Case Details
- Citation: [2023] SGCA 20
- Court: Court of Appeal of the Republic of Singapore
- Date of Decision: 7 July 2023
- Judgment Reserved: 3 April 2023
- Court of Appeal Civil Appeal No: Civil Appeal No 45 of 2022
- Judges: Judith Prakash JCA, Steven Chong JCA and Belinda Ang Saw Ean JCA
- Appellant: Owner of the vessel “NAVIGATOR ARIES” (IMO No. 9403762)
- Respondent: Owner of the vessel “LEO PERDANA” (IMO No. 9363390)
- Proceedings Below: Admiralty in rem actions
- Admiralty in Rem No 170 of 2016: Action against the vessel “LEO PERDANA”
- Admiralty in Rem No 204 of 2016: Action against the vessel “NAVIGATOR ARIES”
- Legal Area: Admiralty and Shipping — Collision
- Core Issue on Appeal: Apportionment of fault for a collision in the Surabaya Strait, focusing on the port sheer of the “LEO PERDANA” and the interpretation/application of COLREGS Rule 9(a)
- Trial Judge’s Apportionment: 70:30 in favour of the “LEO PERDANA”
- Court of Appeal’s Apportionment: 50:50 (both vessels equally to blame)
- Length of Judgment: 81 pages, 21,248 words
- Cases Cited: [2023] SGCA 20 (as provided in metadata)
Summary
This Court of Appeal decision concerns a maritime collision in the Surabaya Strait, Indonesia, between two large vessels travelling on reciprocal courses: the LPG tanker “NAVIGATOR ARIES” (“NA”) and the container vessel “LEO PERDANA” (“LP”). The collision occurred just before midnight on 28 June 2015 and resulted in severe damage to both vessels, including a fire on the NA. Both vessels were under compulsory pilotage and were navigating through a buoyed narrow channel that included a dredged section.
The High Court judge found that the proximate cause of the collision was the LP’s port sheer onto the NA’s path. That sheer was linked to hydrodynamic interaction with a bank on the LP’s starboard side, described as the “bow cushion effect”. On appeal, the Court of Appeal accepted the proximate-cause finding but disagreed with the High Court’s allocation of blame. The Court of Appeal held that both vessels were equally at fault and apportioned liability on a 50:50 basis, reversing the 70:30 apportionment in favour of the LP.
Beyond the factual fault analysis, the Court of Appeal used the case to clarify the proper interpretation of COLREGS Rule 9(a), which governs conduct in narrow channels. The Court examined whether Rule 9(a) should be interpreted consistently with its predecessor in the 1960 Collision Regulations (Rule 25(a)), and how that affects the standard of navigation expected of vessels in such constrained waters.
What Were the Facts of This Case?
The NA was an LPG carrier registered in Surabaya, Indonesia, approximately 160m in overall length and 25.6m in breadth, carrying about 59 tonnes of LPG at the material time. The NA displayed an all-round red light in addition to the regulation lights, indicating the carriage of dangerous cargo. Its drafts at departure were 5.5m forward and 6.2m aft. The LP was a fully cellular container ship flying the Panama flag, approximately 200m in overall length and 32.2m in breadth, laden with containers weighing around 17,200mt. Its drafts at departure were 8.54m forward and 8.65m aft.
The collision occurred within the Surabaya Strait, a roughly 50 nautical mile passage between the northeast coast of Java and the island of Madura. Weather and visibility were good (visibility about six miles). There was a northerly current associated with the ebbing tide, about 1.22m high. The NA was outbound from Gospier jetty in Surabaya to Kalbut at the eastern end of Java, travelling northwards towards the northern entrance of the Strait. The LP was inbound from the northern entrance to Tanjung Perak in Surabaya, travelling southwards. Accordingly, the vessels were on reciprocal courses within the Strait.
Crucially, the collision took place in a buoyed narrow channel governed by COLREGS Rule 9. The buoyage system required northbound vessels to keep the red lateral buoys on their starboard side, and southbound vessels to keep such buoys on their port side. Within the buoyed channel lay a dredged channel (the “Dredged Channel”), deepened and widened by dredging works from 2014 to May 2015. The dredged section was widened from 100m to 150m (excluding slopes), and deepened from 9.5m to up to 13m (chart datum). The dredged section and the adjacent bank geometry were relevant to the hydrodynamic forces at play.
