From frames to method
The previous post traced connections between three uses of "frame" - Fillmore's linguistic frames, Goffman's situational frames, and Dorst's design frames - and concluded by noting that framing as a concept needs methodological grounding if it is to be more than a magic word. Adele Clarke's situational analysis offers one such grounding. Developed as a form of postmodern grounded theory, it provides structured techniques for mapping the heterogeneous elements - human, non-human, discursive, silent - that constitute a situation. Christian Nold's recent work, which I first encountered at Nordes in the summer of 2023, adapts Clarke's method explicitly for design, proposing "mapping situations" and "sensing situations" as pragmatist alternatives to the dominant conversation-with-the-situation and framing metaphors. This post explores what situational mapping offers service design, and asks how the linguistic analysis developed in earlier posts - Fillmore's case grammar, Gärdenfors's conceptual spaces - might inform and enrich the method.
The situation as unit of analysis
Clarke is a sociologist who developed situational analysis (2005; substantially revised in Clarke, Friese and Washburn, 2018) as an evolution of grounded theory, drawing on Dewey's pragmatism, symbolic interactionism, Foucauldian discourse analysis, and Actor-Network Theory's attention to non-humans. Her core insight is that the situation itself is the unit of analysis - not the individual, not the organisation, not the discourse, but the situation understood as a heterogeneous assemblage of elements that together constitute what is going on. She builds on John Dewey's concept of the situation from Logic: The Theory of Inquiry (1938), where Dewey argues that "What is designated by the word 'situation' is not a single object or event or set of objects and events. For we never experience nor form judgments about objects and events in isolation, but only in connection with a contextual whole".
Situations, in this pragmatist sense, are not contexts external to the inquirer; they envelop the researcher alongside a heterogeneous mix of humans and non-humans, physical objects, ideas, discourses, and settings. Critically, situations are agent-relative and practice-relative - their boundaries are set by relevance to the inquiry, not by predetermined scope. As Nold (2023) observes, "This means there are no predetermined elements or scope of a situation. The range of elements and extent of the situation can only be defined by the researcher in relation to the specific inquiry". Dalsgaard (2014) and Stompff, van Bruinessen and Smulders (2022) have both explored the deeper connections between Dewey's pragmatism and design inquiry; Clarke's situational analysis gives these philosophical affinities concrete methodological form.
Clarke's mapping method
Situational analysis employs three mapping techniques - now applied across multiple disciplines (Clarke, Washburn and Friese, 2022) - each addressing a different analytical question. Situational maps lay out all major elements found in the situation as understood by those in it and by the researcher. Social worlds and arenas maps identify collective actors and the arenas within which they interact and negotiate. Positional maps chart the positions taken (and not taken) on key issues. For service design purposes, the situational map is most immediately applicable, since it aligns most closely with Dewey's notion of situation and offers a practical entry point for design inquiry.
Clarke provides structured categories for ordering the initial "messy map" of observations into a more systematic analysis. These categories prompt attention to individual human actors (key individuals and significant unorganised people, including the researcher); collective human actors (particular groups, specific organisations); discursive constructions of those actors as found in the situation; political and economic elements (the state, particular industries, politicised issues); temporal elements (historical, seasonal, crisis, and trajectory aspects); non-human elements and actants (technologies, material infrastructures, specialised knowledges, material things); implicated and silent actors who are affected but not present; discursive constructions of non-human actants; sociocultural and symbolic elements (religion, race, gender, nationality, visual and aural symbols); spatial elements (geographical aspects, local and global spatial issues); major contested issues and debates; and related discourses including normative expectations and mass media narratives.
These categories serve multiple purposes: they prompt attention to elements that researchers might otherwise overlook (non-humans, discourses, silences); they support power analysis by surfacing who and what is implicated but not present; and they create a shared vocabulary for discussing complex situations. The final step is relational: drawing connections between elements to understand how they affect one another. As Clarke (2005) puts it, one should "Center on one element and draw lines between it and the others and specify the nature of the relationship by describing the nature of that line. One does this systematically, one at a time, from every element on the map to every other". This systematic tracing reveals the work that different elements do in the situation - how they enable, constrain, legitimate, or undermine one another.
