Typologising formal school-to-school collaborations—education collectives—in China through the metaphor of Chinese landscape painting

ABSTRACT

In China, formal school groupings known as ‘education collectives’ have become one of the most common forms of school-to-school collaboration, promoted by policymakers to narrow the achievement gap between schools and optimise resource allocation. Previous research has focused on the purposes and achievements of education collectives rather than their structural diversity. This article seeks to address this gap by typologising education collectives in China. We map the landscape of education collectives, illuminating how school-to-school collaboration and education collectivisation have been promoted and operationalised in China. To achieve this, we employ the metaphor of traditional Chinese landscape painting as a methodological tool. Through this, we promote a deeper understanding of the cultural and psychological roots of school-to-school collaboration in China. We conceptualise education collectives from three perspectives inspired by the features of landscape painting: power relations, legal status, and external institutional engagement. These three dimensions also illuminate the homogeneity, exclusion and unsustainability that the education collective may encounter in their development process. This typology places education collectives within the broader policy context of Chinese basic education and system reform and offers insights into the diversity of network and partnership structures among inter-school collaborations.

KEYWORDS:

• Education collectives
• school-to-school collaboration
• education policy
• China
• typology
• mapping

Introduction

China has not been immune from the wave of school-system reform known as the Global Education Reform Movement (Sahlberg, 2015). However, reform there is discursively legitimated through appeals to ‘equity’ and ‘quality’, which feature in China’s national education development plans (Liu, 2021; Qian & Walker, 2020). It is increasingly recognised that it is hard to achieve these two goals by relying only on a school’s individual efforts and qualities, so education policymakers have turned instead to whole-system reform (Qian & Walker, 2020). This involves focusing mainly on narrowing the gap between high-performing and low-performing schools across a system (Fullan, 2004; Hopkins et al., 2014). Schools are increasingly collaborating to meet this challenge, either spontaneously or under government instruction. Accelerating reforms (Yang & Yang, 2020) has produced diversity and intensity in forms of collaboration; inter-school collaboration is no longer limited to superficial resource efficiencies such as teaching resources and technological infrastructure, but extends to the establishment of multi-functional partnerships. Formal school groupings known as ‘education collectives’ (ECs) have gradually become one of the most widely found forms of school-to-school collaboration. ECs are a large-scale multi-level educational organisation based on the optimal allocation of educational resources. Most commonly, a ‘prestigious’ or ‘strong’ school leads several others on a contract basis ultimately to develop the EC in a high-quality and well-balanced way (Meng & Dong, 2020; Zhang, 2015). This new organisational form not only produces changes in the organisational structure and boundaries of the network but also impacts the original school cultures.

Education collectives take many forms. Since the Hangzhou Education Bureau established the first EC, Qiushi Education Collective, in 2002 (Gu et al., 2017), the field’s main source of information about the diversity of collectives and its significance is around 20 years old. A thorough review of the literature on ECs reveals three associated research foci, that is: school improvement, school autonomy and student performance. Previous research has examined the purposes and achievements of ECs (see Gu et al., 2017; Wu, 2013; You, 2021), but there is little acknowledgement of their differentiated forms, making it important to unpack this complexity. Specifically, the field lacks a comprehensive and clear typology of ECs. Therefore, in this paper, we map ECs to illuminate how education collectivisation and school-to-school collaboration have been promoted and operationalised in the Chinese context and what this reveals about the functions of ECs as structures instantiated within a particular time and place (see Courtney & Mann, 2021).

As of 2020, there are 280 ECs in the compulsory education phase in Shanghai (Primetime News, 2022), 160 ECs in Beijing (Fang & Feng, 2022), and 804 ECs in Jiangsu Province (Yu, 2020). Moreover, some provinces (e.g. Guizhou Province, Qinghai Province, Jilin Province) lack statistical information regarding numbers, which makes quantitative typologising difficult. Additionally, inter-school cooperation is complex and variable, which poses specific challenges in categorising collectives (Li et al., 2021). Despite these difficulties, classifying the EC is achievable using qualitative tools and methods.

As McKinney (1966) argued, the main function of typologising is to identify, simplify and order data for comparison. Typologising also transorms data from a complex to a simple state. This simplifying procedure identifies key characteristics and emphasises only a few possible classification combinations (McKinney, 1966). Gunter and Ribbins (2003) insist that generating a typology is to locate oneself in a field, thereby producing knowledge preferentially and partially. To this, we add that typologising is helpful to interpret authoritatively complex information.

