Comparative and Historical Sociology: Lecture Notes

Mathieu Deflem

This is a copy of the lecture notes of my graduate seminar in Comparative and Historical Sociology (SOCY 729) which I first taught at the Unversity of South Carolina in 2007.

Please cite as: Deflem, Mathieu. 2007. "Comparative and Historical Sociology: Lecture Notes." Unpublished paper. Available online via  


       1) The Centrality of the Classics
       2)  Principles of Research Design

       A. On the History of Historical Sociology
       B. Historical Research
       C. Comparative Analysis (The Case)
       Note: Research Proposals (NSF)

       1) Comparisons, Explanation, and Theory
       2) Path Dependence and Explanation
       3) Sociology and History


1) The Centrality of the Classics 

       See my notes on the sociological classics.

2) Principles of Research Design

      See my notes on research design.


See the References section below for some of the sources used in this part.


Introduction: History and Sociology

Doing history is making history. Writing history is an intellectual activity directed by concepts and (more or less explicit) theoretical questions rather than a registration of chronologically ordered facts. Sociologists develop theories and methods in order to study empirical realities. Description and analysis are only analytically distinct (Bonnell 1980). b) The Unique and the General Sociology was traditionally (represented as) nomothetic (generalizing), while history was idiographic (particularizing). 

Historical and/or contemporary topics of inquiry are not by themselves generalizing or particularizing. Instead, data are presented within an analytically relevant model as being unique or as representing a broader pattern. Relatedly, the tension between freedom and causality (choice and determinism) can be conceived in terms of a dialectics of history as contextualized human activity. Also, sociologists as well as historians can be interested in unique causes as they can be treated as manifestations of principles of broad developments across space and time (see Tilly 1997). Today, sociological notions of history as evolution have largely been expelled from modern sociological thought. Historical conceptions that emphasize fluidity and contradictory currents (Marx’ dialectics) appear to have survived better. But a key problem concerns deterministics versus actionism, structure v. agency (e.g., Althusser v. Thompson).

The key is that historical sociology is not merely a sociological study of the past, but an intrinsic part of a sociology of the present: in order to explain the structures of contemporary societies, one must investigate their historical origins and development. It has been remarked that “Sociology is history with the hard work left out; history is sociology with the brains left out” (Cahnman & Boskoff 1964:1). In any case, history and sociology have historically developed into two distinct academic disciplines. Though there may be considerable overlap between both history’s and sociology’s material perspectives (human affairs, interaction, society), they differ formally in manner of approach, theoretically, methodologically, and, not least of all, institutionally. 

The classics were in varying degrees all historical, but mostly evolutionary. Much of post-war sociology was rather skeptical towards historical inquiry. Whereas Durkheim had introduced the distinction between causal explanation and functional analysis, Parsons and others had divorced causal-historical research and functional-synchronic analysis, whereby the latter was often the privileged perspective (survey research). The Chicago School likewise was not historically minded. Most historical perspectives of those days were development and modernization theories (societal differentiation). Likewise, history was dominated by the ideal of neutral-descriptive analysis of concrete historical events, and rather hostile towards abstract conceptual work in the grand theory tradition of sociology. The German historian Leopold von Ranke had delineated history as the chronological representation of the “way things really were” (“wie es eigentlich gewesen ist”). See: Stedman Jones (1976) who questions the equation of history with the past. From the late 1950s onwards, historians directed attention to sociology for conceptual clarification, while sociologists began focusing on history’s sense for empirical detail in order to complement and/or rectify grand theory. Influential, too, were the critique on modernization, inter-actionism’s anti-theoretical empiricism, and the rise of historical Marxism. So here we have the development of the so-called second wave of historical sociologists: they were anti-evolutionists, anti-modernization, and anti-quantification. The emphasis was on social change.

