The Roman Catholic Church in the United States has paid out hundreds of millions of dollars in settlements in respect of abuse claims over the last few years - a fair chunk being funded by insurers.

The US cases are about priests sexually abusing children. In the UK, the focus has been upon child abuse in care homes and private households rather than in churches. In Ireland (like the US) the focus has been on the Church. However, the problem of applying the terms 'occurrence' or 'event' to the facts and achieving a sensible result are the same, regardless of context. How is a word like 'occurrence' which in ordinary language connotes something that happens at a point in time sensibly to be applied to a catalogue of sexual abuse inflicted upon a child or group of children over, perhaps, several years? What if there are 'each and every claim' deductibles and/or aggregate deductibles or cover for some years but not others or different (re)insurers from year to year?

The approach of the US and English courts is very different. The US courts sometimes construe words in a way designed to achieve a socio-political objective. In the context of asbestosis and pollution cases, this approach has given rise to the manifestation theory, the exposure theory, the triple trigger theory and the continuous trigger theory. The basic effect of these theories is to make policies pay when the natural meaning of the words used suggests they should not at all or, at least pay less.

Accordingly, courts in the US have variously found (or, at least it has been argued) in the context of abuse cases that:

  • there is one 'occurrence' for each policy period during which sexual abuse took place (regardless of the number of instances of abuse);
  • there is a single 'occurrence' which triggers only one policy period even if the sexual contact took place on more than one occasion and may have extended over several policy periods;
  • all the victims of abuse by a single offender in a policy period constitute one occurrence; and
  • abuse which took place in a particular year will trigger all policies subsequent to the date of abuse.

Aside from being impossible to reconcile, these 'purposive' approaches to construction and application do not fit with the English authorities on the meaning of 'event' or 'occurrence', at least at present. There is no English authority on the meaning of these terms in the abuse context. However, English law is fairly clear in other contexts:

  • An 'occurrence' should normally mean the same thing as an 'event' (unless, obviously, they both appear in the aggregation clause).
  • An event is something which happens at a particular time, at a particular place and in a particular way: Axa v Field (1996).
  • An event is, accordingly, to be distinguished from a general state of affairs, e.g. a general propensity for a person to behave in a certain way such as negligently or violently. However, identifiable individual negligent decisions can each be individual events (Caudle v Sharp (1995) where the Court of Appeal famously disagreed with the analysis that Mr Outhwaite's underwriting 'blindspot' was an event and instead held that there were 32 events - one for each risk contract he negligently wrote).
  • The event must be something which, as a matter of fact, causes loss: American Centennial Insurance Co v INSCO Ltd (1996).
  • Where a number of individual losses have been suffered, it is necessary for those losses to be sufficiently closely connected to each other in time, locality, cause and motive to be regarded as having resulted from a single event.
  • As well as 'event' based aggregation clauses, there are also 'cause' based clauses. A 'cause' is a broader concept than an 'event' and so necessarily tends to result in greater aggregation of claims.

In English law, there are two principal candidates for 'event' or 'occurrence' in this context: (a) each abusive act perpetrated by a particular adult upon a particular child for which the particular institution is liable in law; or (b) each occasion on which the relevant institution considered (or ought to have done) the position of the particular child or children and failed to act.

Which is correct? The answer (the usual lawyer's answer) is that it depends on the facts. Specifically, where the complaint is that the particular institution is vicariously liable for acts of the abuser, the 'abuse event' analysis fits best. Where the complaint is that the institution failed properly to look after the child by exposing him or her to a dangerous individual, the negligence approach fits best.

How the 'abuse event' analysis might be applied by the English courts is perhaps hinted at by the Court of Appeal in MMI v Sea Insurance Company Ltd (1998). The underlying facts concerned two dragline excavators which were extensively damaged by multiple acts of theft and vandalism (the precise dates of which acts were unknown) over three separate policy periods each with a significant deductible. The reinsurers, Sea Insurance, argued that the relevant deductible applied to each and every act of vandalism or pilferage which occurred during the particular policy period. The court disagreed because the wording of the relevant aggregating clause was cause based (see above). Hobhouse LJ said, however, "If this was a simple "any one event" clause, the defendants would have a powerful argument. They could say that each act of pilferage or vandalism was a distinct event."

However, the problem with such a rigid 'abuse event' approach is that it ignores realities at a number of levels. First, abuse can take the form of neglect rather than identifiable attacks. Second, an approach which requires each individual attack to be identified and valued in terms of damage done to the victim is very difficult to apply. Third, a very large portion of any damage suffered by a victim is likely to be in respect of psychiatric injury as opposed to physical injury. The difficulty with psychiatric injury is identifying when it occurred, what abuse event caused it and when that abuse event happened. The danger in a rigid approach is that it could produce harsh results, i.e. no cover for deserving victims. In the context of direct insurance, it would take a brave court to deny a child victim of long-term sexual abuse a remedy particularly where the perpetrator was insolvent or bankrupt and the victim was bringing a claim under the Third Parties (Rights Against Insurers) Act 1930.

The problems with the 'negligent decision event' analysis are similar. It will be necessary to identify how many negligent decisions were taken by the particular institution during any particular policy period and then to identify which injuries were suffered by each of the victims as a consequence of each negligent decision and what the quantum of those injuries was. This will inevitably be a highly speculative exercise.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.