Following and building on Courtney’s (2015) mapping of English schooling provision, we locate primary and secondary ECs within the policy landscape, which, uniquely, we interpret more literally. Shi (2008) defined landscape as the scene presented in a given area, which reflects the complex of space and matter over the terrain. Moreover, landscape is the imprint of complex natural processes and human activities on the land, and it reflects the living synthesis of individuals and places that is essential to local and national identity. The image of landscape can suggest the large-scale and wide distribution of the EC, but also implicitly indicates its highly decentralised and localised characteristics, as well as how it represents a hybrid of marketising public service and uncertainty.

We borrow the metaphor of China’s landscape painting to conceptualise ECs. This metaphor enables their investigation from various viewpoints. Specifically, some features of landscape painting have informed our overall approach and methodology. Others, however, are used to structure our typology. Furthermore, it enables exploration of the deeper cultural and psychological memes in China. Mapping ECs in China is not simply a taxonomic exercise, but, following Grace’s (1995) policy-scholarship approach, it is also a critical engagement with the history and context of school-to-school collaboration in China. Our contribution is consequently first empirical through the production of a typology which locates ECs within the broader landscape of Chinese compulsory education and system reform, and offers key insights. Our contribution is secondly heuristic, in providing a foundation for continuing questions about network and partnership diversity among inter-school collaboration. Finally, it is conceptual and methodological in proposing the unique lens of Chinese landscape painting to make sense of provision.

Collaboration as an education policy

The establishment of formal collaborative structures as an education policy directive over the last two decades represents a relatively recent shift in Chinese education policy. We argue that the reasons for this are threefold.

The first, and perhaps most significant reason, is the Chinese educational philosophy which, like much of its wider society, has been influenced by Confucianism. The term ‘HE ER BU TONG’ or ‘harmony in diversity’ (Confucius, 1998) is a central facet of Confucianism that values complementarity and diversity in bringing coherence to any human activity (Berthel, 2017). In this sense, Confucius emphasises that individuals should communicate and cooperate on an equal footing with mutual respect so as to progress through learning from each other’s strengths. Confucianism also promotes the idea of harmony as generative (Confucius, 1998), whereby ‘harmony in diversity’ refers to the development and continuous prosperity of human society through diversified mutual collaboration and restraint (Csikszentmihalyi, 2020). For policymakers in China, perceiving unbalanced educational development, this philosophy is attractive because it foregrounds collaboration and promotes the idea that schools of different quality can help each other and develop together.

The second reason concerns the Chinese government’s promotion of its counterpart support policy to address education disadvantage in underdeveloped areas. Launched in 1956, the ‘Notice Concerning Inland Support for Primary School Teachers in Frontier Areas’ was followed by similar policies in 1966, 1974 and 1996 that involved teachers from schools in urban areas delegated to train, cultivate and support teachers in ethnic minority, social and economically disadvantaged regions (Department of Ethnic Education, 2004; National Ethnic Affairs Commission of the PRC, 1996). Although such policies represented a redistribution of human resources rather than structural or cultural changes, they signified collaborative intent through the mobilisation of knowledge and expertise. More recently, the government has introduced a ‘twinning’ strategy whereby ‘key’ schools share capacity and develop schools in need. Despite its drawbacks, including resource implications for key schools and poor implementation in the supported schools, this strategy proved popular, particularly in the early 2000s (see An, 2020; Duan, 2018).

The third reason relates to the growth internationally of school-to-school collaboration as a response to the myriad problems facing education systems (see Hopkins et al., 2008). England is perhaps the most prominent example of a context in which collaborative structures have been promoted as a vehicle for school improvement, despite ambiguities surrounding the purposefulness and tangible influence of such activity (Armstrong et al., 2021). However, numerous education systems internationally are experimenting with such structures and whilst this manifests differentially according to policy, historical and cultural context, it demonstrates the seductiveness of collaboration as a (the) key to educational improvement (see Armstrong & Brown, 2022). Building on a cultural tradition of collectivism, these global shifts have offered further encouragement to education policymakers in China to introduce formal collaborative structures to facilitate school improvement and address inequity in education.

