Chapter 7 · Self-Understanding and the Formal Core

What we discuss in this chapter: Organizational Self-Understanding does not arise from piling up more documents, but through precise consolidation semantics that expose contradictions across cross-source knowledge repositories and make them actionable. The formal Claims model, the conflict_record, the decision-making process of the Four-Zone Review, and a fully worked-through example at Muster AG give this layer its operational sharpness.

Your leverage as a decision-maker: Your organization already contradicts itself in countless places invisible to management—until a strategic decision, an internal audit, or an operational failure painfully exposes these conflicts. Formal conflict semantics transform blind spots into a manageable backlog. This shifts self-understanding from a vague cultural debate into a concrete infrastructure task with a prioritized backlog.

---

7.1 The Weick Connection: Sensemaking as Model and Tension

Karl Weick provided the authoritative scientific description of how organizations arrive at a picture of themselves: through sensemaking—the retrospective structuring of experienced practice into plausible narratives. This perspective brilliantly explains the phenomena we observed in Chapter 1. It clarifies why three different functional departments describe the exact same core process completely differently, each believing its version to be the only plausible one. That organizational self-understanding is constructed rather than pre-existing as objective truth is Weick's central insight, and this theoretical heritage is explicitly acknowledged in our model.

At the same time, Weick's sensemaking approach introduces a fundamental tension for management.

Human sensemaking strives primarily for plausibility, not mathematical precision.

People in organizations require a viable narrative that provides orientation, not a sterile database. An operational infrastructure for self-understanding must neither deny nor mimic this human characteristic.

It does not deny it, because it acknowledges that every source in the body of knowledge represents a perspective interpretation rather than neutral absolute truth. Nor does it mimic it, because it delivers precisely the capability that narrative sensemaking fundamentally lacks: placing differing versions side by side in a structured manner and explicitly highlighting discrepancies. The formal core in this chapter is engineered specifically for this purpose: not as an automated truth machine, but as an uncompromising detector of contradictions.

7.2 The Smallest Unit: The Claim

To make self-understanding operational, existing knowledge must be decomposed into testable building blocks:

Definition 7.1 — Claim (Knowledge Unit).

A Claim is the smallest independently testable statement an organization makes about itself, maintained with at least four attributes: Content (what is asserted), Source (where the statement originates), Context (where it applies), and Validity (status and approval state).

Based on: van der Aalst 2016 (formalization as a prerequisite for analysis; methodological model from process mining research for event data). Boundary of source: Process mining formalizes observed behavior from event logs; declared statements regarding responsibilities, intentions, and formal rules lie outside its scope.

Necessary extension, our position: The same formal rigor is applied to declarative self-knowledge rather than mere technical process traces.

Source: Case B (Ch. 11): Decomposition of a process architecture into testable units; corpus size, benchmark value from ongoing initiatives: approximately 1,400 process elements.

The technical representation of a Claim can be illustrated using a concrete example from Muster AG's repository:

claim MA-0412
  content:    "Class B complaints are decided by site QA."
  source:     QM Manual, Section 8.3 (2021 Version)
  context:    North and South Sites, Series Production
  validity:   approved; last audit 2023

This decomposition into clean individual Claims appears unspectacular at first glance. Its strategic leverage unfolds in the next step, when different source systems addressing the same operational issue are consolidated.

7.3 Consolidation Across Sources: Three Outcomes

When two or more source systems are consolidated into the same knowledge base, exactly three logical outcomes emerge for any given topic.

First: the present versions are completely congruent. In this case, organizational confidence in the statement increases drastically because multiple independent sources confirm the same reality.

Second: the versions diverge in substance. Here, an explicit conflict object is instantiated, remaining managed within the knowledge base until an authorized decision is rendered.

Third: a topic is documented in only a single source. This immediately exposes the documentation gap in the remaining systems.

All three outcomes yield valuable insights; however, the second outcome provides by far the most powerful leverage.

It highlights for management precisely those areas where the organization contradicts itself before severe missteps, audit non-conformities, or quality incidents quietly occur. Edwin Hutchins' research on Distributed Cognition supplies the theoretical vocabulary: the overall system's knowledge exceeds the sum of individual repositories precisely because structured synthesis exposes productive discrepancies invisible within any isolated single source. This requires no magic, only disciplined accounting.