2. Historical Sociology 

Since the Second Wave

Sociology is closely associated with modernity (see Adams, Clemens, and Orloff 2005). But modernity is changing and, hence, so is sociology. Historical sociology is one area where this changing theorizing is taking place. The second wave is changing, though there is no real third wave yet... To the Second Wave Historically, sociology was very attuned to comparative and historical thinking, see e.g., Marx, Durkheim, Weber, Tönnies, de Tocqueville. This is the first wave. But upon the institutionalization of sociology after WW II, things changed. Until the 1960s-1970s, much of US sociology was ahistorical and rooted in modernization theories of development. In reaction, however, some strong individuals emerged (Moore, Lipset, Bendix, Tilly) that led to the second wave of the 1970s and 1980s. Theoretically, much of this work was Marxian and Marxist in orientation. The works focused on political economy, the centrality of the state, the macro-structures of society, the material modes of production, not the cultural level. Modernization theory was discarded in favor of conflict-theoretical perspectives. Historical sociology then also institutionalized in the ASA and in specialized journals. Historical sociologists were accepted, but there were some difficulties. The use of small-N research and the Millean logic were questioned. Other methods were then developed. Mainstream sociologists critiqued historical sociology for not being sufficiently abstract, general and scientific, historians at once critiqued them for not being sufficiently attuned to the particularities of each case.

Beyond the Second Wave?

Theoretically there is no unifying analytical framework. 1) Institutionalism: this challenged the conception of an over-all deterministic structure, e.g., work on social policy. But still the emphasis remains on the political economy (and on politics, as in this book). 2) Rational Choice Theory: emphasis on the utilitarian assumptions of individuals and groups growing out of them. This development is not yet widespread but it will grow. 3) The Cultural Turn: this is very influential presently. The emphasis is on identity, on culturally variable categories, ‘imagined communities’, the self. 4) Feminist Thought: this has opened up many new topics of research and even challenged the center of society such as the state, rather than just the private realm of the family. Yet the masculine themes still resist. 5) Post-Colonialism: sociology is build on the European experience, which is now being challenged... What about the other peoples? So, at present, there is a certain lack of unity...


1. Historical Methods of Observation

a) Content and Document Analysis

Content analysis refers to the quantitative study of written and oral documents. This requires (probability) sampling of the units of analysis in a source, codification of the units, and classification to reveal content. (Krippendorf, K. 1980. Content Analysis. Sage.) Document analysis refers to the qualitative study of traces of the past: it involves the in-depth investigation of sources and aims at hermeneutic understanding.

b) Investigating Historical Sources (Pitt 1972)

See Pitt, David C. 1972. Using Historical Sources in Anthropology and Sociology. New York: Holt, Rinehart and Winston. Historical research is the study of the past through an examination of the traces the past has left behind, the sources of historical work, historical materials or documentary evidence, including: material remnants, written and/or otherwise recorded sources (primary and secondary), and oral history. The issue of accuracy is very relevant, though not exclusively so. Sources -- Historical research // Historical event (the past) -- Portrayal thereof (history)

The procedure of historical research typically involves:

1) Identification and Selection: relevant sources have to be identified and found. Types of sources: public and official archives (the state); social movements sources (cultural groups); business and company documents (economic); letters and diaries (private); literature (fictional); media sources (public opinion); maps and statistics (representations). Problems of availability (materials are lost) and access (sources are available but cannot be accessed, e.g. classified documents). Here are physical and social hurdles.

2) Registration and Classification: on the basis of formal and substantive criteria, depending on the research needs (e.g., the form of the document, time-period, the producer of the source, method of production, contents (intended and actual), audience, etc.). This is very important because it is the first step towards analysis (cf. labor-intensive aspect of much of historical research). Indicate precisely the place where the source originated and where it has been stored. Use computer programs for classifications and electronic indexing. Keep your system flexible so it can serve different purposes.

3) Critique and Confrontation: Are the sources authentic? Do they accurately portray events? Sometimes not many sources are available, so it is difficult to check. This is not very exciting because it mostly centers on details, but it is relevant nonetheless (often assumed). Differentiate fact from interpretation, translate accurately, corroborate with other evidence. If possible, “methodological marriages” with other methods are useful for corroboration to determine who said what to whom, why, how, and with what effect.

