MarIus

3.3 The proper construction of the NMC sections 172 and 172a

It was noted in section 3.2 above that the purpose behind Norway’s reservation was to protect third party interests from having to foot the bill for a shipowner’s pollution. In this section, we shall see that when construing the NMC sections 172 and 172a our focus should not be directed solely at what “type” of costs are being claimed as a loss, but also on the causative event that forms the basis for the claim. This again forces us to pay attention to who is making a claim against whom.

There is no “official” English translation of the NMC, but for our purposes the translation in MarIus 2014 no. 435 suffices.^{(1) One should be aware that the NMC has been amended several times since 2014. A more recent translation, but there only of excerpts, is available through the Norwegian Maritime Authority’s web page: https://www.sdir.no/en/shipping/legislation/laws/norwegian-maritime-code/ (last visited 26 February 2024).} The relevant part of section 172 reads as follows:

“The right to limitation of liability pursuant to Section 175 applies, regardless of the basis of the liability, to claims in respect of:

1. loss of life or injury to persons (personal injury) or loss of or damage to property (property damage), if the injury or damage arose on board or in direct connection with the operation of the ship or with salvage;

2. damage resulting from delay in the carriage by sea of goods, passengers, or their luggage;

3. other damage if it was caused by infringement of a non-contractual right and arose in direct connection with the operation of the ship or with salvage;

4. measures taken to avert or minimize losses for which liability would be limited, including losses caused by such measures. […]”

The relevant part of section 172a reads as follows:

“When the ship’s tonnage exceeds 300 tons, the right to limitation pursuant to Section 175a, regardless of the basis of the claim, applies to claims on the occasion of:

1. raising, removal, destruction or rendering harmless a ship which is sunk, stranded, abandoned or wrecked, as well as everything that is or has been on board the ship;

2. removal, destruction or rendering harmless the ship’s cargo;

3. measures taken to avert or minimize losses for which liability would be limited under this Section, including losses caused by such measures. […]”

First of all we should note that the distinction in the two sections between “claims in respect of” and “claims on the occasion of”, in section 172 and 172a respectively, is not found in the official Norwegian version of the act (where the wording is the same, “krav i anledning av”).^{(2) The same point applies in relation to “regardless of the basis of liability” in section 172 and “regardless of the basis of the claim” in section 172a. In the Norwegian version the wording is the same (“uansett grunnlaget for ansvaret”).} This point might seem trivial, but it illustrates something important: because there is no “official translation” of the act, and as Norwegian courts and arbitrators will therefore have to interpret the Norwegian version, there is generally not much for us to gain from carrying out a careful and detailed analysis of the meaning of the individual words in English. Nonetheless, the construction I argue for in the following still holds, regardless of whether we look at the English or Norwegian version of the act.

An interpretation of the provisions in abstracto is generally both hard to carry out and hard to follow as a reader. Because of this, we shall in the following use the two scenarios set out in section 1.1 above, as examples for our interpretation of the provisions. Before we do this, however, some introductory remarks to the provisions should be made. Firstly, a claim against the shipowner can fall under either section 172 or 172a, regardless of the “basis of the liability [/claim]”, i.e. it does not matter for limitation purposes whether the liability is based on contract or tort (with the sub-division of fault-based liability and strict liability). Secondly, the wording “claims in respect of” is to be understood as meaning that it does not matter whether the claim being made is an “original claim”, or instead a “derivative claim” in the form of a recourse claim.^{(3) NOU 1980: 55 pp. 15–16. The wording here includes what is set out separately in the LLMC art. 2 no. 2.} As we shall see below, however, it is important not to be confused into thinking that a claim for damages for wreck removal costs after a collision is a derivative recourse claim,^{(4) I use “derivative recourse claim” here to make a distinction with how “recourse” can be understood under e.g. English law, cf. Solvang 2016 p. 32 fn. 1 explains this in an instructive manner. A derivative recourse claim is accordingly “narrower” than a recourse claim.} entitling the claimant to a higher limitation amount.

We remember the two scenarios set out in section 1.1 above, one a marine casualty where only one vessel is involved and another with two vessels involved in a collision. Let me flesh them out now, with some further details to make them more illustrative.

Scenario 1: We start with the rather straightforward scenario where the bulker vessel A is not involved in a collision, but due to navigational errors ends up grounding. The grounding causes a bunker oil spill, cargo falls overboard due to the grounding, and, to make the example more illustrative, the grounding happens near shore with a navigational aid being hit and destroyed by the grounding. Public authorities^{(5) The Norwegian Coastal Administration.} carry out a mitigation and clean-up operation of the leaked bunker oil and lost cargo. The costs are claimed reimbursed from A. The authorities order the shipowner to remove the vessel, and the shipowner does this. The destroyed navigational aid is replaced with new equipment and the authorities make a claim for damages against A for the costs.

Scenario 2: The bulker vessel A collides with the much larger oil tanker B. The bow of B tears a hole in the hull of A. To prevent the vessel sinking, the master of vessel A purposefully aims for land where navigational aid is placed. The hole in the hull leads to bunker oil spill, and as in scenario 1, cargo falls overboard and the same navigational equipment is destroyed. As in scenario 1, the authorities carry out a mitigation and clean-up operation of the leaked bunker oil and lost cargo. The costs are claimed reimbursed from A. The authorities orders the shipowner to remove the vessel, and the shipowner does this. The authorities replace the destroyed navigational aid with new equipment and make a claim for damages against A for these costs. A makes a claim for damages against B for the damage to the vessel, the clean-up costs, the costs for the wreck removal, and for the cost of the navigational equipment destroyed. The oil tanker B only suffers a small damage to the bow that is repaired on a repair yard. B makes a counterclaim for damages against A for the damage to the bow.

A fundamental point in this article is that the interpretation of the limitation rules should not be influenced by the shipowners’ fault (or what might intuitively be regarded as “fair”).^{(6) Cf. the point in sections 1.2 and 2 above on “fairness”.} However, for scenario 2 to be any different from scenario 1, we have to assume that B is, at least in part, to blame for the collision with A. This is obvious. Unless B is to blame, there is no liability for B to limit.^{(7) Cf. the NMC section 161.} If A is 100% at fault for the collision, scenario 2 is identical with scenario 1, with the addition that B has a claim against A for the repair costs. In the following therefore, we assume that scenario 2 is a both-to-blame collision,^{(8) When a marine casualty occurs, shipowners usually “declare” their liability limited, before it is determined whether they are liable at all. As a part of this “process”, a global limitation fund is also usually established.} but note that in my view the outcome would not be any different if B were solely to blame.

With the two different scenarios fleshed out, we can move over to categorising the different claims in both scenarios under the NMC sections 172 and 172a.

Under scenario 1, this exercise is not that difficult. Here it is the state making a claim against A. The state’s claim for reimbursement of the cost for the mitigation and clean-up operation is a claim “on the occasion of […] removal, destruction [of] everything that is or has been on board the ship” and “removal [of] the ship’s cargo” and “measures taken to avert or minimize losses for which liability would be limited under this Section”, cf. section 172a nos. 1, 2 and 3. The state’s claim for damages for the destroyed navigational aid, on the other hand, is a “claim in respect of […] loss of or damage to property” that “arose […] in direct connection of the operation of ship”, cf. the NMC section 172 no. 1.

Before we move on to doing the same exercise for the seemingly more complicated scenario 2, we should dwell on scenario 1 for a further moment. Even though the construction given above seems intuitive and straightforward by just reading the wording of the two sections, in principle, the construction could nonetheless be explained in different ways. There is however one approach that is better than the other(s).

One explanation is by focusing on the “type of loss (costs)” that are being claimed. As long as the reimbursement claim consists of costs that relate to clean-up costs and other such measures, they fall under section 172a, with the higher limit in 175a.^{(9) Cf. Solvang 2016 pp. 36ff.} Similarly then, provided the replacement costs of the navigation aid are viewed as a “property damage loss”, they fall under section 172, with the lower limit in section 175.^{(10) If A later can turn around and claim these losses covered by another shipowner (X) with the right to limit liability, the same logic applies. In this “new case”, the new shipowner’s liability is determined by what “type of losses” are put forward by A. Then, in the same way as for A, X would be liable for both one section “175 amount” and one section “175a amount” towards A.} This view can be said to be supported by the use of the words “[…] losses for which liability would be limited under this Section” in section 172a no. 3 and the fact that the “basis of the claim [/liability]” is not relevant for the shipowner’s right to limit liability. I refer to this as the “type of loss” approach.

Another explanation, and in my opinion the better one, is to focus on the causative event^{(11) Best translated into Norwegian as “rettsstiftende faktiske forhold”.} that forms the basis of the claim for which the shipowner declares his liability limited.^{(12) Selvig 2021 pp. 113ff provides a similar, but not completely the same, explanation.} This approach is more in line with how we otherwise “categorise” claims in other areas of the law. The reasoning runs as follows: as we have seen in section 3.2 above, the authorities can make a reimbursement claim against shipowner A, because by owning the vessel, A is “the responsible” party for the pollution, cf. the NPA section 76. For this claim the causative event is simply A’s ownership of a polluting vessel. Similarly, it is A’s ownership that gives the state authority to order the removal of the wreck, cf. section 37, cf. section 28.^{(13) The fact that A’s liability is strict does not affect the validity of the reasoning, the same would have applied if liability additionally required fault.} All heads of losses that are a consequence of this causative event fall under the NMC section 172a, and are thus limited by section 175a.^{(14) Cf. Ot.prp. nr. 79 (2004–2005) p. 42.} However, when we move over to the replacement costs of the destroyed navigational aid, the causative event for the state’s claim is different. Here it is A’s negligent sailing of the vessel that leads to the vessel hitting the equipment that forms the basis of his liability towards the state, cf. the NMC section 151.^{(15) Again, the reasoning would not be any different if the liability for running down the navigational aid were strict.} All heads of losses that are a consequence of this causative event fall under the NMC section 172, and are thus limited by section 175.

With these two different explanations being applied to the categorising of claims in scenario 1, we can move on to the seemingly more complicated scenario 2. As more parties are involved in this scenario, we have to distinguish between the claims made by the state against A, on the one hand, and the claims made by A against B, and B’s counterclaim against A, on the other hand.

When categorising the claims made by the state against A, the exercise is the same as above for scenario 1. In this relationship, nothing has changed by introducing B.^{(16) One could argue that the navigational aid in this scenario is destroyed by an act of necessity. Even if that were the case, A would still be liable towards the state.} As we know from section 3.2 above, under Norwegian law, the state is not in a position to make a claim (directly) against B for the costs related to the pollution from vessel A.

If we start with B’s counterclaim against A for the damage to B’s bow, we can all, I hope, quickly agree that this is a claim “in respect of […] damage to property” that “arose […] in direct connection with the operation of [A]”, cf. the NMC section 172, regardless of whether we prefer one or other of the explanations discussed above. However, when it comes to A’s claims against B, the two explanations lead to different results.

Under the “type of loss”-approach, A’s claim against B for damage to or loss of vessel A, is a “property damage loss” – it is of this “nature” or “character”, cf. the NMC section 172. Similarly, as above, the costs that A had to pay the state for the mitigation and clean-up measures, including A’s own wreck removal costs, have not changed their nature or character just because A is now claiming that these costs be covered by B. Further, still entertaining this approach as the correct one, because it does not matter whether the claim from A against B for these costs is an “original” claim or a “derivative” claim,^{(17) Cf. the comment above on how the LLMC art. 2 no. 2 is implemented under Norwegian law.} since the costs are of the same nature, they then fall under section 172a. The consequence of this is that A’s claim against B for these costs enjoys B’s higher limitation amount in section 175a.

There are however some problems with this approach. In scenario 2, we should not be led into believing that A’s claim against B for the mitigation and clean-up costs is a “derivative claim”, in the sense of a recourse (or subrogation) claim.^{(18) Cf. fn. 60 above.} Under Norwegian law, a key element of a derivative recourse claim is that the claimant has covered another’s obligation.^{(19) This is the case both where the claimant mistakenly thought he was liable, but also in instances where their liability is joint (e.g. surety). Cf. i.a. the Norwegian Supreme Court cases in Rt. 1997 p. 1029 (Mofrakt), HR-2017-2414 (Spiker) and HR-2021-61-A (Sikringsordning).} However, when A paid the state’s reimbursement claim, A did not cover B’s obligation, for as we have seen, B is not liable vis-à-vis the state for the pollution from vessel A. A only covered his own obligation towards the state.^{(20) As noted in section 3.2, there are strong policy considerations behind why A’s limit should be higher vis-à-vis the state (and other third parties affected by the pollution), but these considerations do not automatically transfer to the consequent relationship between A and B.} Furthermore, as long as B cannot be ordered to remove A’s vessel, A’s own costs for the wreck removal is an ordinary claim for damages for A’s consequential losses.^{(21) Selvig 2022 pp. 124–127.}

Some might say that the following reasoning is too abstract to be useful in the concrete analysis of whether a claim falls under the NMC section 172 or 172a. To my mind, however, this admittedly crude outline of some fundamental principles illustrates why we should focus on the causative event, and not on types of losses, when categorising claims.

The fact that A’s claim against B for the wreck removal costs is a claim for damages, illustrates another problem with the “type of loss”-approach. By focusing on the “type of loss”, there is a risk that we overlook the basic but important point: what is it that makes specifically B liable for A’s losses and not somebody else? Further, why is it that A can claim these costs specifically from B? In scenario 2 it is B’s negligent sailing of his vessel that links B to A. The collision caused by B’s negligent sailing is what creates the bond between A and B. This “legally relevant fact”, i.e. this causative event, accordingly explains why specifically these two parties are bound together. For A however, this bond, in and of itself, is without interest if the collision did not cause any damage or loss to A’s vessel. It is when this is the case, i.e. the wrong has caused A to suffer a monetary negative effect, that A is interested in “pulling” on the bond to make B correct this negative effect by paying damages. It is only the negative effects caused by this specific event that A can claim be corrected by B. B’s liability towards A does not however change, depending on the “type” of monetary effect the causative event has had. To the contrary, it is the causative event that determines whether B can be held liable for the particular “type” of loss that A is claiming.

The bond between A and B must, however, be distinguished from the bond between the state and A. In that legal relationship, the legally relevant fact that entitles the state to make a reimbursement claim against A is A’s ownership of a vessel that constitutes pollution.^{(22) Because we do not want the state or other affected third parties to foot A’s “pollution bill”, we have enacted a higher limitation amount for this claim in accordance with the reservation against the LLMC.} The latter causative event that creates a relationship (a bond) between A and the state does not affect the causative event in the relationship between A and B. If that were the case, B’s liability would in principle be expanded, contrary to the rules in the NPA as we have seen above. Quite another issue is of course that of what heads of losses can be included in A’s claim for damages against B. The question of whether A can claim consequential losses from B because of the collision does not however change the claim. It is still a claim for damages.^{(23) Furthermore, the claim for damages does not change whether B is 100% liable or only 5% liable towards A. This only affects how much A can claim from B.} This is why I refer to this approach as the “type of claim” approach.

By isolating the different relationships that are created by the marine casualty in scenario 2, and more precisely, by isolating what it is that creates these relationships, we are hopefully in a better position to categorise properly the different claims under the NMC.

A’s claim against B is based on B’s negligent sailing of his vessel, and this remains the same, whether B is solely to blame for the collision, or only partly to blame. B’s negligent sailing tore a hole in the hull of vessel A, which later caused the bunker oil spill from vessel A, cargo to fall off vessel A, and necessitated the purposeful grounding of vessel A. Even though the marine casualty had dramatic consequences for A, A can only point to the same causative event for his claim against B: B’s negligent sailing. Accordingly, B’s liability towards A “arose […] in direct connection with the operation of the ship [B]”, cf. the NMC section 172, with the limit in section 175 setting out B’s maximum liability.

The state’s reimbursement claim against A is, on the other hand, based as we have seen on A’s ownership of a polluting vessel. Without being too repetitive: this claim, with its different heads of losses, falls directly under the NMC section 172a, with section 175a setting out A’s maximum liability. Strong policy considerations support the higher limits in this relationship, but the same considerations do not apply in the relationship between the two shipowners. Once the state has had its full loss covered by A, the aim of the separate national limitation regime is achieved.

To increase B’s liability, just because the collision caused vessel A to leak oil, lose cargo and become a wreck, compared to a situation where the collision “only” caused damage to vessel A that was then repaired, seems incoherent, given that it is the same reason which makes B liable in both scenarios. With this example, we are, hopefully, also able to accept my point that our sense of “fairness” should not affect our interpretation of the limitation rules. Whether B in scenario 2 is 100% or only 1% to blame for the collision, is neither here nor there when categorising A’s claim against B.^{(27) Provided that B is to blame and, furthermore, as long as B has not caused the “loss deliberately or through gross negligence and with knowledge that such loss would probably result”, cf. the NMC section 174, equalling the LLMC art. 4.} An intuitive sense of “fairness” should however probably make it easier to accept the view argued for in this article, in the scenario where A is mostly to blame for the collision (and the consequences of this), rather than where B is solely to blame for the collision. These intuitions of fairness are however dealt with by our substantive law on whether B can be held liable in the first place. Where this is the case, the limitation rules, through a more or less coherent system, set out a maximum amount for B’s liability. For the individual claimant that has his claim reduced, the impact of this system will obviously feel “unfair”. However, the system is in place so that claimants in as many cases as possible will have their full loss covered through shipowners’ liability insurance.