Civil Law and UAE Epistemic Closure of Legal Decision-Making Loops

1. Meaning of “Epistemic Closure of Legal Decision-Making Loops”

“Epistemic closure of legal decision-making loops” is not a formally named doctrine of UAE civil law. It is an analytical concept describing a legal system in which the court:

facts → evidence → expert analysis → legal interpretation → judicial reasoning → judgment → review/appeal → correction of factual or legal errors

forms a relatively closed decision-making cycle.

The concept becomes especially important when civil disputes involve technical evidence, digital records, artificial intelligence, expert reports, algorithms, financial models, engineering evidence, blockchain records, or complex commercial transactions.

In the UAE, this process is supported by the modern framework of the Law of Evidence in Civil and Commercial Transactions, the Civil Procedure Code, specialist judicial structures, and increasingly technology-oriented court procedures. The UAE's 2026 Civil Transactions Law also emphasizes coherence and clarity in the application of civil-law rules and expressly allows judicial reasoning through specified sources where no applicable legislative provision exists.

In simple terms, epistemic closure asks:

How does the legal system obtain knowledge, test that knowledge, convert it into legal conclusions, and provide mechanisms for correcting mistakes?

2. Epistemic Structure of UAE Civil Adjudication

A simplified model is:

Raw facts
↓
Evidence
↓
Authentication and admissibility
↓
Expert/technical evaluation
↓
Contradictory submissions by parties
↓
Judicial assessment
↓
Application of legal rules
↓
Reasoned judgment
↓
Appeal/Cassation or other review
↓
Final legal determination

The important feature is that the expert, algorithm, database, or technological system does not legally replace the judge.

The information-producing component supplies material for decision-making; the judicial institution retains responsibility for determining the legal consequences.

This distinction is particularly visible in UAE evidence law concerning expert evidence.

3. Evidence as the First Stage of the Knowledge Loop

The UAE Federal Decree-Law No. 35 of 2022 on Evidence in Civil and Commercial Transactions establishes mechanisms through which technical knowledge enters litigation.

Expert evidence is particularly significant because modern civil disputes frequently involve matters outside ordinary judicial knowledge.

For example:

• construction defects;
• engineering calculations;
• accounting;
• valuation;
• banking transactions;
• cybersecurity;
• electronic records;
• medical causation;
• financial loss;
• intellectual property;
• digital assets.

The court can therefore obtain specialist information without transferring its ultimate adjudicative function to the expert.

The legislation also contemplates circumstances in which the court can rely on an expert report from another proceeding and regulates the treatment of expert reports and expert remuneration.

4. Expert Evidence Does Not Automatically Close the Legal Question

A critical principle is the difference between:

technical knowledge and legal judgment.

An engineer may determine:

“The structural defect was caused by inadequate reinforcement.”

But the legal question may be:

“Does this establish contractual breach or civil liability under the applicable law?”

The second question belongs to the adjudicative process.

This distinction was illustrated in Brookfield Multiplex Constructions LLC v DIFC Investments LLC & DIFC Authority [2016] DIFC CFI 020. The DIFC Court discussed the relationship between court-appointed experts and substantive adjudication, observing that an expert's report in the non-DIFC Dubai system was not itself determinative of substantive rights.

Significance

The epistemic loop therefore operates as:

expert knowledge → judicial evaluation → legal conclusion

rather than:

expert knowledge → automatic judgment.

5. Case Law

Case 1 — Brookfield Multiplex Constructions LLC v DIFC Investments LLC & DIFC Authority [2016] DIFC CFI 020

This is an important UAE-related authority concerning the institutional boundary between expert evidence and judicial determination.

The dispute involved technical construction issues and an arbitration agreement. The DIFC Court considered the use of a court-appointed expert and recognized that the expert's technical investigation did not itself determine the substantive rights and liabilities of the parties.

Principle

The case illustrates that:

1. experts generate technical knowledge;
2. courts evaluate that knowledge;
3. substantive liability remains a judicial/arbitral determination;
4. an evidentiary report does not automatically become the legal judgment.

Epistemic significance

The case prevents premature closure of the decision loop.

The system does not treat technical expertise as equivalent to legal adjudication.

6. Case 2 — Fidel v Felecia & Faraz [2015] DIFC CA 002

In Fidel v Felecia & Faraz, the DIFC Court of Appeal considered whether UAE law outside the DIFC should be established through expert evidence.