Nold's adaptation for design
Nold's paper "Mapping and Sensing Dewey's Situations as Design and Social Science Methods" (Nordes 2023) proposes integrating Clarke's situational analysis into design practice. His critique targets the dominant metaphors for how designers engage with situations. Schön's (1983) "conversation with the situation" treats design as iterative dialogue where the designer proposes and the situation talks back, but as Nold observes, this involves translating the complexity of the situation into a simplified problem, potentially losing the nuance and surprise that Dewey's situations can offer. Dorst's framing (2011), explored in the previous post, describes how designers impose frameworks on complex situations, but framing can prematurely cut the inquiry short, accepting a particular definition before fully exploring the situation's heterogeneity.
Nold proposes mapping and sensing as alternatives that preserve complexity rather than reducing it: "Unlike established design approaches to Dewey's situation, the designer does not have total control to frame the situation as they wish, since the complex situation is never fully knowable. Instead, these methods help designers to deal with complexity and ambiguity by sensing and mapping to affect and be affected by a situation" (Nold, 2023).
Nold's adaptation of Clarke emphasises several features that distinguish it from existing design mapping practices. Unlike giga-mapping approaches (Sevaldson, 2011) that embrace visual complexity for its own sake, situational mapping focuses on analysing which elements are impacting one another, thereby providing insights into power dynamics. The method foregrounds silent actors who have been deliberately excluded - a critical move for service design, where the people most affected by services may be absent from design processes.
Clarke's structured categories introduce entities such as collective actors, discursive constructions, and political elements that are typically not part of the designer's vocabulary. Including the researcher or designer in the map requires explicit attention to their own position, agenda, and capacity for transformation. And the maps function not as static representations but as methodological tools for changing the world - instruments for intervention rather than mere description.
Nold's second method draws on Dewey's insight that situations must be sensed or felt as a qualitative whole. This is not flow or intuition but intentional, embodied encounter with the situation's material reality. Drawing on soma design and radical clowning techniques, Nold proposes that designers establish a bodily connection with the situation - encountering its tangible, material entities sensorially rather than purely cognitively. A vignette from a disability design workshop illustrates the approach: a design team stuck on how to engage with "disability in the city" used improvisational games to explore their own physical and emotional constraints, which allowed them to situate the topic with their own bodies and connect with the interpersonal and political dimensions of the situation. The combination of cognitive mapping and affective sensing offers what Nold calls a holistic method for engaging with complexity.
Connecting situational analysis to linguistic frames
How does Clarke's situational analysis relate to the linguistic analysis developed in earlier posts? The connections are more than superficial. Clarke's categories map loosely onto the semantic roles of case grammar: individual human actors correspond to agents and experiencers, collective human actors to collective agents, non-human actants to instruments and patients, spatial elements to locatives, and temporal elements to temporal roles. The parallel is imperfect but instructive. Fillmore's case grammar asks: who acts, what is affected, through what instrument, where, when? Clarke's situational analysis asks similar questions but extends them: who is silenced, what discourses circulate, what power relations obtain? Where Fillmore gives us the argument structure of verbs, Clarke gives us the argument structure of situations.
Fillmore's (1982) insight that words evoke frames - structured background knowledge with roles and relations - also connects to Clarke's method. When we describe a service situation using particular language, we implicitly invoke frames that foreground some elements and background others. Describing hospital discharge as "patient flow management" evokes a logistics frame with roles for flows, bottlenecks, and throughput; describing it as "care transition" evokes a relational frame with roles for carers, receivers, and continuity. Situational mapping can surface these framing effects by asking what elements our current language makes visible and what it obscures.
Gärdenfors's (2017) conceptual spaces framework connects to Clarke's categories in a further way. Each category in the ordered map can be understood as a domain of variation - a space of possible elements. "Individual human actors" is a domain that might include nurses, patients, consultants, family members, social workers, and administrators; "non-human actants" is a domain that might include beds, records, algorithms, buildings, and protocols. Mapping a situation involves identifying which regions of each domain are instantiated - which actors, which technologies, which discourses are present - and the categories provide the dimensions while the specific elements provide the coordinates. This suggests that situational mapping could be enriched by attending to the structure within categories: what dimensions differentiate actors, and what qualities distinguish technologies? Gärdenfors's framework offers vocabulary for this finer-grained analysis.