4) Analysis: The options are open: qualitative or quantitative, interpretation or explanation, structured or unstructured, within the context of theory and research strategy. The most traditional model is as follows: 1) theoretical proposition, 2) conceptualization of the theoretical constructs and formalization of a model; 3) operationalization of the variables of the theory (indicators), and 4) observation and conclusion. The inquiry can be deductive (theory-testing), or inductive (theory-construction). Also consider the various strategies of historical research relative to the objectives of historical sociology (see below).

c) Strategies (Skocpol)

See Skocpol, Theda & Margaret Somers 1980. “The uses of comparative history in macrosocial inquiry.” Comparative Studies in Society and History 22(2):174-197. 1) Parallel investigation of a theory: a theory is applied (or examined) in various historical contexts in order to demonstrate that various particular cases are but different modalities of a more general process (corresponding to natural-science conceptions of laws). 2) Interpretation of contrasting events: different, specific historical events are analyzed in their unique composition (typical for interpretive sociologies in the ‘Verstehen’ tradition). 3) Analysis of causalities at the macro level: based on Mill’s methodology (Skocpol) of the method of difference and of agreement. There is an important debate on the usefulness of Mill’s method for sociological research (see below) and the status of general theory in historical sociology.

d) Advantages and Disadvantages of Historical Methods

The unobtrusive nature of historical research is the main advantage of the method: the research cannot affect its subject matter. Also, several topics can be studied from this perspective, particularly forms of communication. The main weakness of historical research is that one can only reveal the past inasmuch as it is still present today: important documents, for instance, may be lost or destroyed (validity). And because of the often less rigid nature of this method of inquiry, the researcher can (invalidly) affect his/her picture of what has happened. Therefore, corroboration is helpful.

2. Theory and Historical Sociology

a) General Theory (Kiser and Hechter)

See Kiser, Edgar and Michael Hechter. 1991. “The Role of General Theory in Comparative-Historical Sociology.” American Journal of Sociology 97(1):1-30.

Historical sociologists emphasize the accuracy of the descriptive narratives they make about particular events. These events are unique and complex, so that the historical sociologists often taken methodological license, using vague and multiple strategies. In historical sociology, scope (generality) and analytical power have been minimized and replaced by descriptive accuracy. They are not interested in causal relations and the mechanisms that can account for producing these relations. The historical sociologists are inductive.

b) Theory in Historical Sociology (Quadagno and Knapp)

See Quadagno, Jill and Stan J. Knapp. 1992. “Have Historical Sociologists Forsaken Theory?” Sociological Methods & Research 20(4):481-507.

Historical sociologist have a deep concern for theory. Andrew Abbott, for instance, argues that historical sociology can uncover empirical regularities that can then be subjected to more conventional analysis. Also, comparative-historical sociology transcends the boundaries of time and space (temporal and national boundaries). According to Ragin, the use of macro-social units is what is distinct about comparative sociology. This parallels Kiser and Hechter’s emphasis on the mechanisms of causal chains, which lead them to focus on the actors involved. Interpretive sociologists are as theoretical as anybody. They also focus on alternative explanations, although they are not interested in general laws. Kiser and Hechter use a false dichotomy between induction and deduction. Theory can also be a starting point, a catalyst for further research. Theory can be tested, but theory can also lead to questions.


The debate on Mill’s methodology relates to the fact that sociological theories are probabilistic, not deterministic. Millean methododology also assumes that there is only one cause and that there are no interaction effects. How can the study of a case make sociological sense?

1. The Logic of Comparative Sociology (Jack Goldstone)

See Goldstone, Jack. 1999. “ASA Didactic Seminar on Comparative Sociology." Comparative and Historical Sociology 12(1) (with Bibliography).

a) What is comparative sociology?

Comparative sociology is a method, not a subject matter, applying various techniques to units. It involves the use of multiple, detailed observations on a modest number of cases, designed to uncover causal patterns. A case is a detailed understanding of a particular unit.

b) How Does It Work?