The case is particularly useful for understanding the distinction between legal knowledge and technical knowledge.

The Court discussed the difficulty that arises when legal experts move beyond explaining legal principles and begin effectively deciding how those principles should apply to the facts.

Principle

An expert should assist the adjudicator rather than assume the adjudicator's function.

Epistemic significance

The case demonstrates a boundary in the knowledge loop:

expert → informs

but

judge/tribunal → decides.

This is particularly important in technologically complex disputes because an AI system or technical expert may generate highly sophisticated conclusions without possessing legal authority to determine liability.

7. Case 3 — Nihan v Nicholas & Niaz [2024] DIFC CA 012

Nihan v Nicholas & Niaz [2024] DIFC CA 012 provides another useful illustration.

The case involved arguments concerning UAE public policy and arbitrability, supported by expert evidence concerning UAE law.

The Court considered the expert report and examined whether the evidence actually established the proposition being advanced. The first-instance judge had not been satisfied that the expert report demonstrated that enforcement would violate UAE public policy. The Court of Appeal treated the questions of arbitrability and public policy as distinct legal questions.

Principle

An expert opinion does not become authoritative merely because it contains a sophisticated legal analysis.

The adjudicator must examine:

• the legal sources relied upon;
• the reasoning connecting those sources to the proposition;
• whether the proposition actually follows from the law;
• whether competing interpretations exist.

Epistemic significance

This represents a verification stage within the decision loop.

The court does not simply receive information; it tests the information.

8. Case 4 — Korek Telecom Company LLC v Iraq Telecom Ltd [2024] DIFC CA 016

In Korek Telecom Company LLC v Iraq Telecom Ltd [2024] DIFC CA 016, expert evidence concerning UAE law included an alleged UAE “act of state” doctrine.

The Court considered submissions that the expert testimony lacked adequate reasoning or authority and noted the absence of a demonstrated basis for the claimed doctrine in UAE law.

Principle

A legal proposition must be supported by identifiable legal authority and reasoning.

Epistemic significance

This illustrates an important anti-error mechanism:

assertion → source verification → legal evaluation → acceptance or rejection.

Thus, legal decision-making is not supposed to become closed merely because an expert or litigant presents a proposition with confidence.

9. Case 5 — VTJ Limited v Mohammed Ammar Al Hassan [2018–2019] DIFC proceedings

In VTJ Limited v Mohammed Ammar Al Hassan, the DIFC Court considered an application seeking to reopen an appeal based partly upon an expert report and witness evidence concerning an allegedly forged or questionable document.

The Court examined whether the proposed new evidence was sufficiently convincing to justify reopening the final determination and concluded that the necessary threshold had not been established.

Principle

The legal system does not continuously reopen decisions whenever new information is presented.

There must be a legally recognized basis for reopening.

Epistemic significance

This demonstrates the closure stage of the legal loop.

Once a judicial process has reached finality, epistemic correction is constrained by procedural rules.

The system therefore balances:

error correction

against

finality and legal certainty.

10. Case 6 — Eshja v Masri & Others, DIFC CFI 066/2024

A particularly modern illustration is Eshja v Masri & Others, DIFC CFI 066/2024.

The defendants' pleadings had been prepared substantially with the assistance of artificial intelligence and contained false references and misleading material. The Court struck out the relevant defences and counterclaims and imposed costs consequences.

Principle

Technological generation of legal material does not eliminate the responsibility of the human litigant or legal representative to ensure accuracy.

Epistemic significance

This is directly relevant to AI-assisted legal decision-making.

The chain becomes:

AI-generated material → human verification → evidentiary/legal scrutiny → judicial determination

rather than:

AI output → presumed truth.

This is a practical example of a legal system attempting to prevent an unreliable information source from contaminating the judicial knowledge loop.

11. Case 7 — Eshraq Investments PJSC v Shehab M. Gargash & Others [2021] DIFC CFI 077

In Eshraq Investments PJSC v Shehab M. Gargash & Others, the underlying Dubai proceedings involved the appointment of an expert who investigated the substance of a transaction and reported on the parties' positions.

The litigation also demonstrates how expert investigations interact with questions of jurisdiction, evidence, and subsequent judicial determination.

Principle

Technical investigation can assist the court in reconstructing a complicated factual transaction, but the legal consequences remain subject to judicial determination.

