Systems Researcher & Entrepreneur, Softagram Oy
ville.laitila@softagram.com
25 February 2026
Keywords: morphological analysis, formal concept analysis, polyadic tensors, meta-ontology, systems theory, knowledge visualization, AI-assisted research
The challenge of making sense of complex, multi-faceted domains — whether in policy analysis, organizational design, technology assessment, or philosophical inquiry — has generated numerous structured methods. Zwicky's (1969) morphological analysis provides systematic combinatorial exploration. Wille's (1982) formal concept analysis offers mathematical rigor through lattice theory. Beer's (1972) Viable System Model supplies cybernetic organization. Yet each addresses a different facet of the problem, and practitioners must choose between them.
This paper describes a method that integrates insights from these traditions into a single, repeatable framework. The method builds on the GoodReason meta-ontology developed by Eki Laitila (2016), which provides a fixed set of eight archetypal symbols grounded in systems thinking, cybernetics, and innovation theory. The core innovation of the present work is to compose this symbol set with itself across n dimensions (where n ∈ {1, 2, 3, 4}), systematically populate every resulting cell with domain-specific meaning through AI-assisted research, and render the result as an interactive multi-scale visualization.
The method has been applied to domains including cybersecurity governance, municipal strategy (Kempele), geopolitical conflict analysis (Ukraine), business ecosystems (Easor), and architectural design — each time using the same eight symbols, reinterpreted for the domain.
The foundation of PSDC is the GoodReason meta-ontology, developed by Eki Laitila as a practical systems thinking framework (Laitila 2016). GoodReason defines eight archetypal symbols — α (Purpose), β (Structure), φ (Action), τ (Integration), Ω (Feedback), Δψ (Change/Pressure), χ (Information), and π (Theory) — which together constitute a minimal but complete vocabulary for describing any systemic domain. The symbols are drawn from cybernetics (Beer 1972), semiotics, innovation theory, and Laitila's own synthesis of systems science traditions.
The key property of these eight symbols is that they are domain-independent archetypes: while their concrete interpretation varies by domain (see Section 4.2), their structural roles remain invariant. This invariance is what enables cross-domain comparison — the central methodological affordance of PSDC.
Laitila also introduced the seven-level interpretation hierarchy (Micro through Philosophy), in which each symbol receives a level-specific reinterpretation that embeds multiple systems-theoretic traditions (analyzed in detail in Section 3).
Fritz Zwicky developed morphological analysis (MA) as "a method for investigating the totality of relationships contained in multi-dimensional, usually non-quantifiable, problem complexes" (Zwicky 1969, p. 34). The method proceeds by identifying independent parameters, listing possible values for each, and systematically examining combinations within the resulting morphological field.
Ritchey (2013) extended the method with Cross-Consistency Assessment (CCA), which reduces the combinatorial space by identifying internally inconsistent parameter combinations. The Swedish Morphological Society has applied this approach to policy analysis, technology foresight, and defense scenarios (Ritchey 2011).
PSDC shares the combinatorial spirit of MA but differs in a fundamental respect: in classical MA, each axis represents a different parameter with different values. In PSDC, the same eight-element symbol set occupies every axis. This makes the analysis automorphic — the structure maps onto itself. Where Zwicky's box asks "what combinations of different parameters are possible?", PSDC asks "what relationships emerge when the same set of concepts encounters itself?"
This distinction has important consequences. An automorphic morphological field is inherently reflexive: the diagonal elements (where the same symbol appears on multiple axes) represent self-referential concepts, while off-diagonal elements represent inter-symbolic relationships. The structure therefore encodes both identity and relation within a single framework.
Formal Concept Analysis (FCA), introduced by Wille (1982), provides a mathematical theory of concept hierarchies grounded in lattice theory. Wille stated that the aim of FCA is "supporting the rational communication of humans by mathematically developing appropriate conceptual structures which can be logically activated" (Wille 1992, p. 493).
