The measurement of man


Galton’s psychological testing poster F Galton, via Wikimedia Commons

In 1873 the Swiss botanist Alphonse de Candolle published a refutation of Galton’s Hereditary genius, arguing that environmental factors played a much greater part in the creation of eminence than Galton had claimed. In response, Galton devised and distributed one of the first early – and very large – psychological questionnaires to nearly two hundred members of the Royal Society, including questions on the development of their scientific ability, their religious inclinations and assessment of their own characters. In 1874 Galton published the results in English men of science: their nature and nurture, coining the expressions for innate and environmental influences that have characterised the debate ever since. Although Galton considered this work an entirely objective account, many of the questions were designed – either consciously or unconsciously – to elicit particular responses, and his interpretation of the results had a tendency to fit them to his own preconceptions. In spite of the obvious flaws, this was a pioneering piece of work, which led the way in the development of one of the key tools of psychological research. The resultant work also illustrated that Galton would at least allow that environmental factors (or ‘nurture’) played some role in the development of a person: ‘the effects of education and circumstances are so interwoven with those of natural character in determining a man’s position among his contemporaries, that I find it impossible to treat them wholly apart’. Work on this volume also awoke Galton to the value of identical twins for teasing out the effects on an individual of ‘education and circumstances’ from ‘natural character’. Galton was thwarted in his hopes to find numerous sets of twins who had been separated at birth so that he could clearly differentiate between innate and environmental factors. This approach is one that has been developed so that twin studies – in particular the comparison of identical and non-identical twins – have contributed much to our understanding of human heredity.

Later Galton explored ways to measure psychological differences between individuals, creating a new discipline in experimental psychology. His new technique employed introspection and self-analysis ‘allowing the mind to play freely…until a couple…of ideas have passed through it, and then…to turn the attention upon them with a sudden and complete reawakening’. He realised that surprising links might be found, and that beneath the level of day-to-day conscious experience lay a more complex world. Galton conceived a word-association experiment to explore this world, and tested it on himself, realising that many of the associations he made were related to his youth. These experiments preceded by many years Freud’s work attempting to reveal and understand the subconscious.

To test the abilities of others he started with another questionnaire in 1879 – Questions on the Faculty of Visualising – designed to test visual recall and starting with how well participants could recall their morning breakfast table, then moving on to more general ability to visualise people. He found, to his astonishment, that scientists generally performed poorly at these tasks; and people in general were enormously variable. This work was ground-breaking and laid the template for future studies of mental imagery. Galton also discovered that people showed vast differences in the way they perceived numbers, each being highly personalised to the individuals. He further found synaesthesia in some people, who associated different numbers, letters or words with the sensation of colour. He published all these findings in Inquiries into Human Faculty in 1883. It was in this work that he introduced the word eugenics.

Statistics and anthropometrics

Image 08A Anthropometric Laboratory

Galton’s anthropometric laboratory ©UCL Galton Collection

In the mid-nineteenth century there were few ways of analysing statistical data. In 1846 the Belgian statistician Quetelet demonstrated that human measurements such as the heights of French conscripts approximated to a normal distribution, or a bell curve. Galton wondered if this could be applied to human ability, and in the absence of any independent measure analysed the results of the admission exam to the Royal Military Academy at Sandhurst, which he found approximated to a normal distribution. On the basis of this, he divided intelligence into fourteen classes; at the top end of the curve were the subjects of Hereditary Genius and at the lower end ‘idiots and imbeciles’.

In 1884 Galton was given an unprecedented opportunity to indulge his enthusiasm for the measurement of man, or anthropometry. As he said during the delivery of the Rede Lecture at The University of Cambridge in May of that year: ‘The powers of man are finite, and if finite are not too large for measurement’. The International Health Exhibition, held in London in 1884 was the occasion of the launch of Galton’s first anthropometric laboratory; participants paid an entry fee of three pence to undergo a series of tests, including height, weight, strength, breathing capacity, reaction time, hearing and vision, colour perception and judgement of length. Galton had to exercise his flair for invention as he had to design some of the instruments himself. By 1885 when the exhibition closed, Galton had measured 10000 individuals with over 150,000 separate measurements. Galton kept all the results for his records, while the subjects went away with copies of their individual results. At a time when modern statistics was a developing mathematical science, the challenge to analysis this enormous set of data was immense.


