Civil Law and UAE: Knowledge Graphs in Civil Law Reasoning

1. Introduction

Knowledge graphs in civil-law reasoning refer to a structured method of connecting legal concepts, statutory provisions, facts, evidence, contractual terms, judicial decisions, parties, remedies, and legal consequences.

In simple terms, instead of treating a civil case as a collection of separate paragraphs, a knowledge graph represents it as a network:

Facts → Legal Relationship → Legal Rule → Evidence → Judicial Interpretation → Liability → Remedy

This approach is particularly relevant in the UAE because civil disputes may require the lawyer or judge to connect Federal legislation, Civil Transactions Law, Evidence Law, Companies Law, Arbitration Law, contractual provisions, judicial principles, and—in appropriate cases—DIFC or ADGM authorities.

The current UAE framework must also be understood in light of Federal Decree by Law No. 25 of 2025, which repealed the 1985 Civil Transactions Law and brought the new Civil Transactions Law into force on 1 June 2026. 

2. Meaning of a Legal Knowledge Graph

A knowledge graph is a network consisting of:

Simple example

Suppose:

A sells machinery to B.

The knowledge graph may look like:

A → seller → B

B → buyer → A

Sale Contract → creates → contractual obligations

A → obligation → deliver machinery

B → obligation → pay price

If A fails to deliver:

Non-delivery → breach → contractual liability → damage → compensation

The graph therefore converts a factual story into a legally structured chain.

3. Why Knowledge Graphs Matter in UAE Civil Law

Knowledge graphs can assist with five major forms of civil-law reasoning.

1. Classification

The system determines whether the dispute concerns:

• contract;
• tort;
• property;
• unjust enrichment;
• company law;
• employment;
• insurance;
• agency;
• construction;
• arbitration; or
• another legal relationship.

2. Rule identification

Once the legal relationship is classified, the relevant legislation can be connected to the facts.

3. Evidence mapping

Each factual proposition can be connected to:

• contracts;
• emails;
• WhatsApp messages;
• invoices;
• expert reports;
• witness statements;
• electronic records;
• admissions.

4. Precedent mapping

Judgments can be connected according to the legal proposition they establish or discuss.

5. Remedy mapping

The graph can connect the established breach to possible consequences such as:

• damages;
• specific performance;
• termination;
• restitution;
• injunction;
• interest;
• costs.

4. UAE Civil-Law Knowledge Graph Structure

A useful UAE civil-law graph can be organised into six layers:

Layer 1 — Facts

Examples:

• contract signed;
• payment made;
• goods delivered;
• property transferred;
• accident occurred;
• email sent;
• shareholder withdrew.

Layer 2 — Legal relationships

Examples:

• buyer–seller;
• creditor–debtor;
• employer–employee;
• shareholder–company;
• insurer–insured;
• principal–agent.

Layer 3 — Legal rules

Examples:

• contractual obligations;
• good faith;
• interpretation;
• causation;
• burden of proof;
• damages;
• limitation.

Layer 4 — Evidence

Examples:

• written contract;
• electronic signature;
• bank records;
• expert report;
• correspondence;
• admission.

Layer 5 — Authorities

Examples:

• Civil Transactions Law;
• Evidence Law;
• Companies Law;
• Arbitration Law;
• judicial decisions.

Layer 6 — Consequences

Examples:

breach → liability → compensation

or

jurisdiction clause → jurisdictional analysis → forum determination

5. Knowledge Graph and UAE Civil Transactions Law

The new Civil Transactions Law is especially suitable for knowledge-graph reasoning because civil-law disputes frequently require relationships between intention, contractual language, circumstances, custom, good faith, liability and compensation.

Federal Decree by Law No. 25 of 2025 expressly repealed Federal Law No. 5 of 1985 and brought the new Civil Transactions Law into force on 1 June 2026.

Thus, a contemporary UAE civil-law knowledge graph should distinguish:

Old 1985 Civil Code authority → historical legal proposition

from

2025 Civil Transactions Law → current statutory framework

This distinction is important when analysing older judgments.

6. Knowledge Graph for Contract Interpretation

Contract interpretation can be represented as:

Contract wording

↓

Parties' intention

↓

Nature of obligation

↓

Commercial circumstances

↓

Custom

↓

Good faith

↓

Judicial interpretation

↓

Legal consequence

A knowledge graph therefore prevents the lawyer from looking at one contractual sentence in isolation.