Conceptualising collaboration through the metaphor of landscape painting

We make use of the metaphor of Chinese landscape painting as an exploratory methodological tool to conceptualise the landscape of China’s primary and secondary ECs. In Chinese, ‘landscape’ refers to ‘Shan Shui’, which is literally translated as ‘mountain and water’. Mountains and water are two of the most iconic elements in nature, and their combination is considered harmonious and equilibrated (Barnhart et al., 1997). We associate landscape painting with ECs because its rich connotation allows us to capture the seemingly nebulous and ever-changing Chinese educational terrain. Some features of landscape painting have informed the overall approach and methodology in this study, while others are used to structure the typology. We explore these in turn below.

One characteristic of Chinese landscape painting is that it can be used as an ethical scale of aesthetics and benevolence, which is also reflected in the ethos and missions of running the EC and the current purpose of education in China. Chinese landscape painting was not designed for appreciation and decoration, nor does it have a religious purpose (Cai, 2019). It emerged in response to the overall harmony between the system of nature, human, ethics and moral principles (Hawks, 2013). Under the subtle influence of three dominant ideologies of ancient China – Confucianism, Buddhism and Taoism, the view of nature formed by Chinese landscape painting contains unique aesthetic ideals and concepts with Chinese characteristics (The Metropolitan Museum of Art, 2003). For instance, the representatives of neo-Confucianism, Cheng Hao (1032–1085), Cheng Yi (1033–1107), and Wang Yangming (1472–1529) claimed that Tian Ren He Yi (translatable as unity of nature and humans) have become a significant kernel of aesthetic thought in Chinese landscape painting and extend to the order of all objects in nature, becoming an aesthetic system that combines natural laws and ecological ethics (Cai, 2019; Zhou, 2020).

Digging further into the term ‘ethics’, we argue that some landscape painters were accustomed to starting from the standpoint of ethical centralism, particularly for those impacted by the ideological trend of neo-Confucianism. To elaborate, landscape painters attributed the aesthetic value of natural phenomena and art forms primarily to ethics or morality, and combined natural beauty with human moral elements in their aesthetic conceptions (Wang, 2017). There was a tradition of utilising natural objects to symbolise the spiritual character of human beings, which created a correspondence between the beauty of personality and objects (Wang, 2017; Zhou, 2020). Under Confucianism’s ‘Ren (benevolence)’ (see Tan & La Londe, 2023) and the principle of ‘unity of beauty and goodness’, landscape painting creators took ethical and moral concepts as the entry point to appreciate natural and artistic beauty, making landscape painting illuminated by Confucianism possess features of emphasising norms, order and relationships (Wang, 2017; Zhou, 2020). Therefore, Chinese landscape painting can be seen as a profiling of China’s long-lasting social context, which brings us to China’s ideological and cultural background.

Another characteristic of landscape painting is that in the process of its creation, it highlights the subjectivity of the painter in depicting the natural landscape. The Chinese cultural tradition has constantly regarded the nature of mountains and rivers as the original narrative, and landscape is often representative of a certain independent spiritual metaphor (Li, 2021). Furthermore, underpinned by the natural philosophy view represented by Confucianism and Taoism, landscape has separated from social environment, political atmosphere, pastoral geography, living ethics and other aspects, and gradually formed and found its own logic of survival and development (Cai, 2019). However, this is not to say that landscape painting lacks the intimacy of dialogue and communication (Zhang & Wang, 2020), nor does it suggest a split between the self and the world, reason and emotion, subject and object in landscape painting (Jullien, 2009). On the contrary, the painters and their works are in an interactive embrace of giving and receiving, welcoming and being welcomed. Therefore, although Chinese landscape painting does not mirror the world around the painters, in other words, the imagery presented by landscape painting are not necessarily ‘realistic’, but constructions of reality, in exploring what painters ‘thought’ about reality.

We have used this feature of landscape painting methodologically: it enables us to construct the EC landscape through our own subjective analysis of Chinese education, thus establishing a harmonious induction with our imaginary of EsCs. The landscape painting is a pictorial representation of our understanding of the EC and a profound connection to the external terrain of Chinese education. Our approach to constructing the typology may therefore be summarised as ‘landscape painting as methodology’. Like landscape painters, we construct the terrain subjectively, yet are inspired by the formal structures before us. However, even these are amenable to our interpretation. So, landscape painting as a methodology is different from, and supplementary to the three typological dimensions of legal status, power relations and external institutional engagement. These are the ‘what’; landscape painting as methodology supplies the ‘how’.

