A plan is less a bounded document than a temporary stabilisation inside a much longer chain of selective continuities. Long before a proposal acquires a statutory name, history has already distributed capacities: ownership forms, administrative routines, legal competences, professional vocabularies and technical expectations sediment into a field in which some futures arrive as obvious and others require exceptional labour merely to become visible. Sorensen’s historical institutionalism gives this condition temporal thickness. Path dependence is not a claim that the past mechanically determines the present; it identifies positive feedback through which earlier institutional choices alter the cost and probability of later moves. Lindblom adds a more intimate scale of decision.
Because comprehensive search is impossible, actors compare only a limited set of alternatives, often those made familiar by prior controversies and experience. The feasible field is therefore doubly selective: institutions pre-structure what can plausibly be proposed, while practical reasoning narrows what can actually be compared. Forester then shows that this narrowing is never only cognitive. Information enters through asymmetrical communicative situations in which agenda control, technical language, timing and access determine whose knowledge can alter the problem definition. Innes and Booher complicate this further by demonstrating that possibility can also expand relationally. Under conditions of diversity, interdependence and authentic dialogue, networks can generate adaptive innovations that no participant possessed beforehand. Pre-formal planning is thus neither a neutral research phase nor an informal prelude. It is a production environment in which history, evidence, power and negotiation continuously alter the set of credible futures. Digital tools intensify this condition. Vonk and Geertman show that planning support systems have epistemic force only when they fit organisational tasks and learning practices; Potts and Milz demonstrate that even comparable technological resources are taken up differently because professional cultures define what counts as useful, modern, efficient or legitimate; Son and colleagues reveal the rapidly expanding algorithmic repertoire through which urban conditions are classified, predicted and rendered comparable. Taken together, these sources displace the familiar question of whether technology makes planning more rational. A model is never merely added to an existing process. By deciding what can be represented, parameterised and searched, it reorganises the field that decision-makers encounter. AI may enlarge the number of calculable alternatives while narrowing attention to attributes that can be formalised. A collaborative workshop may introduce situated knowledge excluded by a dataset while simultaneously privileging those able to participate. A historical archive may reveal that a supposedly technical constraint is the residue of a political settlement. What precedes the plan is therefore a continuous struggle over addressability: which pasts, risks, actors, scenarios and spatial relations can be called into the present as actionable evidence. The plan appears at one moment as a settlement among these heterogeneous operations, but it never abolishes them. It passes them forward in compressed form, carrying some assumptions explicitly, others tacitly, and still others already forgotten. The decisive test begins when this stabilisation starts moving again. Li and colleagues show why implementation cannot be evaluated simply by overlaying a realised city on an intended map: conformance, performance and decision influence are distinct epistemic objects, and a divergence from the plan may represent adaptation rather than failure. Baek and colleagues make the transition problem more exact. Evidence can survive in the archive while dying operationally; it can remain attached to a planning document yet become unusable for the department that must choose a site, allocate a budget or maintain an intervention. This suggests that planning should be understood as continuity infrastructure: an arrangement for carrying reasons, evidence, classifications and commitments across changes of stage, actor, medium and time without assuming that they remain identical. Such infrastructure includes obvious devices such as repositories, GIS layers, models, minutes, regulations and monitoring indicators, but also less visible relations among professional memory, handover practices, shared vocabularies and institutional responsibility. For spatial research, this changes the status of the corpus. A planning archive is not a warehouse of past representations; it is a record of transformations through which objects became more or less actionable. Metadata must therefore register not only what a document contains but when it entered, who could use it, what decision it addressed, what scale it operated at and whether its meaning remained traceable after transfer. Classification becomes consequential because categories are the joints through which evidence travels: a risk class, zoning code, housing category or performance indicator can preserve comparability across stages, but it can also strip away the context needed for later action. Mapping likewise shifts from depiction to translation. The relevant question is not simply where an object is located but whether its spatial representation can survive movement from research to negotiation, approval, implementation and stewardship. A climate-risk map too coarse for budgeting is epistemically discontinuous even if technically correct; an implementation metric detached from the rationale that generated it may produce administrative compliance while obscuring the original public purpose. Comparability must therefore be designed longitudinally, retaining enough provenance to distinguish legitimate adaptation from silent erosion. This is also where digital planning becomes a governance problem. Algorithms, support systems and platforms can preserve relations at scales impossible for individual memory, yet their schemas may harden provisional categories into infrastructure and make some forms of evidence easy to retrieve while others become computationally opaque. Public legibility consequently depends on selective transparency: not every technical object must be simplified, but the chain by which it affected a decision should remain reconstructable. A robust planning corpus would connect history to current constraint, evidence to argument, argument to commitment, commitment to implementation and implementation to later condition, allowing researchers to inspect where continuity was maintained, translated, deliberately revised or lost. Such a corpus does not restore the fantasy of comprehensive rationality. It accepts Lindblom’s limited comparisons, Forester’s asymmetries, Innes and Booher’s distributed capacities and Baek’s fragile handoffs, but makes their transformations inspectable. Planning then becomes less the authorship of future form than the governance of relations among heterogeneous evidentiary objects over time. Within Socioplastics, this tension condenses as Relational Agency.
Anto Lloveras is an architect and urban researcher based in Madrid. Through LAPIEZA LAB, he develops comparative spatial research on planning, archives, infrastructures, longitudinal fieldwork and the epistemic conditions through which urban evidence becomes operational.
BIBLIOGRAPHY
Baek, S., Kim, J., Park, C. and Lee, T. (2026) ‘Evidence lost in transition: Climate-information continuity failures in South Korea’s local adaptation planning and implementation’, Climate Services, 42, 100649.
Forester, J. (1989) Planning in the Face of Power. Berkeley: University of California Press.
Innes, J.E. and Booher, D.E. (2000) Network Power in Collaborative Planning. IURD Working Paper 2000-01. Berkeley: Institute of Urban and Regional Development, University of California.
Li, G., Wang, L., Wu, C., Xu, Z., Zhuo, Y. and Shen, X. (2022) ‘Spatial Planning Implementation Effectiveness: Review and Research Prospects’, Land, 11(8), 1279.
Lindblom, C.E. (1959) ‘The Science of “Muddling Through”’, Public Administration Review, 19(2), pp. 79–88.
Potts, R. and Milz, D. (2024) ‘A culture of digital planning? An international comparison of culture, planning and technology’, International Planning Studies, 29(4), pp. 327–341.
Son, T.H., Weedon, Z., Yigitcanlar, T., Sanchez, T., Corchado, J.M. and Mehmood, R. (2023) ‘Algorithmic urban planning for smart and sustainable development: Systematic review of the literature’, Sustainable Cities and Society, 94, 104562.
Sorensen, A. (2018) ‘Planning History and Theory: Institutions, Comparison, and Temporal Processes’, in Hein, C. (ed.) The Routledge Handbook of Planning History. Abingdon: Routledge, pp. 35–45.
Vonk, G. and Geertman, S. (2008) ‘Improving the Adoption and Use of Planning Support Systems in Practice’, Applied Spatial Analysis and Policy, 1, pp. 153–173.
