NATURE 



797 



THURSDAY, AUGUST 26, 1920. 



Editorial and Publishing Offices: 



MACMILLAN €r CO., LTD., 



ST. MARTIN'S STREET, LONDON, W.C.2. 



Advertisements and business letters should be 

 addressed to the Publishers. 



Editorial communications to the Editor. 



Telegraphic Address: PHUSIS, LONDON. 

 Telephone Number: GERRARD 8830, 



The Forthcoming Census. 



THE Census Act of 1920 will have one 

 great advantage over previous Census 

 Acts — that it will be a permanent measure, 

 and not, as they have been, limited to 

 the operation of taking the one census that 

 was at the time in contemplation. We have 

 travelled far from the days when the numbering 

 of the people was considered to be an offence that 

 would provoke Divine anger, and it is quite time 

 that the old hesitating policy of passing a new 

 enactment and creating a new staff and machinery 

 every ten years, which doubtless had its origin in 

 consideration for those superstitious scruples, 

 should be definitely abandoned. The system had 

 one indirect advantage while it lasted. For the 

 eleven decenniums since 1801 the eleven separate 

 Acts that have had to be passed have been gradu- 

 ally strengthened and made more workable, as 

 experience has shown what improvements it has 

 been possible to introduce into the practice, and 

 thus have ripened into the materials for a per- 

 manent statute. All the same, the necessity for 

 organising a scratch staff of new men every ten 

 years, and dismissing it as soon as the census 

 work was over, has been a great drawback to the 

 efficiency of the Department, and it is to be hoped 

 that one result of the new Act will be to enable 

 the Census Office so to distribute its work over 

 the whole decennium as usefully to retain the 

 services of an experienced and competent staff of 

 permanent officials. Much credit is due to the 

 successive controllers of the census for the good 

 work they have done under all disadvantages, and 

 it is no disparagement to them to say that they 

 have been hampered by circumstances. 



The Act contemplates, but does not require, a 

 quinquennial census. It enacts that no census 

 shall be required to be taken in any part of Great 

 NO. 2652, VOL. 105] 



Britain in any year unless at the beginning of that 

 year at least five years have elapsed since the 

 beginning of the year in which a census was last 

 taken in that part of Great Britain ; but it leaves 

 to the King in Council to fix the date on which 

 each successive census is to be taken. There can 

 be no doubt that for some statistical purposes the 

 interval of ten years is too long, and that not 

 infrequently in the course of that interval events 

 arise that materially affect the applicability of 

 averages drawn between censuses distant ten 

 years from each other. With careful organisation 

 a quinquennial census might be made the rule, 

 but the Act leaves this question entirely open. 

 It allows, however, of a special local census being 

 made, independent of the date of the last previous 

 census, upon the application of a local authority 

 through the Minister of Health to his Majesty 

 in Council for the purpose of facilitating the due 

 performance by the authority of its statutory 

 duties. 



An important provision of the Act is that which 

 prohibits inquiry at a census into any particulars 

 other than those specified in the schedule to the 

 proposed enactment. These are : — Name, sex, 

 age; occupation, profession, trade, or employ- 

 ment ; nationality, birthplace, race, language ; 

 place of abode and character of dwelling ; educa- 

 tion ; infirmity or disability ; condition as to mar- 

 riage, relation to head of family, parentage, issue ; 

 and "any other matters with respect to which it 

 is desirable to obtain statistical information with 

 a view to ascertaining the social or civil condition 

 of the population." The generality of this last 

 item would no doubt be controlled by the ejusdem 

 generis principle of interpretation, and there need 

 be little fear that any Order in Council would 

 authorise an undesirable extension of it; but care 

 would still have to be taken against the use of 

 the census for indirect or partisan motives. It 

 may be stated as a general principle that the 

 more you increase the number of items of in- 

 formation that you demand, the more you dimin- 

 ish the probability that the information you 

 actually obtain will be accurate. A wise investi- 

 gator, therefore, while naturally anxious to get 

 all the sound information that he can, will care- 

 fully distinguish between that which is essential 

 and verifiable, and that which cannot be relied 

 upon. 



Much light may be expected to be derived from 

 the census returns upon subjects that have re- 

 cently been prominently before the public, such as 

 the diminution in the birth-rate, the extent to 



D D 