Three distances

Landscape painting aims to portray broader patterns of meaning. Although Chinese landscape paintings appear to be a scroll of limited length, they are pictures that encompass objects in the world in a well-ordered manner (Ji, 2014). In other words, there is a special connection between the landscape painting and the real human world. As analysed above, painters influenced by neo-Confucianism regarded natural objects as the embodiments of cosmic experience. When painters organise the mountains and rivers in the scrolls, they are inclined to use concepts such as priority, distance and size, which are analogous to human order, as a way of laying out the landscape scenery (Ji, 2014). They attempt to explore the patterns and principles which underlay the entire cosmos (Zhou, 2020). The principle of ‘three distances’ proposed by Guo Xi (1023–1085) echoes and elaborates on this feature.

According to Guo Xi (Guo, 2010, p. 86), there are ‘three distances’ when drawing and viewing a landscape painting: from the bottom of the mountain to the top of the mountain is upper distance; from the front of the mountain to the back of the mountain refers to deep distance; viewing from the nearby mountain to the far mountain is called level distance. ‘three distances’ is a kind of perspective that puts all aspects of nature seen from various viewpoints (high, deep and plane) in the same work (The Metropolitan Museum of Art, 2022). These three distances are complementary to each other, and each is equally important. Without them, the landscape painting may lose its richness of meaning (Ji, 2014). Differing from the western concept of viewing a scroll from a single point of view, ‘three distances’ emphasise visual diversity and the coherence of the configuration, which fosters viewers immersing themselves in the landscape (Jullien, 2009). ‘Three distances’ inspired us to generate three dimensions which could be employed in this mapping work. To be specific, the three dimensions are power relations, legal status and external institutional engagement. These three dimensions are what we focus on when regarding EC provision as metaphorical and methodological landscape painters, recognising that an understanding of the education collective requires each of the three dimensions to be seen as complementary and equal.

Admittedly, the adoption of landscape painting to depict the EC under the backdrop of China’s contemporary education landscape may be controversial. As some scholars (see Levenson & Schurmann, 2020; Mou, 2008) have argued, ancient Chinese ideology has lost its relevance and practical value and cannot reflect the contemporary Chinese social landscape. The landscape painting analysed in this article and the characteristics behind it serve as a searchlight to empower us to explore Chinese cultural norms. These characteristics, entwined with Chinese ideology and epistemology, are intrinsic to the existence and civilisation of a nation that has been able to endure and develop over a significant period. They directly and indirectly, consciously and unconsciously, influence Chinese policymakers. Using some features of landscape painting, therefore, can inform the overall approach and methodology. More significantly, these features, like a ‘photo album’, record the vast terrain of China’s basic education and enable us to comprehend the reasons that Chinese education policymakers choose and promote education collectivisation, thus allowing us further to elucidate our typology of ECs.

Typologising education collectives in China

In this section, we draw on thinking regarding the ‘three distances’ above to conceptualise education-collective provision in China. We interplay our thinking that is inspired by the ‘three distances’ with a further three dimensions, which we argue are typologically significant here and are all drawn from the academic literature. These three are as follows: legal status; power relations; and external institutional engagement. First, we selected legal status because we see statutory framings as enabling both of understanding state intentions and of comparative analysis between states. Legal status invariably does not capture entirely the terrain (Courtney, 2015) but is arguably an authoritative perspective that we believe merits representation in our, and indeed any, typology.

We selected ‘power relations’ as a typological dimension to illuminate the reasons for partnering, which no other dimension attempts. There is precedent for this in the literature on mapping structures of provision: Chapman et al. (2009) deploy the category of ‘performance federation’ (p. 6) to capture the combination of high-performing/lower-performing federation partnerships in England, which we see as translatable to the context of China. Similarly, our dimension of ‘external institutional engagement’ follows Muijs et al.’s (2011) mapping of school collaborations and networking.

We considered, but rejected, other typological dimensions, including geographical location/spread and duration of collaboration. We omitted these from our typology because the data were not consistently or reliably available.

Education collective types by legal status

Upper distance in the ‘three distances’ method emphasises the perspective of ‘looking up from the bottom of the mountain to the top of the mountain’ (Guo, 2010, p. 98), reflecting a lofty and majestic mountainous terrain that presents at first sight. Legal status refers to the status or position held by an entity as defined by the law (Collins, n.d.) and is illuminated through an exploration from the upper-distance, high perspective. Hence, various parameters such as curriculum, governance, student selection, recruitment policy and staff compensation are definable according to legal status (Courtney, 2015), producing an upper-distance classification judgement.

