Monday, September 21, 2026

AI, Oullim and Validation


Conceptual impressions surrounding this post have yet to be substantiated, corroborated, confirmed or woven into a larger argument, context or network. Objective: To generate symbolic links between scientific discovery, design awareness and consciousness. 

"DAC8" does not appear as a widely published, singular named framework in the existing literature. It is treated here as a speculative-philosophical proposition, synthesized from ontological design theory, consciousness studies, and thermodynamic philosophy of mind. The essay constructs this framework rigorously from established sources. 
* * *

The central issue: Oullim cannot validate itself 
Within DAC8, feedback returning through Oullim should not be treated as automatically truthful, coherent, or morally authoritative. Feedback is always selected, measured, interpreted, and narrated by someone. Consequently, the essential question is not merely: “What did the system report?” 

It is: “Who defined the system, selected its indicators, interpreted its consequences, and possessed the power to exclude competing interpretations?” 

If the institution that designed an intervention also controls its measurement and public narrative, the feedback loop is epistemically closed. It may produce internal consistency, but internal consistency is not necessarily genuine coherence. It can be manufactured through selective measurement, suppressed testimony, narrow system boundaries, or definitions of success that protect the institution. 

In DAC8 terms, this is a failure of the Observer/Source Gate: the institutional observer has mistaken its own perspective for the whole field. 

1. Who has the authority to determine which consequences count? 
No single participant should possess complete authority. DAC8 requires differentiated authority because different judgments require different forms of competence and legitimacy. 

Form of authority: Experiential
Proper holder: People directly affected
What it determines: What was expected as benefit, injury, exclusion, dignity, burden or loss.

Form of authority: Empirical 
Proper holder: Independent researchers and auditors
What it determines: What occurred, with what frequency, magnitude and causal confidence.

Form of authority: Technical
Proper holder: Practitioners and subject specialists
What it determines: Whether the intervention operated according to relevant professional knowledge

Form of authority: Ethical
Proper holder: Affecting communities, ethicists, and rights-bearing institutions.
What it determines: What consequences are permissible, irrespective of efficiency.

Form of authority: Ecological/systemic authority
Proper holder: Long-term and cross-system analysis
What it determines: Effects on environments, future generations and adjacent systems

Form of authority: Democratic
Proper holder: Legitimately constituted public process
What it determines: Which trade-offs society is prepared to authorize

Form of authority: Decision
Proper holder: An accountable governing body
What it determines: What action will  be taken

Form of authority: Review
Proper holder: Independent oversight or appeals body
What it determines: Whether the evidence and decision process were fair and contestable

These authorities overlap, but they should not collapse into one another. Expertise does not confer the right to decide another person’s acceptable level of harm. Personal experience does not, by itself, establish population-wide causality. Democratic approval does not make basic rights disposable. Institutional responsibility does not entitle the institution to evaluate itself without external scrutiny. 

The OECD’s evaluation principles similarly emphasize independence, impartiality, credibility, usefulness, partnership and transparency. Its quality standards require evaluators to be independent from the intervention’s policy, operations and management functions, while conflicts of interest must be disclosed and stakeholders must be able to contribute (OECD, 2010; OECD, 2023). 

Therefore: Those who exercise the greatest power over the intervention should not possess the greatest power to define whether it succeeded. 

2. Which consequences count? 
A consequence counts when it is materially connected to the intervention, not merely when it supports the intervention’s declared purpose. DAC8 should recognize at least five classes of consequence: 

1. Intended consequences ... the results the intervention was designed to produce. 
2. Unintended consequences ... beneficial or harmful effects outside its stated objectives. 
3. Distributional consequences ... who benefited, who paid, and whether benefits and burdens were unequally distributed. 
4. Silenced consequences ... effects excluded because affected people lacked access, status, language, data or institutional credibility. 
5. Delayed and systemic consequences ... effects appearing later, elsewhere, or at another organizational or ecological scale. 

The system boundary cannot be set solely by the designer. Otherwise, the designer can simply define inconvenient consequences as “external,” “anecdotal,” or “outside the project’s scope.” 

The OECD’s broader evaluation framework asks about relevance, coherence, effectiveness, efficiency, impact and sustainability. These criteria are useful precisely because an intervention can be efficient without being just, effective in the short term while unsustainable, or internally coherent while conflicting with other social purposes (OECD evaluation criteria). 

3. Whose interpretation enters the field? 
Every materially affected interpretation should be able to enter the field, but entering the field does not mean that every claim is automatically accepted as equally accurate. 

