Understanding Paradigm Shifts and Asynchrony in Environmental Governance: A Mixed-Methods-Study of China’s Sustainable Development Transition

Abstract

Escalating environmental challenges severely impede global sustainable development, prompting countries worldwide to innovate environmental governance approaches. As the world’s largest developing country, China’s paradigm shifts in environmental governance from “pollution control” to “ecological conservation” embody many inherent complexities. To investigate the evolution and underlying logic of such paradigm shifts, this study introduces a nested asynchrony framework. Employing a mixed-methods approach that integrates qualitative content analysis, Social Network Analysis, and machine learning, this study analyzes China’s environmental planning documents since the 11th Five-Year Plan to clarify the process of the paradigm shifts and their driving mechanisms. The principal conclusions derived from this study are as follows: (1) Environmental planning is uniquely valued as an analytical lens for identifying paradigm shifts in environmental governance. (2) The paradigm shifts in environmental governance are temporally distinct, wherein transformations in value norms precede structural reforms, while shifts in action logic and disciplinary foundations exhibit path-dependent inertia. (3) Inconsistencies within the planning authority framework spanning central and local governments impede the effective allocation and implementation of resources. This study reconstructs the transformation pathway of environmental governance paradigms, validates computational methods in policy analysis, and presents a longitudinal framework for tracking governance evolution. Applicable to other countries or sectors undergoing similar sustainable development transitions, the framework can provide broader utility.

1. Introduction

1.1. Background

Against the backdrop of intensifying global climate change and ecological degradation, sustainable development has emerged as a paramount concern for the international community. Within the 17 Sustainable Development Goals (SDGs) proposed by the 2030 Agenda for Sustainable Development, SDG 11 (Sustainable Cities and Communities) emphasizes the advancement of sustainable human settlements through inclusive urban planning and resource optimization. Concurrently, SDG 13 (Climate Action) demands urgent measures to address climate change and its cascading impacts. As a crucial aspect of national governance, environmental governance entails governments leveraging formal and informal institutions to integrate ideas, theories, resources, and power within environmental governance practices. It also involves coordinating the participation of various stakeholders, including the public and enterprises, in environmental conservation efforts [1]. Developing a new paradigm for environmental governance is not only essential for realizing SDG 11 and SDG 13, but also serves as a comprehensive assessment of a country’s governance capabilities in multiple dimensions such as interest allocation, technological innovation, and institutional reconstruction.

Given the unique convergence of substantial populations, accelerated economic expansion, and ecological vulnerability, developing nations are pivotal to sustainable development. These countries are fundamentally vested in the equity of global resource distribution and the efficacy of climate initiatives. Illustratively, the preservation of the Amazon rainforest in Brazil directly influences global carbon sequestration capacity [2]; concurrently, India’s low-carbon transition is being advanced through extensive investments in renewable energy infrastructure [3].

China, as the world’s largest developing country, presents a particularly compelling case study. Confronted with acute environmental pressures from decades of rapid growth, China has undertaken a strategic pivot toward constructing an “Ecological Civilization” and a “Beautiful China”, aiming to transcend the traditional “treatment after pollution” dilemma. Key policy directives, notably the Guiding Opinions on Building a Modern Environmental Governance System (2020) and the Opinions on Comprehensively Promoting the Construction of a Beautiful China (2023), signal a paradigm shift away from a narrow focus on “pollution control” towards a more holistic vision of “ecological conservation” (see Figure 1). This emerging paradigm emphasizes comprehensive transformation across all fields, multi-level integrated improvements, regionally balanced spatial development, and the mobilization of all sectors of society for collective environmental action. During transformative processes, the principle of “harmony between humanity and nature” must be institutionally embedded within governance systems. However, existing hierarchical frameworks remain constrained by efficiency-oriented path dependence. This fundamental tension between conceptual innovation and institutional inertia induces a temporal disconnect in transformation. Such disconnect not only undermines policy implementation efficacy, but also relegates “Beautiful China” initiatives to symbolic gestures. Consequently, the asynchrony of paradigm shifts constitutes the primary analytical lens for understanding China’s environmental governance transition dilemmas. The unprecedented scale, speed, and ambition of China’s transition offer critical insights for global sustainability efforts, providing valuable lessons for other nations advancing ecological modernization.

1.2. Literature Review

Existing research on the paradigm shifts in environmental governance focus on the following areas: (1) At the theoretical level, scholarly engagement has investigated multiple facets of environmental governance reform, including policy tools [4,5], regional coordination [6,7], and legal development [8,9]; notable gaps persist. Many studies focus on practical implementations without systematically examining the underlying connections and normative principles that bridge conceptual frameworks and on-the-ground actions. Furthermore, the route between the SDGs’ contents and their practical implementation pathways has yet to be comprehensively examined. Although concerned about the role of public participation in environmental governance [10,11], China’s current approach remains predominantly characterized by top-down authoritarian environmentalism [12,13,14], which notably neglects the core tenets of just transition—specifically distributional, procedural, and recognitional justice dimensions [15,16,17]. The implementation of such coercive measures tends to induce policy rigidity, undermining long-term efficacy while failing to address societal needs, particularly those of vulnerable populations. (2) At the methodological level, traditional qualitative and case-based approaches often lack the scalability required to analyze large volumes of policy texts, which are critical for tracing governance evolution over time [18,19,20]. Although the research reveals subnational implementation variations, this limitation impedes the systematic tracking of environmental governance paradigm shifts across macroevolutionary trajectories. (3) At the empirical level, research perspectives remain predominantly anchored in Western governance frameworks—notably the common-pool resources model [21], the environmental Kuznets curve [22], and eco-capitalism theories [23]—which dominate environmental governance scholarship. These research frameworks ignore environmental decolonization discourses while marginalizing traditional knowledge systems and place-based lifeways [24,25,26]. Crucially, China’s unique context, characterized by its socialist state apparatus and accelerated urbanization, necessitates governance mechanisms capable of efficiently coordinating multi-tiered governmental and stakeholder resources [27]. In fact, even in developing countries, Brazil’s authoritarian–populist centralization [28] or India’s postcolonial planning deficits [29] differ from China’s governance model dominated by centralization and hierarchical implementation, thus making it all the more important to strengthen the study of the internal mechanisms of large-scale governance transformations in non-Western countries.

1.3. Research Content and Contributions

To fill these research gaps, this study examines the evolution and underlying logic of China’s environmental governance transition from a “pollution control” paradigm to one centered on “ecological conservation”. Our analysis encompasses a corpus of 176 national and local (Shanghai Municipality and Yangpu District) environmental planning documents spanning 2006 to 2025. Adopting a mixed-methods approach, we combine qualitative content analysis with computational techniques, including Latent Dirichlet Allocation (LDA) topic modeling, Social Network Analysis (SNA), and machine learning-based text similarity calculations. First, we compare the two paradigms in terms of policy objectives, implementation tools, and stakeholder coordination, identifying the defining features of the emerging “ecological conservation” model. We then investigate the driving forces behind the shift, including economic development priorities, policy guidance, and technological change. Finally, we evaluate whether changes in value norms, governance structures, action logic, and disciplinary foundations unfold synchronously. Where asynchrony is identified, we analyze its patterns and implications for the coherence and effectiveness of the governance transformation.