Both vessels used standard nautical charts rather than ECDIS, and both relied on British Admiralty Chart No 975 (26 April 2012 edition). Their primary position monitoring was therefore manual charting supported by radar and other navigational aids. The evidential record included AIS data, voyage data recorders (“VDR”), wheelhouse conversations, and VHF radio communications. Additionally, each vessel had two sets of ARPA radar (X-band and S-band), with radar screengrabs captured at 15-second intervals. The Court of Appeal emphasised that the radar screengrabs and other contemporaneous recordings provided objective evidence of the vessels’ relative positions and headings in the critical minutes before the collision.
In the last minute before the collision, the bridge teams on both vessels were described as relatively calm, and the vessels appeared confident of safely passing port-to-port as agreed. Approximately one minute before the collision, the LP experienced a significant sheer to port, which brought it onto the NA’s path. The LP attempted to use VHF radio to inform the NA that its steering was not responding. The NA noticed the sheer and took avoidance action by altering hard-to-starboard. Despite this, the collision occurred and could not be avoided.
What Were the Key Legal Issues?
The central legal question was how to apportion fault between the two vessels for the collision, given that the proximate cause was the LP’s port sheer. While the parties did not dispute that the sheer resulted from hydrodynamic interaction—specifically, the bow cushion effect—there was a dispute about what each vessel knew or ought to have known earlier, and what navigational errors or rule breaches contributed to the collision.
A second key issue concerned the interpretation and application of COLREGS Rule 9(a) in narrow channels. The Court of Appeal needed to clarify whether Rule 9(a) should be interpreted in a manner consistent with its predecessor provision in the 1960 Collision Regulations (Rule 25(a)). This interpretive question mattered because it affects the standard of navigation expected of vessels when passing through narrow channels, including how far vessels must keep to the appropriate side of the channel and what “safe and practicable” means in context.
Third, the Court had to determine the legal significance of certain alleged breaches by the LP and the NA, including whether specific helm or pilot orders were wrongful, and whether any such wrongful orders were causative or merely coincidental to the collision. The Court also considered the possibility of novus actus interveniens (a new intervening act breaking the chain of causation), which can arise where an independent act is argued to be the true cause of the harm.
How Did the Court Analyse the Issues?
The Court of Appeal began by framing the appeal around fault attribution. It accepted the High Court’s finding that the proximate cause was the LP’s port sheer, and that the sheer was caused by hydrodynamic interaction with a bank on the LP’s starboard side (the bow cushion effect). The decisive question was therefore not whether the sheer happened, but whether the LP’s and NA’s respective navigational conduct contributed to the circumstances that made the sheer possible and/or made collision avoidance ineffective.
On the LP’s side, the Court focused on whether the LP knew or ought to have known, some time before the VHF communication, that hydrodynamic forces were already impeding its ability to respond to helm action. This knowledge question was pivotal because it bears directly on whether the LP’s subsequent actions (including any pilot orders and steering decisions) were reasonable. The Court’s analysis was grounded in objective evidence, including VDR data and radar screengrabs, rather than relying solely on conflicting narrative accounts from the vessels’ personnel.
The Court also examined the LP’s alleged breaches of COLREGS Rules 8(a), 8(c), and 8(d). These rules relate to maintaining proper lookout, safe speed, and taking action to avoid collision. A significant part of the analysis concerned the LP’s failure to detect the bow cushion effect earlier. The Court treated this as a navigational deficiency: if the LP had detected that hydrodynamic interaction was affecting its manoeuvrability, it could have adjusted its navigation earlier to reduce the risk of sheer onto the NA’s track.
In addition, the Court analysed the causative fault of the LP pilot’s “midships” order. The Court considered whether such an order should have been given in the circumstances and, if it was wrongful, what legal significance it had for causation. The Court’s reasoning indicates a careful approach: it did not treat every error as automatically determinative, but instead assessed whether the wrongful order had a meaningful causal contribution to the collision. The Court also addressed the relationship between the relevant COLREGS rules and Rule 14 (which concerns action by vessels in sight of one another and the responsibilities when collision is imminent). Finally, the Court considered whether any intervening act could break causation. The overall thrust was that the LP’s navigational errors were sufficiently connected to the collision to warrant a substantial share of fault.
