Codifying Fault in the Negotiation Phase: Article 121 of the New UAE Civil Code Against the Dutch and Swiss Models

I. The reform and what it actually changes

On 1 June 2026, Federal Decree-Law No. 25 of 2025 (the "New Civil Code") came into force, repealing Federal Law No. 5 of 1985 in its entirety and applying, by virtue of Article 4(1), only to contracts entered into on or after that date. The reform is prospective, not retrospective, and the pre-contractual provisions therefore govern negotiations beginning under the new regime rather than reopening concluded transactions.

Under the 1985 Code, good faith was anchored almost exclusively at the performance stage. Old Article 246 required a contract to be performed in accordance with its contents and consistently with the requirements of good faith. It said nothing about how parties should behave before signature. Pre-contractual fault, where it was actionable at all, was channelled through the general law of tort — the "act causing harm" — rather than through any dedicated doctrine governing negotiations. The result was a grey zone: liability was conceivable but its source, content, and limits were uncertain.

Article 121 closes that gap by importing culpa in contrahendo into Emirati statutory law. It does four things. It subjects the initiation, conduct, and termination of negotiations to the requirements of good faith (121(1)); it confirms that negotiating creates no obligation to conclude (121(2)); it makes a party who negotiates or terminates in bad faith liable for the other party's actual damage, while expressly excluding the expected benefits of the unconcluded contract and the lost opportunities to achieve them, unless otherwise agreed (121(3)); and it deems the deliberate concealment of material information affecting the validity of the contract to be bad faith (121(4)). Articles 122 and 123 round out the framework with a standalone disclosure obligation and a confidentiality duty respectively.

The interpretive question Article 121 leaves open is the one that matters: will the Emirati courts read "bad faith" narrowly, confining liability to genuinely blameworthy conduct, or broadly, treating the negotiation phase as an arena of pervasive duties? The Dutch and Swiss systems — two mature civilian jurisdictions that have lived with the doctrine for decades — frame the available answers, and they frame them very differently.

II. The Dutch model: a staged doctrine that has narrowed under its own weight

Dutch law does not codify pre-contractual liability. It is judicial in origin, rooted in the principles of reasonableness and fairness (redelijkheid en billijkheid) now reflected in Articles 6:2 and 6:248 of the Civil Code, and in the tort provision of Article 6:162. The point of departure is Baris/Riezenkamp (HR 15 November 1957), in which the Hoge Raad held that parties who enter negotiations stand in a legal relationship governed by good faith.

The framework most practitioners know comes from Plas/Valburg (HR 18 June 1982), which divided the negotiation process into three stages:

1. an initial stage, in which a party may break off freely and without liability;

2. an intermediate stage, in which breaking off remains permitted but the withdrawing party must reimburse the counterparty's reliance costs — the negative interest; and

3. a final stage, in which breaking off is "unacceptable," exposing the withdrawing party to reliance damages and, in some cases, lost profits — the positive interest — and even to an order to continue negotiating.

That structure is still taught, but it no longer describes the law as applied. In CBB/JPO (HR 12 August 2005), the Supreme Court restated the doctrine in markedly more restrictive terms. The governing test is now that each party remains free to break off negotiations unless doing so would be unacceptable, judged by reference to the counterparty's justified confidence that a contract would be concluded or by other circumstances of the case — and that this question must be assessed by a "strict standard requiring restraint." Crucially, the Court did not reaffirm the intermediate stage; the live distinction collapsed largely into the first stage (free to withdraw) and the third (withdrawal unacceptable), with the burden on the disappointed party set high.

The trajectory since 1982 has therefore been away from liability and toward negotiation freedom. The "second stage" became a long-running doctrinal controversy precisely because the Hoge Raad never reconfirmed it. The position has only recently been complicated again: in a 2024 decision, following Advocate-General Hartlief's opinion, the Court accepted that a duty to reimburse certain costs may arise even where the breaking off was not unacceptable, grounding that duty in unjust enrichment (Article 6:212) rather than in the unacceptability standard. The practical lesson is that the staged model is best read not as a fixed ladder but as a set of distinct grounds — freedom to withdraw, a residual restitutionary duty, and the high-threshold unacceptability rule — each with its own standard.

Two features of the Dutch system bear directly on the UAE question. First, at its apex the doctrine is genuinely expansive: it can compel lost-profit recovery and even an order to keep negotiating. Second, that expansiveness is heavily disciplined in practice by the "strict and restrained" standard and routinely defeated by reservations. A "subject to contract" or condition-precedent clause (opschortende voorwaarde) prevents legitimate reliance from forming and, with it, forecloses positive-interest liability. The doctrine is broad in theory and narrow in operation.

