12 : Media Discourse about using fish at a local festival in Korea

Seola Joo

Center for Animal Welfare Research, Research Institute for Veterinary Science, College of Veterinary Medicine, Seoul National University Seoul, Korea


Myung-Sun Chun

Center for Animal Welfare Research, Research Institute for Veterinary Science, College of Veterinary Medicine, Seoul National University Seoul, Korea


Growing public awareness of animal welfare has extended ethical concerns on how to treat fish at cultural events.  To investigate the media discourse on using fish for entertainment, we analyzed news articles on the social issues surrounding the Sancheono (cherry salmon) Ice Festival, the largest local fish festival in Korea, with text mining analysis. The articles from the 11 Korean national newspapers with the keyword “Sancheoneo Ice Festival” (2013 to 2020) were analyzed. The main topics on the festival changed from the economic perspectives of this festival to animal welfare and ecological concerns. This shift demands a more profound animal welfare evaluation during the festivals to enhance their sustainability and value.


animal welfare, festival, media analysis, text mining, fish


Using animals in cultural events, such as animal fighting, animal acting, and petting mammals has been strongly criticized for its inhumaneness. Some of these cultural events were discontinued due to public pressure. With growing public awareness of animal welfare, ethical concerns extend to traditional fishing and leisure fishing, especially how fish move and die during the events.

The Sancheoneo (cherry salmon) Ice Festival, held in Hwacheon-gun, Gangwon-do since 2003, is the largest local festival using fish in Korea. It attracts more than 1 million visitors each year. In 2019 the festival organizers released eight hundred million cultivated Sancheoneo (Oncorhynchus masou) into the rivers for ice fishing and bare hand catching (Ji, 2019). However, long-distance transfer and inhumane treatment of the fish are criticized by animal advocates and academics. 

In this study, we analyzed the discourse and implications of the social issues in media surrounding the Sancheoneo Ice Festival. Because news articles report not only the contents of the issues but also the interests of various stakeholders in the issues and public opinions from various perspectives, big data analysis of the news is an effective method for comprehensively identifying major topics that reflect social phenomena (Kim, 2018; Yu, 2017). Analyzing media discourse about the festival revealed a change in public perspectives on fish as live animals, and the way fish are used for human amusements. 





1. Collecting data (Unit of Analysis)

We collected the articles with the keyword “Sancheoneo Ice Festival” from all 11 Korean national newspapers (January 1st, 2003 to July 13th, 2020) using BigKinds (www.bigkinds.or.kr), a big data system for news. The data system offered semantic morphemes from each article. Among the collected 1,180 articles we used 1,102 of them for the analysis after removing duplicate articles. 

The collected data was divided into three stages and analyzed in consideration of the number of articles and important events of the festival to compare trends. Phase 1 was from 2003 to 2011 with 324 articles involving the early stage of the festival. Phase 2 was from 2012 to 2017 with 389 articles in the period of the festival’s success. The last phase was the crisis of the festival from 2018 to 2020 with 389 articles.


2. Data analysis 

Text mining is utilized in diverse fields for discovering meaningful patterns and rules in vast amounts of data. It can be applied to eliciting core topics, presenting changing topics, and analyzing trends (Park, 2019a). Among the text mining technics, we used word cloud and text network analysis to analyze not only the semantic structure revealed but the structure embedded in the context by making up a network on what relationships the words form (Paranyushkin, 2011) with NetMiner 4.0 program.

First, we rearranged the extracted words by removing stop-words, refining similar words, excluding non-relevant words, and defining meaningful phrases (Park, 2018). We analyzed our unstructured data in Korean and translated the final result into English (Park, 2017; Park, 2018). To remove the words commonly found in documents, such as “festival” or “Sancheoneo,” we extracted only words with a 0.2 TF-IDF (Term Frequency-Inverse Document Frequency) value,  indicating how much weight a particular word takes up within a document; therefore, it is more accurate than simple frequencies (Choi, 2019).

To illustrate the most frequently used keywords of each phase, we conducted word cloud analysis by frequency (top 100). The network analysis with visualization was conducted based on the co-occurrence matrices of the keywords and degree centrality. In the sociogram, the larger nodes (red circles) imply higher centrality, and the bolder links (lines) mean a higher connectivity strength between the two keywords, as a higher frequency of co-emergence (Park, 2019b). 




The main issues surrounding the Sancheoneo Ice Festival have changed in each phase. The word clouds showed different words with high frequency (top 100); “fishing,” “experience,” “contest,” and “smelt” in phase 1 (2003-2011), “ice,” “fishing,” “trout,” “international,” and “culture” in phase 2 (2012-2017), and “animal,” “trout,” “management,” “preparation,” and “fishing spot” in  phase 3 (2018-2020) (Figures 1-3).


Figure 1: Phase 1 Word cloud (with frequency for top 100)

Figure 2: Phase 2 Word cloud (with frequency for top 100)

Figure 3: Phase 3 Word cloud (with frequency for top 100)



In the first phase, the main keywords were related with building an identity as a local-based winter festival for families that included fishing or sledding. Additionally, cancellation due to the nationwide FMD (foot and mouth disease) crisis was one of the main issues in this period. The sociograms showed a strong link between the words “fishing,” “smelt,” and “experience,” which indicates the festival’s main focus of catching winter river fish (Figure 4). 