We classify ECs into three legal types: integrated management of one legal entity; joint management of several legal entities; and authorised hybrid management. Among them, the authorised hybrid management, also known as the first-level legal person delegated management, has grown out of integrated management of one legal entity. Table 1 shows the details of categorising ECs under these three legal statuses.

Table 1. Education collectives categorised by legal status.

Education collective type	Number of legal entities	Details
Integrated management of one legal entity	One	The CEO/lead-school principal is legally responsible for all constituent schools of the EC, and the executive principal is responsible for its daily operation. The EC has a unified vision, ethos and management system. Resource coordination among the member schools is relatively strong and closely linked.
Joint management of several legal entities	Equal to the quantity of the EC’s constituent schools	Constituent schools have independent legal status, generally responsibilised by the principals of their respective schools, with independent management systems, clear responsibilities, and distinctive features.
Authorised hybrid management	One	The legal responsibility for the EC remains with its CEO. However, the CEO may delegate legal responsibility to each constituent principal. Or the CEO is maintained as the legal entity under the management of the EC and develops ‘weaker’ constituent schools to become independent and autonomous (Lu & Liu, 2018).

Table 1 illustrates that ECs have diverse leadership and governance structures, which not only collectivise action as school organisations scale up but also cultivate inter-school cooperation (Du & Duan, 2020). This diversity means that typologising by legal status brings challenges, including a lack of informal articulation of legal role and responsibility; this ambiguity may have implications for the joint governing body of, and legal relationships within ECs. This limitation is reflected when employing upper distance to organise landscape painting. ‘Looking up’ to capture the height of the mountain, although the brush stroke is solid, it cannot describe the scenery and details behind the mountain (Ji, 2014). Neither can legal status identify coercion in education collectivisation. Institutional regulations and the profit motive (such as obtaining more financial appropriations) encourage ‘prestigious’ and ‘weak’ schools to advance high-quality education across the collective (An, 2020), yet most ECs are established as a way of collectivising action by local government and education administrations to promote balanced as well as superior development of basic education in the region. This is precisely what legal status as a typological dimension tends to overlook and conceal; that is, one relatively ‘weaker’ school is forced to cooperate with other relatively ‘more reputable’ schools or lose its previous independent legal status after joining an EC, resulting sometimes in homogenisation (Courtney, 2015; Muijs et al., 2011).

Education collective types by power relations

After seeing the high mountains before us, we would like to delve into the scenery behind the mountains. Deep distance expresses the sense of depth of the winding mountain paths (Ji, 2014). In Guo’s (2010) definition of deep distance — ‘peeping from the front of the mountain to the back of the mountain’, the prominence is on the action of ‘peeping’, indicating that the desired scene is elusive or hidden behind. Power relations, too, may be hidden, but can reveal the extent to which the collaborations in ECs are based on equality.

In the process of joint development, the constituent schools in the EC are constantly changing in scale and updating their leadership systems, and their internal governance systems may also be affected. We argue that power relations are a central force in the collective, affecting network dynamics, resource control, and tensions between cooperation and competition. Therefore, they are regarded as the meridian of education collectivisation (Meng & Dong, 2020), tightly embedded in their categorisation. Based on this, we show in Table 2 how we use the dimension of power relations to typologise ECs.

Table 2. Education collectives categorised by power relations.

Education collective type	Power relations	Details
Brokered management	Prestigious school + Weak schools	The ‘weak’ school is brokered to the EC’s management team by the ‘prestigious’ school (or a new principal is appointed by the Local Authority); the original legal entity and administrative affiliation of the ‘weak’ school remain unchanged. The ‘prestigious’ school provides guidance in leadership, learning and teaching. The ‘weak’ school’s teachers receive professional development.
Integrated management	Prestigious school + its Branch schools	The EC’s CEO is its legal representative and is responsible for all within it. The EC operates under the mechanism of ‘one leadership team, multiple branch campuses, unified management and shared resources’ and carries out unified management of personnel, funding and teaching.
Inter-school management	Prestigious school + Weak schools	This kind of EC is an inter-school alliance formed by several schools in the region, led by a ‘reputable’ school, and some ‘ordinary’ or ‘weak’ schools. The EC leadership team is composed of legal representatives of each member school; their original legal person status, leadership team, financial accounting and teachers’ officially budgeted posts remain relatively independent. All constituent schools are coordinated in terms of organisational ethos, governance system and development goals.
Alliance between giants	Prestigious schools + Prestigious schools	Schools with different strengths are combined without being dominated by any participant, and partners cooperate to solve problems on an equal footing. This model is a strong combination of ‘prestigious’ schools and ‘reputable’ schools, using the different advantageous resources of each constituent school to amplify their characteristics and respect differences.

