Our focus here is on time-series regression as a formal analytic tool in social historical inquiry. We have three interrelated purposes. First, we argue that conventional time-series regression is typically ill–suited for social historical inquiry because ahistorical assumptions and conventions regarding time undermine the historical character of social “process–as–analyzed”. Second, we present a modified time-series approach – temporally recursive regression – that takes time seriously and provides a more adequate analytic vehicle for social historical inquiry. Finally, we illustrate the promise of temporally recursive regression by using it to analyze how workplace militancy in post–war America was fueled by massive insurgency waves during successive phases of the civil rights movement.

Historians and social scientists often investigate the conditions that influence the occurrence of particular events. For instance, a researcher might be concerned with the causes of revolutionary action in some countries or the forces that unleash racial rioting in major cities. Or perhaps the researcher wishes to examine why industrial workers decide to strike or what prompts policy-makers to pass new legislation. In each of these examples, a qualitative shift occurs, from a circumstance without racial rioting in a particular city, for instance, to one with racial rioting. Event history analysis can aid researchers in uncovering the conditions that lead to such a shift.

Social historians study social, political, demographic and economic phenomena which take place in geographical space, yet “space” rarely enters historical discourse explicitly as an analytic construct. Given the recent interest among social historians in “localizing” social processes, it is unfortunate that almost all of our efforts have gone toward recognizing the significance of time, to the relative neglect of space. For example, social historians typically assume that events occurring at a given point in time are affected by earlier events. In contrast, it is less widely recognized that events occurring in one location are also affected by similar events in other areas, especially those nearby. In the rare instances in which geographical space has been central, such as in Susan Watkins' treatment of western European fertility decline during the late nineteenth and early twentieth centuries, or in the evidence of the diffusion of democracy in post-war Europe that is emerging from the “Spatial and Temporal Diffusion of Democracy” project at the University of Colorado — Boulder, the payoff has been substantial. Such innovative approaches to historical theorizing give reason to believe that thinking of events in the past as parts of a process moving through time and across space will become more commonplace. If so, then social historians will need an effective and accessible method to account for the spatial distributions of their study phenomena.

This paper illustrates some linguistic and statistical tools that can be profitably used by historians and social historians in the study of events (such as strikes, demonstrations and other types of collective conflict). More specifically, the paper shows that “semantic grammars” provide rigorous tools for the collection of rich event narratives. Semantic grammars structure information around the “canonical form” of the language: noun phrase/verb phrase, or subject, action, object and their modifiers (e.g. time, space). The fact that semantic grammars can be easily implemented in a computer environment using relational database systems (RDBMS) makes feasible the practical application of such powerful coding schemes. The data that computer-based semantic grammars make available are richer, more flexible and more reliable than those delivered by more traditional content analysis methods. They are also very well suited for the application of new tools of data analysis such as network models. Both semantic grammars and network models are fundamentally concerned with actors and their actions, with agents and agency. As such, these linguistic and statistical tools should draw sociology closer to history, traditionally much more concerned with issues of agency. I illustrate the power of both the linguistic and statistical tools using data that I collected from some 15,000 newspaper articles on the 1919-1922 period of Italian history, a period characterized by widespread working-class mobilization (1919–1920, the “red years”) and fascist counter-mobilization (1921-1922, the “black years”).

Social scientists often face a fundamental dilemma when they conduct social research. On the one hand, they may emphasize the complexity of social phenomena – a common strategy in ethnographic, historical and macro social research – and offer in–depth case studies sensitive to the specificity of the things they study. On the other hand, they may make broad, homo genizing assumptions about cases, and document generalities – patterns hold across many instances. Research strategies that focus on complexity are often labeled “qualitative”, “case–oriented”, “small–N”, or “intensive”. Those that focus on generality are often labeled “quantitative”, “variable–oriented”, “large–N”, or “extensive”. While the contrasts between these two types social research are substantial, it is easy to exaggerate their differences and t o caricature the two approaches, for example, portraying quantitative work on general patterns as scientific but sterile and oppressive, and qualitative research on small Ns as rich and emancipatory but journalistic. It is important to avoid these caricatures because the contrasts between these two general approaches provide important leads both for finding a middle path between them and for resolving basic methodological issues in social science Social scientists who study cases in an in–depth manner often see empiri cal generalizations simply as a means to another end – the interpretive understanding of cases. In this view, a fundamental goal of social science is t o interpret significant features of the social world and thereby advance our collective understanding of how existing social arrangements came about and why we live the way we do. The rough general patterns that social scientists may be able to identify simply aid the understanding of specific cases; they are not viewed as predictive. Besides, the task of interpreting and then representing socially significant phenomena (or the task of making selected social phenomena significant by representing them) is a much more immediate and tangible goal. In this view, empirical generalizations and social science theory are important – to the extent that they aid the goal interpretive understanding.

In the past two decades, social network analysis (SNA) has become a major analytical paradigm in sociology and now occupies a strategic place in disciplinary debates on a wide variety of issues. Historians, however, have been slow to adopt the approach for at least three reasons. First, the conceptual orientation of sociologists practicing historical social network analysis (HSNA) remains unfamiliar to the majority of professional historians. Just when SNA was maturing in the late 1980s and 1990s, the interdisciplinary interest in social science theory among historians, so characteristic of the 1970s and early 1980s, began to wane. The subsequent turn toward post modernist thinking in history left the profession increasingly uninformed about both classical and contemporary social theory. Second, those quantitatively-oriented historians who might be predisposed to use SNA's specialized statistical methods constitute less than a quarter of the profession today, thus the risk of SNA finding its way into mainstream historical scholarship is low to start. Third, SNA's data requirements are formidable. SNA demands evidence of social interaction among all members of a social system for a variety of behaviors, and thus necessitates a broad range of high-quality records for the place, time and activities being studied. Because historians are plagued by an incomplete historical record and imperfect understandings of past social relations, HSNA remains an inherently problematic enterprise. Yet despite conceptual, methodological and evidentiary obstacles, SNA possesses real potential for historical analysis.

Event-structure analysis (ESA) is a member of a family of formal analytic procedures designed to analyze and interpret text, in particular the temporal sequences constituting the narrative of a historical event. Its basic purpose is to aid the analyst in “unpacking” an event – that is, in breaking it into constituent parts – and analytically reconstituting it as a causal interpretation of what happened and why it happened as it did. ESA focuses on and exploits an event's “narrativity” – its temporal orderliness, connectedness and unfolding – thereby helping historians and social scientists infer causal links between actions in an event, identify its contingencies and follow their consequences, and explore its myriad sequential patterns. Unlike most other formal analytical techniques, it is completely non-numeric and non-statistical: ESA's value is largely heuristic and centered on how it relentlessly probes the analyst's construction, comprehension and interpretation of the event.