At least three issues are involved in comparative-sociological research: 1) Process tracing: the discovery of particular on-goings and their components over a period of time; 2) Combinations of causes and outcomes must be charted under varying conditions; and 3) Detective work: moving beyond correlation there should be a search for explanation, i.e. one must make sense of a link between variables. Also, attention is to be paid to an efficient picking of cases and to the formulation of a clear research question. Interestingly, a one-case study can be comparative inasmuch it implies an implicit comparison with the universe of cases that back up accepted wisdom. Universal generalization is not a necessary goal of comparative research.

c) Can Comparative Sociology Be Formalized Or Quantified?

Comparative sociology (as any other sociology) is always based on narrative, but a clear structure should also be revealed. For instance, particular combinations (x, y, f, z) of events can take place: A occurring with or without B, and A not occurring with or without B.Or various cases (1, 2, 3) can be compared in terms of certain characteristics (A, B,...). Such a structured presentation is not an explanation, but instead allows for multiple interpretations. Among the strategies of analysis are non-parametric analysis and Qualitative Comparative Analysis (designed by Charles Ragin) that uses Boolean algebra to implement principles of comparison.

d) The 3 C’s Of Comparative Sociology

The essentials of comparative sociology can be summarized as: Cases: to be designed across time, space, and units; Causes: the establishment of plausible connections and/or hypothesis testing; Comparisons: to test hypotheses, to illustrate causal connections, and/or to show variance in conditions and outcomes.

2. The Case for the Case

See Abbott, Andrew. 2001. Time Matters: On Theory and Method. University of Chicago Press.

a) Causal Devolution

In the 1950s, causality became equated with statistical research. Then causality was assigned to association. Path analysis was the crowning moment. Others, however, had attacked methods of association for lacking attention to causality. But the march of associational statistics could not be stopped (see history of statistics). Statistical methods imply theories which sociologists typically do not believe in. Instead, we should recognize that 1) description is a worthwhile objective, and 2) that causal methods are in fact descriptions (p. 123). Social events and processes are always located in concrete times and spaces.

b) What Do Cases Do?

A case can be an instance (a unit of analysis) or an exemplar of some property. In statistical pieces, narratives have variables as their subjects, not the cases. Cases do very little here, beyond a rational calculation. Cases are also undifferentiated. In single-case narratives, first the research subject has to be delineated (boundaries have to be set), such as an event, a social group, a state of affairs. The case can be transformed, e.g. an occupation can become a profession. So now the case becomes very central. After the events have been identified, they are ordered in some way that is explanatory. In multiple-cases narratives, we still have rich description as long as N is not too big (or can be summarized into categories of cases).

3. Example: Comparative History of Criminal Statistics (Deflem 1997)

See Deflem, Mathieu. 1997. “Surveillance and Criminal Statistics: Historical Foundations of Governmentality.” In Studies in Law, Politics and Society, Vol. 17, pp. 149-184.

4. The Merits of Comparative Historical Analysis 

See Mahoney, James and Dietrich Rueschemeyer, eds. 2003. Comparative Historical Analysis in the Social Sciences. New York: Cambridge University Press.

a) History v. Sociology Again

Abrams also argues that Moore shifts from a deterministic to a probabilistic model, adding new variables along the way that merely raise new ‘conditions favorable to...’. Events are a transformation device between past and future. According to Abrams, it also mediates structure and agency. If events would be completely unique, they would be beyond a generalizing analysis. Sociologist tend to move towards a general explanation of historical events by adopting an approach that is both logical and chronological. There is a ladder of conditions that are explanatory. There is a more fundamental general principle that underlies these stages that are held accountable for the happening of certain events.

b) Comparative Historical Analysis

CHA has been around for a long time, addresses many issues (though some such as law are understudied), and engages in theoretical and methodological controversies. CHA focuses very typically on the big questions. Big questions and comparative methods go hand in hand (see Durkheim on why). You need cases that are comparable in some sense, hopefully guided by theory. Practitioners of universalizing theoretical approaches see limits to this strategy, but not their own (ahistorical concepts and theories that are too general). CHA always consists of: 1) causal explanations; 2) processes over time; and 3) comparisons of similar and contrasting cases (such as nation states, their institutions, international constellations, etc.).

