1. Premise¹

Since 2014, basic online searches of the word Hong Kong will quickly reveal an association with the word protest. This association originated in 2014 when large-scale sit-in protests erupted in the wake of an announcement of Hong Kong’s controversial electoral reforms. This occupied international news formed a word association, which began to determine social attitudes towards a host of entangled subjects. Solidarity with these protests in the public sphere had soon pushed 100,000 people to join protests on the streets (Chan 2014). As a means of dispersing the crowds, police fired tear gas and pepper spray at protesters who used umbrellas to shield themselves. Images across international media made the umbrella an international symbol of these protests, which later became known as the Umbrella Revolution (Lee et al. 2015). Nearly a decade later, Hong Kong and its people still find themselves in a state of continued struggle with their identity, which will mark a generation.

Artists and creative practitioners have responded in various ways. When invited to contribute to the edited volume Momentary Glimpses: An Anthology on Contentedness (Chan 2019), the artist Ho Sin Tung (何倩彤) offered six blank pages. Her physical presence on the streets of Hong Kong, she said at the time, was in lieu of any printed response. In its silence, the blank pages represented a loudness elsewhere, an urgency that demanded something more immediate than printed matters: a thesis articulated in action. Yim Sui Fong (嚴瑞芳) offered a series of photographs taken on the streets of Hong Kong, not of the crowds of protesters as depicted in international press, but of quieter moments of an activist’s day: ribbon tied to lampposts, details of the road and pavement. By night, a helicopter is photographed passing overhead, blockades are constructed of fencing and cable ties, and the ribbon is now strewn across the floor. The melancholy of these images is raw and blunt.

Silas Fong (方琛宇) was teaching in Seoul at the time, and his contribution to this volume felt restrained. His pages consisted of a series of sentences each beginning with I marked across lines of yellow school paper. I am an artist, the first one reads. Redefining the form of the anthology offered an opportunity for artists to manifest another form of linguistic expression, transcending the book into action and non-verbal communication; the reverb of language translating from one context into another, embodying with it a live and active concurrent practice elsewhere, and making visible the entanglement of actions and their consequences which simultaneously occur in different social and political situs.

The artistic voices articulate a nuanced imagination of post-colonial Hong Kong and Chinese/Hong-Konger identities, differing in how they respond to the protests. Some focused on the banal everyday interactions that characterise day-to-day city life ––something that would have likely resonated with residents at the time–– whereas others turned inward and reflected on what large-scale societal change meant for them personally and professionally.

While previous studies have shown that the Hong Kong protests have largely been reported in the UK press as a democracy vs. Beijing dichotomy (Sparks 2015), the political and ethnographic identities articulated are less clearly defined. There is also currently little assessment of how arts institutions are representing Hong Kong and Chinese identifying voices. In the present paper, our interdisciplinary approach aims to explore the ways in which these voices have been presented in England’s curatorial imagination since the Umbrella Revolution, and how this impacts the articulation and public reception of these voices. Our critical engagement began with rendering England’s cultural organisations as visible places of intervention and interference (Barad 2007; Chan 2020). We approached the task of exploring this representation in greater detail using corpus-assisted discourse analytical methods.

The journey begins with the exhibition’s curator, whose role is to decide what should be more or less visible as part of an exhibition. Curators hold a high degree of power here; whilst their job is to present an exhibition that will attract visitors, they make a series of decisions in shaping what is visible or hidden to the public. The curated exhibition, then, acts as an apparatus through which audiences develop knowledge. For the present study, we identified press releases as documents which contain an expression of such an apparatus, articulating its intentions and setting out its boundaries. Press releases are issued by museums or galleries to announce new exhibitions and are aimed at the press. In a nutshell, they are tools ultimately used for promotion. We focused on publicly funded institutions through the analysis of press releases from Art Council England’s National Portfolio Organisations (NPOs), which represent a publicly funded body of work available for interrogation. Since we recognise that there is a significant difference in production timescales between producing exhibitions and news reporting, our study covers several years between the end of 2014 and the beginning of 2020. Through a corpus-assisted analysis of this collection of writings (Baker et al. 2008; Partington et al. 2013; Gillings et al. 2023) and by incorporating into the analysis the theoretical notion of ‘diffraction’ proposed by Barad (2007), we can render the images of what is considered Chinese identity from the curatorial articulations presented in the documents and explore the implications of this for artists and for society.

The basic notion of diffraction that many people will have come to know of during elementary science classes, is usually demonstrated by the passing of a white light source through a glass prism. This demonstration renders visible the fundamental elements that make up what we normally see as white light, as a spectrum of different colours; a pattern of differences which collide and overlap to form what we understand as our visibilities and realities. Our realities are formed of many collisions, and when these collisions collide with other matter(s) and form(s) (like the prism), realities are diffracted over and over. Quantum physics invites us to consider the inherent diffractivity of such constituent parts and the understanding that matter comes to matter.

In the arts and the humanities, we come to this invitation by employing diffraction, figuratively, forming methodologies and practices that are inherently difference-attentive. In specific conditions, and with the apparatus of the social sciences, we can see and interpret patterns of difference that are revealed and begin to make sense of how these differences make up our realities. Through the simple demonstration of diffracting light through a glass prism, we may understand diffraction as a means of making visible the constituent parts that make up our experiences of the world. This thinking may help us to map social and cultural phenomena as models of diffraction in order to engage with the genealogies of their realities and the constructions and conceptualisations of what we come to know, understand, and experience in ways which reveal their entanglements. Diffractive methodologies present non-separational models of our social worlds and lead us to consider non-separational models of identity, including identities and imaginations of race, ethnicity, nationality, and politics which we will focus on in this paper. We will discuss our findings in the context of an understanding that cultures and identities not only interfere, influence, and impact the experiences of one another, but they also mutually establish one another through difference making, meaning that individuals are mutually implicated in the lived experiences of all others.

While certainly a relatively small and specific corpus, that is, a “specialised corpus” (Koester 2022: 48), an interrogation of the dataset provides critical insight into the themes, concerns, and areas of practice which have been given a publicly funded platform in England and have inevitably contributed to the public discourse of Chinese identity nationally. This article, thus, aims at answering the following research question: how is Chinese identity represented in publicly funded art gallery press releases in the wake of the Umbrella Revolution?

