Strikes as Sequences of Interaction: The American Strike Wave of 1886

Theories of strikes—or any kind of conflict—are theories of interaction. Marxists envisage class conflict as a dialectical struggle between proletariat and bourgeoisie. Neoclassical economists imagine strikes as an unintended consequence of bargaining between rational actors. The tumult of class conflict is far removed from the quiet of rational bargaining, yet these theories are alike in this respect: strikes ensue from dyadic interaction. Adding political authorities to the dyad of workers and employers extends interaction to three sets of actors.

In theory this interaction is clear. Nevertheless, as Roberto Franzosi (1995: 17) emphasizes, "strategic interactions among the players involved have been lost in the literature." In statistical analysis, strikes are conceived of as the outcome of a linear combination of independent variables, which are interpreted as the determinants of workers' decisions. By implication, strikes are unilateral acts by workers alone. David Card and Craig Olson (1995) are exceptional for modeling the duration and outcome of strikes as the result of decisions on both sides. Labor historians, too, focus on workers. The "consciousness" of the working class is dissected and debated at length, while that of their adversaries remains a mystery. In part, this approach reflects the bias of sources. "The strikers are exposed to the full glare of publicity," observes Michelle Perrot (1987: 250), while employers make decisions in secret.

Interaction between workers and employers is the subject of this article. It analyzes strikes as sequences of interaction in order to capture their essentially diachronic character. How strikes actually unfold over time has attracted little attention. Historians are more interested in larger processes, such as class formation, and they are inevitably hampered by a lack of detailed evidence. From interviews and participant observation, sociologists can provide richly detailed accounts of individual strikes (a wonderful example is Fantasia 1988, chap. 3). What is lacking, however, is any method of systematically analyzing larger numbers of strikes. In statistical analysis, the determinants of strikes operate instantaneously. Within this statistical framework, event-history analysis can introduce a diachronic element; Carol Conell and Samuel Cohn (1995) use this technique to demonstrate how a strike by one group of workers makes it more likely that other workers will go on strike. Moving beyond this framework, sociologists have introduced new methods for analyzing sequences (reviewed by Abbott 1995). Andrew Abbott applies optimal matching to measure the resemblance between different sequences (e.g., Abbott and Hrycak 1990); Larry Griffin (1993) uses event-structure analysis to formalize the causal connections among events. These diachronic methods are completely abstract, and so they can be applied to many kinds of events. By contrast, the method of analysis presented here takes advantage of the elemental structure of a strike, derived from the dyadic relation between workers and employer.

Because conflict can be averted by the decision of either side, strikes are the tip of an iceberg. Submerged beneath the surface (not usually recorded by statistical agencies) are a huge number of interactions—including unfulfilled threats and preemptive concessions. The end of a strike, moreover, does not necessarily terminate the struggle. There may be a series of sequences of interactions, involving the same workers and employer(s), before one side or the other finally accepts the outcome. In situations where employers did not accept collective bargaining, they often granted concessions only to revoke them some months later. Workers then faced the choice between a defensive strike and outright surrender. Therefore some strikes must be explained by prior interactions, rather than by changing economic or political circumstances.

This article examines one particular episode: the strike wave that swept the United States in the middle of the 1880s. It crested in the first week of May 1886, when over 200,000 workers struck for a shorter working day. This was one of the great outbursts of class conflict in American history; relative to the size of the labor force, the number of strikers...