Writer : Oh Jung-shim, PhD
Year : 2021
Next year is the 20th anniversary of the preparation and adoption of the Convention for the Safeguarding of the Intangible Cultural Heritage. It is necessary to ensure the sustainability of intangible cultural heritage (ICH) for the future beyond basic protective activities. Concerning this, it is important to increase higher education and academic systems in ICH. Although it has been almost 20 years since the Convention was prepared and adopted, the results addressing this issue are insufficient. This study collects previous ICH research results and accumulates and investigates the main research subjects, topics and the change and flow of research through the keyword network analysis method based on big data analysis. Through this research, this study contributes to the establishment of the academic system and status of ICH.
research trend, main research topics, knowledge system, academic system, keyword network analysis, knowledge map, big data analysis, big data mining
The Convention for the Safeguarding of the Intangible Cultural Heritage (hereafter referred to as ‘the Convention’) is the first international convention to protect intangible cultural heritage (ICH). It was adopted at the 32nd General Assembly of UNESCO held in Paris in 2003. As of 2020, 180 countries had joined the Convention, an increase from the 30 that had ratified the agreement in 2006. Compared with UNESCO’s other cultural conventions, many countries have joined in a short time, suggesting that the perception of the Convention’s importance has increased. The Convention has a considerable impact on international communities. One of the most distinctive is the universal adoption of the term ICH. Most regions had used their traditional customs and cultural terms as substitutes (Tim 2013; Leem and Roger 2019).
The Convention has similarly revitalised the study of ICH within the field of cultural heritage. Previously, cultural heritage research had been affected by the keystone of UNESCO’s policy and concentrated on monuments, relics and archaeological excavations. After adopting the Convention for the Safeguarding of the Intangible Cultural Heritage (2003), active studies have been conducted concerning ICH, such as human knowledge, beliefs and actions (Shin 2019).
At a recent international conference related to ICH, experts emphasised the necessity of higher education and academic systems for the sustainability of ICH. When the Convention Concerning the Protection of the World Cultural and Natural Heritage was adopted in 1972, the related academic results were collected – and higher education systems were prepared – but not in the ICH field. The issue remains unresolved more than 10 years after the adoption of the Convention, and major students and specialists do not yet understand intangible cultural heritage. The present cultural heritage–related higher education and academic systems are focused on tangible culture. Thus, approaching ICH – a living heritage from the existing system – leads to problems (Galla 2018; Tim 2013).
Because of the importance of collecting previous research results related to ICH and understanding research trends and knowledge systems in the ICH field, this study collects data related to ICH gathered from previous articles since the preparation and adoption of the Convention and analyses the main research subjects, topics and the flow of research. Furthermore, the network analysis uses the keyword network analysis method based on the big data network analysis technique. This method extracts words from massive documents and conducts network analysis of the various characteristics of texts based on the network of the words. Because it allows the analysis of text contexts and structural features quantitatively and objectively, it is often used to supplement existing qualitative research methods (National Information Society Agency 2015; Oh 2020a).
This study aims to collect data from academic papers related to ICH published from 2002, when the Convention was prepared and adopted, to 2020. It analyses the main research subjects, topics and research trends, using the keyword network analysis method. Furthermore, this study will contribute to the establishment of the academic system in ICH research.
This study collected and analysed English abstracts and bibliographical information of a total of 365 articles, searched with the keywords ‘intangible heritage’ in the International Journal of Heritage Studies, Museum International and the International Journal of Intangible Heritage – the representative international journals related to cultural heritage from 2002 to 2020.1 This study considered the following research questions: What subjects are central to the discussion in the ICH field or keys to the organisation of research contents from 2002 through 2020? What are the main topics in the ICH field from 2002 to 2020? How have the main research subjects and the research topics changed and developed during the period? What characteristics appear when the above results are visualised in the form of a knowledge map?
This study was conducted in four steps, including data collection, data preprocessing, data network analysis and synthesis and interpretation (Figure 1). Data preprocessing and network analysis were conducted using NetMiner (4.4.3.b). NetMiner is a software that specialises in big data network analysis. Based on the social network analysis method, an algorithm for big data network analysis combines statistics, graph mining and machine learning.