Galton’s seed experiments ©UCL Galton Collection

Galton also wanted to examine the way that traits such as height were inherited, and he was able to do this with data from 205 sets of parents and their adult children. Plotting the data he found the expected positive association between the heights of the parents and children. However, although very tall parents tended to have tall children, he found that the children tended to be shorter than their parents. Similarly, if parents were short, their children were slightly taller. These data reflected experiments of 1876 with the seeds of sweet peas; using seven groups of seeds, he measured the average diameter of 100 seeds produced by each sweet pea plant. He founded that the smallest pea seeds had larger offspring and the largest seeds had smaller offspring. Galton had discovered the phenomenon of ‘regression to the mean’, where the largest peas and parents had offspring who were smaller and closer to the mean. From this work Galton derived the so-called regression coefficient, and later, for association between any two variables, the correlation coefficient. This would pave the way for the development of statistics as a discipline, through Galton’s follower Karl Pearson and marked the beginning of a new phase of scientific investigation that allowed the objective examination of enormously complex data resulting from biological, psychological and social phenomena.

Unfortunately, Galton was unable to see that his great discovery had undermined his eugenic ideals, which proposed a more straightforward relationship between traits such as height and intelligence in parents and their children.

In 1886 Galton was awarded the Gold Medal from the Royal Society for his work on statistics applied to biology; this recognised not only his work on regression and correlation, but also his quantitative approach to all the fields that lent themselves to measurement and statistical analysis.

Image 10_Pricker gloves

Galton’s ‘pricker’ gloves ©UCL Galton Collection

Galton gave in to his predilection for counting at every opportunity; whilst at Vichy in France, he spent a few hours classifying women into six size categories; from ‘thin’ to ‘prize fat’. This gave way to his project to define a beauty map of Great Britain; to record his observations of women as ‘attractive’, ‘indifferent’ or ‘repellent’. He secreted a piece of card and a ‘pricker’ fixed on a glove which he kept in his pocket, and used the needle to make a mark at one end to signify an ‘attractive’ girl, in the middle for an ‘indifferent’ girl, and at the other end for the unfortunate ‘repellents’. He could later read off and count the results at leisure. He found ‘London to rank highest for beauty, Aberdeen lowest’, showing again that he was blindly unaware of how his prejudices coloured his observations.


Criminology and fingerprints

Image 10B_mugshot

Mugshot of Galton as a criminal ©UCL Galton Collection

Galton’s desire to quantify knew no bounds, and the experiment of the anthropometric laboratory led him to consider how measurement could be used for identification of individuals, particularly in relation to policing and detection of criminals. Galton joined in a project to examine the faces of British criminals to ascertain if ‘consummate scoundrels’ could be identified by special criminal characteristics. He started by creating ‘composites’ by exposing a number of images of different criminals onto a single photographic plate. The effect of this was to create an ‘average’ face, which did the opposite of what was expected, resulting in a very ordinary looking person. But there were other avenues to explore: a French policeman, Alphonse Bertillon, developed a system to measure a number of specific physical dimensions (eg: hands, feet, head) which could be catalogued in such a way to simplify searching for similar types. The individual profile included full-on and side-on mug-shots, which, for Galton, who believed that the psychological character manifested in physical features, were of immense interest.

Image 11_Fingerprints etc

Galton’s fingerprinting equipment ©UCL Galton Collection

Galton recognised that this profiling could be combined with fingerprinting to make a real contribution to the detection of criminals. William Herschel, a civil servant in India, had used hand and fingerprints to identify individuals in his area of administration. Henry Faulds, a missionary who had lived in Japan where fingerprinting had been used for centuries, had spent some time in the 1880s trying to convince Scotland Yard to use the technique in policing. Galton was enthusiastic about the potential of this technique, but he realised that, in order to be accepted by the policing authorities, he needed to establish that a) fingerprints do not change through adult life; b) the variety is extremely great, and c) they can be classified. He spent many years in the 1880s and 1890s, starting with Faulds’ classification system, examining large numbers of fingerprints to establish their unique attributes. He further looked at fingerprints from the same person but separated by decades to pronounce that they were unchanging over time. Classification proved more challenging, but Galton used the major characters of each fingerprint (arch (A), loop (L) or whorl (W)) to come up with a series of ten letters (one for each finger/thumb) that could contribute to Bertillon’s filing system. Gallon published his systematic study in 1892, entitled Finger Prints. This work eventually contributed to an appraisal of the options and the adoption of fingerprinting and Bertillon’s indexing system in criminology. Galton got all the acclaim, giving some of the credit to Herschel, but Henry Faulds, who had tried – unsuccessfully – to convince the police authorities of the value of fingerprinting a decade earlier, was ignored.