This research makes three key contributions. (1) Theoretically, it identifies and offers a preliminary theorization of the “asynchronous” characteristics inherent in the environmental governance paradigm shift and integrates top-down and bottom-up policy coherence, deepening insights into the complex dynamics of large-scale governance transformations and filling the research gap of the integration of inclusive planning (SDG 11) and equitable climate action (SDG 13) within global sustainability frameworks. (2) Methodologically, it demonstrates how integrating methods like content analysis and Social Network Analysis can systematically track and interpret governance evolution through policy documents. (3) Empirically, it provides longitudinal data-driven insights and localized narratives into the understanding of governance challenges in China’s sustainability transition, with implications for other nations advancing similar development agendas.

The paper is structured as follows: Section 2 outlines the theoretical framework. Section 3 details the data and analytical methods. Section 4 presents the research findings. Section 5 discusses the results and their implications. Section 6 concludes with summarizing remarks.

2. Theoretical Framework: Environmental Governance Paradigms

2.1. Defining Governance Paradigms and Shifts

A paradigm denotes a comprehensive framework of shared beliefs, theories, methodologies, and tools within a scientific community [30]. A paradigm shift emerges from theoretical innovation driven by the demands of the times [31]. In the realm of policy and governance, an environmental governance paradigm can be defined as an organic integration of value consensus, theoretical framework, institutional arrangements and practical tools formed by the environmental governance community during a specific historical period [32]. When the existing governance framework fails to address contemporary challenges, achieving a holistic transition of the governance paradigm through conceptual innovation, institutional reconstruction, and tool renewal functions as an inevitable choice for solving complicated issues.

Since the 1970s, environmental governance paradigms in Western countries have followed a process of emergence, rise, rapid advancement, and eventual globalization [33,34,35]. In China, the pursuit of modernization marked by harmony between humanity and nature—characterized by productive growth, elevated living standards, and robust ecosystems—alongside evolving public aspirations for environmental quality, has driven the transformation of its environmental governance paradigm. This evolution has shifted from an early stage characterized by government-led, economy-centered governance to a more systematic and science-based governance framework emphasizing multi-actor collaboration and diverse policy instruments [36,37,38,39]. Such a transition embodies a revolutionary reconstruction of the epistemology and methodology of the governance of social-ecological systems.

The core characteristics of environmental governance paradigm shifts are manifested in governance objectives, governance actors, governance scenarios, governance instruments, and governance values. Scholars commonly summarize these features into four dimensions: value norms, disciplinary foundation, governance structure, and action logic [40,41]. Among them, value norms guide the overarching goals and strategic directions in environmental governance, representing the “conceptual” dimension. The disciplinary foundation serves as both the theoretical basis and the analytical framework construction mechanism, constituting the “theoretical” dimension. The governance structure reflects the operational and interactive processes among governance objects, governance actors, and governance instruments. Action logic encompasses the institutional design and implementation processes, as well as the synergy between individual institutions and institutional frameworks. Together, these form the “practical” level.

2.2. Comparing Key Paradigms

Analyzing the divergent characteristics of China’s “pollution control” paradigm and “ecological conservation” paradigm (see

Table 1. The comparison between the “pollution control” paradigm and the “ecological conservation” paradigm of environmental governance.

Paradigm Composition	“Pollution Control” Paradigm	“Ecological Conservation” Paradigm
Value norm	Passive control	Harmony between humanity and nature
Disciplinary foundation	Interdisciplinary studies in natural sciences	Multidisciplinary integration of natural science and social science
Governance structure	Bureaucratic monopoly	Dynamic game of multiple subjects
Action logic	Administrative enforcement	Value co-creation integrated with market mechanisms, cultural identity

) is critical for identifying the key inflection points and transformation directions of the paradigm shift.

In terms of value norms, the “pollution control” paradigm centers on safeguarding environmental safety thresholds, rooted in the assumption of “weak sustainability” [42]. This assumption holds that as long as we can effectively control the total emissions of pollutants and the upper limits of resource utilization, then ecosystem collapse and threats to human health can be avoided more easily. Under this paradigm, nature is treated as an object to be repaired [43], highlighting the restoration of baseline environmental conditions through technological means, with a more passive and reactive logic of governance [44]. Conversely, the “ecological conservation” paradigm is based on the theory of “strong sustainability” [45], prioritizing the harmony between humanity and nature. It emphasizes natural resource compensation mechanisms and landscape aesthetic evaluation [46,47], transforming ecological services into driving factors for socio-economic growth, which reflects a more proactive and constructive logic of governance.

In terms of disciplinary foundation, the “pollution control” paradigm follows a disciplinary hierarchy from natural sciences to science and engineering, then to environmental science and engineering, and finally to environmental planning and management [48]. Research primarily focuses on pollutant emissions and environmental monitoring, remaining largely within the interdisciplinary domain of science and engineering with a strong emphasis on instrumental rationality [49]. By contrast, the “ecological conservation” paradigm integrates social sciences with natural sciences, drawing from knowledge systems in environmental science, urban planning, public policy, ecology, economics, and sociology [50,51]. It establishes correlation mechanisms and driving factor analysis models that link governance systems, governance capacity, and environmental quality evolution [52], while also prioritizing value rationality in areas like community environmental behavior and ecological aesthetics [53,54].

In terms of governance structure, the “pollution control” paradigm centers on targeted pollution prevention efforts for tangible, material resources such as air, water, and soil. Environmental protection departments act as the leading force, decomposing pollutant emission reduction targets and environmental quality standards through hierarchical bureaucratic systems [55]. Local environmental departments conduct on-site enforcement, with accountability limited to administrative penalties for non-compliant enterprises [56]. Meanwhile, land-use approval for industrial upgrading falls under natural resource departments, while clean technology transformation subsidies require approval from economic and information departments, creating a dilemma where “departments compete and enterprises suffer.” Such intersectoral competition within environmental authoritarian systems can impede both the formulation and implementation of environmental policies, thereby constraining effective civic input [12]. In contrast, the “ecological conservation” paradigm extends to diverse socio-economic activities closely linked to the environment—encompassing intangible, immaterial, and innovative resources like ecosystem services. Also, it emphasizes a multi-level network of “departmental coordination, market embedding, and social co-governance” [57]. Guided by the principles of the “shared responsibilities of Party and government” and interdepartmental cooperation, market actors allocate resources through mechanisms like carbon emission trading [50], while social organizations exercise their supervisory power through environmental information disclosure and environmental public interest litigation [58]. Utilizing mechanisms such as horizontal transfer payments can advance the regional coordination and land–sea integrated governance [59]. Government departments transition from “technical bureaucrats” to “purchasers of ecological services”, enterprises evolve from “regulated entities” to “green technology providers”, and the public obtain the right to participate in environmental decision-making through co-governance. These transformations collectively foster a “rebalancing of power”.

In terms of action logic, the “pollution control” paradigm uses terms such as “control” and “bottom line” to strengthen administrative authority, achieving governance objectives through campaign-style approaches such as assessments, evaluations, and demonstration projects [60]. It ensures the rigorous implementation of pollution prevention goals through terminal-oriented, engineering-based, and technological means. The “ecological conservation” paradigm, by contrast, transcends the traditional administrative control model, stimulates the innovation capacity of local governments through “policy translation”, and uses rhetoric such as “harmonious coexistence” to dilute the concentration of power and shape consensus. It integrates residents’ recognition of ecological culture into the evaluation criteria [61] and promotes civic environmentalism through state-media-guided green consumption norms [13,14]. It emphasizes market mechanisms like third-party pollution control, ecological compensation and carbon trading, leveraging full play to the institutional synergy of “mandatory laws, comprehensive planning, and flexible policies” [62]. It comprehensively employs environmental impact assessments of plans, mid-term evaluations during implementation, and post-implementation evaluations, and integrates emerging technologies such as big data, artificial intelligence, and the blockchain to improve the flexibility, precision, and effectiveness of environmental governance [63,64].