For example:

Clause A → payment obligation

Clause B → delivery obligation

Clause C → termination

Clause D → dispute resolution

These clauses can then be connected to:

performance facts → breach → remedy

7. Case Law 1 — Goel v Credit Suisse

Case

Ashok Kumar Goel & Others v Credit Suisse (Switzerland) Limited [2021] DIFC CA 002

This is a useful authority for knowledge-graph reasoning concerning jurisdiction, contractual wording, party status and surrounding circumstances.

The dispute concerned guarantees governed by Dubai and applicable UAE federal law and contractual language referring to the “Courts of Dubai.” The DIFC Court of Appeal considered the jurisdictional framework and the circumstances relevant to construing that language.

Knowledge-graph significance

The relevant graph is:

Guarantee

→ governing law

→ jurisdiction clause

→ identity/status of contracting parties

→ DIFC Establishment

→ jurisdictional gateway

→ DIFC Court jurisdiction

The important lesson is that jurisdiction cannot always be determined from one isolated phrase. The court may need to connect the contractual language with the surrounding legal and factual structure.

Principle

Contractual wording + legal framework + surrounding circumstances → jurisdictional analysis.

8. Case Law 2 — Rada Trading v Wealth Bridge

Case

Rada Trading LLC FZC v Wealth Bridge Trading Crude Oil and Refined Products Abroad LLC & Cohenrich Energy FZE [2021] DIFC CA 007

This case is particularly useful for a knowledge graph involving electronic communications, contractual variation and evidence.

Graph

Email

→ communication between parties

→ alleged agreement/variation

→ electronic evidence

→ authorship/authentication

→ contractual effect

The reasoning illustrates why digital civil-law disputes cannot be analysed simply by asking whether an email exists.

The graph must connect:

email → sender → authority → content → intention → acceptance → contractual consequence.

Importance

It demonstrates how electronic communications can become part of a broader contractual reasoning network.

9. Case Law 3 — Naveen v Ned

Case

Naveen v Ned [2024] DIFC SCT 068

The dispute concerned an employment relationship and a jurisdictional challenge. The DIFC Court considered the defendant's DIFC status and the contractual/employment relationship in determining jurisdiction. The jurisdictional challenge was dismissed.

Knowledge graph

Employment contract

→ employer

→ DIFC entity

→ employee

→ employment relationship

→ jurisdictional gateway

→ DIFC Courts

The case demonstrates that jurisdiction is a relational concept.

A knowledge graph therefore asks:

Who are the parties?
Where is the entity located?
Where was the relationship performed?
What does the contract say?
What statutory jurisdictional gateway applies?

10. Case Law 4 — Oran, Oaken v Oved

Case

Oran & Oaken v Oved [2025] DIFC CA 004

This case concerned jurisdiction, arbitration and an anti-suit injunction in circumstances involving a contractual dispute-resolution provision referring to DIFC-LCIA Rules after the institutional landscape had changed. The DIFC Court of Appeal allowed the appeals and upheld the jurisdictional objection.

Knowledge graph

The dispute can be represented as:

Contract

→ dispute-resolution clause

→ arbitration reference

→ institutional framework

→ seat

→ court jurisdiction

→ anti-suit relief

→ enforcement consequences

Importance

This demonstrates that one contractual clause can have connections to several legal systems and legal questions.

A knowledge graph helps prevent the common mistake of treating:

arbitration clause = jurisdiction

as an automatic proposition.

Instead:

arbitration clause → interpretation → arbitration agreement → seat → supervisory court → enforcement consequences

11. Case Law 5 — Larmag Holding v First Abu Dhabi Bank

Case

Larmag Holding B.V. v First Abu Dhabi Bank PJSC & Others [2019] DIFC CFI 054

The dispute involved allegations concerning fraudulent inducement and financial transactions. The judgment provides a useful example of connecting representations, conduct, reliance, causation, loss and remedies.

Knowledge graph

Representation

→ made by defendant

→ relied upon by claimant

→ inducement

→ transaction

→ loss

→ causation

→ remedy

This is an important structure for civil-law reasoning because proving a representation alone does not necessarily establish recoverable loss.

The graph must connect every element.

12. Case Law 6 — Credit Suisse Jurisdiction Proceedings

The Credit Suisse litigation is also useful as a separate illustration of multi-node jurisdictional reasoning.

The first-instance proceedings considered whether the DIFC Courts had jurisdiction and examined the contractual arrangements and jurisdictional gateways.