III. The Swiss model: high threshold, fault, and negative interest only

Swiss law begins, like the UAE, from the primacy of freedom of contract: a party is free to enter and to abandon negotiations, and need not justify a withdrawal. That freedom is qualified by the principle of good faith in Article 2 of the Civil Code (ZGB), from which the courts have developed culpa in contrahendo as an uncodified, case-law doctrine. Entering negotiations creates a relationship of trust that imposes reciprocal duties: to negotiate seriously, to disclose genuinely essential information the other party cannot reasonably obtain itself, to show mutual regard, and not to endanger the other's person or property.

Three features distinguish the Swiss approach and make it the conservative pole of the comparison.

The first is the locus of wrongdoing. Swiss courts are explicit that the actionable fault lies not in the act of breaking off, but in having created or sustained the belief that the contract would certainly be concluded — and in failing to dispel that belief in time. Long negotiations do not establish liability. Knowledge that the counterparty has incurred substantial costs does not establish liability. Costs incurred before signature are, in principle, borne at one's own risk. As the Federal Supreme Court confirmed in 4A_313/2019 (judgment of 19 March 2020), the more unreasonable the position taken by the disappointed party, the harder it is to fix liability on the party who walked away.

The second is fault. Culpa in contrahendo requires culpability — at minimum negligence — on the part of the defaulting party, and the claimant must prove it. This is a true fault standard, not an objective unacceptability test.

The third is the measure of recovery. A successful claim yields the negative interest only: the claimant must be placed in the position it would have occupied had negotiations never taken place. That can include wasted negotiation expenses and income lost because the claimant forwent dealings with third parties, but it never extends to the positive interest, and a Swiss court will not order the parties to conclude the contract.

The Swiss model is thus the inverse of the Dutch apex: no compulsion to deal, no lost profits, and a culpability-based threshold that the courts apply with evident reluctance.

IV. Reading Article 121 against the two models

The instinct to ask whether the UAE will follow the Dutch or the Swiss path is sound, but it conflates two questions that Article 121 treats separately: the measure of damages, which the statute settles, and the threshold of liability, which it leaves to the courts.

The remedy question is already decided — and decided narrowly

Article 121(3) confines recovery to actual damage and expressly excludes both the expected benefits of the unconcluded contract and the lost opportunities to achieve them. In comparative terms this is decisive. It adopts the negative interest and statutorily forecloses the two most expansive features of the Dutch apex: positive-interest recovery and any remedy resembling an order to continue negotiating. On the remedy axis, the UAE has aligned itself with Switzerland — and, as it happens, with the codified French position in Article 1112 of the French Civil Code, which likewise excludes recovery of the benefits expected from the lost contract. That lineage is not coincidental. The Emirati civil law tradition descends, through the Egyptian codification, from French civil law, and the resemblance of Article 121(3) to Article 1112 reflects that inheritance rather than a borrowing from Dutch jurisprudence.

The practitioner's conclusion is that the headline risk under Article 121 is bounded. A disappointed counterparty cannot, on the face of the statute, recover the profit it would have made had the deal closed. Its claim is for thrown-away cost and, arguably, for quantifiable opportunities forgone in reliance — closer to the Swiss negative interest than to anything the Dutch doctrine permits at its outer limit.

The threshold question turns on the word "bad faith"

Where the statute is open-textured is on what conduct triggers liability. Here the drafting choice is significant. Article 121(3) does not impose liability for conduct merely falling short of good faith, nor for "unreasonable" or "unacceptable" termination in the objective Dutch sense. It requires bad faith — soo' niyyah. That term carries a connotation of culpability and improper purpose, and it maps more naturally onto the Swiss fault-based inquiry than onto the objective unacceptability standard of CBB/JPO. A court applying Article 121 faithfully to its language should be asking whether the withdrawing party acted with a blameworthy state of mind — feigning serious interest, manufacturing or sustaining a false confidence, exploiting the negotiation to extract value with no intention of closing — rather than whether, viewed objectively, the negotiations had simply advanced too far for withdrawal to be tolerable.

If that reading prevails, Article 121 will operate closer to the Swiss high-threshold model than to the Dutch one: the breaking off of negotiations will not itself be wrongful, however advanced the discussions or however substantial the counterparty's sunk costs, absent conduct that can fairly be characterised as dishonest or improperly motivated. That is the more defensible construction, and the one most consistent with Article 121(2)'s express preservation of the freedom not to conclude.