A Factor of 4.45 Between Two Teams

What happens when two sources implicitly assert incompatible rules in the same context without an institutionalized check was demonstrated with tragic clarity in space exploration in 1999. The ground software from contractor Lockheed Martin calculated thrust data for the Mars Climate Orbiter in imperial pound-force seconds; NASA's navigation software expected metric Newton-seconds according to the interface specification. Both teams performed flawlessly for months—except using figures that differed by a factor of 4.45. On September 23, 1999, the spacecraft plunged too deep into the Martian atmosphere and disintegrated; the orbiter and lander program had cost $327.6 million in total. The investigation report's takeaway: the authoritative specification had existed all along; there was simply no procedure that systematically reconciled the declared assumptions of the parties involved. Exactly this unspectacular check is delivered by a conflict object. The alternative is letting physics perform the review.

— NASA Mars Climate Orbiter Mishap Investigation Board, Phase I Report (1999)

7.4 The Conflict Object

For the targeted management of contradictions, the conflict_record is introduced as a dedicated analysis object:

Definition 7.2 — conflict_record.

A conflict_record is the normalized state of two or more Claims asserting incompatible content in the same context; it captures the involved Claims, discrepancy type, decision status, and, post-decision, the approved version alongside its rationale.

Based on: van der Aalst 2016 (conformance checking as a model: discrepancy as a standalone analysis object). Boundary of source: Conformance checking compares target process models with actual logs; direct comparison of declared target rules against each other and their governed resolution fall outside this method.

Necessary extension, our position: Contradictions transform from ephemeral findings into permanently managed objects with a defined lifecycle (open, under review, decided, obsolete).

Source: Case C (Ch. 11): Conflict density per 100 Claims between two system landscapes, benchmark value from ongoing initiatives: 19.

The fully defined lifecycle represents the crucial advancement over traditional consistency checks. An identified conflict does not quietly vanish by being overwritten; it is resolved exclusively through a documented management decision. The historically superseded version remains traceable in the version history. This exact characteristic guarantees total auditability of the knowledge base.

7.5 The Decision-Making Process: The Four-Zone Review

To process open conflict objects efficiently without time-consuming, endless political debates, a structured evaluation process is applied:

Definition 7.3 — Four-Zone Review.

A review procedure assigning every open conflict_record to exactly one of four zones: Zone 1: trivial discrepancy, resolvable editorially; Zone 2: genuine substantive divergence, subject-matter decision required; Zone 3: justified variance, both versions hold in distinct contexts; Zone 4: resolution impossible or uneconomical, conflict is retained as an explicit risk.

Based on: Argyris/Schön 1978 (distinction between espoused theory vs. theory-in-use as a model for Zones 2 and 3). Boundary of source: Chris Argyris and Donald Schön masterfully diagnose the gap between stated and practiced routines, but offer no operational procedure to decide case by case.

Necessary extension, our position: A repeatable operational procedure featuring clear zones, defined roles, and mandatory approval obligations.

Source: Case B (Ch. 11): Four-Zone Review of program variants with domain owners; zone distribution. Benchmark value from ongoing initiatives: 32% Zone 1, 21% Zone 2, 43% Zone 3, 4% Zone 4.

For the Four-Zone Review to function reliably in corporate practice, the workflow is meticulously structured around roles, decisions, and versioning:

Who sits at the table? The review panel comprises the respective Decision-Owner (process owner or quality manager holding formal authority), domain experts from the affected source systems, and the Governance Lead (architect of the knowledge layer). This panel convenes in fixed cadences.

What is concretely decided across the four zones?

• Zone 1 (Trivial Discrepancy): The Governance Lead resolves linguistic or syntactic differences directly through editorial updates without involving domain leads.
• Zone 2 (Genuine Substantive Divergence): The Decision-Owner makes a binding domain decision favoring one of the variants.
• Zone 3 (Justified Variance): The Decision-Owner confirms that both rules are fully justified, but strictly segregates their applicable contexts (e.g., by facility, product line, or regulatory jurisdiction).
• Zone 4 (Managed Conflict): If a conflict cannot be resolved immediately due to economic or political constraints, it remains in the system as an explicitly flagged risk rather than being concealed.