The methodological paradigm debates surrounding CHA are: 1) Quantitative versus qualitative: the small-N or large-N problem. CHA has been criticized for its small number of cases, bias in selection of the dependent variable, and bias in selection or use of secondary data. But CHA scholars have also addressed all these concerns. 2) Rational choice theory versus CHA: see Goldthorpe and Kiser & Hechter who see CHA as idiographic and anti-theoretical. 3) Interpretive/postmodern versus explanatory CHA: Recent scholars inspired by the cultural turn and postmodernism attack the explanatory approach of CHA, while the latter view that all scholarship contains explanatory elements.

c) Small-N / Large-N

Goldstone argues that large-N studies of revolutions have not been very fruitful, suggesting that the territory of revolutions is far from homogeneous. A revolution in China is not necessarily the same as a revolution in Cuba. So as Ragin says, different conditions can lead to similar outcomes in different contexts. So CHA never assumes a universal causal mechanism. Instead, CHA relies on a prior belief in a theory and then tests cases to see how they can and do shift that belief. Rueschemeyer on the value of a single case study. A single case can inspire theoretical ideas and falsify non-probabilistic propositions. But there are also positive contributions. Case studies can create, test, revise, and retest hypotheses to imply an ever-renewed questioning and testing. Also, case studies and comparative case studies have multiple goals. It makes the conditions of certain theories much more specific. It can also lead to the development of theoretical frameworks.


See the References section below for the sources used in this part.


a) Incommensurability, the Case Study, and Small N’s (George Steinmetz 2004)

Small-N studies are an important method of generating sociological understanding through “critical realism” and theoretically guided comparisons. Small-N comparisons are those research efforts that rely on a few cases (or even a single case) to develop sociological understanding, such as in ethnographic and historical research that seeks to unearth a great depth of understanding of a relatively limited target field.

The criticisms against small-N research come from two camps: 1) Methodological positivism assumes that “causal generative mechanisms are invariant across time and space”. Therefore, only very specifically controlled comparative analysis across large numbers of cases contain scientific value. 2) Nonpositivists cite three issues: Commensuration across observable social phenomena cannot be made because either: a) one cannot make assumptions about the generative mechanisms behind those observations (empiricism); or b) one cannot classify two instances as the same thing when the observed group understands themselves under a different self-description (nominalism). Also, there is an imperfect fit that comes about when making comparisons across languages and cultures. Many concepts cannot be translated perfectly. Lastly, the nonpositivist camp also challenges that some events possess a uniqueness that renders them incomparable to other events.

On the other hand, sociologists who recognize the importance of small-N comparisons defend their work against both camps: 1) Against positivists, small-N sociologists invoke critical realism to contend that the variations of social-causal structures across time and space are not universal but are instead “always and necessarily overdetermined by a plurality of conjuncturally interacting mechanisms.” Experiments are impossible to conduct in the social sciences, so that sociologists must study the generative mechanisms “in the wild.” 2) Critical realism contends that some of the issues raised by nonpositivists are conflating two different issues. If, in fact, two considered social phenomena are incomparable because they are genuinely constructed around different social mechanisms, then this does not invalidate all comparisons everywhere. Instead, it only speaks to the appropriateness of a comparison between those particular phenomena. Some fundamental similarities among the social world are always comparable and it is possible to displace one’s worldview and “see through another’s eyes,” even if in a limited fashion. In conclusion, “any social science oriented toward explanatory accounts will be necessarily involved in the study of specific cases.”

b) Realism, Rational Choice, and Relationality (Margaret R. Somers 1998)

Somers focuses on Kiser and Hechter’s criticism that “historical sociology is subverting the theoretical aims of social science”. Somers explains and places their position within a theoretical realism perspective and offers an alternative, relational realism, as a more conducive theoretical framework for historical sociological work. As Kuhn argued, scientific progress is always situated in particular historical periods. The cultural or historical circumstances that surround scientists at any given time must be taken into account when they propose knowledge-expanding theories. A theory-centric interpretation of Kuhn’s work (theoretical realism) influenced Kiser and Hechter’s view of the importance of rational choice theory, which they see as oriented at a theory-driven logic of social analysis and causal mechanisms. 