Data preprocessing should be undertaken to analyse the keyword network. The computer cannot understand the abstracts and bibliographical information of the articles collected in this study, because they are unstructured data comprised of human language. Thus, it has been necessary to transmute it into a standardised structure for computer analysis (Oh 2020a). This study deconstructed the unstructured texts, using the morpheme network analyser of NetMiner, and extracted noun morphemes in a standardised structure. The data were refined applying the thesaurus functions, such as ‘Synonym’, ‘Directive’, and ‘Exception’. Refining the data using the thesaurus increases the accuracy of the network analysis result. The word ‘Synonym’ in the thesaurus defines the function that unifies words with similar meanings. For example, it unifies similar words such as intangible heritage and intangible cultural asset into ICH. ‘Directive’ is the function that extracts proper nouns with no morpheme segmentation as they are. ‘Exception’ is the function that removes the words generally used in abstracts – such as abstract and study – or unnecessary words (Oh 2020b).
As a result of the data preprocessing of the English abstracts of 365 academic papers concerning ICH collected in this study, 3,867 noun morphemes were extracted. With these words, a keyword network analysis was conducted. The main indices of the keyword network analysis were ‘cooccurrence frequency’, ‘degree centrality’ and ‘eigenvector centrality’.
The phrase ‘co-occurrence frequency’ describes the calculation of how frequently the words appear in a certain range. Words with a high frequency often appeared in several papers simultaneously. Moreover, centrality is the calculation of the degree of the position in the central structure of a network by word, which can be divided into degree centrality and eigenvector centrality. ‘Degree centrality’ is the calculation of the total number of interconnected words. Words with a high degree of centrality are connected to other words that are important and central within a text. ‘Eigenvector centrality’ is the value that measures the number of connections and the influence of words. Words with greater eigenvector centrality are those with important influences on the text composition (Kim 2016). Thus, words with greater degrees of centrality and eigenvector centrality; words with great co-occurrence frequency are vital keywords in the text composition or the main research subjects in the relevant field (Oh 2020a).
Museum International and the International Journal of Intangible Heritage (Table 2). These academic journals are SCI international journals that publish the work of researchers with a high number of academic contributions. In particular, the International Journal of Intangible Heritage, founded in 2006, is the first international journal specialising in ICH. Excluding book reviews, 198 articles have been published at the time of writing, and this study selected 177 articles.
The status of the related academic papers must be examined by year. As seen in Figure 2, for ICH-related papers, four articles were published in 2003 when the Convention was adopted. This number sharply increased in 2004, when 27 articles were published. A considerable number of studies were published to interpret the contents of the Convention immediately after its adoption, such as its origin, purpose and implementation method. Following 2004, there was a decline; in 2006, with 17 related articles published, was another increase, more than 10 articles being published each year by 2012. In 2013, 10 years after the adoption of the Convention, the number of articles sharply increased – compared with the previous periods when 21 articles were published. At the time of writing, over 20 articles are published each year.
Co-occurrence frequency is the value that calculates how often the words appeared in a certain range. The words with a high co-occurrence frequency are those commonly used by the authors while writing related papers. As a result of the network analysis, excluding the search term, intangible cultural heritage,2 the word with the highest co-occurrence frequency was ‘community’ (111 times), followed by ‘culture’ (95 times), ‘practice’ (90 times), ‘process’ (69 times), ‘role’ (65 times), ‘value’ (65 times), ‘form’ (65 times) and ‘safeguarding’ (63 times).
Next, degrees of centrality and eigenvector centrality were analysed. These values calculate the degree of the position in the central structure of a network by word. The words with high centrality are those that have important influences on the text composition. As a result of the network analysis – excluding the search word ICH – the word with the highest degree centrality was ‘community’ (0.2180), followed by ‘practice’ (0.1996), ‘culture’ (0.1793), ‘value’ (0.1730) and ‘knowledge’ (0.1715). The result of the network analysis of eigenvector centrality was similar to that of degree centrality. The order was ‘community’ (0.1830), ‘practice’ (0.1681), ‘culture’ (0.1512), ‘value’ (0.1464) and ‘knowledge’ (0.1440).
The words with high degrees of centrality, eigenvector centrality and co-occurrence frequency are interpreted as vital keywords in the text composition or the main research subjects in the relevant field. The words with high values in all indices were identified in Table 3 to examine the keywords or the main research subjects in the ICH field.
The words measured with high co-occurrence frequency, high degree centrality and high eigenvector centrality included ‘community’, ‘practice’, ‘culture’, ‘value’, ‘knowledge’, ‘process’, ‘role’, ‘place’, ‘safeguarding’, ‘development’, ‘identity’, ‘history’, ‘UNESCO’ and ‘2003 Convention’. These words are the keywords and primary research subjects essential in the composition of
academic papers on ICH.