What is versioned? Every review outcome generates an immutable Decision Record. This logs the selected target Claim, archived former Claims, the Decision-Owner's rationale, effective dates, and sunset clauses for temporary exemptions.

Zone 3 warrants special attention because it protects management from the knee-jerk trap of "harmonization mania." Not every variation between two sites represents an error. Many discrepancies turn out to be highly justified local adaptations. The process forces the organization to document this distinction explicitly rather than bulldozing it during blanket centralization initiatives.

7.6 Muster AG: A Complete Walkthrough

Let us trace the procedure step by step through a concrete walkthrough at Muster AG.

Step 1: Capture of Contradictory Claims. Muster AG's QM manual specifies in Section 8.3: Class B complaints are decided by site Quality Assurance (Claim MA-0412). Conversely, the newer process wiki asserts under Quality: Since the 2024 reorganization, these are decided exclusively by the central Quality Board (Claim MA-1873). Both assertions claim validity within the exact same operational context (series production) and are mutually exclusive.

Step 2: Automated Generation of the Conflict Object. Reconciling both source systems, the formal core detects the incompatible content and automatically creates the object conflict_record CR-0087 with status open.

Step 3: Categorization in the Four-Zone Review. During the review session with the VP of Quality and site QA leads, the case is evaluated. Fact-finding reveals: at the North facility, the central board decides; at the South facility, local QA continues to decide. This is not a linguistic discrepancy, but genuine substantive divergence (Zone 2).

Step 4: Decision and Context Mapping. Quality leadership mandates that the central board rule serves as the company-wide standard. However, a temporary exception is granted for the West facility, which performs custom manufacturing.

Step 5: Versioned Storage in Memory. The knowledge base subsequently stores three elements:

1. The approved standard Claim MA-1873 (validity: enterprise-wide).
2. A new exception Claim MA-1874 with explicit context West Facility, Custom Manufacturing and a twelve-month sunset clause.
3. The historical Claim MA-0412, archived as obsolete (replaced by CR-0087).

To the opening question from Chapter 5 regarding how a complaint at the South facility is actually processed, an unambiguous, verifiable answer now exists—complete with clear status and an assigned Decision-Owner. This is precisely what this theory defines as organizational self-understanding: not metaphysical contemplation, but verifiable information capability in previously contested areas.

Case Study — Case B: A global rolling stock manufacturer operates its modeled process architecture across multiple vehicle programs in countless variants whose actual validity is understood by only a few key individuals. Systematic consolidation with automated conflict detection and subsequent Four-Zone Review transforms paralyzed political harmonization debates into a rapidly actionable backlog. → detailed in Chapter 11.

7.7 Boundaries of the Formal Core

Two essential boundaries of the formal core must be emphasized in conclusion: they are integral components of the theory, by no means deficiencies. First, automated Claim extraction from unstructured repositories using modern AI does not achieve flawless accuracy today. Final human verification therefore remains constitutive (current research on this is synthesized in Chapter 18).

Second, the formal core itself makes no domain decisions whatsoever. It merely renders decision-worthy points unmistakably visible and makes enacted decisions permanently durable. Who holds authority to decide within the enterprise, under what legitimacy, and within what timelines is a question of governance, addressed in Chapter 14 and Chapter 15. Only on this secure foundation can we meaningfully address the third rung: what it concretely means when an organization decides intelligently.

💡 What We Discussed

Organizational Self-Understanding requires a precise structure that decomposes declared rule knowledge into testable building blocks and sharply contrasts divergent assertions.

Using structured records, the process transforms unnoticed friction into manageable objects resolved systematically via a four-zone framework.

This establishes binding decisions with an unalterable history, allowing your leadership team to base complex process decisions on verifiable foundations.

On this secure foundation, the following chapter clarifies how the qualities of Intelligence and Capabilities unfold in daily operations based on reliable self-knowledge.