The disclosure limb is where genuine expansion is most plausible

The one dimension on which Article 121 may prove broader than the cautious Swiss baseline is non-disclosure. Article 121(4) deems the deliberate withholding of material information affecting the validity of the contract to be bad faith, and Article 122 builds a freestanding obligation to disclose information of decisive importance to the other party's consent — an obligation that is non-waivable, that voids any clause purporting to exclude it, and that opens the door to annulment of the resulting contract. Swiss law, by contrast, recognises a pre-contractual disclosure duty only in narrow circumstances, principally where the disadvantaged party cannot reasonably inform itself and the other party knows the information is decisive.

The interaction between 121(4) and 122 will need to be worked out. They overlap but are not co-extensive: 121(4) is tied to information affecting the validity of the contract and feeds the culpa in contrahendo damages claim, whereas 122 addresses information decisive to consent and carries its own remedy of annulment. In high-value transactions this is where the negotiation record will most often become part of the pleaded case — and where the analytical centre of gravity of pre-contractual disputes is likely to settle.

V. Three structural issues the courts will have to resolve

Characterisation. Article 121 liability is best understood as statutory or tortious rather than contractual — consistent with the pre-2026 treatment of pre-contractual fault as an act causing harm. The characterisation is not academic: it governs the applicable limitation period, the rules on causation, and the analytical route to quantifying actual damage. Both comparators have wrestled with the same problem — the Dutch doctrine straddles tort and good faith, while Swiss culpa in contrahendo is treated as a sui generis head of trust-based liability between contract and tort — and Emirati courts will have to take a position.

The dispositive nature of Article 121(3). The words "unless otherwise agreed" make the damages rule a default that the parties may displace. This stands in deliberate contrast to Article 122, whose disclosure obligation is mandatory and cannot be excluded. The asymmetry is consequential. It means the recoverable measure under Article 121 can be enlarged or restricted by agreement — through term sheets, letters of intent, exclusivity undertakings, break-fee provisions, and cost-allocation clauses — whereas the disclosure duty cannot be contracted away. Private ordering therefore becomes the primary instrument of pre-contractual risk allocation, exactly as the Dutch "subject to contract" reservation and standard Swiss practice of disclaiming binding intent operate in their respective systems. Counsel negotiating under onshore UAE law from June 2026 should treat the pre-contractual documents as the place where Article 121 risk is actually managed, while taking care that disclaimers are not drafted so as to appear to waive the non-waivable Article 122 duty.

The onshore / free-zone divide. Article 121 is onshore UAE law. It does not govern transactions subject to the law of the ADGM or the DIFC, which sit in the common-law tradition. The ADGM applies English law directly, under which there is no general duty to negotiate in good faith and a bare agreement to negotiate in good faith is unenforceable (Walford v Miles [1992]); the DIFC regime recognises good faith only in a more limited, codified way and relies on doctrines such as misrepresentation and negligent misstatement to police pre-contractual conduct. The consequence is that the same negotiating behaviour can carry materially different legal risk depending on the governing law and forum. For transactions with an onshore nexus, the choice of governing law and seat is now, on this issue, potentially outcome-determinative — a point of direct importance to any cross-border deal or arbitration touching the UAE.

VI. Conclusion

Article 121 should not be read as an open-ended invitation to litigate failed negotiations. On the question that most exercises commercial parties — what can be recovered — the statute has already chosen the narrow path, excluding lost profits and lost opportunity and limiting the claimant to actual damage. That places the UAE alongside Switzerland and France, and below the Dutch ceiling. On the question it leaves open — what conduct is actionable — the legislator's use of "bad faith" rather than an objective standard of unacceptability points, on the better view, toward a culpability-focused threshold of the Swiss kind rather than the objective, stage-based Dutch inquiry. The Dutch experience is nonetheless instructive in a different way: it shows that even a doctrine that is broad on paper is disciplined in practice by a strict standard and routinely neutralised by well-drafted reservations.

The area in which Article 121 is genuinely capable of broadening liability is disclosure, where the deeming provision of 121(4) and the mandatory regime of Article 122 go further than the Swiss baseline and create a remedy — annulment — with no analogue in the negotiation-breakdown case law of either comparator. That, rather than the breaking off of negotiations as such, is where the contested ground will lie. How narrowly or broadly Article 121 ultimately operates will depend less on the bare text than on whether the Emirati courts hold the line on "bad faith" as a true culpability standard, and on how they reconcile the culpa in contrahendo damages claim with the separate, mandatory disclosure architecture sitting beside it.