Our paper is organised as follows. In Section 2, we look at synergies between art and linguistics, focusing on previous work which has explored art gallery and museum communication. In Sections 3, we outline our corpus, while the keywords and key semantic domains attested in the corpus are presented in Section 4. Section 5 focuses on these key domains in more detail and provides an interpretation of what they mean within the context of our principal research question. In Section 6, we close with some concluding thoughts.

2. Synergies between art and linguistics

2.1. Theoretical similarities

International media, as well as creative and artistic responses, exist as lenses through which we make up our knowledge, stories, imaginations, and world view. The apparatus through which we encounter the world, and which forms our imagination, shapes our reality. Language and behaviour are in constant negotiation with reality, responding, reacting, influencing, and reinforcing. In the words of Fairclough (1992: 3), “discourses do not just reflect or represent social entities and relations, they construct or ‘constitute’ them” (see also Mautner 2016). Discourse is multidimensional, that is, it is a text, but simultaneously also a discursive and social practice (Fairclough 1992).

In these discursive and social practices, we perform “porosity” (Chan 2020: 129). Our bodies and languages take on and give birth to new social worlds. In other words, we are constantly in the process of creation and being created. As the curators encounter worlds in creation, they form imaginations and realities which are undeniably entangled (Barad 2007) in the mutually contaminating teaching and learning relationships of contemporary art (Chan 2020). Whilst Haraway’s (1997: 2) notion of the ‘material-semiotic’ considers the material and the discursive-linguistic in synthesis, thereby breaking down borders between realities, Barad’s (2007: 132) deconstruction of categorical oppositions of realism/social constructionism presents an alternative framework which pushes us towards a relational understanding of the intra-action between phenomena in her notion of ‘agential realism’. Seen through an agential realist perspective, bodies “come to matter through the world’s iterative intra-activity ––its performativity” (Barad 2007: 824). ‘Intra-action’, as opposed to ‘interaction’, involves mutual participation in action (Barad 2007: 141) in which all things are constantly exchanging, diffracting, influencing, and working inseparably. For Barad (2007: 93), diffractive engagement means that something is dialogically read “through one another” rather than employing a hierarchical methodology that would put different texts, theories, and strands of thought against one another (Barad 2007: 93).

We see these perspectives stemming from art, philosophy, and discourse studies as all mutually reinforcing. The current corpus-assisted work is rooted in the foundation of social constructionism: the idea that a shared construction creates reality through discursive formation (Butler 2006: 50; Mautner 2016). Since discourses are created primarily through language, by focusing on how language is used to construct discourses, we can explore how people use language to represent their reality (Gillings et al. 2023). This is amplified even further by taking a corpus-assisted perspective and we can trace how discourses come into being through repeated and incremental usage (Stubbs 2001; Baker 2006: 13).

Whilst we approach this study diffractively, we must acknowledge our own positions. A different configuration of this interdisciplinary approach might lead to different results. Since diffraction asks us to consider all the ways of touching and being in touch (Barad 2007: 72), we must recognise the effects of English language dominance on this knowledge, theory, and method. Our methods are rooted in Anglophone academic practice, having been developed through a British education focusing primarily on Western traditions and philosophies. Our personal and professional contexts have also given us different entry points. Whilst one author identifies as being of Chinese ethnicity, the other is of white British ethnicity. The disciplinary traditions in the academy have guided one author’s approach through data analysis, an understanding of social constructionism rooted in discourse, whilst the other author is guided by artistic research methodologies. We also recognise the importance of critical discourse analytical work such as this in taking an “unabashedly normative” stance (van Dijk 1993: 253), committed to challenging social ills and encouraging a change of discourse to address them. In our discussion of the representation of Chinese identity in England’s art exhibition press releases, we hope to unpack “the role of discourse in the (re)production and challenge of dominance” (van Dijk 1993: 249). After all, it is only through offering a critical commentary that we can truly do justice to the complexity of Chinese identity. As such, we further hope that the results presented here offer not only new insights into the linguistic construction of art press releases, but also insights on the surrounding contexts of political and racial identity constructions.

2.2. Museum communication

This paper is most closely aligned with the area of study of ‘museum communication’, generally taken to mean the wide array of written communication that is produced by museums to strengthen their relationship with audiences (Kotler et al. 2008). We might begin by classifying museum communication as a subfield of ‘organisational communication’. Just like in any study into language use within an organisation, text types under analysis can consist of internal communication ––such as emails between employees, staff memos and newsletters––or they can consist of external communication, such as customer service interactions (e.g., via Instagram or X), advertisements, or press releases. With that in mind, however, museum communication is a very specific type of language use that interacts with several other domains not necessarily covered in traditional definitions of organisational communication. Lazzeretti (2016), for example, opts to approach museum communication as a combination of different, overlapping, discourse types: the first is art discourse, where the main communicative purpose is to describe and evaluate; the second is media discourse, where the purpose is to inform; and the third is promotional discourse, where the purpose is to encourage visitors to visit. This three-way classification makes sense in the context of the present paper, which deals with the art gallery as a specific type of museum.

The first field identified by Lazzeretti (2016), art discourse, is particularly relevant as it refers to discourse which defines the cultural category of art, marking out what lies on each side of an “art/non-art binary” (Irvine 2004–2009, cited in Lazzeretti 2016). This is not necessarily the language of artists or art critics, but instead refers to a whole range of discourses, such as newspaper reviews of art exhibitions, leaflets, and even spoken discourses, such as television interviews or opening galas.

Linguistic studies on art discourse are rare, however, with some exceptions in Blunden (2016) and Boubakri (2023). Whilst Boubakri (2023) focuses on the interaction between the artist and audience in a live-show painting, Blunden’s (2016) work is more closely related to the present paper. Here, she uses systemic functional semiotics and legitimation code theory to analyse two types of data related to two exhibitions: the first being interviews with the teams responsible for organising those exhibitions, and the second being a series of texts produced for them. Blunden’s (2016) findings for the former are especially noteworthy since, in collecting interview data, exhibition teams (including the main curator) were asked about the construction of texts, about who had the authority to tell the story of the exhibition, and on what basis. The results show that the authority to speak was shared and flexible, with the core exhibition text distributed across several team members and as Blunden (2016: 129) notes

Such a process inevitably differs depending on the size of the exhibition and the size of the organisation behind it but, in Blunden’s (2016) case, the text curation process is clearly a democratic and iterative process with a wide range of authors and, importantly, no single person holds all the power.