The Court of Appeal subsequently dismissed the appeal against the jurisdiction ruling.

Knowledge graph

Party A

→ DIFC Establishment

Party B

→ guarantor

Guarantee

→ governing law

Guarantee

→ jurisdiction clause

Jurisdiction clause

→ interpretation

Interpretation

→ surrounding circumstances

Statutory gateway

→ DIFC jurisdiction

This illustrates how legal reasoning can be represented as interconnected propositions rather than isolated rules.

13. Case Law 7 — Larmag and Moral Damages

Larmag is also important when analysing the relationship between loss and remedy.

The case is useful in distinguishing compensatory relief from a purely punitive conception of damages.

A graph can therefore contain:

Wrongful conduct

→ non-economic harm

→ proof

→ causal connection

→ compensatory remedy

This is particularly relevant when building AI-assisted civil-law systems because the system should not automatically infer:

wrongdoing → damages

without intermediate nodes concerning actual legally recoverable harm.

14. Knowledge Graph and Evidence

Evidence is one of the most important components of civil-law reasoning.

A good graph separates:

Fact

“Payment was made.”

from:

Evidence

Bank statement showing payment.

from:

Legal inference

Payment may establish partial performance.

from:

Legal consequence

Depending on the contract, partial performance may affect breach or restitution.

Thus:

Evidence ≠ Fact ≠ Legal conclusion.

This distinction is essential for responsible legal AI.

15. Knowledge Graph for Electronic Evidence

For an electronic transaction, the graph may be:

Email/WhatsApp

↓

Account

↓

User

↓

Identity

↓

Authority

↓

Integrity

↓

Content

↓

Contractual intention

↓

Legal effect

For example:

WhatsApp message

→ sender identified

→ sender had authority

→ message authenticated

→ offer communicated

→ acceptance established

→ contract formation potentially established.

The graph should not simply treat the existence of a message as proof of the entire legal proposition.

16. Knowledge Graph and Causation

Civil liability often requires a chain such as:

Conduct

→ breach/wrongful act

→ actual damage

→ causal connection

→ legally recoverable loss

→ compensation

A knowledge graph can therefore identify missing links.

Example

A contractor delays construction.

The graph asks:

Delay

→ Was there contractual breach?

→ Did delay cause loss?

→ Was the loss actually suffered?

→ Was it foreseeable/naturally consequential under the applicable rule?

→ Is there evidence?

→ What compensation follows?

This prevents the reasoning error:

“There was delay, therefore all claimed damages are recoverable.”

17. Knowledge Graph and Contractual Liability

A contractual liability graph can be expressed as:

Valid contract

↓

Obligation

↓

Performance required

↓

Failure/non-performance

↓

Breach

↓

Damage

↓

Causation

↓

Remedy

This is particularly useful in:

• construction disputes;
• franchise disputes;
• supply agreements;
• employment contracts;
• insurance;
• banking;
• real estate;
• technology agreements.

18. Knowledge Graph and Jurisdiction

Jurisdiction is naturally graph-based.

For a UAE dispute:

Parties

→ domicile/incorporation

→ contractual relationship

→ place of performance

→ location of transaction

→ jurisdiction clause

→ arbitration clause

→ DIFC/ADGM/onshore connection

→ statutory jurisdictional gateway

→ competent forum.

This is especially important because DIFC jurisdiction should not automatically be inferred merely because a transaction has some connection with Dubai.

The Goel decision demonstrates the importance of analysing the relevant jurisdictional gateway and contractual language.

19. DIFC, Mainland UAE and Knowledge Graphs

A sophisticated UAE legal knowledge graph should have separate nodes for:

Mainland UAE

• Federal courts;
• local emirate courts;
• federal legislation;
• UAE Civil Transactions Law;
• UAE Evidence Law.

DIFC

• DIFC Courts;
• DIFC laws;
• DIFC jurisdictional gateways;
• DIFC common-law methodology.

ADGM

• ADGM Courts;
• ADGM regulations;
• common-law-based legal framework.

The graph must not treat all three as one undifferentiated legal system.

For example:

Dubai location

does not automatically mean:

DIFC law

and

DIFC Courts

do not automatically mean:

mainland UAE jurisdiction.

20. Judicial Precedent as Graph Nodes

A case can be represented as:

Case

→ facts

→ legal issue

→ applicable statute

→ interpretation

→ holding

→ reasoning

→ remedy.

