Industrial Research 



271 



of making a case foi- himself, which is exactly the spirit 

 most certain to breed jealousy. Similarly, a salary 

 policy based on dollar returns is essentially unjust, for 

 the money value of various bits of theoretic^al work luis 

 almost no correlation with the scientific acumen which 

 they requhe. This does not mean that a mathe- 

 matician's pay should, in the long run, be independent 

 of the dollar value of his services. It means only that 

 whether he gets a raise this year, and how big it shall 

 be, should properly be based on the size, cliaracter and 

 satisfaction of his clientele, and not upon the commer- 

 cial importance of the questions they saw fit to bring 

 him last year. 



Ked tape is easily avoided by avoiding it. No 

 engineer, whatever his rank in the organization, ought 

 ever need permission to consult a mathematician in 

 the company's employ, and the mathematician in turn 

 ought not need a specific work order or expense allow- 

 ance before giving his advice. In this respect ho should 

 be on the same basis as the free lance investigators 

 who are to be foimd in most large research laboratories, 

 and who are generally known as staff engineers. 



Unavailability is a more serious matter. It is well 

 recognized that in industrial research the urgent job 

 always tends to take precedence over the important 

 one. Left to themselves, fundamental studies give 

 way to the detailed development "which ought to go 

 into production next month." Mathematical studies 

 are no more susceptible than other fundamental 

 research to such interruptions, but the effect upon the 

 career of the mathematician may be more far reaching, 

 for as soon as he is assigned an urgent project of special 

 character his availability as a consultant ceases or at 

 best is temporarily impaired. If his value to the 

 industry is greater as a project man than as a con- 

 sultant this need not be a cause for regret; but to turn 

 a good mathematician into a poor engineer, or an 

 irreplaceable mathematician into a replaceable engmeer, 

 is unfortunate for both employer and employee. 



The Mathematical Research Department 

 of the Bell Telephone Laboratories 



In the Bell Telephone Laboratories, men of this type 

 have been grouped together as a separate organization 

 imit. They have no more specific function than to be 

 helpful to their associates in other parts of the Labora- 

 tories. No engineer is obliged to consult them about 

 any phase of his work; no particular jobs come to them 

 by reason of prerogative; conversely, there is no sort 

 of help which an engineer or physicist may not seek 

 from them if he so desires. No routine need be com- 

 plied with in advance in order to secure their services, 

 and no report is required afterwards, though written 

 reports are frequently prepared when needed for scien- 

 tific record. The expense of the group is distributed 



broadly over the activities of the Laboratories, not 

 charged to specific jobs. Every eflort is made to 

 maintain a spirit of service among the members of 

 this group, and thougli rosponsil)i!ity for engineering 

 projects occasionally descends upon them, it is regarded 

 as an undesirable necessity to be avoided whenever 

 possible and litiuidatcd at the earliest opportunity. 



The group has functioned successfully for a number 

 of years. Its members arc respected by their engi- 

 neering associates, and like their jobs. Information 

 regardmg their activities reaches management almost 

 entirely thi'ough spontaneous acknowledgments made 

 by the engineers they assist. These expressions of 

 appreciation are generous, but rather erratic in that 

 they concentrate attention first on one man, then on 

 another, as the genius and training of the individual 

 happen to click with the important job of the moment. 

 This has not affected the morale of the group adversely, 

 probably because a serious effort is made to avoid 

 erratic salary revisions in which the man who is at the 

 moment in the limelight benefits at the expense of 

 others who are doing equally good but less conspicuous 

 work. 



From the standpoint of the men, tlie principal 

 advantages of being associated together instead of 

 distributed through the engmeering departments, is the 

 stimulus of contact with men of like interests. From 

 the standpoint of management, the advantages are 

 wider availability, greater flexibility in matching the 

 talents of the man with the requirements of the job, 

 and a more uniform appraisal of ability because of 

 supervision by a man of adequate mathematical back- 

 ground. 



So far as is known, mathematicians have not been 

 organized into separate administrative groups in other 

 industries. In most laboratories their numbers have 

 been thought too small to make such an arrangement 

 feasible, and they have been treated as staff engineers 

 distributed throughout the various general departments. 

 It is believed, however, that tliere are a few industries 

 in which this arrangement could be introduced with 

 profit at this time, and that it has sufficient merit to 

 justify its adoption wherever possible. 



The Mathematician in the Small Laboratory 



Wliat has been said above relates primarily to condi- 

 tions in large industries. The qualifications for success 

 in the small industry are not dissimilar, though the 

 relative emphasis to be placed upon them is somewhat 

 different. Matters of personality (gregariousness, un- 

 selfishness, etc.) are not quite so important, because 

 they are oft'set to some extent by the friendly coherence 

 of the small group. On the other hand, a strong in- 

 terest in things as well as ideas, and the ability to 

 translate from the language of concrete experience to 



