Why Felicity Ace's Cause Was Never Proven

Because the evidence burned, sank, and was never recorded in the first place. The Regional Court of Stuttgart dismissed MOL and its hull insurers' roughly €30 million claim against Porsche in May 2026 for one reason: the plaintiffs could not show which vehicle ignited. They could not because Felicity Ace's ~4,000 cars burned for two weeks, the hull sank in about 3,000 metres of water on 1 March 2022 before any investigator boarded, and the ship's detection system recorded a zone, not a vehicle.

The pattern is wider than one ship. Of the car-carrier and ro-ro fires this corpus has examined, every one with a proven cause — Höegh Xiamen, Grande Costa d'Avorio, Corona Seaways, Pearl of Scandinavia, Courage — happened on a ship that stayed afloat, in or near port, where an investigator with jurisdiction could walk the deck. Every deep-sea total loss — Felicity Ace, Sincerity Ace, Grande America, Morning Midas — has no determined cause. For an underwriter that is the difference between a subrogated recovery and a loss that stays where it landed; for an owner it is the difference between a defence and default liability.

What the court actually required

Proof of origin, which is a factual standard the casualty had made impossible to meet. MOL and its insurers argued that a Porsche Taycan's lithium-ion battery self-ignited; the court did not find that it had not — it found that the plaintiffs had not proved that it had. The burden sat with the claimant, the ruling is appealable, and a separate action against Volkswagen Group interests remains before the Regional Court of Braunschweig. The legal point is ordinary; the evidentiary point is the story. A claim built on 'the cargo included EVs, therefore an EV started it' is an inference from stowage, and the court declined to let a stowage list stand in for an ignition record.

The three ways a car-deck fire destroys its own evidence

It consumes the origin, it removes the hull, and it was never recorded to begin with. First, consumption: on a deck stowed 30 centimetres apart, the first vehicle to burn is indistinguishable from the four-hundredth within hours, and the fire-pattern analysis a land investigator relies on — burn direction, low burns, arc mapping — has thousands of overlapping sources. Second, the hull: Felicity Ace sank in roughly 3,000 metres of water, Grande America in about 4,600 metres in the Bay of Biscay, Morning Midas in around 5,000 metres of the North Pacific, and no deep-sea wreck of a car carrier has been examined for cause.

Third, the record: the fixed detection a cargo ship carries identifies a section, not a vehicle; there is no voyage-data-recorder equivalent for a cargo space; deck cameras, where fitted, are not retained; and the 22 crew who evacuated on 16 February 2022 were the only witnesses, and they could not safely reach the deck in question. Each failure alone would weaken a claim. Together they leave nothing to argue from.

Where a cause was proven, and why

Wherever the ship survived and an investigator with jurisdiction boarded. The NTSB proved Höegh Xiamen (Jacksonville, June 2020) to an improperly disconnected battery on a used vehicle on deck 8 because the ship burned alongside, was extinguished, and the deck, the vehicles and the longshoremen were all available to examine and interview. It proved Grande Costa d'Avorio (Port Newark, July 2023) to a 2008 Jeep Wrangler used as a pusher because the vehicle, its recall history and the witnesses were there. The MAIB proved Corona Seaways (2013) to a used vehicle's starter circuit because the ferry reached port.

Pearl of Scandinavia (2010) was a converted car on charge that the crew found. Courage is the corpus's example of a named cause. Jurisdiction is the second condition: Höegh Xiamen and Grande Costa d'Avorio were in US ports and got the NTSB; Corona Seaways was UK-flagged and got the MAIB. Felicity Ace was Panama-flagged, on fire in international waters, and sank before Portugal's or Panama's investigators could reach a deck — the Dutch Safety Board's Fremantle Highway report, on a ship that was saved, still could not establish the ignition source.

What would have proven Felicity Ace

A record that survived the ship. Three data sets would have answered the Stuttgart court's question without a wreck examination: a per-vehicle, time-stamped detection log — which stow position showed the first thermal or gas anomaly, when, and how the signature spread — retained off the vessel; a cargo plan that tied that position to a specific vehicle, its powertrain and its state of charge at loading; and retained video of the space around the first alarm. None of the three needs the hull to be recovered. This is not a novel demand: it is what SSE 12 set out to write into the interim guidelines on video monitoring and detection systems in March 2026, and it is what hull and cargo questionnaires have begun to ask for since the ruling. The evidence problem is a design problem — car decks were built to be watched by a section detector and a patrol, and neither produces a record that outlives a sinking.

What it means for owners and underwriters

Treat the detection record as evidence, not just an alarm. For the hull underwriter who paid Felicity Ace and could not recover, the lesson is that subrogation against a cargo interest runs on an ignition record, and a fleet without one is a fleet whose losses cannot be laid off — which is a pricing fact, not a safety one. For the owner, an unproven cause is not neutral: it leaves liability where it falls by default, closes the recovery route, and means the next charterparty negotiation starts from 'unexplained total loss'. For both, the question to put to any detection scheme is the one the Stuttgart court put to MOL: after the ship is gone, what can you still show? A system that answers 'which vehicle, at what time, with what signature, from a server ashore' is the only kind that would have changed the Felicity Ace verdict.

Sources

• 1. gCaptain — 'Porsche Wins Felicity Ace Lawsuit as Cause of Fire Remains Unproven' (May 2026): Regional Court of Stuttgart dismisses MOL and hull insurers' ~€30 million ($34 million) claim; plaintiffs failed to prove a Porsche vehicle was the source; ruling appealable. The Maritime Executive — 'Court Dismisses Lawsuit Against Porsche Over Felicity Ace Fire'; TradeWinds — 'MOL loses German court battle over car carrier sinking'.
• 2. Contemporary casualty reporting (February–March 2022): Felicity Ace, Panama flag, MOL-operated, fire ~90 nm SW of the Azores on 16 February 2022; 22 crew evacuated; sank 1 March 2022 in ~3,000 m of water with ~4,000 Volkswagen Group vehicles.
• 3. NTSB — Marine Investigation Reports on Höegh Xiamen (Jacksonville, 4 June 2020; adopted December 2021) and Grande Costa d'Avorio (Port Newark, July 2023; adopted 2025): proven causes in a used vehicle's improperly disconnected battery and in a pusher vehicle's overheated transmission respectively. UK MAIB — Corona Seaways (4 December 2013): electrical defect in a used vehicle's engine-starting system.
• 4. Dutch Safety Board — Fremantle Highway investigation; contemporary reporting on Sincerity Ace (North Pacific, December 2018–January 2019), Grande America (Bay of Biscay, 12 March 2019, ~4,600 m) and Morning Midas (North Pacific, sank 23 June 2025, ~5,000 m).
• 5. DNV — 'IMO Sub-Committee on Ship Systems and Equipment (SSE 12)', March 2026: interim guidelines on video monitoring and detection systems for vehicle carriers to precede SOLAS amendments; correspondence group to SSE 13 (March 2027).
• 6. Companion RoRoSAFE analyses — the Felicity Ace incident anatomy (the court case in full), 'Subrogation and the Cargo-Manufacturer Question' (the recovery route the ruling tested), and 'What Causes EV Fires in RoRo Ships?' (the causal record this post draws on).