Editorial on Research on Promoting the Social Sustainability of Urban Neighbourhoods

1. Introduction

The neighbourhood is the smallest unit at which a city can be said to work, or to fail, for the people who live in it. It is where housing stock, public space, service provision and social relations meet, and it is the scale at which residents experience decisions that sustainability policy has taken on their behalf. Sustainable Development Goal 11 commits its signatories to cities that are inclusive, safe, resilient and sustainable by 2030; three of those four adjectives are irreducibly social.

Effort nonetheless remains weighted towards the technical: energy intensity per square metre, embodied carbon, retrofit depth, and green coverage. Those measures matter, and nothing here disputes their importance. The claim behind this collection is narrower. A neighbourhood can improve on every technical indicator and still become a worse place to live in, if the improvement displaces the households who most needed it, if the renewed space quietly excludes older or poorer residents, or if the arrangement that delivered it gave residents no standing in the decision.

Social sustainability carries a definitional problem. The term accommodates cohesion, equity, safety, participation, wellbeing, identity, and access, and elastic concepts are easy to endorse and hard to falsify. The productive response is not to seek one authoritative definition but to specify the concept for a setting and measure it there, accepting that what suits a nursing home does not suit an informal hillside settlement. Readers should expect the thirteen contributions to disagree about what the term denotes; six never use it at all.

Twelve research articles and one review were published, reporting on cases in Saudi Arabia, Qatar and the historic cities of the Levant and the Maghreb, and in China, South Korea, Slovenia, Sweden, Ecuador, Peru, Poland and Egypt. They draw on resident surveys and structural equation modelling, evolutionary game theory and agent-based simulation, spatial and network analysis, historical morphological inquiry, machine learning, and design practice. Three concerns organise them: how social sustainability might be measured, how it is governed, and how it is produced in everyday practice. Section 5 reads across the collection, as an editor’s reading of a self-selected set rather than a systematic review.

2. Measuring Social Sustainability Where It Lives

Three contributions take the measurement problem head-on. Skalicky Klemenčič and Žegarac Leskovar (Contribution 5) apply two tools they developed, Safe and Connected and Well-being and Integration, to four nursing homes: three Slovenian cases from the 1960s, the late 1970s and 2021, and a Swedish benchmark that scores highest. On their combined scale, quality rises from 67 per cent to 81 per cent and then falls to 50 per cent. Two qualifications help in reading that trajectory, and neither diminishes the study. Almost all of the decline sits in the socio-spatial tool, while the safety tool barely separates the cases. And the newest case is the only privately operated one, which is the authors’ own explanation: they attribute the decline not to design fashion but to a shift from public to market provision. With one building per cell, age and ownership cannot be told apart. That such a decline exists and can be detected is established; which cause produced it is not.

Kim and Sim (Contribution 11) surveyed 569 users of ten Urban Regeneration Community Facilities in Daegu and asked whether age integration is associated with residential satisfaction. It is, in every specification, though the coefficient falls from 0.415 to 0.119 once community wellbeing enters, and unadjusted correlations do not differ significantly across three age cohorts. Age integration here is a thirteen-item measure of perceived intergenerational contact and openness of participation, not a demographic mix of who lives where. The respondents are people who have used these facilities, not a probability sample of the surrounding population. Within those bounds the study converts a planning virtue usually asserted into a construct that can be measured and contested.

Kang and Shin (Contribution 10) reconceptualise the university Fab Lab as a commons, reorganising Ostrom’s institutional analysis and development framework into resource, actor and community governance policy domains and testing them against the Idea Factory at Seoul National University. The setting is a campus workshop rather than a residential neighbourhood, and transfer to housing contexts is untested. What travels beyond the case is the counter-finding: demand for continuous access collides with institutional liability rules, and a fully autonomous commons is inevitably constrained by the legal framework of its host. Rules sustain a shared space as much as its floor plan does, but the decisive rules are often not the ones the community writes.