The second field identified by Lazzeretti (2016) is promotional discourse, where the press release is a text type frequently analysed by linguists (e.g., Lazzeretti and Bondi 2012; Lazzeretti 2014). Aside from being generally easily accessible for researchers, they provide an insight into how the organisation wishes to be perceived by a particular audience. And, given that the aim is for these press releases to be picked up by newspapers and transformed into newsworthy pieces, they provide an insight into how they wish to be perceived at their very best.

Finally, the third field is media discourse. Given that the aim of press releases is to be reproduced and recontextualised in the media (such as newspapers or online news sources), they constitute this form of discourse too. Here is where there is a significant overlap with our other fields, though, despite being a media discourse, their aim is to be appealing to journalists to create a certain hype about an exhibition, and so they must also be promotional and are a “hybrid genre” (Fairclough 1992: 207).

Each of these fields is equally relevant because, as discussed in Section 1, we treat the press release as an apparatus ––an artefact–– of the exhibition. It shows how the gallery wishes to describe, inform, and promote their work. Lazzeretti (2016) is by far the work that is most relevant to our current investigation. With the use of diachronic corpus-assisted methods, she explores how museum communication has changed from 1950 onwards. Lazzeretti’s corpus consists of more press releases than ours (430 across both American and British museums), and it also includes stock of newly emerging museum genres, such as e-news, blogs, and social media. Her work promises to be seminal within the field of museum communication, and through this paper, we aim at contributing specificities to that field by focusing on the representation of Chinese identity.

3. Data and methods

We have identified press releases as one source of data (just one of many parts of institutional curatorial practice) which can provide an insight into how art organisations operate. These press releases exist as a written record of the intentions and key subjects for each exhibition, issued by the institution and are then sent into the public sphere. Whilst we recognise that some exhibitions have substantial publications and pamphlets which may present a different picture, press releases are articles which every exhibition produces. As documents, they typically do not differ significantly since they are designed to express the most salient information about the exhibition.

The present study analyses a corpus of press releases collected from Art Council England’s NPOs between 2014 and 2020 (hereafter, exhibitions corpus) in order to provide a broad national picture of this particular discourse. Given our interest in the representation of Chinese identity in the curatorial imagination since the Umbrella Revolution, press releases were included if they were written for exhibitions related to several identified key themes, such as Chinese identity, China, Hong Kong, and the Umbrella Revolution. In total, the data consist of 148 press releases, totalling 74,408 words, and was collected in the summer of 2020 when it was gathered from both physical and online archives kept by institutions or by the artists themselves. It is worth noting that 44 out of the 148 press releases were published by the Centre for Chinese Contemporary Art (CFCCA) in Manchester.² Whilst this is unproblematic as it is reflective of the full range of appropriate sources for the study, it is something that should be borne in mind in the analysis.

An example of a press release included in our corpus can be seen in Figure 1. This is a press release found on the 20–21 Visual Arts Centre website,³ which is advertising The Other Mountain, an exhibition focusing on the gradual internationalisation of Chinese jewellery as design students travel the world, learn new crafts and styles, and bring their new-found designs back to China to integrate them into their work. Given our method of analysis, we copied only the core text of the press release and discarded boilerplate text or any accompanying images (although we do acknowledge that meaning is achieved via different semiotic resources, and our analysis of discourse is limited as a result).

Figure 1: Screenshot from the 20–21 Visual Arts Centre website, promoting The Other Mountain

In this study, we also consider the implications of such an identity construction and examine how the creative agency of these publicly funded institutions might be contributing to the social and political embodied experience of individuals who identify as Chinese, or those who are assigned a Chinese identity both formally and informally in the UK. This second question leads us to propositions and provocations for creative praxis in response to these findings.

In our study, we used a corpus-assisted discourse method (Baker et al. 2008; Partington et al. 2013; Gillings et al. 2023). The corpus study consists of systematically analysing large amounts of texts, whilst the discourse analytical perspective offers both a theoretical and methodological set of tools allowing us to study language use in society (Taylor and del Fante 2020). The aim is to uncover the non-obvious or hidden meaning in the type of discourse under study (Partington et al. 2013). To do so, we employ four widely used corpus analysis techniques: 1) keyword analysis; 2) key semantic domain analysis; 3) collocation analysis, and 4) concordance analysis. Keyword and key semantic domain analysis allow us to determine which words and themes are the most salient within our dataset (Potts and Baker 2012; Potts 2015), and thereby to get a general overview of the press releases to consider what makes them distinctive from everyday written language. Our collocation and concordance analyses draw upon the lessons learnt in the general keyword and key semantic domain analyses, and then examines those linguistic features within their immediate co-text (Gillings and Mautner 2024).

4. Corpus-assisted discourse analysis

Our keyword and key semantic domain analyses follow the methodology outlined in Dayrell et al. (2020). We use Wmatrix 5 (Rayson 2008) to compare our exhibitions corpus to the Written British National Corpus Sampler (Written BNC Sampler),⁴ which comprises 1,000,000 words of everyday written English, to examine which linguistic items are being overused in our data. Working from the assumption that the most salient words and semantic domains are meaningful in some way, this initial analysis allows us to get a broad overview of the contents of the dataset before focusing on our specific areas of interest. Keywords and key semantic domains are identified using a three-part filtering procedure, as in Archer and Gillings (2020). Firstly, they are filtered using a statistical test of significance (log-likelihood), whereby we impose a minimum critical value of 6.63 (p < 0.01, 1%). Secondly, we impose an effect size measure (LogRatio),⁵ whereby those keywords with a LogRatio lower than 1.5 are discarded. Thirdly, we impose a minimum frequency cut-off of 50. In practice, this means that we can be sure that any keyword identified in the exhibitions corpus is frequent, distinct, and statistically meaningful. This presented us with 40 salient keywords within the exhibitions corpus, when compared to the Written BNC Sampler. Table 1 lists the top 20 of these keywords.