The results were compared by analysing co-occurrence frequency and degree centrality to examine the difference between the main research subjects depending on the academic journal. The analysis revealed that ‘community’ commonly appeared as the main research subject in the three academic journals. Additionally, the contents were summarised and compared (Table 4) to the results of the analysis of the International Journal of Heritage Studies (hereafter, IJHS) and the International Journal of Intangible Heritage (hereafter, IJIH), which identified significant differences. This is the summary of the top 30 words by grading them based on the analytical results.
In a comparison between the two journals, the following differences were found in the main research subjects: ‘conservation’, ‘preservation’, ‘site’ and ‘context’ appeared in IJHS, while ‘hand’, ‘safeguarding’, ‘person’, ‘life’ and ‘activity’ appeared in IJIH. Consequently, the journals seem to differ in their perceptions of ICH.
Meanwhile, NetMiner’s PFNet is the function utilised to keep the key nodes intact and leave important links only. This function is usually used to draw a network map. A PFNet keyword network map was drawn to examine how the words displayed as the main research subjects, including ‘community’, ‘practice’, ‘culture’ and ‘value’, were connected to other words and used as research topics. The topics are identified by arranging the words connected to the keywords (marked in yellow) by looking at the picture. As seen in Figure 4, the main research topics in the ICH field included ‘ICH community (form, resources and government)’, ‘heritage practice’, ‘heritage value’, ‘ICH knowledge’, ‘ICH knowledge and museum’, ‘heritage role and challenge’, ‘heritage process’, ‘UNESCO convention’ and ‘heritage history’.
The words with high co-occurrence frequency, degree centrality and eigenvector centrality are interpreted as keywords essential in generating texts or the main research subjects in the relevant field. In Table 6, the words with both high frequency and high centralities by period include ‘ICH’, ‘value’ and ‘community’ in Period 1; ‘ICH’, ‘museum’ and ‘community’ in Period 2; ‘ICH’, ‘practice’ and ‘community’ in Period 3; and ‘ICH’, ‘safeguarding’ and ‘community’ in Period 4.
It is interesting to note that network analysis was conducted on ‘community’ as a main research subject in all the periods. This suggests that ‘community’ has been an important research subject in the ICH field irrespective of the period. The Convention commentary book, published by UNESCO, also specifies that the role of the community is important in the protection of ICH. UNESCO emphasises that regular performance and learning should be undertaken among generations in the community, so the intangible can come to life (ICHCAP 2019; UNESCO 2010).
Next, a PFNet keyword network map was drawn to examine the connection between the keywords, as the main research subjects, and the other words, to show how they were used as topics. The examination revealed that the main research topics in Period 1 (2002–2006) included ‘ICH description (definition, preservation)’, ‘community value’, and ‘community role’. Period 1 is the period during the adoption of the Convention. This suggests that the studies were mainly conducted on the definition of the concept of ICH and the direction of the interpretation of the Convention. It is important to note that, instead of ‘safeguarding’, ‘preservation’ is connected to ‘ICH description’.
The main research topics in Period 2 (2007–2011) can be examined using the same method. As seen in Figure 6, the main research topics were ‘ICH preservation concept’, ‘Convention practice’ and ‘museum theory’. In Period 2, studies related to the Convention’s execution method and the implementation of the Convention through museum theory were conducted.
The main research topics in Period 3 (2012–2016) were ‘ICH development’, ‘ICH community form’ and ‘ICH policy’. As seen in Figure 7, compared with Periods 1–2, the number of words connected to ‘ICH’ increased in Period 3, and their meanings were diversified. From this information, studies of ICH on various topics were clearly conducted in Period 3. Of note, ‘development’ appeared, which had not been seen in the previous periods, and ‘preservation’ was not among the words connected to ‘ICH’. ‘ICH development’ can be interpreted as the development and use of ICH or the sustainability of ICH. UNESCO announced that it would also aim at the sustainability of ICH and the future protection of ICH – at the sixth session of the General Assembly of the States Parties to the Convention for the Safeguarding of the Intangible Cultural Heritage held in 2016 (ICHCAP 2017).
The main research topics in Period 4 (2017–2020) were also examined (Figure 8). First, the words that were excluded in the previous periods, including ‘person’, ‘process’, ‘life’ and ‘heritage context’, appeared in connection with ‘ICH’ in Period 4. The main research topics were ‘ICH person’, ‘ICH process’, ‘ICH and heritage context’, ‘community practice’ and ‘community culture system’.