Therefore, the “pollution control” paradigm represents a linear, reactive governance model centered on pollutants, whereas the “ecological conservation” paradigm embodies a coupled, collaborative governance model grounded in ecosystem services and human aesthetic perception to respond to the goals of SDG 11 and SDG 13. It reconstructs a positive cycle of “natural restoration–economic benefits–cultural identity”, emphasizing systematization, prevention, multi-subject engagement, and multi-method application. This paradigm not only coordinates the relationships between humanity and nature, high-quality development and high-level protection, and government–enterprise–society interaction at the national level, but also illustrates the variety of “beauty in difference” and the inclusiveness of “shared beauty” [65].

However, the transition from “pollution control” to “ecological conservation” inevitably encounters multiple challenges, including goal conflicts, governance structure mismatches, and asynchronous transformations. Therefore, the need to analyze the internal mechanisms and influencing factors of this paradigm shift is urgent in order to overcome transition bottlenecks.

2.3. Analytical Framework: Driving Forces–Elements–Structure

The “Driving Forces–Elements–Structure” analytical framework originates from Marx’s theoretical proposition that “productive forces determine relations of production, and the economic base determines the superstructure.” Here, “driving forces” are the “starters” of transformation, propelling the expansion and upgrading of elements; “elements” reflect the specific constituent units of natural relations and serve as the “carriers” of transformation, with their compositional changes directly shaping structural forms; and “structure” denotes the organizational form of social relations and functions as the “stabilizer” of transformation, forming an adaptive response to driving forces through institutional solidification, action feedback and theoretical integration. This framework emphasizes the interplay between natural and social relations [66], making it particularly suitable for analyzing the interactive dynamics among various components of environmental governance—a complex system that bridges nature and society. This framework systematically uncovers the dynamic relationships among the internal driving forces, constituent elements, and organizational structures within such complex systems (see Figure 2).

The “driving forces” are composed of critical events, value norms, and technological logics which stem from the profound interplay of social pressure mechanisms, political priority realignments, and technological revolutions [67]. Firstly, critical events in society generate social pressure through public opinion, amplified by social media, compelling environmental concerns onto the policy agenda. Secondly, international commitments such as the “Dual Carbon Goals” (carbon peaking and carbon neutrality), coupled with the evolution of the principal contradictions in the country’s development, have reshaped the ecological priorities within national strategies. Meanwhile, digital technology advancements have propelled a transformation of environmental governance from experience-driven approaches to data- and intelligence-driven approaches. Critical events have been found to play a significant role in reshaping value norms. For instance, the cyanobacteria outbreaks in Taihu Lake prompted the adoption of integrated watershed management, while the convening of COP29 facilitated the integration of global ecological governance responsibilities into China’s national planning. Conversely, value norms can also mitigate the impact of critical events. As public environmental awareness increases, pollution incidents are more likely to be scrutinized and to trigger environmental protests.

The “elements” refer to the reorganization of governance resources and capabilities—comprising governance actors, governance objects, and governance instruments—within the governance framework. Their profound transformation stems from the structural breakthrough of production factor boundaries spurred on by the “driving forces”. As governance actors deploy governance instruments upon governance objects, knowledge elements reshape the cognitive and participatory scopes of environmental issues; spatial elements redefine the allocation principles of “development rights vs. protection rights”; technological elements fortify the precision governance cycle; and capital elements unleash market dynamics and social capital. These changes collectively catalyze a transition from governance dominated by single administrative resources to a “government–market–public” mosaic of resource distribution. In the process of expanding and integrating various elements, a just transition should be placed at the core—one that ensures not only the effectiveness of environmental governance but also the preservation of social equity and cultural diversity. Distributional justice involves the equitable allocation of resources across different levels and regions. Procedural justice emphasizes meaningful participation by diverse stakeholders in decision-making processes, thereby preventing the policy rigidity that may arise from hierarchical monopolies. Recognitional justice calls for the acknowledgment of local knowledge and the respect for pluralistic values.

The optimized integration of “elements” subsequently drives the synergistic transformation of institutional structure, action structure, and disciplinary structure—collectively comprising the “structure” dimension. The institutional structure frames the behavioral boundaries through laws, regulations, policy systems, and assessment mechanisms; the action structure evaluates the effectiveness of institutions through practical feedback from diverse actors; and the disciplinary structure offers a systematic disciplinary theory and methodological framework for both. However, the non-synchronization of these three structures’ transformations may trigger structural tensions within the governance system, resulting in a shortage of synergy among rule design, practical implementation, and theoretical supply, thereby constraining the effectiveness of the paradigm shift.

Therefore, the “Driving Forces–Elements–Structure” analytical framework forms a closed-loop mechanism, which will guide the subsequent empirical analysis of environmental planning documents. Unlike the multi-level governance framework, which emphasizes the spatial distribution of authority and responsibilities across government levels, or the policy coherence framework, which focuses on the static alignment of cross-sectoral policy objectives, this framework retains the hierarchical coordination strengths of multi-level governance through a “spatio-temporal nesting” mechanism. It further offers an innovative solution to the temporal management of policy divergence. As a result, it serves as a replicable analytical tool for governance transformation, characterized by both theoretical inclusiveness and strong practical explanatory power.

3. Research Design: Data and Methods

3.1. Data: Environmental Planning Documents

Environmental planning, defined as the process of “planning for the environment”, involves organizing and regulating diverse socio-economic activities and the production and living behaviors of different actors. It encompasses not only the plans formulated by ecological and environmental departments but also those developed by other relevant sectors whose activities are inherently linked to the environment and exert significant environmental impacts [68].

As a critical component that guides and supports the practice of environmental governance, environmental planning, characterized by continuity and long-term vision, integrates the diverse public affairs in environmental governance across temporal, spatial, and relational dimensions into a holistic and systematic planning framework. This enables the understanding of the development trajectories of diverse governance actors, the evolution of environmental policies across different periods, and the effectiveness of environmental governance practices [69]. Therefore, the selection of environmental planning texts as the analytical foundation for examining the paradigm shift in environmental governance was both reasonable and feasible in this study.

Environmental planning is an organic whole comprising dimensions such as planning nature, planning actors, and planning content. Specifically, the guiding ideology establishes the overall direction of the plan and reveals the value norms of environmental governance; planning nature and objectives define concrete tasks and target actors; the core tasks propose feasible pathways to achieving these objectives; and planning actors engage in collaborative responsibility allocation within this framework, thereby offering a partial reflection of the governance structure. In addition, safeguard measures offer resources and institutional support to the implementation of planning tasks, aligning with the auxiliary action logic. Collectively, these components encapsulate the evolutionary trajectory of environmental governance’s disciplinary foundations (see Figure 3).

3.2. Research Samples

Considering the pivotal role of planning in China’s governance framework (“governing the country through planning”) [27], we utilized environmental planning documents as core evidence, as these texts formalize governance objectives and track their evolution over time. At the Fifth Plenary Session of the 16th Central Committee of the Communist Party of China, the construction of a resource-conserving and environmentally friendly society was, for the first time, established as a medium- and long-term strategic task for national economic and social development. Since the 11th Five-Year Plan period (2006–2010), China has officially rebranded its Five-Year Plans as “Plans”. A notable milestone during this period was the State Council’s first-ever issuance of the 11th Five-Year Plan for Environmental Protection, which signaled a significant step forward in China’s environmental governance framework.

