奎克的藏经阁 · 原书印刷页

Labor and Monopoly Capital The Degradation of Work in the Twentieth Century · 第 139 页

PDF 物理页索引 82 · Acknowledgments

书名
Labor and Monopoly Capital The Degradation of Work in the Twentieth Century
作者
未记录
印刷页
139
PDF 物理页
82(从 0 开始)
章节
Acknowledgments
UUID
bde76793-4c65-415a-a237-6ccacc2e0802
来源类型
original_book
页面 SHA-256
c4f68cfd00004275df64c53290a8695e741021c844baeb48ad1e429e174ea555

Numerical control is thus used to divide the process among separate operatives, each representing far less in terms of training, abilities, and hourly labor costs than does the competent machinist. Here we see once more the Babbage principle, but now in a setting of technical revolution. The process has become more complex, but this is lost to the workers, who do not rise with the process but sink beneath it. Each of these workers is required to know and understand not more than did the single worker of before, but much less. The skilled machinist is, by this innovation, deliberately rendered as obsolete as the glassblower or Morse code telegrapher, and as a rule is replaced by three sorts of operatives.

First, there is the parts programmer. The process of taking the specifications of an engineering drawing and recording them on a planning sheet is essentially the same work as that previously done by the machinist when, drawing in hand, he approached a given job. But the parts programmer is not required to know everything else the machinist knew: that is to say, the actual craft of cutting metals in its execution on the machine. He learns instead merely the shadow of the process in tabulated and standardized form, and with that his learning ceases. He is taught to work this information up in a manner suitable for coding.

“The planner," in the words of one description, “simulates the machining done in the shop. . . . He goes through every step in very much detail, leaving no decision to be made later at the machine. The planner determines the feeds and speeds, the cutters required, and even the miscellaneous functions such as when coolant is to be on or off. Importantly, he determines the feed rates and the depths of cut to be taken."l6 To convert the specifications on the drawing into a planning sheet requires, in the case of much of the equipment used for numerical control, only a knowledge of blueprint reading such as the craftsman acquires in his first months of apprenticeship, plus the basic arithmetic of adding and subtracting, plus the use of standard data on machine capabilities. This may be made more or less difficult by the type of equipment used and the complexity of the job in hand. In a recent development, the machining specifications are themselves stored on computer tape and the programmer need only make a description of the part (both rough form and finished form) by simply converting the engineering drawing into a listing of dimensions, using nothing but simple machine-shop terms. The computer delivers a machine control tape, a printout of what is on the tape, a tool list, and the computed job cycle time. This system, it is claimed, reduces program time for a part that once required four or five hours to only twenty or thirty minutes. One company advertises another programming system that uses a“10-word program vocabulary" and says: "“Within a week a shop man can program effectively."17 An article in Monthly Labor Review says: "Most of the functions of the skilled machinist have been shifted to the parts programmer. Consequently, skilled machinists often staff these positions."18 This may have been so at the outset, but the job of parts programmer now increasingly falls to the technical college graduate (often from the two-year junior college) who fits the “labor profile" of this desk job more closely than the machinist, especially in being cheaper.

The next job is that of converting the planning sheet into machine-readable form—usually a paper tape punched on a simple coding machine. Here the candidate is immediately selected by unanimous choice: the “girl" machine operator who learns her job in a few days, attains optimum efficiency in a few weeks or months, and is drawn from a large pool at hardly more than half the pay of the machinist.

So far as the machine operator is concerned, it is now possible to remove from his area of competence whatever skills still remain after three-quarters of a century of “rationalization." He is now definitively relieved of all the decisions, judgment, and knowledge which Taylor attempted to abstract from him by organizational means. The true “instruction card"—Lillian Gilbreth's “self-producer of a predetermined product"—is at last fully revealed in the program tape. “Numerically controlled machines," we are told by one authority, “are fundamentally easier to operate. The skills required of an operator are less than with conventional machines, where he must often be a trained machinist. With numerical control equipment, the operator must, of course, know his equipment. He must have the training and intelligence required to perform several rather straightforward prescribed routines, but he does not possess the technical skills of the experienced machinist. The intelligence corresponding to these latter skills is on the tape in numerical control."

The difference between the trained machinist—even one so limited in craftsmanship that he can operate only a single machine—and the operator of a numerically controlled machine is often understated, both by managers, who are prepared (in public at least) to conceal the downgrading in the interest of a smoother transition and for public relations reasons, and also unions, for whom the exaggeration of the “demanding nature” and “increased responsibility" of the new process is a routine part of contract bargaining. But some idea of the manner in which managers view this difference can be seen in the following response to a University of Michigan survey of companies using numerical control: "Cost of developing and training an operator to produce identical parts by conventional methods and machines compared with NC machining system is approximately 12 to 1.20 This would mean that if it takes four years to give a machinist his basic training, an operator of the sort required by numerically controlled machine tools may be trained in four months. Experience bears this out.\*

138 Labor and Monopoly Capital

Machinery

139

\* This is not to say that, in unionized situations, the pay of machinists is immediately reduced to operator levels the moment numerical control is introduced. In some exceptional instances, where very few numerically controlled machine tools have