The trajectory from Hoselitz through Southall, Friedmann, Pickvance and Nijman reveals successive enlargements of the comparative object. Hoselitz still works through developmental typology, asking how different kinds of cities contribute to or inhibit economic and cultural transformation. His generative/parasitic distinction belongs to the mid-century search for regularities capable of explaining urbanisation across divergent societies, yet it already refuses to treat city growth as valuable in itself: what matters is the relation between urban concentration and wider social transformation. Southall’s cross-cultural anthropology makes the challenge more empirical and more plural. Urbanisation appears through kinship, ethnicity, caste, migration, household form, transport, locality and supralocal institutions across Africa, Asia, Latin America and the Pacific. The city becomes comparable only by decomposing the supposedly universal urban condition into social relations whose combinations vary historically and geographically. Friedmann then changes scale. Building on the Marxian urban turn associated with Castells and Harvey, the world-city hypothesis places urban formation inside the world economy. A city’s internal structure is no longer analytically sufficient; its economic functions and position within transnational accumulation become constitutive. Comparison therefore shifts from cities as bounded societies to cities as differentiated positions in global systems. Pickvance subsequently formalises the epistemological problem that these enlargements create. If different contexts can produce similar outcomes and similar contexts can produce different outcomes, comparison cannot rely on a single causal grammar. His four varieties of comparative analysis make plural causation explicit and show that explanatory comparison is a research design for analysing causal configurations rather than a descriptive juxtaposition of places. Nijman carries this insight directly into urban geography. Comparison must operate between universalism and irreducible place-particularity, incorporating historical sequence and geographical context as explanatory dimensions. His “deep analogies” provide the most precise synthesis of the historical lineage: the object of comparison is neither superficial resemblance nor total urban form, but a process that may recur through very different spatial, institutional and temporal configurations. Miami can therefore be read through Amsterdam, Hong Kong, Shanghai and Dublin without becoming equivalent to any of them. The comparison strengthens specificity while also producing theory. Historical comparative urbanism emerges from this sequence as an epistemology of bounded generalisation: it constructs concepts capable of travelling, but subjects their reach to the resistance of place, history and causal complexity. For contemporary urban research, this historical lineage matters because many supposedly new methodological problems were already visible in earlier comparative work. The current insistence on relational cities, planetary connections, situated theory and globally distributed conceptual production is not a rupture with comparison but a transformation of its scale and assumptions. Hoselitz demonstrates the productivity and danger of typology: categories can organise difference, but they can also harden developmental hierarchies. Southall demonstrates that comparative range can destabilise those hierarchies by bringing multiple cultural and institutional forms into the same analytical field. Friedmann shows that external relations are not contextual supplements but internal conditions of urban formation. Pickvance makes causal heterogeneity methodologically legitimate, while Nijman turns geographical and historical difference into an active check on abstraction. Together they establish a powerful principle for comparative urban epistemology: generalisation should emerge from carefully constructed relations among cases rather than from the prior assumption that cases belong to a common universal type. This has direct consequences for how an urban research corpus should be organised. Cases should remain richly situated, but their relations should be explicit enough to permit comparison across scale, period and geography. The archive must preserve not only outcomes but causal conditions, temporal sequences, institutional structures and external connections. A city can then be compared through migration regimes, state relations, infrastructural connectivity, financial roles or social forms without requiring overall similarity. This makes comparison generative: each new case can reveal a deep analogy, expose a false equivalence, split an inherited category or identify a causal pathway previously hidden by a dominant model. The historical value of these texts therefore exceeds disciplinary genealogy. They provide a methodological architecture for producing urban knowledge that is simultaneously mobile and resistant to simplification. Within Socioplastics, this condition condenses as GenealogyGradient.
Anto Lloveras is an architect and urban researcher working across comparative urbanism, spatial epistemology, archives and relational infrastructures through LAPIEZA LAB and Socioplastics.