For example:

Goel

→ jurisdiction

→ DIFC JAL

→ contractual wording

→ surrounding circumstances

→ jurisdictional conclusion.

Rada Trading

→ electronic communication

→ contractual variation

→ electronic evidence

→ legal effect.

Oran/Oaken

→ arbitration

→ jurisdiction

→ dispute-resolution clause

→ institutional consequences.

This allows lawyers to search cases by legal proposition, rather than merely by case title.

21. Knowledge Graph and Legal Research

Traditional legal research often works like:

Search topic → read cases → identify principle.

Knowledge-graph research can work as:

Legal issue → connected statutory provision → connected cases → factual similarities → contrary authorities → evidentiary requirements → remedies.

For example, for contract termination, the graph may retrieve:

termination

→ contractual right

→ statutory right

→ breach

→ notice

→ cure period

→ materiality

→ causation

→ damages

→ restitution.

This produces a more complete legal analysis.

22. Knowledge Graph and Judicial AI

Knowledge graphs can be used in AI-assisted judicial systems, but they should be treated as reasoning-support infrastructure rather than an autonomous decision-maker.

A properly designed system should show:

1. Fact identified
2. Evidence supporting fact
3. Legal rule connected
4. Authority supporting rule
5. Contrary authority
6. Inference made
7. Reason for inference
8. Potential remedy

This improves explainability.

23. Explainability Example

Suppose an AI system concludes:

“The defendant may be contractually liable.”

A transparent knowledge graph should allow the lawyer or judge to trace:

Contract exists

↓ Evidence: signed contract

Obligation exists

↓ Evidence: Clause 5

Obligation not performed

↓ Evidence: delivery records

Breach

↓ Applicable legal rule

Loss occurred

↓ Evidence: expert report

Causation

↓ Evidence + reasoning

Compensation potentially available

This is much more transparent than an unexplained AI-generated conclusion.

24. Knowledge Graphs and Conflicting Authorities

A sophisticated system should also represent disagreement.

For example:

Legal proposition X

→ Authority A supports X

→ Authority B limits X

→ Authority C distinguishes A

→ Later authority modifies interpretation

Therefore, the graph should contain relationships such as:

• supports
• distinguishes
• overrules/replaces where legally applicable
• limits
• follows
• questions
• applies
• does not apply

This is important because UAE civil-law reasoning cannot safely be reduced to a simple list of “precedents.”

Dubai Courts itself describes the Court of Cassation as the highest judicial level in Dubai and notes the role of its technical office in preparing and publishing judicial principles and rules. Dubai Courts also provides a system for accessing judicial principles and precedents.

25. Knowledge Graphs and Legal Precedent in UAE

The UAE legal environment requires careful treatment of precedent.

A knowledge graph should therefore identify:

Court

→ Federal Supreme Court

→ Dubai Court of Cassation

→ Abu Dhabi Court of Cassation

→ DIFC Court of Appeal

→ ADGM Court

and separately identify:

Jurisdiction

→ mainland

→ DIFC

→ ADGM.

This prevents an AI system from treating a DIFC judgment as automatically controlling an onshore UAE civil dispute.

26. Knowledge Graph for a Civil Case

A complete civil dispute can be represented as follows:

PARTIES   ↓ LEGAL RELATIONSHIP   ↓ CONTRACT / TORT / PROPERTY / COMPANY   ↓ FACTS   ↓ EVIDENCE   ↓ LEGAL ISSUES   ↓ STATUTORY RULES   ↓ JUDICIAL AUTHORITIES   ↓ APPLICATION OF RULES   ↓ BREACH / LIABILITY   ↓ CAUSATION   ↓ DAMAGE   ↓ REMEDY   ↓ ENFORCEMENT

This is essentially a legal reasoning graph.

27. Example — UAE Construction Dispute

Assume:

Developer → Contractor

Contract provides:

• completion date;
• extension-of-time mechanism;
• liquidated damages;
• arbitration clause.

The knowledge graph becomes:

Construction contract

→ contractor obligation

→ completion deadline

→ delay

→ cause of delay

→ contractual extension

→ entitlement to extension

→ actual delay

→ contractual breach

→ liquidated damages

→ evidence

→ expert determination

→ arbitration.

The graph can also identify a competing chain:

Employer variation

→ additional work

→ extension entitlement

→ revised completion date

→ no delay during extension period.

Thus, knowledge graphs are useful for handling competing causal explanations.