DAC8 needs two simultaneous commitments: 
Interpretive inclusion: no relevant voice is excluded merely because it is inconvenient or lacks institutional power. 
Evidentiary discipline: claims remain open to corroboration, contradiction, contextualization and revision.

 This distinction protects DAC8 from two opposite failures: 

- Technocratic closure: only institutional experts are considered credible. 
- Indiscriminate relativism: every interpretation is treated as equally supported regardless of evidence. 

Participatory evaluation includes stakeholders in defining questions, interpreting evidence and judging results, not merely as respondents from whom data are extracted (BetterEvaluation). Yet participation must carry real influence. Consultation without the ability to contest indicators, inspect evidence or change a decision is ceremonial participation rather than shared authority. 

4. Must the decision be synchronistic, synergetic, neither or both? 
It need not be synchronistic, and it cannot be justified by synchronicity alone. 

Synchronicity 
Synchronicity may identify a meaningful correspondence: an unexpected convergence among testimony, events, symbols or patterns. Within DAC8, it can function as a signal for inquiry. It should not function as proof of causation or as a mandate for policy. A synchronistic event can suggest: 

“This relationship may deserve attention.” It cannot, by itself, establish: “Therefore this intervention is correct, ethical or effective.” 

Synergy 
Synergy concerns the behavior of the whole that cannot be adequately inferred from isolated parts. In evaluation, it helps reveal interactions, emergent benefits and compound harms. 

But synergy is not automatically good. A coercive institution can operate synergistically; its surveillance, incentives, messaging and enforcement may reinforce one another extremely effectively. Synergy describes integrated systemic performance, not moral legitimacy. 

Oullim 
Oullim should name the ethically governed relation between differentiated elements. It is not mere harmony, agreement or efficiency. It requires that the elements remain distinguishable and capable of answering one another. 

Thus the preferred DAC8 relationship is:

Synchronicity may disclose a possible meaning; synergy may reveal a systemic interaction; Oullim tests whether that interaction can become truthful, ethical, adaptive and mutually sustainable. 

The decision may therefore involve both synchronicity and synergy, but it must be warranted by neither alone. Its authority must come from evidence, participation, ethical limits, accountability and revisability. 

5. An Oullim protocol for determining direction 
The returning feedback should pass through all eight DAC8 gates: 

DAC8 gate                     Governing question 

ONTOLOGY What and who actually belongs within the system? Who or what has been omitted? 
EPISTEMOLOGY How do we know? Who collected the evidence, and what evidence could disconfirm the preferred conclusion? 
CREATIVITY What alternatives exist beyond continuing or terminating the present intervention? 
CAUSALITY Did the intervention cause the result, merely accompany it, or benefit from external conditions? 
TEMPORALITY Are apparent benefits immediate while harms are delayed, cumulative or transferred to the future? 
DYNAMICS How are power, incentives, resistance, adaptation and feedback changing the system? 
SEMIOSIS Who controls the language of success, failure, normality, risk and evidence? 
STRUCTURE Are measurement, governance, funding, appeal and correction institutionally independent? 

The Observer then does not pronounce a solitary verdict. It integrates the differentiated findings and selects among five provisional directions: 

- Continue ... evidence supports benefit and no unresolved threshold harm exists. 
- Modify ... the basic intervention remains warranted, but its design, scale or safeguards must change. 
- Pause ... uncertainty or emerging harm is too significant for responsible continuation. 
- Reverse or terminate ... harms, rights violations or structural defects outweigh plausible benefits. 
- Escalate to independent review ... the evidence or decision-making structure is compromised. 

The direction should always include: 
1. The evidence on which it rests.
2. The values and priorities used. 
3. The burdens accepted and by whom. 
4. The uncertainty that remains. 
5. A record of unresolved dissent. 
6. A date or condition for reassessment. 
7. A defined authority capable of stopping or reversing the intervention. 

Contemporary evaluation guidance similarly recommends an explicit theory of change, examination of underlying assumptions, consideration of negative causal pathways, attention to feedback loops and continuing revision as new evidence emerges (HM Treasury, 2026). 

