The Purpose of a System Is What It Does

December 28, 2025

Examples of applying the POSIWID lens.

Introduction

In management cybernetics, there is a well-known formulation: “the purpose of a system is what it does” (POSIWID). It is commonly used as a practical way to analyze complex systems—organizations, institutions, social practices, and recurring patterns of behavior.

The idea is to set aside declared goals, missions, and explanations, and instead observe the actions and effects a system reliably produces over time. Not isolated decisions or stated intentions, but stable patterns—especially those that become visible under stress, growth, or conflicting incentives.

With sustained observation, these patterns tend to reveal a system’s actual direction more clearly than its language does.

Some of the examples below may feel uncomfortable or overly familiar. This is not accidental.

POSIWID tends to become most visible in systems where expectations are high, language is carefully managed, and outcomes matter personally to many people. The purpose of these examples is not provocation, but clarity: they make the underlying mechanics easier to see.

In practice, this often means looking at systems we would prefer to trust at face value.

Organizational Systems

Example 1: User-centered design

An organization declares that user well-being is its primary principle. In theory, internal processes—from product design to performance evaluation—are aligned around this goal.

When examining recurring actions, a different pattern may emerge: mechanisms that encourage impulsive decisions, friction in reversing commitments, frequent prompts optimized for engagement, and interface changes that increase usage without proportional increases in user benefit. These mechanisms persist over time and are actively refined.

Through the POSIWID lens, the system appears oriented toward optimizing measurable business outcomes. The language of user focus remains present and highly visible, but it functions as accompanying justification rather than as a controlling constraint.

This gap is often most noticeable to users precisely because they are repeatedly addressed in the language of care, trust, and alignment.

Example 2: Equal opportunity policies

An organization formally commits to equal opportunity across demographic and cultural dimensions. Policies are documented, compliance is tracked, and official communication remains neutral.

Over time, however, team composition may converge toward greater similarity. Informal networks form, and hiring and promotion decisions increasingly rely on trust, familiarity, and perceived reliability, while formal criteria remain unchanged.

At the system level, this produces a redistribution of influence toward those who are already embedded. At the level of individual decision-makers, the logic is pragmatic: reduced uncertainty, lower coordination cost, and improved predictability. The stated principle remains intact, while operational decisions follow a different optimization path.

Example 3: Employee feedback loops

An organization regularly conducts employee surveys. High workload and poor work–life balance appear as recurring themes. These issues are acknowledged, discussed, and accompanied by commitments to improvement.

At the same time, core structures—staffing models, incentive systems, delivery timelines—remain unchanged.

Taken together, surveys and discussions form a stabilizing cycle. They allow tension to be expressed and redistributed without altering the underlying configuration. Through POSIWID, this suggests that workload reduction is not a primary objective. Maintaining performance under sustained pressure functions as a means toward higher-level goals such as growth or output stability.

Political and Institutional Systems

The following examples are not presented to rank systems morally, but to illustrate how the same analytical lens applies across very different domains. The goal here is not judgment, but consistency of observation.

Example 1: Historical authoritarian regimes

Some authoritarian regimes have declared goals related to social welfare or national renewal. When observing repeated actions, however, patterns such as militarization, concentration of power, suppression of dissent, and systematic violence become dominant—even when these actions undermine stated objectives.

Through POSIWID, ideology appears less as a goal and more as a legitimizing language accompanying a system oriented toward control and domination. For participants, this produces stable effects: redistribution of responsibility, a sense of participation in a larger mission, and personal significance derived from alignment with power.

At a structural level, this dynamic resembles individual forms of coercion, where denying another’s agency becomes a source of perceived control.

Example 2: Progressive governance environments

Some regions, particularly large urban and economically dynamic ones, explicitly describe themselves through the language of equality, care, and social justice. These values are consistently present in public communication and policy framing.

At the same time, long-term outcomes may include rising costs of living, limited housing accessibility, and regulatory structures that disproportionately benefit those already established within the system.

Through the POSIWID lens, this does not necessarily indicate hypocrisy. Rather, it suggests that the system is optimized to preserve stability, limit disruption, and protect existing allocations of resources. The value language helps maintain legitimacy and reduce social tension, while underlying processes continue to follow their own constraints.

Example 3: Symbolic continuity and power concentration

Political actors often appeal to foundational narratives or historical continuity while introducing structural changes that alter the distribution of power. Symbols, rituals, and language remain familiar, while decision-making authority becomes increasingly centralized.

From a POSIWID perspective, symbolic continuity functions to stabilize legitimacy, while repeated actions reveal a shift in the system’s operative purpose—from stewardship to consolidation.

The following examples move away from institutions and toward direct human interaction. Here, the same lens applies, but the mechanisms become visible in a more immediate form.

Religious practices

In some religious contexts, punitive practices are justified through moral or doctrinal explanations. When focusing on form and effect, these practices often involve public, demonstrative control over bodies and behavior.

At the system level, such practices function to constrain autonomy and enforce conformity. At a deeper level, they generate stable experiences of hierarchy and moral superiority. These effects persist independently of the specific justifications offered.

Sexual violence

Contemporary research indicates that sexual violence is weakly associated with sexual desire and strongly associated with the pursuit of control. Repetition, denial of consent, and asymmetry of power are its consistent features.

Viewed through POSIWID, sexual violence can be understood as an attempt to assert causality and agency by overriding another person’s autonomy.

School bullying

Bullying rarely consists of isolated incidents. It is typically public, repetitive, and structured around persistent power imbalances. Over time, violence functions less as an end and more as a means.

At a surface level, it reinforces status. At a deeper level, it restores a sense of visibility and significance through eliciting reaction. The response of the target confirms the interaction’s social reality.

Self-isolation and avoidance

Some forms of self-isolation appear as protection or moral stance. In stable form, they may serve a different function: avoidance of vulnerability and responsibility, preservation of control through non-participation.

In this configuration, moral reasoning becomes a mechanism for maintaining predictability while minimizing exposure to relational risk.

On Basic Needs and Narratives

When explanations are set aside, many stable system behaviors reduce to basic human needs: safety, predictability, access to resources, status, risk management, and a sense of significance.

Decisions are typically made at this level—quickly and contextually. Narratives emerge afterward, integrating those decisions into a coherent worldview and reducing internal tension.

From this perspective, explanations are secondary. They do not guide action; they accompany it. Observing which needs are reliably satisfied—and for whom—often provides a clearer understanding of a system than its stated intentions.

Practical Implications

Recognizing a system’s actual operating goals does not necessarily imply opposition. More often, it removes misplaced expectations and enables more precise interaction.

It also makes it harder to mistake alignment of language for alignment of incentives.

When declared values diverge from observed outcomes, influence through rhetoric is limited. Actions that align with the system’s real optimization targets are more likely to be noticed and supported.

The same applies within organizations. If advancement depends on reliability and risk reduction rather than formal principles, efforts aligned with those dynamics will be more effective.

Conclusion

Viewed through the POSIWID lens, words do not determine direction. They matter only when they align with sustained action and downstream consequences. Otherwise, they coexist with processes without shaping them.

Underlying these processes are basic human needs. They require no justification and no evaluation. They simply find forms in which they can be repeatedly satisfied.

A system does what it is actually built to do. Understanding this makes it possible to see its movement clearly—and to choose how to engage with it.