304 



SCIENCE 



[N. S. Vol. XLI. No. 1052 



democracy that we do not lose sight of the 

 first principle in democracy, which is to 

 let the control of policies and affairs rest 

 directly back on the people. 



Some enterprises should be much cen- 

 tralized, whether in a democracy or else- 

 where ; an example is the postal service : 

 this is on the business side of government. 

 Some enterprises should be decentralized; 

 an example is a good part of the agricul- 

 tural service: this is on the educational 

 side of government. 



Whether there is any danger in our new 

 nationalized extension work, which we are 

 all so glad to have and from which we ex- 

 pect so much, I suppose not one of us 

 knows. But for myself, I have apprehen- 

 sion of the tendency to make some of the 

 agricultural work into "projects" at Wash- 

 ington. If we are not careful, we shall not 

 only too much centralize the work, but 

 we shall tie it up in perplexing red-tape, 

 official obstacles, and bookkeeping. The 

 merit of the projects themselves and the 

 intentions of the officers concerned in them 

 are not involved in what I say; I speak 

 only of the tendency of all government to 

 formality and to crystallization, to ma- 

 chine work and to arm-chair regulations; 

 and even at the risk of a somewhat lower 

 so-called "efficiency" I should prefer for 

 such work as investigating and teaching 

 in agriculture, a dispersion of the initia- 

 tive and responsibility, letting the coordi- 

 nation and standardizing arise very much 

 from conference and very little from arbi- 

 trary regulation. 



In the course of our experience in democ- 

 racy, we have developed many cheeks 

 against too great centralization. I hope 

 that we may develop the checks effectively 

 in this new welfare work in agriculture, a 

 desire that I am aware is also strong with 

 many of those who are concerned in the 

 planning of it. 



Twice I have spoken as if not convinced 

 that the present insistence on "efficiency" 

 in government is altogether sound. That 

 is exactly the impression I desire to con- 

 vey. As the term is now commonly ap- 

 plied, it is not a measure of good govern- 

 ment. 



Certain phrases and certain sets of ideas 

 gain dominance at certain times. Just now 

 the idea of administrative efficiency is 

 uppermost. It seems necessarily to be the 

 controlling factor in the progress of any 

 business or any people. Certainly, a peo- 

 ple should be efficient; but an efficient 

 government may not mean an efficient 

 people — it may mean quite the contrary 

 or even the reverse. The primary purpose 

 of government in these days, and partic- 

 ularly in this country, is to educate and 

 to develop all the people and to lead them 

 to express themselves freely and to the 

 full, and to partake politically. And this 

 is what governments may not do, and this 

 is where they may fail even when their 

 efficiency in administration is exact. A 

 monarchic form may be executively more 

 efficient than a democratic form; a despotic 

 form may be more efficient than either. 

 The justification of a democratic form of 

 government lies in the fact that it is a 

 means of education. 



The final test of government is not exec- 

 utive efficiency. Every movement and 

 every circumstance that takes starting- 

 power and incentive away from the peo- 

 ple, even though it makes for exacter ad- 

 ministration, is to be challenged. It is 

 specially to be deplored if this loss of 

 starting-power affects the persons who 

 deal first-hand with the surface of the 

 planet and with the products that come 

 directly out of it. 



If it is important that the administra- 

 tion of agricultural work be not over- 

 much centralized at Washington, it is 



