THE APPROACH TO THE SCIENTIFIC INVESTIGATION OF ACCIDENTS
DR C.D. POSTER
DR J.H. BURGOYNE PARTNERS
39A BARTHOLOREW CLOSE,, LONDON EC1A 7JN
1. INTRODUCTION
Most of us take out insurance to cover us against financial embarrassment resulting from an accident of some kind or another. Insurers need to know the basic details of the circumstances of an accident giving rise to a claim in the very least, and in a large number of cases it is necessary to engage specialists to investigate specific aspects of the accident so that insurers may be fully appraised of their position before they take any decision which may affect their liability.
Thus, there is considerable scope for the application of a scientific investigation of accidents to the investigation of insurance losses. In particular, to assist in the determination of liability whether it be of the insurer to his insured where fraud, non-disclosure or warranted exclusions may be relevant, or where evidence may be required to assist insurers to determine their subrogation rights. Assistance of a scientific nature may also be required to investigate the veracity of an employee's claim against his employers as a result of an industrial action.
Finally, the loss adjuster may require assistance in quantifying a claim, especially where merchandise is damaged by fire, as a means of assessing the amount of stock originally present.
The ostensible purpose of the investigation of an accident is usually to establish the circumstances that led its occurrence - in a word, the cause. Presumably, the object implied is to avoid its recurrence. In practice, our investigations are of course diverted or perhaps distorted to serve other ends. This occurs, for example, when it is sought to blame or exonerate certain people or things as is frequently the case with accidents giving rise to insurance claims. This may lead to bias, where only those aspects are investigated that are likely to strengthen or defend a position taken up in advance of any evidence. This of course, represents the very antithesis of true investigation.
However, when an accident forms the subject of litigation, arbitration or public enquiry, its various facets are looked at - often at great length an in great, although uneven detail - but the various aspects are looked at by the different parties concerned and the process is cumbersome and costly, extended as it is by partisan manoevering for position. And in the end the conclusions, technically speaking, are only as good as the original investigations. No amount of argument and discussion of "cold" evidence can make up for the observations not made or noted, or for questions not asked when the evidence was "hot" and undisturbed.
Method of Investigation
Ideally, the investigation of an accident giving rise to an insurance claim should be undertaken as a research project with a specified objective. The method may be expressed in different ways but may be summarised as: enquiry, deduction, induction, validation and finalisation. These steps work out somewhat as follows.
- The first step is to collect all immediately available evidence, but a prerequisite is that it is safe to do so. On the site of an accident, there is often a particular liability to further accident because normal controls have lapsed. During the first gathering of evidence, nothing should be disturbed. This rule is for the benefit of the enquirer himself, and also for the benefit of others who may have a responsibility for investigation. Photography of the undisturbed scene or situation is invaluable, but it should be carried out according to a plan, so as to yield a systematic and comprehensive survey. If this is done, recording of the photographs themselves is greatly simplified.
Because of the "no disturbance" rule, the examination of physical evidence at this stage may have to remain superficial. Some deeper insight may be obtained, however, by questioning the available witnesses regarding the situation immediately before, during and immediately after the event. Many witnesses lack the power of unambiguous description and of holding clearly in the imagination what they have seen. Questioning on the site of the event being investigated has a number of advantages, including a more accurate indication of what was seen, where and when. Nevertheless, witnesses are commonly not reliable in describing unexpected, rapidly-occurring, or violent events. Cross-checking between witnesses is therefore particularly important. It should not be concluded, as is sometimes done, that witnesses are useless. Nevertheless, their evidence has to be evaluated and interpreted rather than accepted literally.
- Having gathered the immediately available evidence, this may be reviewed and a tentative view of the accident developed. The dangers of premature theorising are often emphasised, but as a guide to the next steps a working hypothesis is necessary. It should be constantly borne in mind, however, that the hypothesis is for the purpose of working and since it is at this stage based upon a partial evidence only, it is by no means a theory to be defended.
A consideration of the hypothesis, which may contain alternative elements, will suggest further questions that should be asked, further indications that should be sought amongst the physical evidence, and perhaps certain calculations that should be made or tests that should be carried out. All this provides the programme for the next stage in the investigation.
It is in the formulation and carrying out of this programme that detailed expertise is brought into play. Experience is of course of great value in the preliminary investigation, but it is in the deeper probing that a detailed knowledge of the disciplines concerned is fully employed, whether they be concerned with the. physics, chemistry, metallurgy, one of the branches of engineering, or the technologies of fire, explosion and other accidental happenings.
- The next step involves the collection and assessment of the less accessible evidence now considered to be relevant. This may or may not support the hypothetical view of the accident already formed. If it does not do so the hypothesis has to be re-shaped, or discredited or alternatively discarded.
The testing and adjustment of the working hypothesis is likely to be in the iterative process rather than a singlestep operation. It is continued until diminishing returns suggest that it is time to stop. In a matter of complexity, there will usually be some "loose ends" however far the matter is pursued, and it is a matter of judgement how far the pursuit should be pressed of physical evidence that is expensive to uncover, witnesses that are hard to find or unwilling to testify, and tests that are costly to carry out. The last may often present a problem because the only tests that are useful are those that reproduce the circumstances of the accident closely, including its scale. Tests that do not fulfil this condition rigorously may well carry little conviction.
4The final step, having assembled and evaluated all the evidence that will be obtained is to formulate a final view of the accident and to report it. The report will contain an account of the accident, the investigation and the conclusions reached. If a single comment may be allowed, it is that a chronological account of the investigation is often best, both for the writer and the reader, and because it allows an independent judgement to be formed as to the way in which the investigator has arrived at his conclusions.
Insurance Investigations
As I have already indicated, these are usually concerned with the relevance of insurance policies or with the possible transfer of liability to another party. Special interests therefore tend to prevail. If however, the matter is to proceed to litigation or arbitration - and such a possibility is often difficult to exclude in the early stages of enquiry - no investigation can be too thorough. Cost must, nevertheless, be weighed against the prospect of success and there are many cases where the latter prospect appears so slight early on, that the investigation is terminated after the preliminary stage. Much skill and insight can be exercised in reaching such a decision on the basis of an overall but superficial review of the evidence.
In my experience, after competent experts have made a thorough investigation and have vied with one another in court, regarding their interpretations, the final conclusions reached are not only legally effective, but of ten technically satislactoty -as technically satisfactory if an agreement is reached by bargaining or a judgement by legal technicality.
Most of our investigations relate to the occurrences of fire and explosion, although aspects such as tool marks and glass analysis in cases of forcible entry, material damage to structures, chemical analyses and contamination claims, electrical defects and failures, and mechanical failures frequently fall for our consideration.