (vocational) junior high school, (general/vocational) senior high school, junior college or above, and military/police school or college.
Third, we controlled for the respondents’ own level of education in order to see the extent to which school is influential as an agent of political socialization compared to the influence of father and mother. We re-categorized the levels of education in the same manner as we did for parents’ education.
Fourth, we controlled for gender to observe whether there is a gendered effect on the attitude and behaviour of mother and father towards the respondent in conveying their value of identity. Fifth, the number of years of education is controlled in addition to level of education as an alternative indicator for schools as agents of socialization. Controlling for two indicators of the same concept can help to corroborate the extent to which parents are truly influential as the primary agents of political socialization. Lastly, we controlled for respondents’ sense of pride in the nation, to check that their sense of Taiwanese-ness has been established as a result of a unidirectional influence from primary agents of socialization rather than from their own calculation or evaluation of various national phenomena with which they were nurtured by other agents of socialization. We derived the variable from a set of questions that asked about the respondents’ sense of pride regarding ten aspects of Taiwan's development on a scale of 1 to 4, where 1 represents “very proud” and 4 “not proud at all.”Footnote 54 Assuming that each aspect is equally important, we first reversed the scale, so the greater the pride, the higher the score. Then, we averaged the responses to obtain the mean score and rounded it to the nearest number between 1 and 4 to indicate respondents’ sense of pride in being Taiwanese.
Analysis model
For the analysis, we deployed logistic regressions because the dependent variable (Taiwanese-ness) is binary. We first included the core independent variables: father's identity, followed by mother's identity, then identities of both mother and father. Next, we added the levels of education attained by father and mother. The controlled variables were added last.
Empirical Findings and Analysis
Descriptive analysis
Table 1 shows the cross tabulation of father's and mother's identities on respondents’ sense of Taiwanese-ness. As shown, 70 per cent of all respondents reported being “absolutely Taiwanese” while 30 per cent had reservations about making such a claim. Among those whose fathers are Taiwanese, the majority (66 per cent) are “absolutely Taiwanese,” which is almost triple the number who chose “Otherwise” (24 per cent). Among those whose fathers are mainlanders, the number choosing “absolutely Taiwanese” is only two-thirds that of those choosing “Otherwise.” The same pattern applies for the cross tabulation of mother's identity and respondent's sense of Taiwanese-ness: 30 per cent of the respondents chose “Otherwise” and 70 per cent chose “absolutely Taiwanese.” Similarly, in cases where the mother is Taiwanese, the majority of respondents (67.5 per cent) defined themselves as being “absolutely Taiwanese,” and the rest (26.7 per cent) claimed “Otherwise.” The number of those whose mothers are mainlanders and who identify themselves as “absolutely Taiwanese” is about three-quarters of that of those whose mothers are mainlander and who identify themselves as “Otherwise.” Therefore, there is a preliminary association between parents’ identity and respondents’ sense of Taiwanese-ness. To determine the separate effects of the identity of father versus mother, logistic regression is needed.
Table 1: Frequencies for Extent of Taiwanese Identity and Father's Identity and Mother's Identity
Logistic regression analysis
Table 2 presents the results of a logistic regression of parents’ identities on respondents’ sense of Taiwanese-ness, without separating them according to whether they were born before or after compressed modernity, and controlling for other variables.Footnote 55 The identity of both father and mother, when regressed separately (in Models 1 and 2 respectively), is shown to be statistically significant. Moreover, respondents whose fathers are mainlanders, when compared to those whose fathers are Taiwanese, are more likely to consider themselves to have an identity other than just Taiwanese (-1.199). The result is the same for the mother's identity (-1.112). However, when we control for the identities of both father and mother, as presented in Model 3, mother's identity becomes insignificant. Even after controlling for mother's level of education in Model 5, mother's identity remains insignificant. On the other hand, father's identity remains significant even after we control for his education as well as all the other control variables. In addition, the variable of residing in an urban or rural area is consistently significant throughout all ten models presented. Respondents who reside in rural areas are more likely than those residing in urban areas to report being “absolutely Taiwanese.”
Table 2: Logistic Regression Model for Respondents’ Sense of Taiwanese-ness and the Identity of Father versus Mother
Notes: *** p < 0.01, ** p < 0.05, * p < 0.1.