Theoretical realism assumes that theories survive to the extent that they are true pictures of what is real. Somers’s interpretation of Kuhn’s work challenges rational choice theory and “provides a renewed appreciation for problem-driven research and the centrality of history in the construction of knowledge.” Theory growth depends on facts turning into evidence only after and in response to a particular problem or research question. In doing so, historical elements of time and space are needed to help theory bloom. “Why a theory is confirmed at one time and not another”, she asserts, “requires a historical and causal explanation, rather than a strictly logical one.” 

Causal narrative and path dependency are methods in which theory can be linked successfully to real-world empirical cases in order to explain how and why phenomena occur within a particular time and space. With relational realism, “one can believe in the reality of a phenomenon without necessarily believing in the absolute truth or ultimate reality of any single theory that claims to explain it. Belief in a phenomenon or an outcome instead depends on evidence of its causal, practical, and relational significance in time and in space.” Historical sociology can complement and help to forge theory in a way that allows for causal mechanisms to be properly placed and explained in spatial-temporal contexts.


a) Initial Conditions, General Laws, Path Dependence (Jack Goldstone 1998)

Goldstone argues that both Somers and Kiser & Hechter sides are using faulty logic. The difference between these two camps lays in their emphasis on initial conditions (Somers) and general laws (Kiser and Hetcher). On the one hand, Kiser and Hetcher rightly argue that an over-reliance on initial conditions “will lead to purely narrative explanations of particularly sequences of events.” Yet, on the other hand, they also confuse issues of “general law, mechanisms, and desirable principles of explanation.” Goldstone states that good historical analysis choses “the method of explanation best suited for its explanatory goal.” A historical phenomenon that occurs often and in a variety of different settings and initial conditions would benefit from a general-theoretical approach. A historical phenomenon that occurs only occasionally and from specific initial conditions should be approached with a theory linking initial conditions to outcomes. Lastly, for the historical phenomenon that happens only once, despite similar initial conditions found elsewhere, should be approached as a path-dependent system.

b) Path Dependence in Historical Sociology (James Mahoney 2000)

Path dependence has been misconceived by historical sociologists as simply meaning that past events influence future events. Mahoney suggests a more specific definition of path dependence, which accounts for how “process, sequence, and temporality” affect social explanations of historical events.Path-dependent analyses minimally have three defining features: 1) the study of causal processes that are especially sensitive, in a sequence, to early historical events, which are more important than later events; 2) these events are contingent occurrences that cannot be explained by prior events or initial conditions; and 3) that once contingent events take place, the path dependent sequence becomes a deterministic pattern.

These three features are utilized in understanding two different types of path dependent sequences: self-reinforcing and reactive sequences:

1) Self-Reinforcing Sequences: Economists characterize self-reinforcing sequences appropriately by the feature of increasing returns. Important early events establish the sequence (feature one), which lead to contingent events which further the continuation of the sequence irrespective of initial conditions or expectations (feature two), and, finally, increasing returns lead to an accountable path that continues until an equilibrium is reached (feature three). Example: institution persistence.

2) Reactive Sequences: In a reactive sequence, each event in the sequence is both a reaction to antecedent events and a cause of subsequent events. Such sequences are characterized as “backlash processes” where important initial contingent events (features one and two) trigger reactionary events, which in turn trigger counter-reactionary events in a continuing sequence (feature three). With reference to chaos theory, Mahoney suggests that the final outcome of these sequences is extremely unpredictable. The purpose of path dependence analysis is to examine sequences of events that do not fit predictions of general social theory. As long as this divergence from theory can be established, the use of path analysis is completely appropriate. In examining such sequences of events, path analysis can provide much needed explication and inform future general theory, but otherwise it is just a fad.