As a globally integrated metropolis confronting complex governance challenges, Shanghai offers a paradigmatic case for analyzing urban environmental planning. Additionally, its Yangpu District, with the task of ecological transformation from a traditional industrial base, serves as a valuable local-level example. The combination of Shanghai and Yangpu District provides dual cases of authority and representativeness for examining urban environmental planning challenges and vertical governance mechanisms.

Consequently, this study incorporated environmental-protection-related planning documents issued during the 11th Five-Year Plan period (2006–2010), 12th Five-Year Plan period (2011–2015), 13th Five-Year Plan period (2016–2020), and 14th Five-Year Plan period (2021–2025) at the national level, the municipal level (Shanghai), and the district level (Yangpu) as its research samples. Using keyword searches, content screening, and consistency testing, the study identified a total of 121 national-level environmental planning documents, 42 documents from Shanghai, and 13 documents from Yangpu District (see

Table 2. Number of environmental planning documents issued during 11th to 14th Five-Year Plan periods.

Period	National Level	Shanghai Municipal Level	Yangpu District Level
11th Five-Year Plan period	35	3	4
12th Five-Year Plan period	29	9	2
13th Five-Year Plan period	25	6	3
14th Five-Year Plan period	32	24	4

). More details can be found in the Supplementary Materials.

3.3. Methods of Analysis

This study used a mixed-methods approach integrating qualitative and quantitative analyses to examine the environmental planning documents and trace the evolution of environmental governance paradigms. This approach comprised qualitative content analysis, Latent Dirichlet Allocation (LDA) topic modeling, Social Network Analysis (SNA), and machine learning-based text similarity analysis.

3.3.1. Qualitative Content Analysis and Critical Discourse Analysis

Qualitative content analysis served as the foundational interpretive framework for interpreting the keywords, themes, and underlying meanings embedded in the planning documents. This method helps in understanding the implicit information and nuances conveyed. In this study, it was specifically applied to analyze selected sections such as “Guiding Ideologies”, “Main Tasks”, and “Safeguards” within the planning documents. This approach enabled a qualitative exploration of the evolution of value norms, governance objectives, and institutional frameworks across diverse periods and administrative levels.

As an interdisciplinary research approach, Critical Discourse Analysis (CDA) uncovers underlying power relations and ideological constructs by deconstructing linguistic structures, metaphors, and narrative strategies within texts [70]. In the context of environmental governance, CDA can illuminate the instrumental nature of policy texts, revealing how they may serve as tools for dominant actors to legitimize their interests, while also exposing deeper contradictions within policy practices. In this study, CDA was employed across three dimensions—textual analysis, discursive practice, and social practice—to reveal asymmetries of power and deconstruct ideological tendencies embedded in policy discourse.

3.3.2. Latent Dirichlet Allocation (LDA) Topic Modeling

To augment the qualitative findings and conduct a systematic quantitative analysis of thematic evolutions, this study deployed Latent Dirichlet Allocation (LDA) topic modeling. As a sophisticated generative probabilistic model, LDA was used to categorize planning documents by discerning the probabilistic distribution of latent topics. This method not only enabled the classification of documents but also facilitated the hierarchical clustering of thematic content by examining the content of documents sharing the same theme. Renowned for its advanced sampling algorithms and robust generalization capabilities, LDA has been widely validated in this field of research [71]. Text segmentation and cleaning were conducted using Python’s jieba library, followed by hyperparameter optimization through cross-validation to identify the optimal model based on topic perplexity and coherence. Finally, the effectiveness of the LDA method in policy text analysis was ensured through the use of pyLDAvis for visualization and manual semantic calibration.

This study leveraged LDA to conduct dimensionality reduction on the text semantics within the “Main Tasks” section. This approach enabled a quantitative examination of how governance topics and institutional priorities had evolved over time and across levels. The optimal number of topics for the LDA models was systematically determined through perplexity and coherence metrics. Thematic coherence was assessed as a primary criterion, followed by the evaluation of perplexity. Under the premise that the thematic semantics were reasonable, perplexity was then used to determine whether the model satisfied basic fitting requirements, thereby achieving a dual assurance of interpretability and statistical robustness. The mathematical formulations for these metrics are provided below.

$$Perplexity(T)=\exp \left\{-{\displaystyle \frac{{\displaystyle {\sum }_{t=1}^M\log (p(w_t))}}{{\displaystyle {\sum }_{t=0}^M{N}_t}}}\right\}$$

(1)

$$Coherence(T)=\log ({\displaystyle \frac{p(w_i,w_j)}{p(w_i)\ast p(w_j)}})$$

(2)

where T represents the word segmentation vocabulary set; M represents the total number of texts; N_t and p(w_t) signify the number of words and generation probability of text t, respectively; p(w_i) and p(w_j) indicate the probability of occurrence of single words w_i and w_j, respectively; and p(w_i,w_j) represents the probability of two words w_i and w_j co-occurring.

A Sankey diagram illustrating topic evolution across different levels, generated using the LDA topic model, visualized the correlations and degrees of closeness among topics within various tiers of environmental planning. The LDA topic model was first applied to extract core topics from planning texts at different levels. Subsequently, cosine similarity and the proportional representation of each topic within the documents were used to assign weights, thereby calculating the strength of the thematic associations. Only connections with a similarity or weight greater than 0.3 were retained for further analysis. Finally, the Sankey diagram was produced using Python 3.9.7’s pySankey library, with a total of 30 valid connections visualized.

3.3.3. Social Network Analysis (SNA)

This study applied Social Network Analysis (SNA) to examine the governance structure, focusing on the interaction patterns and relationship strengths among the government departments responsible for formulating environmental planning documents [72]. SNA utilizes relational data, graphical visualization, and quantitative metrics to analyze network structures. Although communication, collaboration, and even conflict occurring outside formal structures play a significant role, the effectiveness of informal cooperation mechanisms is difficult to measure in a standardized manner. In contrast, formal, documented cooperation serves as the baseline framework for intersectoral collaboration, providing traceable and verifiable data that enables large-scale, longitudinal comparisons.

In this study, the planning and issuing departments were regarded as “actors” within a social network, and an undirected network was constructed based on the number of jointly issued documents and instances of inter-governmental cooperation, representing the relational ties among these actors. This study utilized Ucinet 6.0 and NetDraw 2.184 software to conduct SNA, visualizing and quantifying the collaboration intensity among government departments involved in environmental plan formulation across various periods. By mapping these interdepartmental interactions, the analysis sheds light on the evolving synergy and potential conflicts within the governance network. Key network metrics included the number of nodes (N), representing distinct departments; the number of edges (E), representing the frequency of joint document issuance between departments; network density (D), capturing the overall strength of collaboration; the clustering coefficient (C), showing the extent to which departments form tightly cohesive groups or rely on central coordinating agencies; and betweenness centrality (B), measuring the extent to which a department serves as a bridge or intermediary in the network, with visual representation through node size. In general, higher network density and clustering coefficients indicate a more tightly connected network, thereby enhancing the overall network’s influence on the attitudes and behaviors of individual actors. The mathematical formulations for these metrics are provided below.

$$D={\displaystyle \frac{2E}{N(N-1)}}$$

(3)

$$C={\displaystyle \frac{3\times {{\rm T}}_{∆}}{{{\rm T}}_{triplet}}}$$

(4)

$$B(v)={\displaystyle \sum _{\mathrm{s}\ne v\ne r}\frac{{\sigma }_{s\mathrm{r}}(v)}{{\sigma }_{sr}}}$$

(5)

$$B_{normalized}(v)={\displaystyle \frac{B(v)}{(N-1)(N-2)/2}}$$

(6)

where $T_{∆}$ represents the number of closed triangles connecting three nodes in pairs, $T_{triplet}$ denotes the number of non-closed structures involving at least two edges among three nodes, B(v) stands for the betweenness centrality of node v, B_normalized(v) indicates the normalized betweenness centrality of the node, σ_sr represents the total number of shortest paths from node s to r, and σ_sr(v) represents the number of shortest paths passing through node v.