28. Example — Online Civil Transaction

For an online transaction:

Website

→ terms displayed

→ user account

→ click acceptance

→ electronic record

→ contract

→ payment

→ performance

→ dispute.

The graph must then separately ask:

Which court?

Which law?

Was there an arbitration clause?

Where was the contract performed?

Where are the parties located?

This is why an online transaction should not be analysed simply as:

website located in UAE = UAE court jurisdiction.

29. Benefits of Knowledge Graphs in UAE Civil Law

A. Better issue identification

They reveal connected legal issues that might otherwise be missed.

B. Better precedent research

Cases can be connected by legal proposition rather than merely keywords.

C. Better evidence management

Each factual proposition can be linked to supporting evidence.

D. Better legal drafting

Lawyers can construct pleadings around a structured chain of:

fact → rule → evidence → conclusion.

E. Better explainability

An AI-generated result can be traced back to its underlying authorities.

F. Better conflict detection

The system can identify apparently conflicting rules or judgments.

30. Limitations and Risks

Knowledge graphs are not automatically legally correct.

1. Incorrect extraction

If the system incorrectly extracts a proposition from a judgment, the entire graph may become misleading.

2. Context loss

A case decided on a particular factual basis may not apply to different facts.

3. Outdated legislation

The graph must identify whether an authority relates to the former 1985 Civil Transactions Law or the current 2025 law.

The 2025 legislation expressly repealed the former Civil Transactions Law from 1 June 2026.

4. Jurisdiction confusion

DIFC, ADGM and mainland authorities should not be treated as interchangeable.

5. False certainty

A graph may show that two propositions are connected without proving that the legal conclusion necessarily follows.

6. Missing evidence

A legal conclusion is only as reliable as the evidence supporting its factual nodes.

31. Recommended UAE Civil-Law Knowledge Graph

A comprehensive UAE system could contain these principal node categories:

32. Six Essential Relationships in the Graph

For examination purposes, remember these six:

1. FACT → RULE

A factual event is connected to an applicable legal rule.

2. RULE → AUTHORITY

The rule is supported by legislation or judicial authority.

3. FACT → EVIDENCE

The factual proposition is supported by documentary or other evidence.

4. AUTHORITY → INTERPRETATION

A judgment explains or applies the legal rule.

5. BREACH → LIABILITY

Established breach connects to legal responsibility.

6. LIABILITY → REMEDY

Legal responsibility connects to the appropriate remedy.

33. Practical Legal-Research Method

When analysing a UAE civil case through a knowledge graph, use this sequence:

Step 1 — Identify parties

Who is claimant and who is defendant?

Step 2 — Identify relationship

Contract? Tort? Property? Company? Employment?

Step 3 — Identify facts

What actually happened?

Step 4 — Identify evidence

What proves each factual proposition?

Step 5 — Identify legal issues

What questions must the court answer?

Step 6 — Identify applicable legislation

Which current statutory provisions govern?

Step 7 — Identify cases

Which authorities interpret or apply those provisions?

Step 8 — Identify conflicting authorities

Are there distinctions or different factual contexts?

Step 9 — Apply rules

Connect facts to legal rules.

Step 10 — Determine possible consequences

Damages, termination, injunction, restitution, enforcement, etc.

34. Key Case-Law Lessons

The underlying judgments support these case-specific propositions.

35. Conclusion

Knowledge graphs can provide a structured model for UAE civil-law reasoning by connecting facts, evidence, statutes, contractual provisions, judicial authorities, legal issues, liability and remedies.

Their central value is that they transform legal analysis from:

“Find a case containing the right words.”

into:

“Identify the legal relationship, connect the facts to evidence, connect the facts to the governing rule, connect the rule to authoritative interpretation, and then trace the consequences.”

For UAE civil law, this is particularly important because the legal researcher must distinguish current legislation from historical legislation, mainland UAE courts from DIFC/ADGM courts, statutory rules from judicial interpretation, and factual evidence from legal conclusions. The 2025 Civil Transactions Law, effective from 1 June 2026, makes keeping the statutory layer current especially important.

Quick Revision Formula

Knowledge Graph in UAE Civil Law =

Facts + Evidence + Legal Relationship + Statute + Case Law + Interpretation + Causation + Remedy

Exam line:

A legal knowledge graph represents civil-law reasoning as an interconnected network in which factual propositions, evidence, legal rules, judicial authorities and remedies are linked so that each legal conclusion can be traced back to its supporting facts and authorities.