The relational analysis step - drawing lines between elements and specifying their nature - also connects to verb semantics. Relations between elements are often best described verbally: the nurse updates the record, the protocol constrains the discharge, the family advocates for the patient. Each relation verb evokes its own frame with its own semantic roles. The case grammar post developed dimensions of verb meaning - tense, aspect, modality, polarity - that apply directly to relational analysis. Does the relation hold now or is it historical? Is the relation necessary, possible, or merely permitted? Is the relation positive or negative? Enriching relational lines with this grammatical detail would make situational maps more precise about the nature of connections.
Situational mapping applied: hospital discharge
To illustrate what this method surfaces, consider its application to hospital discharge - a situation I have been working with through the TCH weeknotes. A messy map of the situation might include patients waiting for discharge, ward nurses, discharge coordinators, bed managers, consultants, social workers, family members, patient transport, care home staff, community nurses, the discharge lounge, the bed board, the electronic patient record, the discharge checklist, patient notes, medication lists, "medically optimised" as a category, "stranded patient" as a label, "flow" as an organising metaphor, length of stay targets, weekend discharge rates, NHS England policy, local commissioning arrangements, the patient's home, care home availability, winter pressures, and historical bed closures.
Ordering this through Clarke's categories surfaces elements that an unstructured brainstorm would likely miss. Under implicated and silent actors: patients who lack family advocates, care home staff who are rarely consulted about discharge timing, community nurses who receive the patient downstream, and patients' previous selves - their pre-admission identity, which the institutional frame tends to erase. Under discursive constructions: "medically optimised" renders the patient as a complete product ready for output; "stranded" implies the patient is stuck, carrying a blame-adjacent connotation; "flow" renders patients as liquid moving through pipes; and "bed blocker", though deprecated, persists in informal use.
Under temporal elements: the length of stay target creates artificial urgency, weekend discharge rates drive policy focus, winter pressures introduce seasonal crisis dynamics, and each patient's trajectory from admission to discharge follows its own temporal arc.
The relational analysis reveals how elements interact. The bed board represents the patient as a status (medically optimised, waiting for transport), which is a relation of reduction that enables coordination but obscures individuality. The "medically optimised" category enables discharge planning but silences the patient's own sense of readiness. Family members advocate for the patient to discharge staff, a relation that depends on family presence - which immediately surfaces the implicated actors: patients without advocates. The discharge checklist constrains the consultant's sign-off through a necessary-condition relation with bureaucratic authority. The discharge lounge spatially separates patients from the ward, enabling bed turnover but creating liminal discomfort.
Linguistic observations sharpen the analysis further. The dominant verb forms in discharge discourse are passive and perfective: patients "are discharged", "have been medically optimised", "will be transferred". Agency is diffused; the patient is grammatically patient. The modal structure is deontic: patients "must" meet criteria, staff "must" complete checklists, and possibility modality is rare - the situation is framed as obligation rather than opportunity. The polarity is contested: "discharge" is officially positive (the patient gets to go home) but often experienced negatively (rushed, uncertain, unprepared).
These observations surface design opportunities that a conventional journey map or blueprint would not reveal. How might patients without family advocates be better supported? The bed board represents patients as statuses; could a different representation foreground patient readiness and preference? "Medically optimised" frames discharge from the system's perspective; what would a patient-centred frame look like? The target-driven urgency shapes everything; what would a trajectory-sensitive approach look like that attends to patient time, not just system time? The discharge lounge is a space of waiting; what does it feel like to wait there, and what would humane waiting look like? These questions emerge from attending to the situation's heterogeneity rather than flattening it into a linear journey.
Method as probe
Drawing on Clarke, Nold, and the linguistic connections developed above, the practical approach moves through several phases. The first is messy mapping: surfacing all elements that might matter without premature filtering - people, things, technologies, organisations, ideas, debates, places, times, including the researcher or designer themselves - and explicitly asking who is affected but not present and what is assumed but not stated.
The second phase orders these elements using Clarke's categories to ensure comprehensive coverage, paying particular attention to non-human actants, discursive constructions, implicated or silent actors, and temporal elements. Categories function here as prompts rather than containers; they surface elements that the initial brainstorm missed.
