468
anthropometry
every other means in their power. In the case of small com- ing through photographic albums and voluminous records, munities and local crime personal identification is usually at°the cost of much time and labour, and with a doubtful the authorities can, theoretically at least, lay their a simple matter and presents little difficulty to the police, result, hands with certainty on the document required within a who are more or less familiar with all the persons with minutes even among 90,000 others. The man is whom they are called upon to deal; but it becomes few and his head-length is found to be over 191 difficult or impossible with persons who travel from place measured, millimetres. He therefore falls under the “longs,” and to place, as the more important professional criminals do, two-thirds of the registration cabinet, or 60,000 records, and in great cities by sheer excess of numbers. Previously at once eliminated. The head-breadth in like manner to the introduction of the Bertillon system the means are eliminates 20,000 more; and so on until the search comes relied upon were (1) the memory of warders and policemen, who often had to travel to distant places to view down to a bundle of six or so, among which the suspect’s prisoners, (2) photographs, (3) written descriptions, with card is found, with photograph, description, &c. The thing is entirely independent of names. The actual special reference to marks or other peculiarities. These whole identification is made by means of the photograph and things have their uses still, but by themselves they formed distinctive marks, if any. Some additional measurements, a very imperfect means of identification. The chief including the dimensions of the ear, are taken by M. Berdefects in practice were (1) frequent failure to identify, (2) liability to mis-identification—not very common but tillon, but they are not essential. The system is still of the greatest importance to the innocent, (3) time, labour, undergoing modification and improvement by the introand expense involved in searching records and circulating duction of new features. An important addition is the finger-print process, described below. descriptions. The weak point in the system is the margin which has Anthropometry was designed to overcome all these defects. The great advantage claimed for it lies in the to be allowed for error in the measurements. M. Bertillon systematic recording on a uniform plan of the distinctive himself allows 1 millimetre, but with less skilful or less features of each criminal, and the classification of these careful officers than those at the Paris Prefecture it may be records on the numerical basis afforded by measurement. necessary to allow more. This introduces a doubt into the The principle of classification maybe explained as follows:— classification, rendering it uncertain to which of two secFive dimensions are taken as a basis, namely (1) length of tions measurements near the dividing line really belong, and head, (2) breadth of head, (3) length of left middle finger, so increasing the area of search to a corresponding degree. (4) length of left foot, (5) length of left forearm. These In practice, however, the system has proved highly were chosen by M. Bertillon chiefly on the ground that successful in France. A striking result is the great they are the most constant in the individual and the diminution in the number of habitual criminals who easiest to take accurately; but they may be varied with- attempt to evade recognition; the inference is that they out affecting the principle. Each of these is then divided have learnt to regard the endeavour as futile. In 1893 a departmental committee was appointed by into three classes :—(1) long, (2) medium, (3) short. Thus heads upwards of 191 millimetres in length are classed the British Home Office to inquire into the anthropometric “long,” those of 185 to 190 millimetres are “medium,” system and also into finger-prints. It reported in the foland those of less than 185 are “short.” Similarly with lowing year, recommending the adoption of anthropometric the breadth of head, length of middle finger, and so on. classification by means of the five primary measurements This division of the five dimensions into three classes described above; but for the further classification and each gives 35 or 243 classes, which are assumed to be actual identification it recommended the . finger - print approximately equal. Supposing, now, it is required system elaborated by Mr Francis Galton in preference to keep a register of 90,000 criminals, as contemplated to the Bertillon secondary measurements, as being more by M. Bertillon, each represented by a card containing scientific and more accurate. These recommendations were personal details, photograph, and other particulars. The provisionally adopted with some modifications, and have first division by length of head will divide the 90,000 been in use in England ever since. The actual measurecards into three sections of 30,000 each; these are sub- ments now taken are the following :—(1) head length, (2) divided by breadth of head into nine subsections of head breadth, (3) face breadth, (4) left middle finger, 10,000 each, and again by the remaining measurements (5) left cubit, (6) left foot, (7) height. These are entered successively into 27 sets of 3300 each, 81 of 1100 each, on a form together with other details, including age, and 243 of 370 each. The cards are kept in a cabinet, name, complexion, hair, eyes, and distinctive marks. which is divided into sections and subsections corre- The prisoner’s photograph is also affixed in a space left sponding to the measurements described, and eventually vacant for the purpose. The reverse side of the same into drawers, representing the last subdivision. There piece of paper is devoted to the finger-prints. Impresare, therefore, 243 drawers, each containing about 370 sions of the thumb and four fingers of each hand are cards. Then a further subdivision of the cards in each taken, both with the fingers together and each in a drawer is given by the height, the length of the little separate space. The papers are classified and arranged finger, and lastly by the colour of the eye. The height in sections to form a register in the manner already gives three subdivisions of 124 each, which are explained. It is to be noted, however, that owing to further divided into groups of 41 by the little finger, difference of judicial procedure the system is carried out and finally into packets of 6 by the colour of the in a somewhat different way in Great Britain, and is used eye, of which seven varieties are distinguished. The to a much less extent than in France. The French law practical working of the system can be readily understood. pertnits the arrest, detention, and interrogation of persons A prisoner is brought into the Prefecture; the police on mere suspicion, and the identification of “recidivists ” require to know his record, and whether previously con- at this stage is of great importance, because it places a victed or not. He gives a name which is not found in powerful weapon in the hands of the police and of the the alphabetical register, kept in addition to the anthro- examining magistrate in working up the case. It is also pometric register. Let us suppose that the name is false, important for the protection of the innocent. All persons and that he is really an old offender. In order to find out arrested are therefore measured at once by the police as it is necessary to search the records. It is here that the a matter of routine, and this is done by specially trained anthropometric classification comes in. Instead of search- and skilled officers at the identification bureau. The