As Table 2 illustrates, power relations is one of the most complex dimensions in the typology because it illuminates opaque internal governance structures which shape the vision and reproduce interests. However, power relations focuses more on the interior of the EC, while the power dynamics between the EC and local government remains occluded. There is limited evidence (see You, 2021) that brokered management has redistributed school autonomy under the influence of China’s bureaucratic model. We note further that this dimension is internally diverse, including concerning the autonomy and independence of constituent schools in different collectives of ‘prestigious school + weak schools’. Also, whether the brokered ‘weak’ school is regarded as a level of the governance structure, or just a subject that needs to be helped, needs to be further explored.

Furthermore, the ‘weak’ schools in an EC are expected to develop rapidly, notwithstanding the lack of capacity that networks consisting of only/mostly weak schools often have to collaborate (Muijs et al., 2011). We suggest that the competition between constituent schools is intentionally weakened in ECs, regardless of type. Just like Guo Xi’s further elucidation of deep distance — ‘The colour of deep distance is heavy and dark, and the artistic conception of deep distance is overlapped’ (Guo, 2010, p. 69), these are all hidden in the bureaucratic context.

External engagement

The extent to which external institutions are involved in the establishment and governance of the EC is also typologically significant. Here, ‘external institutions’ mainly refer to university and state-owned enterprises. We have excluded from our typology other commonly involved sorts of external institutions such as profit-making organisations (Liu & Zhu, 2018) because their involvement goes beyond the public-education sector. In China, universities and state-owned enterprises participate in ECs mainly through funding and permitting their name to be associated with them.

We divided this category into two periods, namely, before and after local government participation, and identified different types of affiliated ECs from these two historical periods. Before local government participation, affiliated ECs reflected the wider context in which ‘universities and state-owned enterprises run the society’ (Wang & Shen, 2021, p. 26), with limited participation from local governments.

The second phase of the development of the affiliated ECs comprises local governments taking the lead and universities supporting the development of relatively weak local and secondary ECs. External institutions provide only professional or resource support to affiliated ECs, mostly through consultancy. Meanwhile, some universities obtain enrolment quotas from affiliated ECs to coordinate the enrolment of their staff’s children (Beijing Municipal Education Commission, 2022). External institutions and the EC are independent of each other, and their relationship is government-oriented, yet cooperative rather than subordinate.

Strengthened through the level distance perspective, Table 3 details three types concerning external engagement. Level distance is viewing from the nearby mountain to the far mountain (Guo, 2010), which means that the creators need to put different scenes in the same scroll and paint the vast space between the front and back (Cahill, 2009), thus reflecting a more expansive landscape. This encouraged us to simultaneously investigate the external institution’s engagement in the EC both before and after local government’s involvement. However, since ‘before’ and ‘after’ co-exist chronologically in different collectives, these are typological constructs rather than objective indicators. When we stood before this landscape painting and examined this terrain, we found that the cooperation between the local government and local universities to jointly build affiliated ECs has gradually become a vital type and trend among external institutions to participate in the construction and governance of ECs. The most significant reason is that this kind of cooperative relationship has not changed the relatively independent relationship between ECs and universities, and policymakers suppose that the involvement of universities can promote the expansion of the total number of high-quality educational resources in the region (S. Zhang, 2021).

Table 3. Education collectives categorised by external engagement.

External institution	Details
University	High engagement. This type of affiliated EC is an internal institution of the university. The university is responsible for its recruitment, funding, staffing, leadership, premises and management. The EC’s organisational philosophy, resources, and teacher training are key areas of support from the university.
State-owned enterprise	High engagement. ECs affiliated with state-owned enterprises develop in accordance with its parent enterprise (Chen, 2007). Advantage: these ECs combine flexibility with the public interest (Lin & Chen, 2004) and are strongly dependent on enterprises (Chen, 2007). When the enterprise has to survive independently in the market, this kind of EC loses the relative stability required as an educational entity. If the state-owned enterprise declines, their affiliated EC becomes unsustainable.
Mainly local government, supplemented by a university	Low engagement. Government-affiliated and oriented ECs aim to achieve quality education in the region. They are also in partnership with universities but are not administratively affiliated. The ‘pairing’ is facilitated by a co-established coordinating body (Wang & Shen, 2021). Universities gradually import brand resources (primary and secondary collectives are renamed as university-affiliated collectives), curriculum resources (course design developed with university research institutes and staff), teaching resources (professors give lectures in affiliated ECs, and universities professionally develop teachers.