The classical FCA context is dyadic: a binary relation between objects and attributes. Lehmann and Wille (1995) extended this to triadic contexts, motivated by Peirce's pragmatic philosophy and its three universal categories. A triadic context is defined as a quadruple (G, M, B, Y) where G, M, B are sets and Y ⊆ G × M × B is a ternary relation; the elements of G, M, and B are called objects, attributes, and conditions respectively.
Voutsadakis (2002) generalized further to polyadic (n-adic) concept analysis, establishing concept-forming operators and a basic theorem for arbitrary dimensions. This theoretical framework provides the formal underpinning for PSDC's multi-dimensional structures:
| PSDC Dimension | FCA Analog | Cell Count | Structure |
|---|---|---|---|
| 1D (symbols) | Monadic context | 8 | Vector |
| 2D (squares) | Dyadic context | 64 | Matrix |
| 3D (triplets) | Triadic context | 512 | 3-tensor |
| 4D (quads) | Tetradic context | 4,096 | 4-tensor |
Table 1. Correspondence between PSDC dimensions and polyadic FCA contexts.
In PSDC, however, the sets G, M, B, (and C for the tetradic case) are all the same set — the eight symbols. This makes each cell an automorphic n-adic concept: a point where n instances of the same symbolic vocabulary intersect.
Charles Sanders Peirce's three universal categories — Firstness (quality, possibility), Secondness (reaction, actuality), and Thirdness (mediation, law) — provide a philosophical basis for the centrality of triadic relations (Peirce 1903; see Houser & Kloesel 1992).
Peirce argued that triadic relations are irreducible: any genuine three-place relation cannot be decomposed into dyadic relations without loss of essential information (Peirce, CP 1.345). This principle is directly observable in PSDC: the triplet cell (α, β, π) carries semantic content that is not derivable from the three constituent pairs (α,β), (β,π), and (α,π). The 512-cell cube contains emergent meaning beyond what the 64-cell matrix encodes.
Empirically, this can be verified by comparing the squares data (64 dyadic concepts) with the cells data (512 triadic concepts) in any PSDC application. The triadic concepts consistently contain information — contextual modulations, three-way interactions, emergent patterns — absent from the dyadic layer.
The challenge of visualizing four-dimensional data is addressed in PSDC through a strategy analogous to Tucker decomposition (Tucker 1966; Kolda & Bader 2009). A 4-tensor T(d1, d2, d3, d4) ∈ ℝ8×8×8×8 is decomposed into a macro-level 8×8 grid indexed by (d1, d3) and a micro-level 8×8 grid indexed by (d2, d4):
T(d1, d2, d3, d4) → MacroGrid(d1, d3) → MicroGrid(d2, d4 | d1, d3)
This is a visual implementation of the matricization operation in tensor algebra: the 4-tensor is unfolded into a matrix of matrices. The user navigates the macro grid to select a (d1, d3) slice, then explores the micro grid within that slice.
This approach aligns with Tufte's (1983) small multiples principle — "series of graphics, showing the same combination of variables, indexed by changes in another variable" — extended to two levels of nesting: small multiples of small multiples.
George Kelly's Personal Construct Theory (Kelly 1955) and the associated Repertory Grid technique share structural similarities with PSDC. In Kelly's triadic elicitation method, a subject considers three elements and identifies how two are similar but different from the third, thereby surfacing personal constructs (bipolar dimensions of meaning).
PSDC inverts this process. Where Kelly's method is inductive — constructs emerge from the subject's experience — PSDC is deductive: a fixed set of eight constructs (the symbols) is imposed on the domain, and every possible combination is systematically explored. This aligns with deductive content analysis (Elo & Kyngäs 2008), where a pre-existing theoretical framework guides the coding of empirical material.
The distinction is methodologically significant. Inductive approaches risk incompleteness (not all relevant constructs may surface); deductive approaches risk forced fit (the framework may distort the domain). PSDC mitigates the latter risk through the multi-level interpretation system (Section 3), which allows each symbol to be calibrated to the domain.
