"NUTS AND MEN."
[To THE EDITOR OP THE uSPECTATOR.1 Six,—In your article last Saturday, on a lecture that I gave at the Royal Institution, you dwell on the difficulty of classifying men according to their faculties in the way I proposed. It is easy to range nuts in the order of their bigness, but you think it impracticable to range men in order of merit, as regards any quality, or still less, any group of qualities. With your permis- sion, I will shortly-state the grounds of my contrary belief, having been precluded by want of time from doing so when I lectured, for I had to touch on many points, and I explained that these would be more fully examined in a future publication.
I appeal to the experiences of official life. Here is one of them : —An imaginary case, typical of many with which I myself and probably not a few of your readers are familiar. The executive officers of some association want a confidential clerk at a fair salary, and receive applications from more than one hundred applicants. The choice is referred to a committee, who meet and quickly despatch three-fourths of the list, seeing at once that they have no chance of being elected ; they consider about the rest, and reserve a dozen for further inquiries, which they direct to be made, and they adjourn. At the next meeting, having additional facts before them, the choice narrows to six, whom theysee and speak to ; then, after further discussion, the choice is reduced to three. A second set of inquiries are directed, and the committee again ad- journ. At the third meeting the result of the latest inquiries is well -considered, and one of the three candidates is elected by a nearly unanimous vote. The qualifications needed for a confidential clerk are numerous, and no two applicants possess those qualifications in the same proportion, yet, as a matter of fact, committees are found to come to a pretty unanimous estimate of their value in the cross, and I have no reason to doubt that independent committees would arrive at nearly the same conclusions. I want biographers and critics to make a practice of taking the same pains to rank the men they write about among their fellows and competitors, as our imaginary committee took in respect to the three men among whom their choice finally lay. Their proceedings, with variations of detail, necessarily characterise every selection to a coveted post, whether it be that of the foreman to a factory, the representative of a borough, the commander of an army or his staff, a bishop or the nominee to a living, a diplomatist, a Royal Academician, a -scientific medallist, or any other of an almost endless number of cases.
In days like these of competitive examinations, it is superfluous to speak of the precision with which candidates are classed in order of merit with respect to the special subject of competi- tion, whether athletics, rifle shooting, or scholastic ability, of any particular kind. Nay, I have lately heard, and this instance is -nearer to my present purpose, of a party of distinguished examiners in their vacation rambles appraising everything by marks, on a prearranged system. They severally made their estimates, and were surprised at their accordance, whether in respect to beauty of scenery, attractiveness of lady acquaintances, or the provender at their tables d'hotes.
Above all, I would point to the extraordinary accuracy of estimate by experts of the money value of commodities. Inde- pendently of the estimate of quality, money value is a function of two variables,--supply and demand ; yet even when both of these
vary through wide ranges, as in bijouterie, china, horses, and pictures, auctioneers and dealers make very shrewd guesses, confirmed by the results of the sale. When the demand and sup- ply are both pretty constant, its to the greater articles of com- merce, corn, iron, cotton, tea, &c., experts differ by the merest shade in their estimates. I was astonished last autumn, during the Bradford meeting of the British Association, to observe the wool-sorters at work. They rapidly seized one fleece after another, and plucked and tore each of them into half-a-dozen or more por- tions of wool of different classes of fineness. A trifling uncertainty in the estimating powers of these men would make a serious differ- ence in the yearly profits of the undertaking. It is not too much to say that we are each of us obliged to act on our estimates on many occasions every day of our life. We are guided by them in selecting our purchases, and the shops where we make them ; we select our amusements, our associates, even our dishes at dinner. The practice is universal of comparing merits, but it has not pre- viously been shown that a series of simple comparisons supplies the data for a standard scale of measurement of the quality in question.
I show, first, that, by a succession of comparisons, we obtain all that is necessary for a complete classification as regards any speci- fied quality or group of qualities. Then I show that, having got a classification, never mind of what, it gives by implication a standard scale of measurement of the quality in question. Lastly, I point out that by proceeding in a converse direction to that followed by Quetelet, we may so graduate our scale as to make it approximately a scale of equal parts, and more nearly so than by any other general system of graduations that can be mentioned. The " Common Statistical Scale" which I propose is a purely natural system, applicable to all classifications, and has remarkable advan- tages, in addition to its being a uniform system and approximately a scale of equal parts. Into these I cannot enter now. There are also some objections, as yet unmade, which I have considered, and which, if made before I have occasion to write more fully on the subject, I am prepared to answer ; but I have no right, unasked, to inflict this description on your readers. I can only say that I am ready to defend my position.—I am, Sir, &c.,
FRANCIS GALTON.
[We can only express our wonder, and repeat our belief that what Mr. Galton has succeeded in doing, is in exposing the utter inapplicability of physico-scientific methods to intellectual and moral subjects. Does he suppose that when a Committee agrees on the superior practical fitness of a particular person over many others for a post involving given duties, such a Committee is agreed, or could agree, on the relative amount of the separate moral elements involved in that fitness which the various candi- dates possess ? We can imagine no more profitless or idle task than the attempt to draw out a Statistical Scale (say) of Candour or of Power of Repartee, and to arrange the public men of this generation in it, except indeed doing the same thing for a con- siderable number of qualities, and giving the reasons for the place assigned in the biographies which would be rendered unreadable by the process.—En. Spectator.]
Previous page