3. Governing the Transition: Who Pays, and Who Decides

A second group asks how the social dimension is shaped by policy instruments. Zhang and Xin (Contribution 13) build an agent-based model of a heterogeneous neighbourhood and test four instruments—incentives, feedback, participation, and compliance—alone and in selected combinations, against an indicator structure covering emissions, household energy burden and its dispersion, adoption and compliance dynamics, and cumulative net public cost, with an exogenous price shock added for resilience. Participation-centred mixes generate the strongest early uptake pulse, and the incentive-and-participation mix attains the highest end-of-horizon adoption and lowest emissions in the baseline runs; participation alone finishes below incentives alone. Compliance dampens voluntary uptake and can register net fiscal revenue where fine income exceeds enforcement outlays, though the authors decline to read that accounting result as welfare gain. Two properties govern how far the numbers travel, and the authors supply both: the model is parameterised in stylised rather than city-calibrated units, and a two-hundred-draw sensitivity analysis shows distributional rankings to be comparatively more sensitive than adoption and emissions rankings.

Zhao, Ma and I (Contribution 3) reach a compatible conclusion from a tripartite evolutionary game of consumers, enterprises and government, calibrated on Shenzhen and Beijing data. Incentives shift consumer behaviour strongly but have limited purchase on enterprises, whose strategies evolve quickly. Market stability can be maintained through staged subsidy reduction provided technology-driven cost decline keeps pace; where it does not, reductions generate negative incentives and market regression. The prescription is explicitly sequential: consumer subsidies first, then technology cost-reduction alliances as enterprise subsidies are phased out, and consumer support tapered last. The model assumes fully rational actors, covers three stakeholder groups only, and leaves equilibrium analysis to further work; the paper says so. Contribution 13 arrives independently at a three-phase sequence of its own, beginning with participation and means-tested incentives and adding a compliance backstop last. Two unrelated formalisms converge on the claim that instrument order matters as much as instrument choice.

Marey, Kozma and Szabó (Contribution 12), the collection’s one review, works a layer below policy. Their framework maps twenty-seven green concrete sustainability indicators against the seventeen Sustainable Development Goals for the Egyptian residential sector, scoring each pairing from a documentary evidence base of peer-reviewed studies, Egyptian codes and the national rating system, with values normalised before scoring. The strongest contributions are to responsible consumption, sustainable cities and clean water. Because the scores summarise an evidence base rather than measured project outcomes, they indicate where the potential lies rather than what has been realised. The authors are candid that the social sustainability consequences for residential buildings require further investigation, which marks the boundary this Special Issue was convened to work on.

4. Form, Memory and Everyday Practice

The largest group attends to the fabric of the neighbourhood and to what residents do with it. Hammad, Li and Vrcelj (Contribution 1) combine content analysis of historical sources with historical inquiry, tracing privacy, spatial hierarchy and community resilience from the enclosed quarters of Damascus, Aleppo, Cairo, Tunis, Fes and Rabat to two contemporary Gulf gated developments. The argument is more nuanced than what a summary can easily convey: the contemporary form is not presented as a lineal descendant of the historic one, and the historic city is not presented as egalitarian, since it too arranged residents by class and confession. The claim is that the older form graded social distance while keeping the quarter permeable, whereas the contemporary form makes exclusivity the design objective and disrupts urban connectivity. Walls are not the variable; what the walls are for is. The authors note, and the point recurs below, that a study built on secondary sources cannot include residents’ firsthand perspectives.

Cardenas-Condori and colleagues (Contribution 9) triangulate cartography, a visual study of 548 of a settlement’s 589 lots, non-participatory observation and 72 semi-structured interviews in an informal hillside settlement in Arequipa. Stairs, platforms, retaining walls and setbacks built for stabilisation and access are progressively appropriated as semi-communal space supporting encounter and collective identity, and the resulting sociability, as the authors put it, challenges the public and private dichotomy on which formal planning rests. The paper refuses a celebratory reading of its own material: these spaces evidence an unequal distribution of urban resources, improvised stairs and unreinforced walls on unstable slopes below a volcano are sources of risk, and half the dwellings lack full basic services. Alongside that, it makes a comparison rarely made between formal parks that stand unused because they are disconnected from daily life and self-produced space that thrives because it is attached to housing and cared for by its occupants.