A distinctive feature of PSDC is that the eight symbols are not fixed in meaning but are reinterpreted across seven hierarchical abstraction levels. Each level corresponds to a different scale of analysis, and each symbol receives a level-specific interpretation.
| Level | Name | Scale | Focus |
|---|---|---|---|
| 1 | Micro | Individual components | Actors, modules, tools |
| 2 | Process | Operational dynamics | Resources, delivery, feedback |
| 3 | System | Organizational wholes | Coordination, control, negotiation |
| 4 | Organization | Institutional structures | Epistemology, governance, competence |
| 5 | Ecosystem | Inter-organizational networks | Strategy, transformation, cognition |
| 6 | Society | Societal-scale systems | Mission, lifecycle, disruption |
| 7 | Philosophy | Foundational principles | Time, ethics, limits, truth |
Table 2. The seven abstraction levels of PSDC.
Analysis of the level-specific symbol interpretations reveals that each symbol independently traverses a hierarchy drawn from a different systems-theoretic tradition:
β (Structure) encodes Beer's Viable System Model. The progression S1 (operations, Level 2) → S2 (coordination, Level 3) → S3 (management, Level 4) → S4 (development, Level 5) → S5 (identity, Level 6) directly maps the five subsystems of the VSM (Beer 1972, 1979). Beer's key insight — that viable systems are recursive, containing viable systems modelable with identical cybernetic descriptions (Beer 1979, p. 118) — is preserved but recontextualized: instead of the same S1–S5 repeating at each recursion, the entire 8-symbol framework is reinterpreted.
τ (Integration) encodes Bronfenbrenner's Ecological Systems Theory. The progression Microsystem (Level 2) → Mesosystem (Level 3) → Exosystem (Level 4) → Macrosystem (Level 5) → Lifecycle/Chronosystem (Level 6) follows Bronfenbrenner's (1979) nested environmental systems. Each outer layer contextualizes the inner, and the sixth level explicitly references the chronosystem — the temporal dimension that Bronfenbrenner (1994) added later.
χ (Information) encodes Boulding's emergent complexity. The progression Boundaries (Level 1) → Control (Level 3) → Cognition (Level 5) → Limits to Growth (Level 7) mirrors Boulding's (1956) hierarchy of system complexity, where each level introduces qualitatively new properties irreducible to lower levels. The same informational principle — understanding one's environment — manifests as boundary-setting at the micro level and as collective intelligence confronting growth limits at the philosophical level.
α (Purpose) encodes Wilberian holarchic development. The progression Actor → Object → Context → Epistemology → Challenge → Concreteness → Time Perspective follows the logic of Wilber's (1995) holons: each level transcends and includes the previous. The actor (who does?) becomes aware of the object (for whom?), then the context (in what situation?), then the epistemological basis (how do we know?), escalating to temporal perspective (where is this all heading?). This is holarchic depth in Wilber's terminology: increasing levels of consciousness and encompassment.
Δψ (Change/Pressure) encodes Holling's panarchic adaptive cycle. The progression Sensitivity → Dynamism → Criticality → Recovery → Transformation → Visioning capacity → Homeostasis traces the adaptive cycle phases (Holling 1986; Gunderson & Holling 2002): exploitation (growth) → conservation (rigidity) → release (creative destruction) → reorganization (innovation). The cross-scale dynamics of panarchy — where "revolt" travels upward and "remember" travels downward — are implicit in the seven-level structure: disruption at one level cascades through the hierarchy.
The seven-level system thus functions as a metatheoretical weave: five major systems-theoretic traditions (cybernetic, ecological, complexity-theoretic, integral, and resilience-theoretic) are integrated not by reducing one to another but by assigning each to a different symbolic dimension and allowing them to co-evolve across shared abstraction levels.
This makes the total interpretive space of PSDC considerably larger than any single application suggests. A complete PSDC analysis, across all dimensions and levels, encompasses:
In practice, a single domain application typically operates at one level, producing 8 + 64 + 512 = 584 concepts (or 4,680 with the tetradic layer). The full multi-level analysis remains a theoretical horizon.