Turning to the LP’s compliance with Rules 2 and 9(a), the Court examined the interpretation of Rule 9(a) of the COLREGS. Rule 9(a) requires vessels in narrow channels to keep to the appropriate side of the channel so far as is safe and practicable. The Court clarified that this “safe and practicable” standard is not purely subjective; it is assessed by reference to what a competent navigator should consider in the circumstances, including the vessel’s handling characteristics, the channel geometry, and the likelihood of hydrodynamic effects. The Court also considered whether it was safe and practicable for the LP to have gone further to starboard past the western edge of the Dredged Channel. The Court’s conclusion was that the LP’s navigational choices did not meet the expected standard, contributing to the conditions leading to the sheer.
As for the NA, the Court analysed alleged breaches of COLREGS Rules 5, 7, and 6, as well as a breach of Rule 9(a). While the proximate cause was the LP’s sheer, the Court did not treat the NA as blameless. It assessed whether the NA’s navigation and speed management were consistent with the COLREGS requirements for safe speed and proper lookout, and whether the NA’s actions were appropriate given the narrow-channel context and the need to anticipate potential manoeuvrability issues in constrained waters.
In weighing the vessels’ respective faults, the Court adopted a structured approach: it identified each vessel’s breaches relevant to apportionment, evaluated their causal contribution, and then assessed relative blame. The Court ultimately concluded that the High Court had over-attributed fault to the LP. The Court of Appeal held that the NA’s breaches were not merely technical, and that the LP’s breaches were not so dominant as to justify a 70:30 split. The result was a 50:50 apportionment.
What Was the Outcome?
The Court of Appeal allowed the appeal in part by disagreeing with the High Court’s apportionment. It held that both vessels were equally to blame for the collision and ordered liability to be apportioned on a 50:50 basis, instead of the 70:30 split in favour of the LP.
Practically, this meant that each vessel’s owner bore an equal share of the collision-related liability determined in the admiralty in rem proceedings. The decision therefore affects not only the legal reasoning on COLREGS compliance but also the financial allocation of damages and costs arising from the collision.
Why Does This Case Matter?
This case is significant for maritime collision law in Singapore because it provides a detailed, evidence-driven framework for apportioning blame where a collision is triggered by a proximate physical event (here, the LP’s port sheer) but where the legal inquiry turns on navigational responsibility and rule compliance. The Court’s emphasis on objective contemporaneous evidence—AIS, VDR, VHF communications, and especially ARPA radar screengrabs—reinforces that collision litigation will often be won or lost on how well the parties can demonstrate what was known or should have been known in the critical minutes before impact.
From a doctrinal perspective, the decision clarifies the interpretation of COLREGS Rule 9(a) in narrow channels. By addressing whether Rule 9(a) should be interpreted consistently with the 1960 Collision Regulations’ predecessor, the Court offers guidance on how “safe and practicable” should be applied in constrained waterways. This matters for practitioners advising shipowners, pilots, and navigators on compliance standards and for litigators assessing causation and fault where channel geometry and hydrodynamic effects are implicated.
For practitioners, the case also illustrates how courts treat pilot orders and steering decisions in the causation analysis. The Court did not simply label an order as “wrong”; it examined whether the order should have been given and whether it had legal significance for causation. This approach is useful for future cases involving complex manoeuvring, where multiple decisions in the lead-up to collision may each be argued to be causative.
Legislation Referenced
- Merchant Shipping Act (Cap 179, 1996 Rev Ed)
- Merchant Shipping (Prevention of Collisions at Sea) Regulations (Cap 179, Rg 10, 1990 Rev Ed), in particular reg 3 incorporating the COLREGS
- International Regulations for Preventing Collisions at Sea 1972 (COLREGS), including Rules 2, 5, 6, 7, 8(a), 8(c), 8(d), 9(a), and 14 (as discussed in the judgment extract)
- International Regulations for Preventing Collisions at Sea 1960 (1960 Collision Regulations), including Rule 25(a) (as discussed for interpretive comparison)
Cases Cited
- [2023] SGCA 20
Source Documents
This article analyses [2023] SGCA 20 for legal research and educational purposes. It does not constitute legal advice. Readers should consult the full judgment for the Court's complete reasoning.