Keyword	LogRatio	Log-likelihood	Frequency
Artists	6.05	1285.88	294
Exhibition	5.85	1225.00	287
Artist	5.84	916.73	215
Chinese	5.63	1253.20	302
Art	5.20	1432.17	369
China	5.11	1020.79	267
Gallery	4.84	475.88	131
Works	4.61	623.15	130
Hong Kong	4.57	421.83	123
Culture	4.37	268.69	82
Arts	4.35	244.21	75
Online	4.34	292.73	90
Cultural	4.25	219.46	69
Image	3.82	169.08	60
Practice	3.63	142.70	54
Media	3.62	244.92	93
Project	3.34	176.25	74
Performance	3.15	158.60	72
UK	3.14	275.03	125
Sound	3.07	140.69	66

Table 1: Top 20 keywords in the exhibitions corpus compared to the Written BNC Sampler

The key semantic domain analysis presented us with 13 categories which are overused within the exhibitions corpus when compared to the Written BNC Sampler. These categories are automatically determined by the Wmatrix tool, with each word being assigned one of 232 semantic labels (Archer et al. 2002). Table 2 lists those 13 key semantic domains and the five most frequent lemmata within each domain. Lemmata refer to “a base form of a word together with its inflected forms” (Collins 2019: 197), and they are in SMALL CAPITALS to indicate their various forms:

Key semantic domain	LogRatio	Log-likelihood	The five most frequent keywords
Alive	4.49	482.25	LIFE, live, alive
Measurement: Area	4.25	384.75	SPACE, spatial, stretches
Arts and crafts	3.63	5079.78	ART, ARTIST, gallery, photography
The Media: TV, radio, and cinema	3.58	1471.16	FILM, video, documentary, cinema, animation
Green issues	2.77	194.37	ENVIRONMENT, nature, pollution, conservation, ecosystems
Information technology, computing	2.54	729.46	online, digital, internet, software, screen
The media	2.08	259.05	media, published, censorship
Attentive	2.03	111.78	FOCUS, HIGHLIGHT, attention, mindful, pay attention
Industry	2.02	216.33	factory, INDUSTRY, workshops, mining
Drama, the theatre and show business	1.94	208.66	PERFORMANCE, scenes, theatre
Evaluation: False	1.74	69.53	FICTION, fantasy, surreal, imaginary
Open, finding, showing	1.66	742.87	EXHIBITION, open, FEATURE, found, shown
Science and technology in general	1.54	116.12	TECHNOLOGY, SCIENCE, lab, experiments, scientist

Table 2: Key semantic domains in the exhibitions corpus compared to the Written BNC Sampler

5. Results

When interpreting results from a keyword and key semantic domain analysis, one must be careful to draw a distinction between those findings that are unique to a particular dataset and those findings which may be representative of the genre as a whole. In other words, given that keywords were calculated by comparing our corpus to the Written BNC Sampler, we will find words that are unique to art exhibitions in general, and words that are specific to the topic of this dataset (i.e., representations of Chinese identity). It is therefore to be expected that words such as exhibition, Chinese, China, Hong Kong, and gallery are listed in the top ten keywords. It is also to be expected that the key semantic domain alive is attested, as the word life is frequently used in the exhibitions corpus as part of the relatively banal constructions everyday life, public life, and so on. However, those words less immediately related to the key themes or art exhibitions in general, such as references to digital culture and the UK, are of additional interest. As for key semantic domains, one might expect to find Arts and Crafts, but the categories Green Issues and Evaluation: False are more specific and may offer additional insights into different curatorial contexts. In each case, we explored concordance lines at length to determine their relevance to our research question, allowing us to make this distinction in the findings.

Based on the data in Table 1 and Table 2 alone, we had some idea of what might be found in relation to specific concerns of contemporary artists practicing through the period: these were international concerns of the environment and the Anthropocene, as well as the internet, digital culture, and disinformation (e.g., keywords such as fiction and censorship). Not only are these topics flagged as keywords and key semantic domains, but they are also general themes which are becoming increasingly consolidated public concerns of our time.

The following sections have been organised by key semantic domain or theme, and they house opportunities for diffractions which are symptomatic of the corpus analysis we carried out. For each key semantic domain under analysis, we returned to the concordance lines within the data to explore what was written in each context. In line with Barad’s (2007) notion of diffraction, we should reiterate that each section cannot be read independently.

5.1. Colonial history

We started our analysis by eyeballing the list of keywords and key semantic terms, as listed in Tables 1 and 2 respectively. Given our interest in the representation of Chinese identity in press releases from publicly funded art institutions, we began by looking at the keywords China, Hong Kong, and UK in detail. To some readers, it may be unsurprising to find these three place names as key, given that the socio-political life of Hong Kong has complex and entangled relationships with the British Empire which continue to influence the present. More specifically, the British have a very specific and complex interest in Hong Kong, as the 1997 handover ceremony signalled not only the end of British occupation in Hong Kong, but the end of the British Empire as a whole. We hypothesised, that this colonial past may now find itself in the ecology of today’s contemporary art (Grant and Price 2020). After all, marking out the self/other is an oft-studied area in Critical Discourse Studies,⁶ and it is one way through which interlocutors create a distinction between groups and thus construct identities. To explore whether this was the case in our dataset, we looked at the collocates of China, Hong Kong, and UK. However, as we will see, these discourses overlap significantly, and the results show that the collocates of Hong Kong and UK do not point towards clear findings. For the collocation analysis, we looked at co-occurring words within a span of three words to the left and right of the node word, ranked by the LogDice metric (score of eight or above) and with a minimum frequency of five.

China is found 392 times in the corpus, and it seems to collocate with other place names, such as Hong Kong, UK, Shanghai, Hangzhou and Shenzhen. Further examination of these collocates via a concordance analysis, shows that these place names are either attested on their own or as part of a longer list of place names. They are typically used when listing where a particular artist or art academy are based, as illustrated in (1) and (2) respectively.

1. … Jiu Society, an artist group based in Shenzhen, China.


2. … Department at China Academy of Art, Hangzhou, China (2008) and has had solo exhibitions …

When part of a longer list of names, it is notable that UK, China and Hong Kong are found separated and, given that they occur frequently as a list of three, this is why these three terms are attested as keywords in their own right. In these lists, China references the People’s Republic of China, yet Hong Kong is seen as a separate entity even though it became a special administrative region of the People’s Republic of China in 1997. Not only does this highlight the special status of Hong Kong in these press releases, but it also has political implications when considering exactly which imagining of China is featured in England’s publicly funded exhibitions. Whilst this appears to promote political autonomy, there is danger that it may also prompt narratives of difference; again, something which has implications for a specific type of discursive representation. We found only one example where Hong Kong was explicitly referred to as part of the People’s Republic of China.

