The Switchboard Problem: Scale, Signaling, and Organization in Manual Telephone Switching, 1877–1897

Abstract

The Switchboard Problem: Scale, Signaling, and Organization in Manual Telephone Switching, 1877—1897 MILTON MUELLER A telephone exchange is a radical rearrangement of'social space. It brings any two speakers together on demand, regardless of where they are located. By thus collapsing social space, it also vastly expands its scale, making millions of people who would otherwise be inacces­ sible to each other capable of instant, real-time conversation. The very radicalism of this act of social integration led the early developers of telephone switching into one particularly troublesome dilemma. In 1881, the manager of the fledgling Milwaukee Bell telephone exchange complained that “the general impression among subscribers is that if an exchange of 100 subscribers can be run at [rates of] $12 a month, then an exchange of 1,000 ought to be run for about 40 cents. You can’t make them believe anything else.” In fact, as he was painfully aware, telephone exchanges became more expensive to run (per subscriber) as the number of subscribers rose. The “economies of scale” eagerly anticipated by customers did not exist. On the contrary, growth brought only rate increases, and the large exchanges in New York, Chicago, and Boston charged three or four times the rates of smaller cities. The locus of the problem was the switching process, and the problem was not technical so much as it was organizational. As the number of subscribers to a telephone system grows, the number of possible connections among them grows much faster—roughly as the Mr. Mueli.er is a Ph.D. candidate in the Annenberg School of Communications at the University of Pennsylvania. He thanks the members of the AT&T historical archive; stimulating discussions with Alan Gardner and the support and encouragement offered by Robert Garnet and Robert Lewis are especially appreciated. He also thanks Dr. Thomas Hughes, whose University of Pennsylvania graduate seminar led to the exploration of the topic, and his adviser, Dr. Carolyn Marvin, for permitting one more independent study project.© 1989 by the Society for the History of Technology. All rights reserved. 0040- 165X/89/3003-0008$01.00 534 Manual Telephone Switching, 1877—1897 535 square of the number of subscribers. Consequently, switchboards became increasingly expensive to construct, and the operations needed to make connections became increasingly complex and slow, as more people joined the exchange. This diseconomy of switching was the most important “reverse salient” encountered during the early years of telephone development. More than any other single factor, it constricted progress by linking system growth to slower service and higher costs. It also was unusually persistent. It took the Bell system twenty years and three generations of switching technol­ ogy to come up with a long-term solution to the problem. Not until the introduction of automatic lamp signals, the common battery switchboard, and the development of a science of traffic engineering between 1892 and 1897 did switching cease to be a constraint on telephone system growth. This article differs from other works on the history of switching technology both in terms of its subject and its approach. The account of manual switching technology in volume 1 of the History of Engi­ neering and Science in the Bell System series,1 while containing much useful information, fails to identify the central role played by the diseconomies of growth in the development of the technology. Other works on the history of switching tend to separate automatic from manual switching in order to concentrate on the former. A. E. Joel’s history of switching in the Bell system, for example, begins in 1925, after Bell had committed itself to automation.2 Robert Chapuis, in an otherwise thorough treatment of the first 100 years of switch­ ing, devotes only a few pages to the manual era.’ In effect, develop­ ments before 1910 are consigned to the prehistory of switching. This bias is understandable; from a purely technical viewpoint, automatic switches are more interesting than their predecessors. But this article is not about the technical apparatus of switching per se. Its subject is the confrontation with the multiplying possibilities of an expanding network, a problem that was most visible during the manual era. This confrontation is of interest for two...