Vidal-Domper and co-authors (Contribution 6) subject Jacobs’s eyes on the street thesis to empirical test in Quito, operationalising density, land use mixture, contact opportunity, and accessibility through nine indicators. The design is worth stating precisely, because a composition model is easily mistaken for a volume model: the estimation asks whether a recorded robbery was committed against a person rather than against a vehicle or goods, so the results describe what kind of robbery occurs where, not how much. Six of the nine indicators are significant. A richer mix of commercial and civic facilities and higher population density are associated with less person-directed robbery; a more even residential to non-residential balance, larger blocks, proximity to transit and the presence of street lighting with more. The authors read the lighting result as lighting following pedestrian activity rather than causing crime, call the study exploratory, note the absence of socio-demographic and perceptual controls, and warn against treating Jacobs as universally valid.

Shen, Tan and Li (Contribution 7) examine sensory stimulation in community public art, building a stimulus-organism-response framework from sensory marketing theory and testing it with structural equation modelling on 332 valid questionnaires, with eight installations in central Shanghai as case references. Participation willingness is driven directly by perceptual response, and cognitive memory is the strongest single path, shaped more by olfactory than visual stimulus. Observations and interviews further suggest that age groups differ in how they participate, which is reported qualitatively rather than tested. The paper does what few sensory-design studies attempt: it puts a mechanism, from stimulus through cognitive memory to participation, into a form carrying estimated path weights, and therefore into a form that can be disputed.

Chen and colleagues (Contribution 4) pursue heritage-consistent renewal by other means, fine-tuning a diffusion model with low-rank adaptation on 150 typology-guided image and label pairs from thirty buildings in Krakow’s Wesoła district, and reporting a 24.6 per cent improvement in Fréchet inception distance over the base model, corroborated by 68 expert evaluators. The work is confined to two-dimensional facades, and the trained model has been released publicly, which is a service to reproducibility. The paper’s own discussion of authenticity and of the irreducibility of human judgement anticipates a question that belongs to the field rather than to this study; Section 6 returns to it.

Chu, Liu, and Li (Contribution 8) combine morphological spatial pattern analysis, connectivity indices, field surveys with analytic hierarchy process weighting, minimum cumulative resistance modelling, and GIS overlay to construct composite ecological, heritage, and recreation corridors across the six provincial-level regions of the Qinling–Daba Mountains, yielding 69 multifunctional nodes and 141 corridors in four node categories and three corridor types. It is the contribution that organises its whole design around Goal 11, and it is admirably self-critical: the authors record that their resistance weights rested on expert judgement, which, in their phrase, reduces the social legitimacy of the model.

Huang and colleagues (Contribution 2) report a design-practice case at the Wuyistar tea garden in Wuyishan, Fujian, built on field survey, GIS and plant resource inventory and organised around restoring mountains and rivers, rereading cultural context, ordering style and appearance, and improving function. Only part of the scheme was built before the pandemic halted construction, but the paper reports post-completion use through study tours, tea-culture events, and enterprise outcomes, and sets the design against the reserve’s community participation and collective forest governance arrangements, which it finds largely passive and organised as though the community were a counterparty. It supplies what the measurement studies do not: an operative design repertoire, and an account of the institutional conditions under which it is deployed.

5. Reading Across the Collection

Five observations follow, and four concern what is missing. This is not a complaint about the contributions, which are individually careful. It describes where a field stands when its central concept has not yet been made comparable across studies.

First, newer is not automatically better, but this collection cannot yet show that it is worse. Contribution 5 finds that the most recent nursing home scores below a building half a century older, with the confounds noted above. Contribution 13 finds, in its baseline runs, that the mix maximising adoption and minimising emissions carries slightly higher burden inequality than no policy at all, yet under the same paper’s sensitivity analysis that mix is the one most often ranked best on inequality. The proposition that technical optimisation can worsen distribution deserves to be taken seriously, and neither result settles it. Saying so seems more useful than converting two suggestive findings into one conclusion.