Interestingly, China also collocates with words that signify some form of spatial placement (e.g., realm, sphere, rural, Province, and beyond). Likewise, SPACE, spatial and stretches were also identified via the key semantic domain analysis, listed in the Measurement: Area domain in Table 2. Realm, collocating seven times, is attested in the phrases China’s online realm or China’s digital realm, and sphere also refers to these online spaces. This so-called realm is described as having an unruly topography, a messy vitality, and filled with commodity fetishization and self-posturing. These press releases (and the exhibitions themselves) hint at China’s restrictions on internet usage, with one aiming to destigmatise preconceptions about China’s online realm by reasserting the value of vernacular creativity. Again, art is seen here as a creative outlet in an otherwise heavily restricted space.

Sticking with the collocates related to space, beyond (collocating five times) is found as a reference to the country’s borders. Some of these are again related to where art has been showcased, as shown in (3), but others are more metaphorical in usage and refer to, for example, China’s expanding tech industry, as illustrated in (4). Such references (via words such as realm, sphere, and beyond) might index an increasingly softer border between China and the world in some examples. In fact, though, their usage in and of itself (particularly in the latter example) reaffirms ideas associated with a border between China and the rest of the world, suggesting that China is separated and difficult to approach in space, place, and context. The motivations for this are discussed in more detail in Sections 5.3 and 5.4.

3. … featured at galleries across China and beyond, including Germany …


4. … exceed anything envisaged beyond China’s borders, ensuring that the vast …

The lemma SPACE, which occurs 115 times and is identified via the key semantic domain analysis, seems to point towards something slightly different. Space is found to collocate with time (i.e., as part of the phrase time and space), indicating that Chinese art is a way through which visitors can deconstruct and traverse different realities, as shown in (5). Space also collocates with virtual, indicating certain exhibitions’ focus on the boundary between real and virtual spaces, as in (6), and with gallery, when focusing on the physical aspect of visiting an exhibition, as in (7).

5. … traversing the boundaries between time and space to explore the computers of the early electronics industry in China.


6. … emotions and feelings generated in such virtual space are real …


7. … wanted to create a space within the gallery which had more in …

5.2. Foregrounding ethnicity

We also found a trend in how the artists themselves are (re)presented. In the dataset, we identified a tendency to highlight differences between artists by ethnicity in ways that suggest an otherness from being British. In other words, the press releases use linguistic bordering between being British and not.

Whilst at first glance this may seem a banal finding (the corpus is, after all, made up of press releases included specifically because they relate in some way to Chinese identity), there is something more complex going on regarding identity construction. In fact, a closer look at the data recognises that such identity construction is unlikely to be a random choice but is rather a deliberate act. The examples below emphasize and foreground identity in a way that presents ethnicity as integral to the artists’ professional identity, often describing artists as Chinese artists despite the information of their ethnicity being presented elsewhere, sometimes in the same sentence. That said, we do not know whether this identity marker is imposed on the artist by the gallery through their press release, or whether individual artists choose to be described as such in the same way as when researchers submit a short author bio to a journal paper. If the identity is ascribed, that may tell us something about the galleries’ promotional technique (as discussed in Section 2.2), whereas if it is the artists’ own words, this tells us something about how they wish to be perceived in the construction of their professional identity. The reader is unaware of this construction and sees only the label, contributing to the wider discourse around the representation of Chinese identity. This is shown in our analysis since the phrases Chinese artist (17 times) and Chinese artists (40 times) are frequently attested, as illustrated in (8)–(10) below. Hong Kong artists in turn is only attested one time in a press release promoting the CFCCA’s 30–year anniversary programme.

8. Highlights include solo shows by Chinese artists Ma Qiusha and Shen Zin; the group show History Repeats Itself …


9. ... dialogue on female Chinese contemporary artists. NOW aims to readdress the marginalisation of Chinese female artists; presenting new viewpoints and exploring how female artists navigate …


10. … sculptures by internationally acclaimed Chinese artist Ai Weiwei.

Furthermore, we also notice that whilst white artists are often presented as raising questions and concerns surrounding a wide range of social and cultural observations, non-white artists are often presented in ways which foreground their colour or ethnicity, presenting race and ethnicity as the primary, if not sole, subject matter of their practice.

Foregrounding can also happen in physical spaces through performative semiotics, such as objects and decor which signal Chinese identity. As an example, at the CFCCA, the interior design references tropes of stereotypes of Chinese design with elements taken from generic ideas of the Hutong courtyard. Likewise, its small gift shop is stocked with ‘Lucky Cat’ memorabilia, pandas and chopstick-shaped pencils. Such foregrounding sets a foundation upon which audiences encounter and understand the exhibition and the voices of the artists within. As such, it contributes to an orientalising which upholds harmful racial stereotypes. This form of linguistic bordering ––both in exhibition spaces and in press releases–– presents Chinese artists as perpetually foreign (Wu 2002) in a way that can also suggest an outsideness to the sphere of contemporary art.

Our analysis found 14 uses of the phrases British Chinese or British-born Chinese referring to both artists and audiences of Chinese ethnicity in Britain, but there was only one occasion, illustrated in example (11), where an artist of East and Southeast Asian descent was described explicitly as British. In (11), the explicit assertion of British identity marks a different intention, and it is used to avoid an assumption of ethnic identity, which may have resulted from pre-formed assumptions and associations with the artist’s name. The lineage of their name appears to have necessitated an attempt to debunk those assumptions. In all these examples, the foregrounding of identity exists to distinguish one racialised body or culture from another.

11. Gayle Chong Kwan is a British artist whose photography, installations, and public realm works …

5.3. The media

The analysis also revealed a focus on the media as a core concern within the subject matter of exhibitions, which has also been a theme featured heavily in UK news on China. The word censorship occurred 21 times within the exhibitions corpus and was part of the key semantic domain the media. Examples are shown in (12) and (13).

12. … civilians attempt to take back power from restrictive government in light of growing censorship and surveillance online.


13. … the prevailing image of China’s internet as a barren wasteland; a place censored to the point of sterility and therefore devoid of any meaningful creative expression?