This suggests that the understanding of ICH as living heritage was broadened and that studies on this broadened understanding were actively conducted. Moreover, ‘preservation’ is notably not included in the words connected to ‘ICH’ in Period 4. Instead, ‘safeguarding’ is connected. ‘Safeguarding’ and ‘preservation’ appear to be similar words, yet they have completely different meanings. ‘Preservation’ is a word with static and passive implications, observing culture as an object to preserve. Meanwhile, ‘safeguarding’ focuses on the process of human-based holistic understanding (ICHCAP 2019). In summary, almost 20 years after the adoption of the Convention, the understanding of ICH as living heritage has expanded, and, in place of ‘preservation’, ‘safeguarding’ is universally used.
This study collected the article data in the ICH field from 2003 – when the Convention for the Safeguarding of the Intangible Cultural Heritage was prepared and adopted – to 2020. It analysed the main research subjects, research trends and the change and flow of research, using the keyword network analysis method. Furthermore, this study contributes to the establishment of an academic system in the ICH field.
This study collected the English abstracts and bibliographical information data of 365 articles, searched with the keywords ‘intangible heritage’, from three representative international cultural heritage–related journals, the International Journal of Heritage Studies, Museum International and the International Journal of Intangible Heritage. It also involved a keyword network analysis. The study results can be summarised, focusing on the research questions presented in section 2, as follows.
First, as a result of the co-occurrence frequency analysis, ‘community’ had the highest frequency, followed by ‘culture’, ‘practice’, ‘process’, ‘role’, ‘value’, ‘form’ and ‘safeguarding’. This suggests these words are the most commonly used when the authors prepared ICH-related articles.
Second, this study found and examined the words with high co-occurrence frequency, degree centrality and eigenvector centrality. The words with high values of the three indices were identified as the main research subjects in the ICH field, or the essential keywords in preparing the related texts. The examination revealed that ‘community’, ‘practice’, ‘culture’, ‘value’, ‘knowledge’, ‘process’, ‘role’, ‘place’, ‘safeguarding’, ‘development’, ‘identity’, ‘history’, ‘UNESCO’ and ‘convention’, were keywords – or the main research subjects.
Third, to examine if the main research subjects differed depending on the academic journal, cooccurrence frequency analysis and centrality analysis were conducted to compare the results. According to the analysis, ‘conservation’, ‘preservation’, ‘site’ and ‘context’ appeared in the International Journal of Heritage Studies. Meanwhile, ‘safeguarding’, ‘person’, ‘life’ and ‘activity’ appeared in the International Journal of Intangible Heritage. This suggests differences in the perceptions of ICH between the two journals.
Fourth, to examine the change and flow of research in the ICH field, a keyword network analysis of data was conducted and compared by breaking down the period into Period 1 (2002–2006), Period 2 (2007–2011), Period 3 (2012–2016) and Period 4 (2017–2020). As a result of the co-occurrence frequency and centrality analysis, the value of ‘community’ was high in all periods. The analysis indicated that ‘community’ was an important research subject in the ICH field.
Fifth, the change and flow of the research results of keyword network analysis were compared by drawing a PFNet network map for each period. The examination revealed that the main research results of keyword network analysis in Period 1 (2002–2006) and Period 2 (2007–2011) were ‘ICH description’, ‘convention practice’ and ‘museum theory’. Studies – such as the direction of the interpretation of the Convention and the implementation method – were conducted. In Period 3 (2012–2016), compared with Periods 1 and 2, the number of words connected to ICH increased and their meanings diversified. Consequently, in Period 3, studies of ICH were actively conducted with various results of keyword network analysis. Last, in Period 4 (2017–2020), ‘ICH person’, ‘ICH process’, ‘ICH and heritage context’, ‘community practice’ and ‘community culture system’ appeared without precedent. This suggests that the understanding of ICH as living heritage has deepened and studies related to this deeper understanding have been conducted.
Sixth, a change in the perception of ICH occurred throughout the periods. In Period 1, ‘preservation’ was connected to ‘ICH description’; however, in Period 3, ‘preservation’ disappeared, and in Period 4, instead of ‘preservation’, ‘safeguarding’ appeared and was connected to ICH. ‘Preservation’ implies passivity, viewing culture as an object to be preserved. Meanwhile, ‘safeguarding’ focuses on the process and people involved in ICH, based on a holistic understanding of heritage. In summary, almost 20 years after the Convention was adopted, understanding ICH as living heritage has deepened and – instead of preservation – safeguarding has been used universally.
This paper facilitates the understanding of research subjects – the result of keyword network analysis – and the change and flows in the ICH field by collecting and analysing academic data previously accumulated in the ICH field. This study evaluated the contents of the abstracts intuitively by presenting the research results in the form of a knowledge map. The research methodology of this paper can be used to understand and investigate future global issues related to ICH.