Domain material is gathered using AI-assisted systematic research (cf. the emerging literature on LLM-assisted systematic reviews: Fabiano et al. 2024; Wang et al. 2024). In current practice, Google Gemini Deep Research is used to produce comprehensive domain surveys in markdown format.
The prompt design for this phase is methodologically critical. The researcher must: (1) define the domain scope and boundaries; (2) request coverage across all eight symbolic dimensions; (3) ensure sufficient depth for triadic/tetradic interpretation; and (4) specify the target abstraction level(s).
The eight archetypal symbols are calibrated to the domain. Each symbol receives a domain-specific label, description, and interpretation that preserves its archetypal character while making it semantically meaningful in context.
For example, α (Purpose/Direction) might become:
Every cell in the chosen tensor dimension is systematically populated with semantic content. For the triadic case (512 cells), this means assigning a concept name and description to each (x, y, z) coordinate where x, y, z ∈ {0, 1, ..., 7} each index a symbol.
The population process produces structured data in the format:
{"triad": ["α", "β", "π"], "title": "Concept Name", "desc": "Description of the triadic relationship"}
Beyond the primary tensor, three derived 2D layers are generated:
These layers provide a dialectical perspective (cf. Hegel's thesis–antithesis–synthesis) on the dyadic relationships: base state, negative pole, positive pole.
The populated data is rendered as an interactive 3D visualization using WebGL (Three.js via React Three Fiber). The visualization employs:
PSDC is not a statistical method. It does not discover structure in data through algorithmic means (unlike, e.g., principal component analysis or clustering). It is not a grounded theory approach where categories emerge from data (Glaser & Strauss 1967). It is not a pure visualization technique, though visualization is an essential output.
PSDC is a structured interpretive method — a systematic framework for producing interpretations of complex domains. It belongs to the family of deductive qualitative methods where a theoretical framework precedes empirical analysis (cf. Elo & Kyngäs 2008). Its closest methodological relatives are:
The distinctive contribution of PSDC is the combination of:
The principal strength of PSDC is its structural comparability. Because the same eight symbols and the same dimensional framework are used across all domains, analyses of different domains are structurally isomorphic. This enables cross-domain pattern recognition: a pathological dynamic identified in cybersecurity governance may have a structural analog in municipal strategy, discoverable because both occupy the same cell coordinates.
The fixed eight-symbol framework is both the method's strength and its primary limitation. Domains that do not map naturally onto these eight dimensions may require forced interpretations. The quality of the analysis depends critically on two human judgments: the calibration of symbols to the domain (Phase 2) and the population of combinatorial cells (Phase 3). These remain interpretive acts, not algorithmic ones.
The combinatorial scale — particularly at the tetradic level (4,096 cells) — raises questions of cognitive tractability. Can a human meaningfully engage with 4,096 distinct concepts? The visualization layer partially addresses this through progressive disclosure (macro → micro navigation), but the fundamental challenge of combinatorial explosion remains.
Promising directions include: automated consistency checking of cell content (analogous to Ritchey's CCA); machine learning approaches to detecting cross-domain structural patterns; integration with knowledge graph technologies; and empirical studies of how users navigate and derive insight from the multi-scale visualizations.
Polyadic Symbolic Domain Cartography represents a synthesis of morphological analysis, polyadic formal concept analysis, hierarchical systems theory, and interactive visualization into a repeatable method for structured domain interpretation. Its automorphic character — the same symbols meeting themselves across multiple dimensions — produces reflexive knowledge structures that are at once combinatorially rigorous and semantically rich. The seven-level interpretation system, which weaves together Beer's cybernetics, Bronfenbrenner's ecology, Boulding's complexity hierarchy, Wilber's integral theory, and Holling's panarchy, positions the method not merely as an analytical tool but as a metatheoretical integration framework.
Corresponding author: ville.laitila@softagram.com