6. When is coherence genuinely restored? Coherence is not restored merely when conflict subsides. Silence may mean fear, exhaustion, exclusion or resignation. Nor is coherence proven by unanimous agreement. In DAC8, genuine coherence is restored when the following conditions exist together: 

- Epistemic coherence: claims are consistent with openly inspectable evidence.
- Experiential coherence: the official account remains recognizable to those who experienced the consequences. 
- Ethical coherence: benefits were not purchased through hidden or impermissible harm. 
- Structural coherence: those who report failure are protected, heard and able to appeal. 
- Temporal coherence: improvement persists beyond the reporting period. 
- Systemic coherence: apparent success in one domain has not merely displaced costs elsewhere. 
- Adaptive coherence: the institution demonstrates an actual capacity to change in response to valid feedback. 
- Narrative coherence: minority reports, uncertainty and negative findings remain visible rather than being absorbed into a triumphant institutional story. 

A concise DAC8 test would be:
 
"Genuine Coherence"="Evidence"+
"Inclusion"+"Ethical Legitimacy"+"Corrective Capacity"
-"Coercion"-"Suppressed Consequences" 

This is conceptual rather than a literal quantitative equation. 

7. How would everyone know coherence had been restored? 
They might not all agree, and universal agreement should not be the test. The more defensible standard is publicly contestable recognition. Participants should be able to see: 

- the original baseline and success criteria; 
- raw or sufficiently disaggregated findings; 
- who selected the indicators; 
- who funded and performed the evaluation; 
- contrary evidence and minority interpretations; 
- changes made because of feedback; 
- unresolved harms and uncertainties; 
- the next review point and available appeal mechanism. 

Coherence becomes credible when even critics can say: 
“I may not endorse the final decision, but my experience was represented accurately, the evidence was not controlled by one interest, the reasoning is visible, and the decision remains open to correction.” 

That is stronger than consensus. It is legitimate disagreement within a trustworthy process. 

8. When the designer controls measurement and narrative 
This arrangement creates a circular authority structure: Such a loop can imitate Oullim while preventing genuine return. Feedback returns only after being translated into terms acceptable to the institution. The remedy is not necessarily to exclude the institution; it possesses relevant operational knowledge. The remedy is to separate powers: 

- Indicators are co-defined before implementation. 
- Baseline conditions and failure thresholds are recorded in advance. 
- Data are accessible to independent evaluators, subject to privacy protections. 
- Affected groups can introduce their own indicators and evidence. 
- Evaluation funding and appointments are insulated from program management. 
- Results include dissenting or minority reports. 
- Material harms trigger independent review rather than internal public-relations management. 
- Decision makers publish a response explaining what changed, what did not, and why. 
- Whistleblowers and participants are protected against retaliation. 
- Periodic reassessment examines delayed and displaced consequences. 

OECD standards explicitly call for cross-validation of information, disclosure of evaluation-team independence, open acknowledgment of conflicts of interest and freedom from interference (OECD DAC Quality Standards).

* Not to be confused with the DAC8 Model of Energy in Motion, C.G. Garant

Conclusion 
In a mature DAC8 system, the Observer is not the sovereign voice at the center. The Observer is the responsible integrator of a field it does not own. 

Oullim therefore does not mean that everything has returned to agreement. It means that differentiation has been preserved, power has been made visible, evidence can circulate without institutional capture, harms cannot be narrated out of existence, and the system remains capable of changing itself

Where the designer controls the intervention, measurements and narrative, coherence has not yet been demonstrated. At most, the institution has demonstrated coherence with its own account of itself. Genuine Oullim begins only when the institution risks discovering, through independent evidence and empowered participation, that its account may be wrong. 

References 

- BetterEvaluation. (2021). Participatory evaluation. https://www.betterevaluation.org/methods-approaches/approaches/participatory-evaluation - HM Treasury. (2026). The Magenta Book: Central government guidance on evaluation. UK Government.  
- Organisation for Economic Co-operation and Development. (2006). DAC evaluation quality standards.  
- Organisation for Economic Co-operation and Development. (2010). Quality standards for development evaluation
- Organisation for Economic Co-operation and Development. (2021). Applying evaluation criteria thoughtfully. OECD Publishing.  Organisation for Economic Co-operation and Development. (2023). Evaluation systems in development co-operation 2023. OECD Publishing. 

The author generated some of this text in part with ChatGPT 5.2 OpenAI’s large-scale language-generation model. Upon generating draft language, the author reviewed, edited, and revised the language to their own liking and takes ultimate responsibility for the content of this publication.

* * *


* * *


"To believe is to accept another's truth.
To know is your own creation."
Anonymous





Edited: 
Find your truth. Know your mind. Follow your heart. Love eternal will not be denied. Discernment is an integral part of self-mastery. You may share this post on a non-commercial basis, the author and URL to be included. Please note … posts are continually being edited. All rights reserved. Copyright © 2026 C.G. Garant. 

No comments: