7.1 The Principle of Proximate Cause

What Is Proximate Cause?

Proximate cause (also called the dominant effective cause or, from Latin, causa proxima) is the legal principle that an insurer is liable under a contract of insurance only for a loss that is proximately caused by an insured peril. Where a loss has a chain of causes, some insured and some not, the law looks for the real, dominant, efficient cause — not simply the event nearest in time to the damage. The doctrine is the cornerstone of causation in English insurance law and is heavily tested in IF1.

The principle was codified for marine insurance by section 55(1) of the Marine Insurance Act 1906:

"Subject to the provisions of this Act, and unless the policy otherwise provides, the insurer is liable for any loss proximately caused by a peril insured against, but, subject to the like provisions, he is not liable for any loss which is not proximately caused by a peril insured against."

Although the Marine Insurance Act applies directly only to marine insurance, English common law has carried the same test into every class of insurance — property, casualty, liability, and personal accident. The IF1 exam expects you to know the principle and its statutory root.

---

The Dominant Effective Cause, Not the Last Event

The phrase "proximate" can mislead candidates into thinking the test is temporal — the cause nearest to the loss in time. It is not. The test is qualitative: which cause was the real, efficient, dominant cause that brought about the loss? Lord Shaw of Dunfermline in Leyland Shipping Co v Norwich Union [1918] AC 350 captured the idea: to look only at the event nearest in time is "to look at the thing too microscopically"; the court must step back and ask what was the efficient cause that set the chain in motion and dominated the outcome.

This means that:

• The first event in a chain can be the proximate cause, even if several later events follow.
• The last event before the damage can be the proximate cause, but only if it is genuinely the dominant cause.
• An event in the middle of the chain can be the proximate cause if it dominates the others.

---

Distinguishing Proximate from Remote Causes

A cause that plays only a minor, incidental, or historical part in the loss is treated as a remote cause, and the insurer is not liable for damage proximately caused by it. Remote causes are sometimes described as causa remota — the distant cause the law declines to follow.

The distinction is not always obvious, and the courts have repeatedly warned against a mechanical approach. Two helpful tests:

1. The "but for" test is necessary but not sufficient. A peril that the loss would not have happened without may still be remote. "But for" the insured peril the ship would not have been at sea; but that does not make the insured peril the proximate cause of a later, independent fire.
2. The dominance test is decisive. Ask which cause, in common sense and on the facts, was the operative cause that dominated the loss. If an insured peril sets in motion an unbroken chain that naturally leads to the loss, the insured peril is the proximate cause. If a new, independent event intervenes and becomes the dominant cause, that intervening event becomes the proximate cause (see section 7.2).

---

Worked Scenarios

Scenario A — Fire causes collapse (insured peril is proximate cause)

A warehouse is insured against fire. A fire breaks out and weakens the structural steel. Two hours later the building collapses. The collapse is a separate physical event from the fire, but it is the direct, natural consequence of the fire. The fire is the proximate cause, and the collapse loss is covered, even though the policy word is "fire" and the immediate damage is "collapse."

This mirrors the logic of Leyland Shipping: a torpedo (war, an excepted peril) struck a ship; the ship limped to port; later, outside the harbour, heavy seas caused her to sink. The House of Lords held the torpedo — the dominant efficient cause — was the proximate cause, not the perils of the sea nearer in time. Where an insured peril plays the equivalent dominant role, the claim succeeds on the same logic.

Scenario B — Excepted peril is the dominant cause (claim fails)

A shop policy insures against fire but excepts war and kindred risks. During an armed conflict, a shell strikes the shop and starts a fire that destroys the building. The fire is the immediate mechanism of destruction, but the dominant efficient cause is the hostile act of war — the excepted peril. The claim fails, because the loss was proximately caused by an excepted peril, not by an insured peril. This is exactly the result in Leyland Shipping: the excepted war peril (torpedo) dominated, so the insured peril (perils of the sea) was not the proximate cause.

Scenario C — Uninsured peril is the dominant cause (claim fails)

A property policy covers fire and theft but not flood. A river overflows its banks and floodwater damages the building. There is no suggestion of fire or theft. Flood is the sole and dominant cause. Because flood is an uninsured (but not excepted) peril, the insurer is not liable — the loss was not proximately caused by an insured peril. Note the contrast with the concurrent-cause rule in section 7.2: where an insured peril operates alongside an uninsured peril, the result can differ.

---

Section 55(2) MIA 1906 — Specific Excluded Losses

Section 55(2) of the Marine Insurance Act 1906 lists losses that are not recoverable even if an insured peril is involved in the background:

These statutory exceptions illustrate a wider idea that recurs in non-marine policies: gradual deterioration, inherent vice, and delay are not insured perils, and a policy does not respond merely because an insured peril sits somewhere in the background of the chain.

---

The Burden of Proof

The burden of proof rests on the insured to show, on the balance of probabilities, that an insured peril was the proximate cause of the loss. Where the evidence is genuinely evenly balanced and no dominant cause can be identified, the claim fails — the insured has not discharged the burden. This is the lesson of Rhesa Shipping Co SA v Edmunds (The Popi M) [1985] 1 WLR 948: courts must not invent a cause simply to fill the gap. The insured's evidence must point to a real, identifiable insured peril as the dominant cause.

---

Why Proximate Cause Matters in Practice

A single loss will often have multiple causes stretching back in time. Without a proximate-cause rule, every policy would degenerate into arguments about how far back the insurer must trace the chain. The rule gives a workable test: find the dominant, efficient cause and ask whether it is an insured peril. If it is, the insurer pays (subject to the policy terms and the measure of indemnity). If it is not, the insurer does not pay — whether the dominant cause is an excepted peril or simply an uninsured one.

The rule also protects the principle of indemnity: the insurer responds only to the risks it has agreed to underwrite, and the insured cannot expand the scope of cover by pointing to an insured peril that played only a remote part in the loss.

---

Common Exam Pitfalls

• Do not equate "proximate" with "nearest in time." The test is dominant efficiency, not chronology.
• Do not assume that because an insured peril appears somewhere in the chain, the claim succeeds. The insured peril must be the proximate (dominant) cause.
• Do not assume that an excepted peril needs to be the last event to defeat the claim. A dominant excepted peril earlier in the chain is enough (Leyland Shipping).
• Remember the burden of proof is on the insured; if no dominant cause can be proved, the claim fails (The Popi M).
• Remember that delay, wear and tear, inherent vice, and rats/vermin are excluded by s.55(2) MIA 1906 even if an insured peril is in the background.

---

Key Takeaways

• Proximate cause is the real, dominant, efficient cause of the loss, not merely the last event in time.
• Section 55(1) of the Marine Insurance Act 1906 codifies the rule for marine insurance and the common law applies it to all classes.
• The insurer is liable only where an insured peril is the proximate cause; where an excepted peril is dominant, the claim fails even if an insured peril is nearer in time to the damage.
• The burden of proof is on the insured to identify the insured peril as the dominant cause; if no dominant cause can be proved, the claim fails.
• Section 55(2) MIA 1906 excludes wilful misconduct of the assured, delay, ordinary wear and tear, inherent vice, and rats/vermin even where an insured peril sits in the background of the chain.