TRANSACTIONS OF THE SECTIONS. 123 



individual instances. In dealing with the individual human being every thing is 

 uncertainty : in dealing with man in the aggregate, results may be calculated with 

 the precision and accuracy of a mathematical problem. To take a familiar instance, 

 — the length of a single life can never be known beforehand ; but by the accurate 

 keeping of returns the aggregate length of ten thousand or a hundred thousand lives 

 is easily ascertained. This aggregate length, the conditions of life being generally 

 the same, approximates to a constant quantity', however often the experiment be 

 repeated ; and from that quantity, thus obtained, we deduce an average, which, as 

 the experience of every insurance office shows, is near enough to the truth for ordi- 

 nary purposes of calculation. Accidental diversities, whether of internal constitution 

 or of external circumstances, tend to neutralize one another. Their influence 

 diminishes as the area of investigation increases, until, if that area be sufficiently 

 extended, we are justified in disregarding them altogether, and in admitting as 

 approximately, if not as absolutely true, the general inference to which our suc- 

 cessive trials point. I will not lead you into those strange and startling conclusions 

 to which Quetelet has come, when comparing some of the averages obtained with 

 one another, and representing them in mathematical form ; he finds in the laws thus 

 discovered a close resemblance to, perhaps an actual identity with, those which 

 operate in physics ; as, for instance, when he lays it down that the obstacles which 

 oppose the increase of population act in a manner exactly the same as does the 

 resistance of the medium in which a body moves to the motion of that body. Wide 

 as is the field of thought which such a suggestion opens, it must probably be, for 

 many years, premature to enter it : the laws as yet made known to us by statistical 

 research are too few to allow of generalization relative to their mutual inter-con- 

 nexion. Enough to cite the dictum of Quetelet, confirmatory of what was said 

 above, "All observation tends to confirm the truth of this proposition, that that 

 which concerns the human race, considered collectively, is of the order of physical 

 facts : the greater the number of individuals, the more completely does the will of 

 individuals disappear, and allow the series of general facts, which depend upon the 



causes by which society exists and is preserved, to predominate We 



must admit, that on submitting to careful experiment unorganized bodies, and the 

 social system, we are unable to say on which side causes act in their efi^ects with the 

 greatest regularity." 



This, then, is the first characteristic of statistics as a science : that it proceeds 

 wholly by the accumulation and comparison of registered facts ; — that from these 

 facts alone, properly classified, it seeks to deduce general principles ; and that it 

 rejects all a priori reasoning, employing hypothesis, if at all, only in a tentative 

 manner, and subject to future verification. It starts from the assumption, verified 

 by many trials, that human action, fluctuating as regards the human unit, is 

 approximately invariable as regards the masses which make up society. But there 

 is another aspect in which it may be considered. As a rule, the degree of certainty 

 which attends any science is exactly proportioned to the extent to which such science 

 admits of the application of numbers. We know what has been done for chemistry 

 by the discovery of a single numerical law — the theory of definite proportions — 

 turning, by one stroke,' into a science, what was before little more than a collection 

 of important, but detached observations. And what we aim at in statistics is, to 

 substitute for vague phrases, intended to express certain qualities, arithmetical 

 formulae, by which the came idea may be conveyed with a precision to which 

 language alone cannot attain. For instance, the uneducated man, speaking of a 

 climate or season of the year, will say only that it is warm, hot, or very hot ; the 

 statistician registers the temperature of each day, strikes an average, and gives his 

 result, in numerical form, extending, it may be, over a period of several years, and 

 calculated, accordingly, with the most absolute accuracy of which human investi- 

 gation is capable. Again, the traveller, in describing a nation which he has visited, 

 writes that offences of violence are exceedingly common, probably more so than in 

 any other country ; the statistician obtains returns of convictions, distinguishes the 

 different classes of crime, ascertains the per-centage of murder, or assaults per head, 

 on the total population, allows for the probable amount of undetected criminality, 

 and finally compares these results with others similarly obtained in other parts of 

 the world. 