However, this cooperative relationship appears to be one-sided, with primary and secondary ECs merely affiliated to universities, and the affiliation sustained through external forces. Simultaneously, ECs affiliated with state-owned enterprises have received little scholarly attention, despite their potential significance. For instance, the focus on this type of EC has enabled state-owned enterprises to examine how they can integrate the EC’s educational programmes into their operations and thus contribute to the development of human resources. Furthermore, China has introduced policies to encourage the transfer of some of the non-collectivised primary and secondary schools run by the state-owned enterprises to the local government (Ministry of Education of the People’s Republic of China, 2001). This transfer process raises vital questions about the financial arrangements for retired school employees, particularly regarding pension payments (Ministry of Education of the People’s Republic of China, 2017).

Additionally, the most suitable sites for such external institutions to participate in ECs are concentrated in areas with abundant higher education resources. When other, less endowed regions transplant the developed-region model, they do not undergo the necessary remediation and adaptation. Therefore, the participation of external institutions becomes more diluted, which is overlooked by policymakers and academic advisers.

Conclusion: thinking with ‘blank-leaving’

This mapping work is looking at the most widely adopted form of school-to-school collaboration practice in the Chinese context – education collectivisation (Liu, 2022), and how it is categorised. We argue that ECs, to some extent, represent one good vision of Chinese policymakers for school improvement in the compulsory education field, that is, to realise the flow and reallocation of high-quality educational resources in a certain area in a relatively moderate and low-cost way (Zhang, 2015). The process of moving from a single governance model to a collective governance model involves the transformation of educational ideas, a change in the governance structure, an adjustment in the pattern of interests, a reshaping of the school culture, and an innovation of organisational methods.

We have innovatively employed the metaphor of Chinese classical landscape painting and its main features to inform and structure this typological work. This article makes pioneering use of eastern concepts to think through and elucidate the process of typologising, which is a significant conceptual and methodological contribution to this field. In this final section, we borrow another concept, blank-leaving, in Chinese landscape painting to conclude this article.

Blank-leaving in China’s landscape painting, also known as middle void, usually refers to white and unpainted space (Trouveroy, 2003). It can be understood as employing brushwork to create narrow gaps between objects, such as clouds between mountains, sunlight, moonlight and cracks between rocks and trees (T. Y. Zhang, 2021). Blank-leaving is a distinctive visual style in landscape painting, a space intentionally created by the creator (Yue, 2014), and for emotional integration (Li, 2021). There is also a blank-leaving space in this landscape painting of the education collective; we have therefore drawn on the concept of blank-leaving to summarise this article. This unpainted space is the research void which cannot be explored by the above three dimensions. At the same time, it also symbolises the potential of this research.

Fundamentally, the three viewpoints discussed above cannot fully demonstrate whether the EC is cross-regional. For example, the geographical spread of ECs may be within the same region or across regions. Moreover, under the backdrop of rural–urban differences in China’s primary education, although schools in the EC are located in the same area, there are still variations in that the constituent schools are all in the city and the countryside, or a hybrid of urban and rural areas (Yang & Tian, 2018). Currently, ECs are mainly concentrated in urban schools, but urban-rural co-construction ECs can also be found in many provinces in China, that is, a collective-running form composed of a famous urban school and one or more rural schools (Duan, 2018). The education collectivisation of township schools, especially small-scale rural areas schools, not only has the feature of low coverage but also lacks the internal drive for sustainable development (Yu, 2020). In the process of promoting collectivised school management, the principals of some rural schools, confined to the management system of township central schools, have not realised the positive significance of education collectivisation to enhance the integration of urban and rural education (Jiangsu Agency for Educational Evaluation, 2019). Hence, township schools are generally underpowered in relation to the substantive stimulation of education collectivisation.

It is challenging for the aforesaid three dimensions to reflect whether the schools in the collective are in the same stage. To be specific, all member schools in some ECs are primary schools or secondary schools, while some ECs adopt a mixed model, namely there are both primary schools and secondary schools. However, when schools at different stages participate in one collective, it seems that it is more challenging to reconstruct organisational boundaries (An, 2020). In terms of deepening collaboration and linking different stages of education, a mixed model seems to be desirable, which enables building an open and cooperative educational atmosphere and achieving the connection within the compulsory education stage (Yang & Tian, 2018).