Epistemic significance

The case illustrates the movement from:

transactional data → expert reconstruction → judicial interpretation.

12. AI and Epistemic Closure in UAE Civil Law

The concept becomes particularly significant because the UAE judicial environment increasingly incorporates digital technologies.

The DIFC Digital Economy Court framework expressly covers disputes involving:

• artificial intelligence;
• digital assets;
• blockchain;
• substantial databases;
• cloud systems;
• e-commerce;
• digital payment systems.

Its rules also permit electronic dynamic systems and AI-driven decision-tree forms to collect information necessary for the conduct and disposal of claims.

This does not mean that AI independently exercises judicial power.

Instead, it creates an additional information layer:

human evidence + digital evidence + AI-assisted processing → judicial evaluation → legal decision

13. The Risk of “False Epistemic Closure”

One of the most important problems is false closure.

Suppose an AI system produces:

“The defendant caused the loss with 94% probability.”

That statement does not itself establish civil liability.

Several questions remain:

1. Where did the underlying data originate?
2. Was the data authentic?
3. Was the dataset complete?
4. Was the algorithm appropriately designed?
5. Can the parties challenge the methodology?
6. Can the result be independently reproduced?
7. Were alternative explanations considered?
8. Does the applicable legal rule actually use the relevant probability standard?
9. Who is responsible for an erroneous output?

Thus:

algorithmic certainty ≠ legal certainty.

14. Evidence, Contradiction and Procedural Fairness

A healthy epistemic loop requires both sides to have an opportunity to challenge the information used against them.

The UAE evidence framework provides a structured environment for expert evidence, including mechanisms concerning expert reports and expert participation.

This is important because information becomes legally reliable not merely through its production but through adversarial testing.

The basic sequence is:

Evidence presented

↓

Evidence disclosed

↓

Opposing party challenges it

↓

Expert or witness responds

↓

Court assesses competing explanations

↓

Court gives reasons

↓

Review becomes possible

That is epistemic closure with procedural safeguards.

15. Reasoned Judgment as the Closing Mechanism

The judicial judgment is the principal point at which factual knowledge is transformed into a legally binding conclusion.

A properly functioning loop therefore requires a connection between:

evidence → findings of fact → applicable legal rule → legal conclusion → remedy.

If the judgment simply states the conclusion without demonstrating the connection, the corrective function of appeal becomes more difficult.

Consequently, judicial reasoning performs two functions:

Internal function

It assists the judge in organizing the evidence and legal rules.

External function

It allows the parties and reviewing courts to understand why the result was reached.

16. Appeals and Cassation as “Feedback Loops”

Epistemic closure does not mean that the first decision is necessarily the final truth.

Rather, appellate mechanisms provide feedback.

A simplified model is:

Trial court

↓

decision

↓

Court of Appeal

↓

correction / confirmation

↓

Cassation

↓

legal consistency and finality

This is analogous to a control system:

decision → review → correction → stabilized decision.

The legal system therefore incorporates institutional mechanisms for correcting certain factual, procedural, or legal errors while ultimately preserving finality.

17. The 2026 Civil Transactions Law and Legal-System Coherence

The UAE's new Civil Transactions Law, issued in 2026, expressly emphasizes the creation of a comprehensive and integrated civil-law framework, simplification of legal rules, coherence of legal references, and clarity in their practical application.

It also expands judicial reasoning where no applicable legislative provision exists, allowing reference to specified principles of Islamic Sharia and permitting the judge to select an appropriate solution according to the circumstances, justice and public interest within the statutory framework.

From an epistemic perspective, this is significant because the legal system must provide a method for moving from:

known law → legal gap → permissible interpretive source → judicial reasoning → legally justified outcome.

The legal system therefore attempts to avoid an uncontrolled decision-making vacuum.

18. Relationship Between Human Judges and Computational Systems

The future UAE civil-law decision architecture can be represented as follows:

The critical point is that automation can increase informational capacity without necessarily acquiring legal authority.

19. Epistemic Closure and Digital Evidence

Digital evidence creates additional problems because information may be:

• automatically generated;
• continuously updated;
• distributed across jurisdictions;
• stored in cloud systems;
• encrypted;
• generated by algorithms;
• dependent upon third-party platforms.