Second, participation is this collection’s most confidently asserted and least directly measured commitment. Contribution 13 gives participation a diffusion signature distinct from the other three instruments, but that signature follows from a coordination switch the model wires to participation alone: the simulation shows what happens if participation works that way, not that it does. Contribution 10 shows where such arrangements meet institutional limits, in a campus rather than a residential setting. No contribution measures the effect of participation in an actual neighbourhood. The corollary is sharper still. Most contributions derive their evidence without asking residents anything: the nursing homes and the corridor weights are scored by experts, the Fab Lab from administrative records and management, the green concrete pairings from documents, the gated communities from secondary sources, and the two formal models, my own co-authored paper included, from parameter assumptions. Contributions 1 and 8 name the gap themselves, and Contribution 8 puts a price on it, in reduced social legitimacy. A field that recommends participation should be readier to practise it inside its own instruments.

Third, distributional outcomes are almost absent. Only Contribution 13 reports the dispersion of a welfare-relevant quantity, and it is a simulation in which mean energy burden barely moves across scenarios. Contribution 11 reports that income predicts satisfaction while age does not, which is a hint of heterogeneity it does not pursue. Studies reporting only average improvement cannot support claims about social sustainability, because the concept concerns distribution at least as much as level; the collection makes that case by omission rather than demonstration.

Fourth, and most consequentially, not one of the thirteen contributions measures displacement. Housing cost, rent, tenure security and residential mobility appear nowhere as variables. This matters because the first target under Goal 11 concerns adequate, safe and affordable housing and security of tenure, and because displacement is the outcome with which this editorial opened. A collection assembled under the banner of socially sustainable neighbourhoods has produced no evidence about who leaves. That absence also explains something usually blamed on measurement technique: neighbourhood-scale social indicators do not connect to city-level reporting against Goal 11, partly because they are not measuring what Goal 11 asks about.

Fifth, the collection does not share a concept. Six contributions never use the phrase social sustainability; among those that do, the referent ranges from crisis resilience in a care home, to burden dispersion in a retrofit programme, to corridor accessibility across six provinces. Calls for the field to converge on a definition have been issued regularly for two decades and have not been answered, and this Special Issue is evidence of that failure rather than a remedy for it. A more modest request may fare better, because every contribution here could meet it: that each study state, in one sentence, whose welfare its claim is about. For the second edition I intend to ask authors for a four-field statement covering the population studied, the dimensions measured, the distributional statistics reported if any, and the scope of the claim, and to check it at submission. A standard with a gate behind it has a different history from a standard without one.

6. Conclusions and Outlook

These contributions support a modest but consequential proposition: social sustainability at neighbourhood scale is neither a by-product of good buildings nor a residual to be handled once the technical work is done. It is an outcome that has to be specified, measured, governed, and, when the evidence says so, revised. Read together, they also show how much of the necessary measurement has not been attempted.

Several directions follow. Almost all the empirical evidence here is cross-sectional or single-period, while the questions that matter most are longitudinal: whether age-integrated regeneration sustains satisfaction over a decade, whether participatory retrofit coalitions survive the tapering of subsidy. The cost of participation, in resident time and administrative capacity, is almost never priced, even by studies that recommend it. Transferability of assessment frameworks across institutional contexts is more often programmed as future work than executed. And the arrival of generative methods in heritage renewal, now with publicly released models, raises a question of authorship that the field has the tools to formulate but has not joined to its participation literature: who is authorised to decide which typology a neighbourhood is given back.

A second edition of this Special Issue is open for submissions until 31 October 2026. Four kinds of contributions would be particularly welcome: studies that track who leaves a neighbourhood and at what housing cost; studies that follow neighbourhoods over time; studies that report distributional statistics rather than means alone; and studies that put residents inside the measuring instrument rather than only inside the case.