The question now is whether it is always the case that only the father's identity matters in shaping his children's sense of Taiwanese-ness. We first analyse the pattern of the effect of the identities of mother and father for respondents who were born during the Chiang Kai-shek era before Taiwan experienced compressed modernity. The result, shown in Table 3, demonstrates again that the effect of the father's identity remains consistently significant for respondents born during the Chiang Kai-shek era. The result also is compatible with the results shown in Table 2 that respondents whose fathers are mainlanders, compared to those whose fathers are Taiwanese, are significantly more likely to consider their identity as not solely Taiwanese. Furthermore, mother's identity plays a significant role for those born in the Chiang Kai-shek era. The effect is consistently significant even after controlling for all other variables. Respondents whose mothers are mainlanders are more likely than those whose mothers are Taiwanese to report a weaker sense of Taiwanese-ness.
Table 3: Logistic Regression Models for Respondents Born during Chiang Kai-shek Era and the Identity of Father versus Mother
Notes: We used a vif to check for multicollinearity and did not find any collinearity issue between independent variables and other controlled variables. *** p < 0.01, ** p < 0.05, * p < 0.1.
To compare with those who were born during compressed modernity, we ran another set of logistic regression models solely for those born during or after the Chiang Ching-kuo era. The results are presented in Table 4. Again, the effect of father's identity is consistently significant even after controlled variables are included in the model. Compared to respondents whose fathers are Taiwanese, those whose fathers are mainlanders are more likely to regard themselves as not solely Taiwanese. This result is congruous with the results in Tables 2 and 3. However, for those born during or after the Chiang Ching-kuo era, the identity of their mother is rendered insignificant throughout the regression, even when we regress only the effect of mother's identity on the respondents’ own sense of Taiwanese-ness.
Table 4: Logistic Regression Models for Respondents Born during or after Chiang Ching-kuo Era and the Identity of Father versus Mother
Notes: No high VIF scores were found, indicating no major collinearity issues among variables. *** p < 0.01, ** p < 0.05, * p < 0.1.
Consequently, our first and second hypotheses are supported by the empirical data. We believe that the experience of compressed modernity has had an influence on the ways in which fathers and mothers socialize their children's sense of Taiwanese-ness. Moreover, the respondents’ level of education is not consistent throughout the regressions in Tables 2 to 4. Thus, the school, as an agent of socialization, probably matters a lot less than parents in Taiwan.
To corroborate the argument of compressed modernity, we delve further into the comparison between residents of rural and urban areas. However, we encountered a lack of cases reporting to be living in rural areas, so we were unable to compare the groups. Nonetheless, we compare respondents residing in urban areas who were born during the Chiang Kai-shek era (Table 5) with those born during or after the Chiang Ching-kuo era (Table 6). In both periods, father's identity remains significant throughout all models, and those whose fathers are mainlanders still are more likely than those whose fathers are Taiwanese to report their identity as not solely Taiwanese. For urban respondents born during the Chiang Kai-shek era, mother's identity still is somewhat consistent in its significance except for instances when father's identity and level of education are controlled without controlling for mother's level of education. It is clear that mother's identity still has some significant influence and that the effect is consistent with father's identity. In Table 6, when we look only at urban respondents born during or after the Chiang Ching-kuo era, the effect of mother's identity becomes totally insignificant. Thus, the effects of compressed modernity and cultural lag still are salient, although we cannot completely prove Hypothesis 3 without having another regression just for rural respondents.
Table 5: Logistic Regression Models for Urban Respondents Born during Chiang Kai-shek Era and the Identity of Father versus Mother
Notes: *** p < 0.01, ** p < 0.05, * p < 0.1.
Table 6: Logistic Regression Models for Urban Respondents Born during or after Chiang Ching-kuo Era and the Identity of Father versus Mother
Notes: No high VIF scores were found, indicating no major collinearity issues among variables. *** p < 0.01, ** p < 0.05, * p < 0.1.
Discussion and Conclusion
Although Taiwan has developed into an advanced capitalist society like the US and Japan, the identity socialization effect of mothers is not greater than that of fathers. Similarly, despite sharing the same Confucian culture and influence as parents in China, parents in Taiwan do not share the same degree of influence on the national identity of their offspring. The results of our study indicate that it is the father, rather than the mother alone or father and mother together, who has significant influence on the self-identification of individuals. Therefore, our findings have ruled out the macro developmental and cultural perspectives for understanding identity formation and transformation.
Our study delves into a historical process rather than a state-of-being, which would be the developmental perspective, or a habitual phenomenon, to understand why Taiwan presents a different pattern from the US, Japan and China. Taiwan's unique modernization process is termed “compressed modernity,” whereby there is rapid development in the socioeconomic and political state to catch up with the West, but a cultural lag remains. This process is the key to understanding how the pattern of primary agents in political socialization in Taiwan differs from the patterns in Western societies and East Asian societies like China and Japan.
