The Generational Effects on Collective Memories in the Context of International Migration: The Collective Memories of Koreans in South Korea and the U.S.

Abstract

This article examines generational effects on collective memories of Korean history, while taking into account international migration. We asked 216 subjects in South Korea and the United States to name three important events in Korean history and to provide reasons for their selections. We found generational effects in both countries in a similar pattern. This is a remarkable social achievement of the U.S. emigrant subjects. The current study adds a cross-cultural perspective to the literature on collective memories, which has focused predominantly on U.S. and Western case studies. By comparing memories of people who share a national origin but live in different cultural contexts, the current study also intersects collective memories studies in other fields such as transnationalism and diaspora. Our findings suggest that future studies can benefit from a transnational approach to collective memories, which may or may not circulate across borders.

View correction statement:

Corrigendum

Acknowledgments

We would like to express our deepest gratitude to Eviatar Zerubavel, Howard Schuman, Amy Corning, and Barry Schwartz.

Notes

¹In fact, there have been scholarly works that demystify assertions of the Korean national homogeneity (e.g., Em 1999; Grinker 1998; Kwon 2000; Lie 1998; Seol 2004, 2006; Seol and Skrentny 2004, 2009). These works highlight the long history of division on the Korean peninsula, the multi-ethnic or labor immigrant demographic “realities” in Korean society. Thus, the common belief in the “long history as [a] one people nation” and “homogeneity” reflects the nationalist perceptions of homogeneity in South Korea and this should not be misunderstood as demographic/historic realities. In this vein, we want to make it clear that the notion of cultural homogeneity has little to do with actual homogeneity. Whether the belief in self-homogeneity is supported by evidence of history is not our problem in this article. What we mean by Korea as a culturally homogeneous country is not historical facts but the belief commonly held in public while we acknowledge the problematic nature of the belief.

²We recruited research subjects by convenience sampling. We acknowledge the limited generalizability of findings drawn from the sampling method. Thus, we make no claim that this sample is an accurate representation of Koreans in South Korea and the United States. However, we also note that the limitation in the sampling does not diminish this study's contribution to the literature. This is the first study to empirically test how generations affect the collective memories of an emigrant group and compare them to their homeland counterparts. As evidenced by the findings in this exploratory study, this issue warrants further study.

³Logistic regression revealed that generational effects are unaffected by respondents' level of education.

⁴Population statistics were drawn from the 2005 Census of Korea. Retrieved May 20, 2010 from http://kosis.kr/nsportal/abroad/abroad_01List.jsp.

⁵ Statistics regarding Korean immigrants in the United States are from the 1-Year Estimates of the 2008 American Community Survey. Retrieved May 20, 2010 from http://factfinder.census.gov/servlet/IPCharIterationServlet?_lang=en&_ts=298775404218.

⁶Later studies have modified the original time frame of the questionnaire designed by Schuman and Scott (1989). The original 50-year period was changed to “the past 60 years” in surveys conducted in the United States and Britain (Scott and Zac 1993) and Israel (Schuman et al. 2003). In the German case, the question states “about 1940” instead of the original “about 1930” in consideration of German history related to World War II (Schuman et al. 1998). In the Japanese case, the time frame was given as “the beginning of the Showa Era” because “the chronology used in the West is not so well known in Japan” (Schuman et al. 1998:431).

Note. Each of three event mentioning responses is not mutually exclusive.

Note. Respondents were asked to name three events, and the event choices were not mutually exclusive.

*p < .05, **p < .01, ***p < .001 (t-test).

Note. Respondents were asked to name three events, and the event choices were not mutually exclusive.

*p < .05, **p < .01, ***p < .001 (t-test).

Note. Each respondent's reason could be counted in as many as in three categories, thus each reason category of an event is not mutually exclusive.

*p < .05, **p < .01, ***p < .001 (t-test).

⁷Because respondents' reasons varied in length and complexity, reasons were not coded discretely. It was rather straightforward to code responses about how the Korean War divided “our nation,” “our territory,” or “our people” as Division of Our Nation. We coded respondents' interpretation of the war as a sad, horrible, horrific, tragic, unforgettable, painful, or regrettable historic event as Emotional Impact. The respondents also recognized the Large Consequential Impacts of the War (e.g., the war perpetuates the division of our nation, the division caused by the war has slowed down our nation's growth, or the intense hatred against socialism and communism limits intellectual and cultural growth). The code of Ideological Conflicts between Democracy and Communism indicates the respondents' perception of the war as the invasion of democracy (represented by South Korea) by communism (represented by North Korea) or as the triumph of democracy over a communist attack. The older respondents also mentioned their Personal War Experiences. The code of War Damages includes the respondents' statements of great loss of life and property, such as “the war destroyed the entire country.” Lastly, Tragedy of Fratricide was coded verbatim. The term is one of the most common phrases used in South Korea to characterize the war.

Note. Each respondent's reason could be counted in as many as in three categories, thus each reason category of an event is not mutually exclusive.

*p < .05, **p < .01, ***p < .001 (t-test).