The curatorial interest on censorship not only responds, but also contributes to the wider public discourses in the UK and North America surrounding the heightened anxiety around surveillance and Chinese technology, such as Huawei Telecoms infrastructure (Siu and Chun 2020). Criticism of surveillance and lack of personal privacy are presented as Chinese problems. Such narratives perpetuate notions of the ‘Yellow Peril’ and positions ‘Whiteness’ as an international jurisprudence. The Yellow Peril is characterised by a fear to East Asia, which is considered as an existential danger to the western world (Siu and Chun 2020). In the Yellow Peril imagination, China exists as a psycho-cultural jeopardy waiting to happen, an evil force infiltrating borders. As Crean (2023: 2) points out, the Chinese culture is positioned as “alien and threatening, particularly to civilizations based upon European culture,” and, whilst the danger is communicated as a Chinese one, the prevalence of surveillance in technology and its impact on privacy is a global concern, and an increasingly evident one in the UK.

Storytelling in relation to censorship is not always destined to fall back into narratives of Yellow Peril. For instance, in the work by Ho Sin Tung, Yim Siu Fong and Silas Fong (see Section 1; Chan 2019), the artists, who presented work outside of the institutional curatorial framework, were able to offer highly critical engagement with the discourse on surveillance without falling back into these narratives. Contrary to being censored, their artistic gestures reflected the loudness of holistic socially engaged practices. Specifically, Ho Sin Tung’s blank pages challenge the passivity of readers: it asks them to look beyond their encounter with the work and the exhibition and question why some artists might be better acknowledged as absent.

Unfortunately, we are unable to discern whether the curatorial approaches attempted to dispel or affirm the narratives mentioned. What can be however claimed is that there is a strong and active associating of themes of censorship with Chinese identity which inevitably contributes to social discourse and entangled global politics. In future research, a closer cross-disciplinary analysis of other elements of the curatorial presentation ––such as the selection of artworks, labels and other sources–– could reveal trends in storytelling around this subject matter. Potential biases could be explored in an analysis with a similar dataset associated to another racial identity. This would allow us to identify whether censorship is being curated in England as a primarily Chinese issue.

5.4. Fantasy

The data also revealed a trend of interest in practices which are concerned with the boundary between fiction and reality. This was identified through the key semantic domain Evaluation: False, where 26 instances of fiction and 13 of fantasy were attested. Whilst the word fiction (cf. examples 14 and 15), used in relation to reality, seems to suggest something is not real, fantasy ––as a specific mode of fiction–– implies an alien, otherworldly conception (cf. examples 16 and 17).

14. a radical exploration of photography when the boundaries between truth and fiction, machine and human are being increasingly called into question.


15. featuring artists using speculative fiction as a productive medium …


16. At the intersection between fantasy and critical observation


17. In Whose Utopia, factory workers roleplay their fantasy lives within the confines of an industrial environment.

Whilst many artists across the globe are concerned with the construction of tomorrow’s social world, to conceptualise this as fantasy remains a specific form of framing. Whilst fiction conceives of possibility, fantasy conceives of a specific subcategory of fiction which conjures images of the impossible, of monsters and beasts, of magic, smoke and mirrors, which have long been part of the stereotyped British imagination of Chineseness. As Mayer (2013) argues, this is exemplified in manifestations of the fictional character Fu Manchu. Beginning in 1912, the character was introduced in a series of novels and short stories written by English author Sax Rohmer, the most prominent example being The Mystery of Dr. Fu-Manchu. The character has been later used in films, television shows, and comic books, and is now widely considered an archetypal supervillain: an evil criminal genius wrapped up in mystique. Crean (2023) argues that the prominence of the Fu Manchu stereotype, even to this day, is inherently linked to the Yellow Peril. Characters and tropes such as these are continually envisioned in ways that occupy spaces of visibility to the exclusion of the invisible, and such discourses naturally affect Chinese identity and the way it is perceived. As such, and pointed out above, further investigation of datasets related to other racial identities, or indeed any other marginalised identities, would allow us to identify any bias towards the association of fantasy (as opposed to fiction) towards some groups and not others.

5.5. Green issues

Green Issues was the fifth most common key semantic domain. Environmental concerns made up a significant portion of the discourse here, with the lemma ENVIRONMENT being attested 63 times in the corpus. Whilst the noun environment is used in a variety of contexts ––for example, the socio-political environment and the digital environment–– it is most prominently used when discussing environmental issues of our time, such as climate change, destruction, and rapid urbanisation, as shown in (18)–(19) below. Not only are these issues of importance to the artists represented in our dataset, but their prominence also gives us an insight into the types of art on display in publicly funded galleries across England. The press releases included in our corpus were collected between 2014 and 2020, a time at which the public began to accept human-induced climate change as a scientific reality, and we can identify a shift in policy by many news organisations to stop giving airtime to climate sceptics in the name of balance. This marked change in discourse was in large part due to significant world events occurring at the time, such as the 2015 Paris Agreement or the subsequent election of Donald Trump as President, who promised to take the United States out of international mandates. When reporting on the fluctuation of climate change discourse over time in the UK press, Gillings and Dayrell (2023: 128) point out that “from that point onwards [2015], the ‘balanced’ coverage of climate change issues seems to have lost force and the focus shifted to calls to action.” Especially in the UK, the public began to reject scepticism and rather focused on adaptation and (where possible) mitigation. Art is, naturally, one way through which such calls to action can be mediated.

18. … changes can be made to protect the environment and to minimise our impact, both …


19. … or escape from the crowded urban environment and hidden anxieties about excess …

This discourse is further confirmed when looking at the other words making up the Green Issues semantic domain: nature, pollution, ecosystems, and polluting. In (20), the artist seeks to highlight the devastating effects of climate change on indigenous communities in Taiwan who are experiencing a range of environmental damages. Likewise, in (21), the exhibition features manipulated photographs of waste debris scattered along 30 beaches in Hong Kong, with the aim of raising local and global awareness of the need for stricter waste management.

20. … typhoons, landslides, flooding, pollution and other environmental damages.


21. … about the unsettling truth of waste pollution in the world’s oceans and beaches, …

The noun nature is attested 20 times in the exhibitions corpus. However, in 16 of the occurrences, the word does not refer to environmental concerns but rather to the nature of being. Whilst the lemma ENVIRONMENT points towards a marked concern for issues around the Anthropocene, nature points towards discourses around the nature of one’s being, as in the nature of modern society, the nature of Hong Kong, and what it means to be human. This is illustrated in (22)–(24). Nature, in this usage, refers to the innate qualities of the subject. Twinned with the discourse around the Anthropocene, this is an interesting find within the exhibitions corpus, as it points towards an ongoing meta-discourse around identity construction both at the level of the artist and at the level of nationhood and nationality, which we intend to explore further in a future study.