The empirical results show that before Taiwan experienced compressed modernity, both fathers and mothers were significant primary agents of political socialization and were predictors of respondents’ sense of Taiwanese-ness, although fathers in general had greater influence. Nevertheless, since the experience of compressed modernity, mothers have become completely irrelevant to predicting respondents’ sense of Taiwanese-ness, while fathers remain significant. Even after controlling for the temporal differences in the bitter memories of the 228 Incident in 1947, the findings remain consistent. This situation is expected as a result of compressed modernity, because Taiwan remains highly influenced by the Confucian culture of patriarchy and, although many aspects of the society have developed further, that cultural belief remains deeply ingrained in the hearts of the people. Thus, the father is the only primary agent of political socialization. Our findings are consistent with the research conducted by Lee and Yang, which shows that the 228 Incident was a more significant event for older than for younger Taiwanese generations in forming and reinforcing their sense of Taiwanese-ness.Footnote 56
We are aware that the contextual background of the official household registration system between 1945 and 1992, when newborns were registered according to the province of their father's birth, may explain how respondents’ Taiwanese identity is associated with their father's identity.Footnote 57 However, our findings only partially support this contextual factor. As demonstrated by those who were born during or after the Chiang Ching-kuo era, while there is a significant relationship between father's identity and respondents' self-perceived identity as Taiwanese throughout all six models, mother's identity is not significant. However, in Table 3, which demonstrates the logistic regression model for respondents born during the Chiang Kai-shek era, the identities of both father and mother are significantly associated with respondents’ perceptions of themselves as Taiwanese or otherwise. If the institutional context is influential, we should expect only father's identity to be significantly associated with respondents’ self-identification as Taiwanese or otherwise. The partial finding for the institutionally contextual factor of household registration also strengthens our argument regarding the effect of compressed modernity in Taiwan on the shift in the pattern of parental political socialization of Taiwanese identity.
The findings from respondents residing in urban areas point to the same result – the identity of both parents had significant effect on the identity of their children before Taiwan experienced compressed modernity, and only the father's identity had significant effect afterwards. This shift corroborates further our central argument that the macro historical process affects the micro interactional level of the political socialization of identity.
Our study echoes the view that the condition of compressed modernity exists in Taiwan, although not specifically in the way that parents of different class “had differential access to globalization and uneven relations with the modernity projects.”Footnote 58 Moreover, our results support Dominic Yang's historical finding that parents are the main agents of political socialization of identity formation through memory production.Footnote 59
Although some studies point to the importance of school as an agent of political socialization, our findings suggest otherwise when we consider the effect of fathers and mothers, the primary agents of political socialization.Footnote 60 We suggest that future studies explore and juxtapose the extent to which primary and secondary agents of political socialization are significant, as our findings cannot provide an answer.
One limitation of our study is the lack of variation among rural respondents to demonstrate how a weaker degree of compressed modernity affects the pattern of parents’ political socialization. As our study focuses only on Taiwan, a comparison between residents of urban and rural areas is essential to further corroborate the theory of compressed modernity. Therefore, we suggest that future studies should either include societies which have experienced different patterns of modernity or else collect data with a greater sampling number to derive enough variation to execute analyses.
Our study demonstrates a preliminary conclusive attempt to address the question of why different societies exhibit different patterns for fathers and mothers transferring identity to their offspring. We point to the importance of a macro perspective to explain micro-level differences. Specifically, a macro perspective is not just a matter of state-of-being but a dynamic historical process – of modernization in our case. Future research may consider expanding on the modernity perspective to interpret the changing patterns of the primary agents of political socialization in other societies, or even consider comparing different societies with various modernizing trajectories to examine their resilience and applicability.
Overall, our study contributes to the literature on ethnic identity formation and transformation in which the political dimension of nation-building is often neglected. The findings from our study also imply that the culture of patriarchy is still active in Taiwan despite the modernization of socioeconomic and political aspects. Therefore, there continues to be a latent gender polarization in politics regarding the shaping of individual political identity. More attention should be paid to gender equality in political influence in Taiwan.
Acknowledgements
We are thankful for the suggestions from Charles Fung Chi Keung and the financial support of the HKU Presidential Scholarship of the University of Hong Kong.
Competing interests
None.
Yun-Tzu CHANG is a PhD student in politics and international studies at the School of Oriental and African Studies (SOAS), University of London. Chang completed her bachelor's and MPhil degrees in sociology at The Chinese University of Hong Kong. Her current research interests include identity and diasporic politics, and civil society.
Eric FONG is chair professor of sociology at the University of Hong Kong. He is the current president of the Hong Kong Sociological Association. Fong has published widely in the areas of immigration and race and ethnicity. His most recent book is entitled Segregation and is published by Polity Press.