3.3.4. Machine Learning Text Similarity Analysis

To evaluate the collaborative efficiency within the governance action framework, this study conducted text similarity analysis grounded in machine learning techniques. Specifically, it integrated the Term Frequency–Inverse Document Frequency (TF-IDF) algorithm with cosine similarity calculations. As a well-established statistical approach, TF-IDF quantifies the significance of each word in a given document by measuring its frequency relative to an extensive corpus [73]. In this study, the scikit-learn library was selected for model construction due to its core advantages, including the efficient implementation and integration of traditional statistical learning algorithms. It offers a standardized text processing workflow, a wide range of high-performance similarity computation tools, compatibility with topic modeling techniques, and features a low coding threshold alongside high reproducibility.

To evaluate policy alignment and coherence within the governance action structure, this analysis quantified the textual consistency of environmental plans across diverse policy scopes (e.g., comprehensive vs. sectoral) and administrative levels (e.g., national vs. municipal). A total of 121 national-level documents and 8 local-level documents were utilized in the analysis. The process began with word segmentation and text cleaning using the jieba library in Python, followed by the conversion of the processed corpus into TF-IDF vectors through the TfidfVectorizer function from the scikit-learn library. Following this, a TF-IDF model was constructed and trained on the dataset. Cosine similarity calculations were then performed on the TF-IDF vectors of distinct plans to measure their textual resemblance. This metric captures the cosine of the angle between two vectors representing different plans, reflecting their content similarity independent of length or scale. Cosine similarity values range from 0 to 1, with higher values indicating stronger textual alignment. The mathematical formula underpinning this calculation is presented below.

$$vertical=\cos (A,B)={\displaystyle \frac{A\cdot B}{‖A‖\cdot ‖B‖}}={\displaystyle \frac{{\displaystyle \sum _{i=1}^n(A_i\cdot B_i)}}{\sqrt{{\displaystyle \sum _{i=1}^n{(A_i)}^2}}\sqrt{{\displaystyle \sum _{i=1}^n{(B_i)}^2}}}}$$

(7)

where n is the dimension of the vector and A_i is the value of the vector at the i-th position. The closer the cosine value is to 1, the more similar the two vectors are.

4. Findings: Analyzing the Paradigm Shift Through Planning Documents

Guided by the “Driving Forces–Elements–Structure” analytical framework, this study undertakes an empirical examination of environmental planning texts. The research endeavors to uncover the impelling factors and influence mechanisms and identify the persisting challenges that underpin the paradigm shift.

4.1. Coordination of Multi-Objective Conflicts

This section delves into the coordination mechanisms of multi-objective conflicts within the paradigm shift in environmental governance. Specifically, it investigates the evolution of value norms, the transformation of governance objectives, and the effectiveness of goal alignment across diverse administrative levels in the national, Shanghai municipal, and Yangpu District environmental planning documents. Additionally, the study identifies challenges such as the adaptability of indicator-based assessments to local practices.

From the 11th Five-Year Plan to the 14th Five-Year Plan, China’s environmental planning guiding ideology underwent a profound evolution, shifting from a “bottom-line guarantee” approach focused on resolving prominent environmental problems to the dual emphasis on “high-quality development” and “high-level protection” characterized by the promotion of ecological conservation and green development. This reflects the evolutionary trajectory of value norms. During the 11th Five-Year Plan period, the primary focus was on safeguarding basic environmental security through strengthened legal measures and intensive pollution control efforts. The 12th Five-Year Plan witnessed a strategic pivot towards promoting green development while ensuring social equity. The 13th Five-Year Plan further advanced this agenda by advocating for systemic governance that addressed both symptoms and root causes through systematic governance. Finally, the 14th Five-Year Plan introduced the modernization of harmony between humanity and nature. This transformation is realized through integrated strategies of carbon emission reduction, pollution control, green development promotion, and economic growth stimulation. Inspired by the philosophy of “community of life between humanity and nature”, the public has gradually evolved from passive recipients or beneficiaries of environmental governance to active contributors and participants. The vision of building a “Beautiful China” has emerged as a collective aspiration and conscious endeavor of the entire population. This evolution reflects a shift from pure pollution control to ecosystem restoration, and from end-of-pipe governance to source prevention, gradually alleviating the conflict between environmental protection and economic development objectives.

As shown in

Table 3. Topic names of China’s national environmental planning from 11th to 14th Five-Year Plan periods.

Period	Optimal Number of Topics	Topic Name
11th Five-Year Plan period	6	Solid waste and air pollution control; river basin water pollution control; rural environmental improvement; ecological function protection area; biological species resource protection; environmental technology management system
12th Five-Year Plan period	9	Safe treatment and disposal of solid waste; comprehensive control of air pollutants; prevention and control of groundwater pollution; water ecological restoration; river basin water pollution prevention and control; reduction in major pollutants; ecological protection and supervision; development of environmental protection industry; environmental management technology system
13th Five-Year Plan period	8	Solid waste pollution prevention and control; river basin water pollution prevention and control; comprehensive water environment governance; ecological space control; ecological economic belt development; regional green coordinated development; supply-side structural reform; green scientific and technological innovation
14th Five-Year Plan period	5	Comprehensive governance of river basin water environment; regional joint prevention and control; integrated protection and systematic governance of mountain, water, forest, farmland, grassland, and desert ecosystems; reform in ecological conservation; green technology innovation system

, the results of the LDA topic modeling reveal that early-stage environmental governance predominantly centered on pollutants, targeting tangible, materialized resources such as land, water, and air. End-of-pipe control measures, including total pollution control and emission permitting systems, were implemented to tackle overt environmental issues, enabling swift responses to ecological degradation. Over time, with the gradual improvement of air pollution control, solid waste management, and watershed water pollution conditions, the focus of environmental governance has expanded to encompass wetland conservation and ecological space control. Concepts such as integrated governance, systematic approaches, regional co-governance, and green technological innovation have emerged as pivotal trends in contemporary environmental governance [74,75]. Notably, since the 14th Five-Year Plan, environmental governance has extended to all facets of social and economic activities, driving the green transformation of ecological systems (e.g., watersheds), emerging green technologies (e.g., big data), natural capital (e.g., ecological products), and multifunctional spaces (e.g., mountains, rivers, forests, farmlands, lakes, grasslands, and deserts). This expansion has significantly enhanced both ecological environmental quality and social welfare. While the topic “reform in ecological conservation” conveys the authority of institutional change through the use of the term “reform,” the text lacks explicit reference to the conflicts of interest inherent in the reform process. This omission may obscure underlying power struggles and downplay the structural issue of insufficient public participation. Shanghai now incorporates multidimensional indicators, including carbon emission intensity, fine particulate matter (PM2.5) concentration, and the proportion of water bodies with good water quality. In terms of governance measures, it not only focuses on traditional pollution control but also places greater emphasis on the integrity and synergy of ecosystems, thus achieving a “beauty-oriented” upgrading of environmental governance.