22. … artists in the exhibition explore the interconnected nature of the human spirit and the habitat …


23. … while simultaneously questioning the very nature of those desires …


24. … the exhibition aims to identify with the evolving nature of Hong Kong as it enters a phase of rapid development.

6. Discussion and conclusions

Our analysis has explored constructions and representations of Chinese identities in England’s curatorial imagination, that is, it focused on the ways through which art galleries, their curatorial, programme and marketing teams, articulate Chinese identities through their press releases. Using a corpus of 148 press releases collected from Art Council England’s NPOs between 2014 and 2020, we were able to explore how Chinese identity is linguistically constructed, and thus offer provocations for considering what publicly funded image is portrayed throughout England’s publicly funded art galleries.

Our analysis concludes that, in the wake of the Umbrella Revolution, Chinese identity is represented in a range of entangled ways, but most notably by the promotion of themes which are, in some cases, related to Chinese stereotypes. Firstly, by examining the collocates of China, we find that it is often listed alongside the words UK and Hong Kong, pointing towards its complex socio-political history, and alongside spatial words. Second, we find that Chinese artists’ ethnicity is foregrounded much more prominently, especially when contrasted to that of British artists, which are not foregrounded. Thirdly, we find discussion of the media, which has relevance to ongoing political issues across the world. Fourthly we find frequent references to fantasy. This is interesting, not least for the specific connotations conjured by the term, as opposed to an alternative such as fiction. Finally, we find frequent discussions of environmental issues and the nature of being.

We argue that through curatorial selection (conscious or otherwise), the general public’s understanding is being impacted in very specific ways. After all, the key themes examined above are the key representations of Chinese identity being showcased by England’s publicly funded art institutions. These institutions hold a particularly important place in the exchange of knowledge. Generally, we regard our museums as trustworthy, and often the curator, unaware of the extent of their power, does not see their construction of embodied realities as a construction at all. When Deleuze (1990: 226) grapples over Spinoza’s question about what a body can do, he argues that “we do not even know of what affections we are capable… nor the extent of our power.” Here, Chinese identity has been formed through the curatorial apparatus, constructing the borders of Chinese identities and continually shaping social, cultural, political, and both physical and imagined borders. These are the realities which come to form specific embodied experiences.

Furthermore, our corpus data highlights how even small elements of standardised curatorial practice, such as the writing of the press release are 1) a significantly agential creative practice, 2) apparatus through which public knowledge and socially-coded modes of thought and behaviour are created, and 3) carry a significant entangled responsibility for the social worlds which they co-create. We must be cautious of how we create and of the words and language we use in the task. As Haraway (2016: 12) points out, “it matters what matters we use to think other matters with.”

Whilst our study focused on a small and specific dataset and explores discourse specifically related to curatorial presentations of Chinese identity immediately after the 2014 Umbrella Revolution, the methodologies for analysis can be applied to a variety of other datasets to further understand the social agency of such materials and their impact on the embodied experiences of racialised people in the UK. Likewise, future work may also wish to give increased attention to other semiotic modes. Stereotypes are also reinforced via non-verbal means, such as the ‘Lucky Cat’ memorabilia discussed in Section 5.2, The examination of these tropes may also allow to unpack the complexity of Chinese identity construction in other non-verbal gesture. Importantly, we reiterate our understanding that art institutions involve not only artistic but also personal bodies as intra-active agents.

Our study reveals a very specific representation of Chinese identity through one element of standardised curatorial practice in England during the 2014–2020 period. In some cases, this representation may have caused unintended harm or negatively impacted Chinese identifying people or those identified as Chinese by others, both nationally and internationally. This study thus raises important questions for institutions in England and the rest of the UK on the position of their practice in discourses on anti-racism, decolonisation, inclusive language, the politics of the exhibition, and their curatorial roles in forming knowledge. Undoubtedly, artists included in the exhibitions we surveyed have been impacted to varying extents by the Umbrella Revolution and the contested representation of Chinese identities as they are encountered physically, digitally, socially, politically, symbolically, and personally. Whilst we find only minimal direct references to the Umbrella Revolution in our exhibitions corpus, we also acknowledge that NPOs do not reflect the total range of artworks being made or what the full range of Chinese artistic voices in the UK may be attempting to articulate through the dissemination of individual artistic practices in discourse. Artists may have found other ways to speak that are not captured via examination of these press releases. For instance, through non-verbal means that are inaccessible to the corpus-assisted discourse analyst, such as in the artwork itself.

Against the background of the Umbrella Revolution, our study has revealed how often the curatorial imagination can continue to linger within colonial perspectives and troublesome imaginations. Indeed, old tropes inherited from the projects of empire oftentimes reinforce division regardless of political positioning. This imagination and its rendering in the creative arts is often rehearsed, reinforced, and replayed without critical awareness of one’s own capacity to influence the social and cultural ecology of the present and future.

We hope that this article can provide some provocations for both artistic and curatorial practice, as well as for further considerations of the ways corpus linguistics and the arts can be read through one another, diffracting and revealing what might otherwise be hidden in plain sight.

References

Archer, Dawn and Mathew Gillings. 2020. Depictions of deception: A corpus-based analysis of five Shakespearean characters. Language and Literature: International Journal of Stylistics 29/3: 246–274.

Archer, Dawn, Andrew Wilson and Paul Rayson. 2002. Introduction to the USAS category system. http://ucrel.lancs.ac.uk/usas/usas guide.pdf

Baker, Paul. 2006. Using Corpora in Discourse Analysis. London: Continuum.

Baker, Paul, Costas Gabrielatos, Majid KhosraviNik, Michal Krzyżanowski, Tony McEnery and Ruth Wodak. 2008. A useful methodological synergy? Combining critical discourse analysis and corpus linguistics to examine discourses of refugees and asylum seekers in the UK press. Discourse & Society 19/3: 273–306.

Barad, Karen. 2007. Meeting the Universe Halfway: Quantum Physics and the Entanglement of Matter and Meaning. Durham: Duke University Press.

Blunden, Jennifer. 2016. The Language with Displayed Art(efacts): Linguistic and Sociological Perspectives on Meaning, Accessibility and Knowledge-Building in Museum Exhibitions. Sydney: University of Technology Sydney.