The third phase enriches understanding through semantic analysis. This means identifying the semantic roles of key human actors - are they agents, patients, beneficiaries, experiencers? - and asking what non-human actants afford, constrain, or enable. It also means attending to the frames that discursive constructions evoke, and noting the verb forms used to describe the situation: what tense, what modality, what polarity.
The fourth phase maps relationships between elements. Each relation has a nature best expressed as a verb (enables, constrains, legitimates, undermines), a direction (unidirectional, bidirectional, asymmetric), a modality (necessary, possible, contested), and a visibility (explicit, implicit, hidden). Specifying these dimensions makes the relational analysis more precise than simply drawing lines between elements.
The fifth phase complements cognitive mapping with embodied encounter - Nold's sensing. This involves identifying material elements that can be directly encountered, designing encounters that engage multiple senses, attending to felt qualities, and noting where the situation resists or exceeds expectations.
The sixth phase synthesises toward intervention: identifying leverage points where intervention might shift the situation, surfacing design opportunities from unmet needs, silent actors to include, and relations to reconfigure.
The method has limitations that should be acknowledged. A map is not the territory; situational maps are representations that inevitably simplify and select, and the structured categories can become templates rather than prompts, leading to formulaic analysis. Situations evolve while maps capture moments, and single maps risk freezing what is actually in flux. Despite the structured categories, the researcher's perspective shapes what is seen; including oneself in the map is necessary but not sufficient for reflexivity. And identifying "silent actors" is itself a power move - the analyst decides who is silent and speaks for them, which requires humility and, ideally, participatory methods that give voice to those implicated.
Situational mapping nonetheless complements existing service design methods in specific ways. Journey maps capture temporal sequence while situational maps capture heterogeneous elements; stakeholder maps identify actors while situational maps add non-humans, discourses, and silences; system maps such as those of Jones and Van Ael (2022) show relationships while situational maps enrich these with linguistic precision about relation types; and blueprints separate frontstage and backstage while situational maps question that separation by surfacing what is excluded from both.
The strongest use may be at the fuzzy front end - before the brief is fixed, before the problem is framed, when the situation's heterogeneity can still surprise and redirect. As Nold suggests, situational mapping operates as a probe that preserves complexity rather than reducing it too soon. Dewey's pragmatist insight remains: inquiry transforms situations from indeterminate to determinate. But the path from indeterminacy to determinacy matters, and situational mapping offers a path that takes heterogeneity seriously, attends to what is silenced, and keeps the designer reflexively aware of their own situation-shaping role.
References
Clarke, A.E. (2005). Situational Analysis: Grounded Theory After the Postmodern Turn. Sage.
Clarke, A.E., Friese, C. and Washburn, R. (Eds.). (2018). Situational Analysis: Grounded Theory After the Interpretive Turn (2nd ed.). Sage.
Clarke, A.E., Washburn, R. and Friese, C. (Eds.). (2022). Situational Analysis in Practice: Mapping Relationalities Across Disciplines. Routledge.
Dalsgaard, P. (2014). Pragmatism and design thinking. International Journal of Design, 8(1), 143-155.
Dewey, J. (1938). Logic: The Theory of Inquiry. Henry Holt.
Dorst, K. (2011). The core of 'design thinking' and its application. Design Studies, 32(6), 521-532.
Fillmore, C.J. (1982). Frame semantics. In Linguistics in the Morning Calm (pp. 111-137). Hanshin Publishing.
Gärdenfors, P. (2017). The Geometry of Meaning: Semantics Based on Conceptual Spaces. MIT Press.
Jones, P.H. and Van Ael, K. (2022). Design Journeys Through Complex Systems. BIS Publishers.
Nold, C. (2023). Mapping and sensing Dewey's situations as design and social science methods. In Nordes 2023: This Space Intentionally Left Blank. https://doi.org/10.21606/nordes.2023
Schön, D.A. (1983). The Reflective Practitioner: How Professionals Think in Action. Basic Books.
Sevaldson, B. (2011). Giga-mapping: visualisation for complexity and systems thinking in design. In Nordes 2011 (pp. 137-156).
Stompff, G., van Bruinessen, T. and Smulders, F. (2022). The generative dance of design inquiry: exploring Dewey's pragmatism for design research. Design Studies, 83, 101136.