The main topics in phase 2 described factors for the huge success of the festival, such as “international,” “local government,” “management,” “culture,” and “representative.” The emerging words “management” and “preparation” in the sociogram linked with the main topics indicate that the festival was not a simple traditional winter festival with a local specialty anymore, but an artificial event with massive cultivated fish under the control of the regional government (Figure 5). 

In the third phase, the newly emerged main topics showed criticisms about animal welfare status. The schedule delay due to the warm weather was another issue. In addition, the conflict over the festival between the Environment Minister's concerns about animal welfare and the Hwacheon county governor advocating for the local economy was also importantly debated. 

The word “animal” in the sociogram was linked to the word “group” and “experience” (Figure 6). The former showed active criticism of animal advocate groups, and the latter included what pain the animals in the festival could experience.



Table 1: The Top 20 Keywords by Centrality Index

Rank Phase 1 Phase 2 Phase 3
 1 fishing ice animal
2 smelt fishing preparation
3 experience preparation schedule
4 contest management management
5 village open fishing spot
6 sledge culture international
7 family Korea Pyongchang
8 tourism experience experience
9 join local government event
10 international international group
11 progress host governor
12 product smelt foreigner
13 Pyongchang Pyongchang Choi Munsun
14 nationwide trout delay
15 local government representative plan
16 use progress minister
17 culture product trout
18 fishing spot contest weather
19 FMD tourism culture
20 host street  representative

Figure 4: Sociogram with the Top 10 Keywords of phase 1

Figure 5: Sociogram with the Top 10 Keywords of phase 2

Figure 6: Sociogram with the Top 10 Keywords of phase 3



The findings of the analysis showed that debates on ecological effects, animal welfare, and human-animal interaction among stakeholders, authorities, and the public recently became more intensive than ever before. The media reflected the change of public perception of the Sancheoneo into living animals that could experience pain. The public awareness toward animal welfare has been strengthened not only for pets and farm animals but also for fish. This shift demands a more profound animal welfare evaluation during the animal using festivals to enhance their sustainability and value. 

From a food ethical standpoint, the animals, used in the festival, had been the “absent reference.” (Adams, 2003) The Sancheoneo used for the festival were mostly considered only an object (tool) consumed for entertainment or food. Little was known where they come from and how they are raised. A Sancheoneo is a fish that lives only in a clean environment. They are considered as native species in the Hwacheon area where the festival is held. However, this species cannot inhabit in that area naturally. The cultivated Sancheoneo from all over the country were transferred to artificial iced streams in the festival area. The 90 percent of the whole domestic cultivated Sancheoneo were used for the festival. (Lee, 2017). During the adapting process to the cold water, which is not suitable for them, many of Sancheoneo died. They are suffering from a lack of food and the fear of contact with humans. With the growing recognition of fish as sentient animals, the welfare status of the fish became an animal issue. In addition, the current government's animal-friendly policies and strengthened animal laws are strengthening public awareness of animal welfare. The Sancheoneo case may be an evidence of the public belief on ethical value of responsibility in using animals for humans.  





This research was partly funded by the Ministry of Education (2019S1A5A2A03047987) and was supported by the National Research Foundation of Korea (NRF). 



  • Park, Gyung-Yeol and An, Hee-Ja. 2019a. The Topic Modeling Analysis of The DMZ Tour Issues Using Text Mining. Journal of Tourism and Leisure Research 31(4): 143-159 
  • Park, Chan Sook. 2019b. Using Text Network Analysis for Analyzing Academic Papers in Nursing. Perspectives in Nursing Science 16(1): 12-24.
  • Choi, Jeong Sil and Kim, Youngji. 2019. A Study on Research Trend for Nurses’ Workplace Bullying in Korea: Focusing on Semantic Network Analysis and Topic Modeling. Korean Journal of Occupational Health Nursing 28(4): 221-229.
  • Park, Eun-Jun, Dae Woong Ahn, and Chan Sook Park. 2018. Text Network Analysis of Newspaper Articles on Life-Sustaining Treatments. Journal of Korean Academy of Community Health Nursing 29(2): 244-56.
  • Paranyushkin, Dmitry. 2011. Identifying the Pathways for Meaning Circulation Using Text Network Analysis." Nodus Labs 26. http://noduslabs.com/research/pathways-meaning-circulation -text-network-analysis Accessed 30 July 2020.
  • Kim, TaeJong, Park, SangOk. 2018. Topic Analysis of Local Lifelong Education Using News Big Data - From 2003 to 2018 local media articles. Journal of Cybercommunication Academic Society 35(4): 37-102. 
  • Park, Chankook and Yong. 2017. Taeseok. Prospect of Korean nuclear policy change through text mining. Energy Procedia 128: 72-78.
  • Yu, Ye-Lim. 2017. Analysis of Media Coverage on 2015 Revised Curriculum Policy Using Big Data Analysis. Doctoral dissertation, Seoul National University.
  • Ji, Hwan. 2019. YTN. https://www.ytn.co.kr/_cs/_ln_0115_201901051536542051_005.html. Accessed 26 July 2020. 
  • Adams, C. J. 2003. The rape of animals, the butchering of women. The animal ethics reader, 209-15.
  • Lee, Seokjong. 2017. BBS. https://news.bbsi.co.kr/news/articleView.html?idxno=858808. Accessed October 6, 2020.