Boubakri, Awatef. 2023. Visual art, discourse, and cognitive linguistics: The live-show painting as a triple-scope conceptual integration network. Cognitive Linguistic Studies 10/1: 227–245.

Butler, Judith. 2006. Gender Trouble: Feminism and the Subversion of Identity. New York: Routledge.

Chan, JJ. 2019. Momentary Glimpses: An Anthology of Contentedness. London: Folium Publishing.

Chan, JJ. 2020. Performing porosity: Is there some method? Performance Research 25/5: 129–134.

Chan, Johannes. 2014. Hong Kong’s Umbrella Movement. The Commonwealth Journal of International Affairs 103/6: 571–580.

Collins, Luke. 2019. Corpus Linguistics for Online Communication: A Guide for Research. New York: Routledge.

Crean, Jeffrey. 2023. The Fear of Chinese Power: An International History. London: Bloomsbury.

Dayrell, Carmen, Ram-Prasad Chakravarthi and Gwen Griffith-Dickson. 2020. Bringing corpus linguistics into religious studies: Self-representation amongst various immigrant communities with religious identity. Journal of Corpora and Discourse Studies 3: 96–121.

Deleuze, Gilles. 1990. Expressionism in Philosophy: Spinoza. New York: Zone Books.

Fairclough, Norman. 1992. Discourse and Social Change. Cambridge: Polity Press.

Gillings, Mathew and Carmen Dayrell. 2023. Climate change in the UK press: Examining discourse fluctuation over time. Applied Linguistics 45/1: 111–133.

Gillings, Mathew and Gerlinde Mautner. 2024. Concordancing for CADS: Practical challenges and theoretical implications. International Journal of Corpus Linguistics 29/1: 34–58.

Gillings, Mathew, Gerlinde Mautner and Paul Baker. 2023. Corpus-Assisted Discourse Studies. Cambridge: Cambridge University Press.

Grant, Catherine and Dorothy Price. 2020. Decolonizing Art History. Art History 43/1: 8–66.

Haraway, Donna J. 1997. Modest₋Witness@Second₋Millennium.FemaleMan₋Meets₋OncoMouse: Feminism and Technoscience. New York: Routledge.

Haraway, Donna J. 2016. Staying with the Trouble: Making Kin in the Chthulucene. Durham: Duke University Press.

Irvine, M. 2004–2009. Approaches to the Art Media: Modes of Art Talk, Discourses, and the Construction of Art as an Object. http://www9.georgetown.edu/faculty/irvinem/CCTP738/ArtMediaTheory.html.

Koester, Almut. 2022. Building small specialised corpora. In Anne O’Keeffe and Mike McCarthy eds. The Routledge Handbook of Corpus Linguistics. New York: Routledge, 48–61.

Kotler, Neil, Philip Kotler and Wendy I. Kotler. 2008. Museum Strategy and Marketing: Designing Missions, Building Audiences, Generating Revenue and Resources. San Francisco: Jossey-Bass.

Lazzeretti, Cecilia and Marina Bondi. 2012. ‘A hypnotic viewing experience’. Promotional features in the language of exhibition press announcements. Pragmatics 22/4: 567–589.

Lazzeretti, Cecilia. 2014. A landscape never goes out of style. Diachronic lexical variation in exhibition press announcements. Journal of Language and Communication in Business 27/52: 107–124.

Lazzeretti, Cecilia. 2016. The Language of Museum Communication: A Diachronic Perspective. London: Palgrave Macmillan.

Lee, Paul, Clement York Kee and Louis Leung Wing Chi. 2015. Social media and Umbrella Movement: insurgent public sphere in formation. Chinese Journal of Communication 8/4: 356–375.

Mautner, Gerlinde. 2016. Discourse and Management. London: Palgrave.

Mayer, Ruth. 2013. Serial Fu Manchu: The Chinese Supervillain and the Spread of Yellow Peril Ideology. Philadelphia: Temple University Press.

Partington, Alan, Alison Duguid and Charlotte Taylor. 2013. Patterns and Meanings in Discourse: Theory and practice in Corpus-assisted Discourse Studies (CADS). Amsterdam: John Benjamins.

Potts, Amanda. 2015. Filtering the flood: Semantic tagging as a method of identifying salient discourse topics in a large corpus of hurricane Katrina reportage. In Paul Baker and Tony McEnery eds. Corpora and Discourse Studies: Integrating Discourse and Corpora. London: Palgrave Macmillan, 285–304.

Potts, Amanda and Paul Baker. 2012. Does semantic tagging identify cultural change in British and American English? International Journal of Corpus Linguistics 17/3: 295–324.

Rayson, Paul. 2008. From key words to key semantic domains. International Journal of Corpus Linguistics 13/4: 519–549.

Siu, Lok and Claire Chun. 2020. Yellow peril and techno-orientalism in the time of COVID–19: Racialized contagion, scientific espionage, and techno-economic warfare. Journal of Asian American Studies 23/3: 421–440.

Sparks, Colin. 2015. Business as usual: The UK national daily press and the Occupy Central movement. Chinese Journal of Communication 8/4: 429–446.

Stubbs, Michael. 2001. Words and Phrases: Corpus Studies of Lexical Semantics. Oxford: Blackwell.

Taylor, Charlotte and Dario del Fante. 2020. Comparing across languages in corpus and discourse analysis: Some issues and approaches. Meta 65/1: 29–50.

Tekin, Beyza. 2010. Representations and Othering in Discourse: The Construction of Turkey in the EU Context. Amsterdam: John Benjamins.

Van Dijk, Teun. 1993. Principles of critical discourse analysis. Discourse & Society 4/2: 249–283.

Wu, Frank H. 2002. Where are you really from? Asian Americans and the perpetual foreigner syndrome. Civil Rights Journal 6/1: 16–22.

Notes

¹ We would like to thank our research assistant on this project, Pui Kan, for aiding with data collection. [Back]

² The CFCCA is now known as the East and Southeast Asian Contemporary (ESEA). [Back]

³ https://www.2021visualartscentre.co.uk/ [Back]

⁴ https://ucrel.lancs.ac.uk/bnc2sampler/sampler.htm [Back]

⁵ http://cass.lancs.ac.uk/log-ratio-an-informal-introduction/ [Back]

⁶ For instance, Tekin (2010: 113) explores how Turkey is represented as an ‘Other’ in contrast to the ‘European self’. [Back]