It is therefore limiting to adopt a single dimension to categorise ECs and, to a certain extent, the process of categorising is simplified by adopting only these three dimensions. However, we propose that the blank-leaving space, which is considered the creator’s and viewer’s ‘eye of the gaze’ (Trouveroy, 2003, p. 36), as well as the ‘space of communication’ (T. Y. Zhang, 2021, p. 69), allows us to focus more on these three dimensions. Further, legal status, power relations and external engagement enable us to avoid getting lost in front of a huge landscape or being overwhelmed by other lofty peaks. These three dimensions, respectively, emphasise the relationship between the legal person and the boundaries of constituent schools in one EC, the relationship between inter-school cooperation and the differential development of constituent schools in the collective, and the relationship between external institutions and the value of establishing connections between external institutions and the EC. More significantly, they individually illuminate the homogeneity, exclusion and unsustainability that the EC may encounter in its development process.

The first is to explore how ‘weak’ and ‘new’ constituent schools cultivate independent campus cultural characteristics and community cultural resources in the type of integrated management of one legal entity and brokered management in ECs. In terms of the purpose, training goals, and the vision, it not only reflects the cultural continuity of the EC, but also develops the differences and characteristics among constituent schools. The second is when a school is ‘collectivised’ and its internal power governance collides with the EC’s power position, how the leadership team obtains the school’s independent decision-making power, and how hierarchical relationships are delicately balanced in a more complex organisation. The third is how to attract external engaged institutions to provide sustainable support and impetus to affiliated ECs which are artificially designed or generated by policies. Given that the unique value of affiliated schools is becoming increasingly obvious, it is essential to explore the internal motivation of external institutions to participate in the improvement of ECs at the practical level. The interweaving of these three dimensions plays a vital role in our construction of the landscape painting of ECs. It breaks the limitation that ordinary paintings can only observe the landscape from one angle, that is, focusing on perspective, and provides the possibility of mapping the ECs landscape and the potential research perspective of China’s educational field.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Pinyan Lin

Pinyan Lin is in the final year of a PhD programme at The Manchester Institute of Education, having previously graduated with the Master’s degree in Educational Leadership. Her research interests lie in educational leadership, school-to-school collaboration and critical policy studies. Pinyan’s main research seeks to critically investigate and conceptualise governance structures and leadership models of education collectives in China. Currently, she is the assistant editor of International Journal of Leadership in Education, and a member of the Critical Education Leadership and Policy group.

Steven J. Courtney

Steven J. Courtney is Professor of Sociology of Education Leadership at the University of Manchester. He carries out research in areas including education policy, system leadership, charisma, structural reform, depoliticisation and education privatisation, particularly in relation to the identities and practices of those constructed as educational leaders. He is currently Director of Research in the Manchester Institute of Education (MIE), Co-Editor-in-Chief of the journal Critical Studies in Education and co-convenor of the MIE research group, Critical Education Leadership and Policy (CELP). His most recent book (2021) is ‘Understanding educational leadership: Critical perspectives and approaches’ with Bloomsbury, co-edited with Helen Gunter, Richard Niesche and Tina Trujillo. His latest two books are in press: ‘Keywords in education policy: A conceptual toolbox’, co-authored with Andrew Wilkins and Nelli Piattoeva (Policy Press) and ‘Critical education policy and leadership studies: Essays in honour of the intellectual contributions of Helen M. Gunter’, co-edited with Tanya Fitzgerald (Springer).

Paul Armstrong

Paul Armstrong is Senior Lecturer in Education at the Manchester Institute of Education, University of Manchester. His research interests concern contemporary forms of educational leadership and management and professional collaboration between educators. Paul has over 18 years of experience in education research leading on national and international projects and publishing across a number of areas including educational improvement, networking, leadership, management and policy. He has also led a number of Masters programmes for teachers and aspiring school leaders. Paul is currently Programme Director for the MA Educational Leadership in Practice at the University of Manchester and Editor-in-Chief of the Sage journal Management in Education. His latest book, an edited collection with Professor Chris Brown entitled School-to-School Collaboration: Learning across international contexts, is now available through Emerald Publishing.

Amanda McKay

Amanda McKay is a Senior Lecturer in the Manchester Institute of Education, at the University of Manchester. Her research explores the contemporary challenges of principals’ work and how to better attract, support, and retain school leaders within the profession. She is Co-Editor of the Journal of Educational Administration and History.