Turning to the literature of the subject we find a vast number of variegated opinions on the frequency of defined, alcohol-based behaviour patterns in road traffic, as also for the incidence typical accidents for drunken drivers. thus we find the following percentages for typical failures resulting from alcohol: for zigzagging across the road and back between 3.3% and 19.0% driving on the wrong side of the road between 0.78 and 11.4% and for leaving the road altogether-between 10.46% and 40.0% all from the literature data. Of course these discrepancies can be attributed to differences in the factual material on which was research was based in the fort place. What is more striking, though is that the approach and the terminology different markedly throughout the various papers on the subject. Quite apart from this, as shown above, different phases in the accidents are taken so that no real comparison of the results is possible. For this reason what is needed is a systematic recording of traffic accidents which would make it possible so to arrange the factor established as to permit their juxtaposition with those from other sources. In this case we cannot - as so often happens - use as our allocation factor the nature of the collision, for this is often purely a matter of chance. Behaviour of the zigzagging kind - so typical a feature of drunken drivers - would not serve in every case, for it is often not possible to prove this kind of behaviour in retrospect. It is for these reasons that a phenomenological recording of traffic accidents is proposed, based on the nature of the traffic pattern prior to the accident as the forest allocation factor and using the course of the accident as the second factor. As this requires purely objective appraisal of the affair, and does not involved, the result should be shorn of all too common sources of error.
Abstract