Consequently, the court must distinguish between:

data existence

and

data reliability.

For example, a blockchain record may demonstrate that a transaction was recorded on a particular ledger, but questions can remain concerning:

• who controlled the relevant wallet;
• whether the underlying transaction was authorized;
• whether the digital identity corresponds to the legal person;
• whether fraud occurred outside the blockchain;
• what contractual obligation the transaction represented.

Thus, technological immutability does not necessarily answer the legal question.

20. Epistemic Closure and Expert Accountability

The UAE's modern evidence framework also recognizes that expert evidence itself requires regulation.

The Evidence Law provides consequences where an expert's negligence or fault affects the report, including possible repayment of amounts received and potential disciplinary and compensation consequences.

This is important because the knowledge loop would be defective if:

expert → unquestionable authority

were the model.

Instead:

expert → regulated participant → scrutinized evidence → judicial evaluation

is the more appropriate model.

21. Main Legal Principles

The UAE framework can therefore be understood through the following principles:

1. Evidence must be legally processed

Raw information does not automatically become legally established fact.

2. Experts assist rather than replace adjudicators

Technical expertise informs the court but does not ordinarily determine legal liability.

3. Legal propositions require authority

Expert opinions concerning UAE law must be tested against legislation and applicable jurisprudence.

4. Parties must be able to challenge information

Epistemic reliability is strengthened through adversarial testing.

5. AI output requires verification

AI-generated pleadings, research, references, or analysis cannot simply be presumed accurate.

6. Judicial reasoning connects evidence to outcome

The court must transform evidentiary material into legally justified conclusions.

7. Appeals provide feedback

Review mechanisms permit correction of legally cognizable errors.

8. Finality eventually closes the loop

Unlimited reconsideration would undermine certainty and enforceability.

22. Major Challenges

A. Algorithmic opacity

If an AI system cannot adequately explain how an output was produced, parties may have difficulty challenging it.

B. Data contamination

Incorrect or biased input can produce apparently sophisticated but unreliable conclusions.

C. Automation bias

Judges or litigants could give excessive weight to computational outputs because they appear mathematically objective.

D. Expert overreach

Experts may unintentionally cross the boundary between technical assistance and legal adjudication.

E. Fragmentation

Federal courts, local courts and specialist jurisdictions such as the DIFC can operate under different procedural frameworks.

F. Rapid technological change

Legal rules can become outdated more quickly than conventional civil-law doctrine.

G. Accountability gaps

Where an AI system contributes to an erroneous legal decision, responsibility may be divided among:

• developer;
• vendor;
• expert;
• lawyer;
• litigant;
• institution;
• judicial user.

23. UAE Model of Epistemic Closure

A useful conceptual model is:

FACTS
↓
EVIDENCE
↓
AUTHENTICATION
↓
EXPERT/TECHNICAL ANALYSIS
↓
CONTRADICTORY TESTING
↓
JUDICIAL EVALUATION
↓
LEGAL INTERPRETATION
↓
REASONED JUDGMENT
↓
APPEAL/CASSATION
↓
CORRECTION OR CONFIRMATION
↓
FINALITY

The UAE's digital-court developments add another layer:

AI / algorithmic processing

but that layer should be understood as an evidentiary or procedural instrument rather than an autonomous source of legal authority.

24. Conclusion

Epistemic closure of legal decision-making loops in UAE civil law describes the way the legal system transforms uncertain information into an authoritative legal determination through evidence, expertise, judicial reasoning and review.

The concept is particularly valuable for understanding modern UAE civil litigation because disputes increasingly contain AI-generated information, digital evidence, blockchain records, complex financial models, engineering evidence and algorithmic analysis.

The cases discussed—Brookfield Multiplex, Fidel, Nihan, Korek Telecom, VTJ, and Eshja, together with Eshraq Investments—illustrate different points in the same broader process: technical information must be distinguished from legal judgment; expert opinions must be scrutinized; legal propositions require authority; new evidence does not automatically reopen final decisions; and AI-generated legal material remains subject to human verification and judicial control.

The central conceptual proposition is therefore:

UAE civil adjudication can incorporate increasingly sophisticated technological and expert sources of knowledge without treating those sources themselves as the final source of legal authority.

That distinction is what keeps the decision-making loop epistemically controlled, procedurally contestable, legally reasoned, and ultimately capable of reaching institutional finality.