In the environmental planning documents across the three administrative levels—the national level, Shanghai Municipality, and Yangpu District (see Figure 4)—the planning topics generally exhibit strong vertical consistency. While each level of planning inherits the objectives and requirements set by higher-level plans, they also make appropriate adjustments and expansions based on local contextual features. However, during the 14th Five-Year Plan period, the theme of the “Optimization of Industrial Space Layout” in Shanghai’s environmental planning exhibits a dual disconnect: it neither fully integrates the strategic guidance of the national plan nor provides a clear direction for localized implementation at the Yangpu District level. In terms of distributional justice, Shanghai’s resource allocation has failed to account for regional environmental costs and benefits and has not sufficiently aligned with the goal of the fair distribution of resources within cities as outlined in SDG 11. Consequently, areas such as Yangpu District bear disproportionately high transformation costs. Regarding procedural justice, the top-down decision-making process lacks meaningful participation from multiple stakeholders, resulting in one-size-fits-all policies. In terms of recognitional justice, inadequate attention has been paid to the geographic appropriateness of optimizing industrial spatial distribution. This gap conflicts with the requirement to integrate climate risks into planning under SDG 13, thereby weakening community recognition and constraining the potential for effective and just transition.

4.2. Adaptation of Governance Structure

This section examines the adaptation and implementation of governance structures amid the paradigm shift in environmental governance. Specifically, it investigates the interdepartmental coordination mechanisms in the formulation of environmental planning documents across national and local governmental tiers. The analysis further dissects the structural contradictions encountered by governance actors at various levels, focusing on both horizontal interdepartmental collaboration and vertical policy transmission.

The issuance of environmental planning documents typically takes three forms: documents issued by a single department, joint documents issued by multiple departments, or comprehensive documents issued by the State Council or local governments. Meanwhile, diverse actors such as market entities and the public participate in the protection of environmental and natural resources and pollution control through formal or informal institutional mechanisms. As a research simplification, this section restricts its focus to the government as the principal actor in environmental planning formulation and implementation. As illustrated in Figure 5 and

Table 4. Overall network analysis of China’s national government departments during 11th to 14th Five-Year Plan periods.

Period	Number of Network Relationships	Network Density	Clustering Coefficient	Percentage of Co-Issued Documents
11th Five-Year Plan period	18	0.43	1.19	28.57%
12th Five-Year Plan period	98	0.41	1.49	41.38%
13th Five-Year Plan period	174	0.46	1.16	52.00%
14th Five-Year Plan period	362	0.48	1.25	62.50%

, from the 11th Five-Year Plan to the 14th Five-Year Plan period, the number of departments involved in issuing environmental planning documents has been increasing, and the proportion of jointly issued documents has steadily grown. The National Development and Reform Commission (NDRC) and the Ministry of Ecology and Environment (MEE) have consistently served as the core “brokers” in the network, while since the 14th Five-Year Plan period, the frequency of participation by the Ministry of Natural Resources (MNR) has significantly elevated. However, despite the expansion of the planning network and enhanced data sharing enabled by technological advancements, the overall network density and clustering coefficient have exhibited minimal variation. This suggests that the exchange of resources and information and the extension of connections between departments have not become more efficient. For example, although the Ministry of Natural Resources has enhanced its collaboration with the Ministry of Ecology and Environment in areas such as ecological red line delineation and land spatial planning, and occupies a central position within the inter-ministerial cooperation network, it faces structural challenges. These include conflicting functions with the Ministry of Ecology and Environment regarding land approval authority, as well as insufficient spatial data sharing. Such issues reflect the Ministry of Natural Resources’ conflicting dual roles as both policy enforcer and regulator, alongside shortcomings in procedural coordination. The issue of sectoral fighting is frequently obscured by the discourse of “collaborative governance”, yet, in practice, management remains predominantly controlled by hierarchical authority, further marginalizing market and social actors.

As depicted in Figure 6, the practice of jointly issuing environmental planning documents in Shanghai only began during the 13th Five-Year Plan period, with a notable surge in the number of planning documents during the 14th Five-Year Plan period. The departments responsible for issuing these documents in Shanghai largely mirror their national-level counterparts, underscoring the city’s adherence to the action framework set by higher-level authorities when formulating its plans. Many important planning documents are issued independently by the Shanghai Municipal People’s Government. While this centralized issuance model can safeguard the coherence and authority of the planning system, it may also lead to insufficient interdepartmental collaboration and communication, potentially undermining the effectiveness of collaborative governance. Before the 14th Five-Year Plan period, local environmental protection agencies in Shanghai were entirely under the jurisdiction of the local government regarding financial appropriations, personnel management, and organizational structure. This hierarchical relationship frequently resulted in environmental governance being subordinated to the strategic priority of economic development. Since 2016, the Central Committee of the Communist Party of China and the State Council have launched reforms to implement a vertical management system for the monitoring, supervision, and law enforcement functions of ecological and environmental agencies at the sub-provincial level. This reform requires local environmental functional departments to not only effectively transmit policies vertically but also to enhance horizontal interdepartmental coordination. While the institutional standing of the Shanghai Ecological Environment Bureau has been notably elevated, the reform has also imposed stricter requirements on its policy adaptability.

In the environmental planning documents issued by Yangpu District, the District People’s Government assumes a central governing role, with no established crossdepartmental joint issuance mechanism. The allocation of environmental governance authority is closely related not only to administrative levels but also to contextual factors such as regional area, population density, and economic scale. These variables collectively shape local governments’ resource allocation patterns and governance strategies in environmental planning and management. Notably, during the 11th Five-Year Plan period, the World Expo, a major international event, served as a catalyst for Yangpu District to implement targeted environmental remediation initiatives.

4.3. Asynchrony of Transformation

This section investigates whether the action structure, institutional structure, and disciplinary structure within the paradigm shift in environmental governance have undergone a synchronized transformation. It aims to identify and analyze the structural characteristics and key bottlenecks in the transformation dynamics across these three dimensions. In the context of the action structure, “horizontal translation” denotes the process of integrating national-level macro-strategic planning with the practical development needs of specific sectors to formulate special plans that align with national strategic orientations and operationally feasibility. “Vertical translation” refers to the process by which local governments, within the bureaucratic system, adapt national strategic plans with local realities, emphasizing regional characteristics and needs during policy implementation, thereby forming planning texts with local adaptability [27].

As illustrated in Figure 7, based on scikit-learn, the cosine similarity comparison of other national environmental special plans in the same period with the corresponding contents in the comprehensive environmental plans of the same period (notably, as the full text of the 14th Five-Year Plan for National Ecological and Environmental Protection has not yet been published, this study uses relevant sections from the Outline of the 14th Five-Year Plan for National Economic and Social Development and Long-Term Goals for 2035 as a proxy for analysis) showed that the average cosine similarity of national-level special planning texts remains around 0.5. From the 11th to 13th Five-Year Plan periods, the average cosine similarity between the national environmental special plans and national comprehensive environmental plans exhibited a gradual upward trend, but this reversed during the 14th Five-Year Plan period. This is because the “pollution control” paradigm primarily emphasizes the consistency of pollution reduction targets and measures, along with centralized actions, which exhibit strong task-oriented and emergency response characteristics. However, this approach is associated with high governance costs and unstable results. In contrast, the “ecological conservation” paradigm prioritizes customized governance that is tailored to the unique characteristics and needs of different regions and sectors. Furthermore, by comparing the cosine similarity between comprehensive environmental plans across diverse administrative levels within the same period, the planning text similarity between Shanghai and the national level, as well as between Yangpu District and Shanghai, has generally remained above 0.8. This indicates that subnational governments place significant emphasis on aligning with higher-level strategies. Notably, during the 14th Five-Year Plan period, the similarity of Yangpu District’s planning texts to higher-level documents further increased, suggesting that the concept of “ecological conservation” has not yet deeply penetrated the district level, revealing a certain lag, and that the capacity of district-level governments to formulate environmental planning according to local conditions still needs to be improved. An analysis of the “Safeguards” in environmental planning texts across different periods reveals an evolution in public participation. During the 11th Five-Year Plan period, public involvement was relatively rudimentary, primarily focused on raising awareness and encouraging behavioral changes through environmental information disclosure, with oversight largely dependent on government administrative supervision. In the subsequent 12th and 13th Five-Year Plan periods, the public’s rights to information, participation, and supervision were progressively strengthened, alongside the increased promotion of social organization involvement. By the 14th Five-Year Plan period, public participation had become a recognized and integral component of environmental governance. Governments and enterprises at various levels were explicitly mandated to actively advance the socialization and publicization of environmental protection policies. Additionally, mechanisms for environmental public interest litigation and the protection of environmental rights were further developed. However, these initiatives remain in the process of deepening and have yet to be fully realized.

In terms of the institutional structure, an analysis of the “Safeguards” sections in environmental planning texts across different periods reveals that during the 11th Five-Year Plan period, environmental governance was predominantly government-driven, relying on traditional instruments such as environmental law enforcement, fiscal subsidies, and administrative control, with market mechanisms playing a marginal role. Beginning in the 12th Five-Year Plan period, the pollution discharge fee system was strengthened, and market-oriented tools like environmental taxation and green credit were actively promoted. Concurrently, an enterprise environmental credit evaluation system was established to increase the economic costs of pollution enterprises. During the 13th Five-Year Plan period, policies such as carbon emission trading pilots, ecological compensation mechanisms, and green finance were deepened, while the third-party governance of environmental pollution was promoted. These measures encouraged enterprises and social capital to pursue green transformation through market mechanisms. The 14th Five-Year Plan period further advanced the market-oriented reform of resource and environmental elements, expanding the scope of market-based allocation of carbon emission rights, energy use rights, water use rights, and pollution discharge rights. Also, efforts were intensified to strengthen rule-of-law governance, environmental credit regulation, and the system of paid use for natural resources. Nevertheless, these reforms remain ongoing and require sustained implementation.

An analysis of the disciplinary structure, based on the examination of planning texts and relevant literature, reveals that during the 11th Five-Year Plan period, environmental governance relied predominantly on environmental science and engineering technology, resulting in relatively single-disciplinary support. In the 12th Five-Year Plan period, the gradual introduction of market-oriented regulatory instruments such as pollution discharge fees, environmental taxes, and green credit policies facilitated the integration of environmental economics into the disciplinary framework of environmental governance. In the 13th Five-Year Plan period, the significance of ecology in environmental governance was markedly elevated, with concepts like ecological compensation and ecological conservation red lines becoming integral components of the policy framework. Environmental economics witnessed further maturation, as market mechanisms like carbon trading and green finance were actively promoted. Concurrently, sociology and environmental law expanded their influence, accelerating the legalization of ecological civilization initiatives. This era also marked a phase of adjustment and exploration in the restructuring of the disciplinary foundation underpinning environmental governance. During the 14th Five-Year Plan period, technologies including artificial intelligence, big data, and remote sensing have been gradually integrated into environmental planning. Meanwhile, the objectives of carbon peaking and carbon neutrality have propelled climate and energy economics to emerge as critical supporting disciplines. However, the application of disciplines such as paid use for natural resources, social governance, and smart technology in environmental governance has not yet solidified into a stable theoretical system. Existing research on environmental planning faces several bottlenecks: the theoretical foundation remains underdeveloped, the core technologies are largely in a catch-up phase, the talent cultivation system is inadequate, and the disciplinary positioning remains ambiguous [48,76,77,78]. As a result, the disciplinary foundations remain incompletely established.

5. Discussion and Suggestions

5.1. Results

Our findings uncover a distinct pattern of asynchrony in the environmental governance paradigm shift under investigation. Value norms, representing the conceptual dimension, transform earliest. This shift gained substantial traction during the 13th Five-Year Plan period, particularly with the consolidation of high-level national philosophies prioritizing ecological imperatives. These evolving values, anchored in concepts such as harmony between humanity and nature and ambitious national environmental objectives, necessitate translating abstract ideals into measurable governance targets. This process is critical for resolving inherent conflicts among multiple policy goals.

Guided by new value norms and supported by technological advancements (e.g., digital tools) and critical events (e.g., institutional reforms, major international events hosted locally), the constituent elements of governance, including actors, objects, and instruments, all undergo expansion and reorganization. Digital technologies, for instance, help break down traditional data silos, integrating information and knowledge into the governance system. Exogenous drivers like departmental reforms reshape actor responsibilities and resource allocation, disrupting path dependencies. This “element expansion,” driven by these forces, fosters optimized combinations that, in turn, influence governance “structure” changes.

Consequently, transformations in the governance structure (encompassing institutional, action, and disciplinary structures) typically lag behind the evolution of value norms. Our analysis indicates that the 13th Five-Year Plan period served as a transition phase for structural change, with the 14th Five-Year Plan period emerging as a more profound turning point. The action structure, reflecting how plans are translated horizontally (across sectors) and vertically (across administrative levels), also shows a shift, particularly accelerating in the 14th Five-Year Plan as the central government balances strategic guidance with latitude for local innovation. However, institutional arrangements, requiring deep legal and market reforms, evolve at a slower pace and remain in a phase of “deep exploration”. The disciplinary foundation, which requires time to integrate practical experience and guide institutional development, lags the furthest, lacking the systematic theoretical framework needed to fully underpin innovative practices.

At the core of the asynchrony of transformation lies a fundamental tension between authoritarian institutional structures and the current stage of development. On one hand, the division of powers and responsibilities, along with the polarization of sectoral interests within a hierarchical system, has intensified resistance to transformation through mechanisms of “fragmented authoritarianism”. On the other hand, the central government has actively sought to reconstruct governance synergies through instruments such as policy experimentation and digital governance, demonstrating the distinctive capacity of authoritarian regimes to drive large-scale institutional change. This dynamic interplay among driving forces, elements, and structure, and the resulting asynchronous transformation, aligns precisely with our analytical framework, as depicted in Figure 8. This conceptual model helps visualize how value norms act as catalysts for change, instigating the reorganization of governance elements, which in turn reshapes governance structures in an uneven manner. The observed asynchrony connects to broader theoretical concepts: the role of problem and political streams in initiating policy agendas [79]; the reconstruction of governance networks through resource sharing among actors [80]; and the gradual, path-dependent nature of institutional adaptation [81]. In contrast to the European Union’s Green New Deal, which ensures policy coherence through legislation and supranational institutions while emphasizing multi-stakeholder participation, China’s approach is characterized by centralized policymaking. It relies on vertical administrative pressure to gradually reconcile economic growth with ecological protection within a hierarchical structure of power and responsibility.

5.2. Existing Problems

Despite the strategic shift at the top level, substantial practical hurdles hinder the effective implementation of environmental governance initiatives. Current top-down performance evaluation and accountability mechanisms often lack effective positive incentives. Local governments, constrained by inflexible indicator assessments and facing multiple overlapping tasks with limited resources, struggle to generate sustained momentum for collaborative improvement. This gives rise to an “involution” scenario, where efforts are concentrated on short-term, superficial measures to meet upper-level targets, resulting in formalism and inefficiency at the grassroots level. In an effort to rapidly comply with the requirements of higher-level inspections, local governments often resort to one-size-fits-all measures such as blanket shutdowns and production restrictions on enterprises. While these actions may temporarily satisfy assessment criteria, pollution levels tend to rebound significantly once production resumes. In some cases, grassroots officials, under pressure to meet performance targets, have resorted to fabricating inspection records using outdated photographs or repeatedly reporting the same locations for inspection. As a result, on-the-ground inspections and substantive problem rectification remain inadequate, failing to address the underlying environmental issues.

Furthermore, national ecological and environmental governance functions often suffer from significant fragmentation, with responsibilities divided among different government bodies focusing on pollution control, ecological protection, and overall coordination. While such specialization may enhance problem-specific efficacy, these functional divisions can fragment governance authority, conflicting with the systemic demands of holistic ecological governance [82,83].

The observed asynchrony highlights a critical disconnect in the interactive chain linking concepts (value norms), theory (disciplinary foundation), and practice (action and institutional structures). Value norms change first because they serve the construction of state legitimacy, guiding practical actions (action structure) which in turn gradually shape institutional arrangements; ideally, these experiences should inform theoretical development (disciplinary structure). The lag in institutional transformation reflects the resistance of entrenched interests, while the delay in disciplinary restructuring stems from a deeper systemic disconnect between the disciplinary framework shaped by the industrial era and the evolving modes of knowledge production in the digital age. First, the pace of knowledge generation often lags behind the rapidly changing demands of governance practice. Second, the traditional tree-like disciplinary classification struggles to accommodate the growing need for non-linear, crossdisciplinary research. Third, the resource allocation model dominated by administrative authority constrains the organic emergence of interdisciplinary academic communities. Currently, however, a persistent gap remains: “actions running ahead, institutions watching, and theory lagging”. Abstract concepts are not effectively translated into operational variables by theory, leaving institutional design dependent on broad value statements. Conversely, fragmented practical experiences struggle to coalesce into robust explanatory paradigms due to a lack of theoretical integration, leaving the disciplinary foundation unanchored.

5.3. Policy Recommendations

To address the challenges of limited local autonomy, insufficient innovation, and weak public participation, a management model that balances central direction with local autonomy, clarifies responsibilities, and delegates appropriate authority is needed to refine top-down coordination, manage transaction costs effectively and overcome the systemic rigidity associated with authoritarian environmental governance. This aligns with the principle of common but differentiated responsibilities outlined in SDG 13. For instance, initiatives such as the establishment of pilot ecological compensation zones have encouraged local innovation in environmental governance while maintaining alignment with national ecological protection objectives. Simultaneously, efforts to promote environmental awareness have been strengthened, alongside expanding the public’s rights to access environmental information, participate in decision-making, and exercise oversight. These rights include participation in hearings, inquiries, and plan revisions, as well as the ability to report environmental violations. These measures contribute to the realization of the right to public participation enshrined in SDG 11, thereby enhancing both public awareness and engagement in environmental protection efforts.

To address the inefficiency of intersectoral cooperation, closer coordination mechanisms (e.g., joint meetings, enforcement) are critical to ensuring consensus and information sharing across departments on fundamental aspects like planning scope, objectives, and monitoring methods. Additionally, joint training programs should be provided for officials involved in ecological protection, economic development, and data management to enhance cross-sectoral understanding and cooperation.

To address the lagging development of the disciplinary foundation, reconstructing the interdisciplinary knowledge system for environmental governance, including its academic foundations, disciplinary structure, and discourse practices, and cultivating crossdisciplinary talent is essential for closing the gap between theory and practice. Emphasis should also be placed on the localization of knowledge systems. These initiatives would enable theoretical advancements to guide governance practices while allowing real-world experiences to inform and fortify disciplinary growth.

Therefore, to surmount transformation bottlenecks, strategic efforts should prioritize strengthening top-down coordination, enhancing multi-actor collaboration mechanisms, and fostering a more robust disciplinary foundation. These approaches offer valuable insights for socialist-system countries, such as Vietnam, seeking to implement environmental governance reforms. For other countries embarking on analogous large-scale governance shifts, these insights underscore the importance of pursuing adaptive governance reforms through gradual element optimization, grounding transformations in local conditions while engaging with global norms, and avoiding over-reliance on purely technological or institutional fixes.

5.4. Limitations and Prospects

However, we must acknowledge several limitations. Although planning texts articulate policy intentions, a notable gap often persists between these intentions and implementation outcomes, thus rendering it challenging to predict governance effectiveness solely from textual analysis. Social Network Analysis based on joint document issuance captures formal relationships but may not fully reflect the substantive strength or effectiveness of interdepartmental collaboration. Additionally, deploying computational text analysis methods like LDA and TF-IDF/scikit-learn in the Chinese language context presents challenges, including potential semantic instability and ambiguity.

Future research should seek to address these limitations. Developing a unified measurement framework for the “ecological conservation” paradigm through crossnational comparative studies and aligning benchmarks with SDGs would add significant value. Local environmental quality data should be integrated to evaluate the actual outcomes of environmental governance. Additionally, emerging technologies such as natural language processing (NLP) can be employed to identify cases of informal collaboration by combining semi-structured interviews and participatory observation within the framework of SNA. Methodologically, by integrating LDA and scikit-learn models with semantic web frameworks, a hybrid BERT-LDA model could be employed to unlock deeper semantic insights from policy texts.

6. Conclusions

This study examined the paradigm shift in China’s environmental governance through an analysis of environmental planning documents from 2006 to 2025. Our analysis yields several key academic findings:

(1)

Environmental planning documents serve as a valuable analytical lens, offering unique insights into the trajectory and nature of national environmental governance paradigm shifts.

(2)

The transition from “pollution control” to “ecological conservation” demonstrates pronounced asynchrony: changes in value norms lead, followed by transformations in governance structures, while action logic and the underlying disciplinary foundation exhibit considerable lag. The pursuit of “beauty” (ecological conservation) is driving the reconstruction of the knowledge system, but this concept remains incompletely theorized within the governance framework.

(3)

Notable challenges endure at the local level, including an absence of explicit criteria for balancing higher-level mandates with local planning autonomy, and institutional ambiguity in planning authority allocation, both of which impede efficient resource distribution and implementation effectiveness.

Building on these findings, this study analyzes China’s environmental governance within the global framework of SDG 11 and SDG 13, adopting a comprehensive and comparable approach in order to uncover the complex interplay of driving forces, elements, and structures involved in the paradigm shifts in environmental governance. It deepens the understanding of the asynchrony of this transformation and lays a foundation for future comparative research on governance paradigm shifts across different national contexts.

Ultimately, navigating the transition towards sustainable development necessitates a profound grasp of the dynamic interplay between driving forces, elements, and structures in environmental governance. Systematically addressing the resistance points and releasing the synergistic potential inherent in this transformation process are pivotal for establishing effective action coordinates for a sustainable future.