c) Revisiting General Theory in Historical Sociology (James Mahoney, 2004)

Mahoney argues for the merits of general theory in empirical research. He defines general theories as “postulates about foundational causes,” involving a particular “causal agent” (unit of analysis) and “causal mechanisms” (i.e., properties of casual agents that produce outcomes and associations). Rather than referring to universal application, general theory is general in the sense of “its use of an abstract causal mechanism that exists outside space and time.” Among the benefits: 1) general theory forces analysts to be explicit about causal agent and mechanism; 2) it allows a broad scope of research agendas with a single causal mechanism; 3) it can generate new hypotheses; 4) it can explain inconsistencies of a causal mechanism by that causal mechanism (e.g., rational choice explains collectively irrational outcomes). To benefit empirical research, propositions are derived from postulates: “propositions are testable hypotheses and predictions about the occurrence of outcomes,” whereas “postulates are assertions about conditions in the world and the logical relationships that govern these conditions.” Postulates are assumptions used to generate logically deductible propositions. Historical outcomes are explained on the basis of general theory.

Five Examples of General Theory In functionalist theory, the causal agent is the social system, and the causal mechanisms are needs/requisites (equilibrium). It is a macro theory. Rational choice theory is a micro theory with the individual as the causal agent, and instrumental rationality as the causal mechanism. Power theory analyzes at the meso-level of investigation. The collective actor is the causal agent, and resources are the causal mechanism. The collective actor is seen as social groups and organizations rather than individuals. Neo-Darwinian theory is ‘radically micro’ in that the gene is the causal agent, and contribution to fitness is the causal mechanism. Cultural theory is a meso-macro level of analysis. The causal agent is collectivity and the causal mechanism is semiotic practices, which are “systems of meaning and associated practices.”


a) Narrative, Event-Structure Analysis (Larry Griffin 1993)

Griffin argues that event-structure analysis (ESA) should be used to examine causal relationships with narratives. Griffin claims that an event can be understood “both as a historically singular event and as an instance of a class of historically repeated events” and that this analysis can move from the particular to the general. Griffin points out that narratives are sequences of social action constructed by the narrator in a particular temporal order for a particular purpose that include a particular series of actions. Narratives allow for “deep theoretical knowledge about the mutually constitutive interplay of agency and social structure”, because the events are examined within their historical context. He does warn, however, that because narratives often blend both descriptions and interpretations, it is often easy to obscure causality of the events described within them. Griffin argues that one should consider counterfacts or “what if” questions. This type of analysis forces the researcher to ponder what did not happen thereby contextualizing what did, but this should be done within a historical and factual framework of objective possibilities. Causal interpretations can be arrived at by asking a series of questions based on facts and counterfacts about historical sequences.

Event-structure analysis (ESA) was developed by David Heise to analysis “cultural routines and the subjective representation reality.” The computer program ETHNO is the software application used to uncover causal and interpretative connections within the narrative. ESA is influenced by cognitive anthropology as well as rational choice theory and because of these features Griffin recommends that this analysis be used on narratives. ESA places emphasis, according to Griffin, on “temporal ordering and sequencing of action, not historical scope” (p. 1125). These he argues means neither micro or macro events limit the usefulness of ESA. ESA does not model social structure, but rather social actions are placed in logical sequence within the social structure. Griffin advocates the use of ESA in historical sociology because it uses temporal sequencing to explain social processes.

b) How Is Sociology Informed by History? (Larry J. Griffin 1995)

Sociologists disagree on the value of history. For some, history is only the “storehouse of samples” which serve as the base for the development of social theory. For others, history is important by itself. In any case, Griffin asserts that the use of history is increasingly visible in the discipline, as indicated by the rise of historical sociology articles in prestigious general journals; an increase in number and visibility of journals that specifically integrate history and social science (e.g., Comparative Studies in Society and History, the Journal of Historical Sociology, and Social Science History); the methodological incorporation of narrative, event, and biography as analytical tools; and the increase of books that integrate history and sociology. History is now more than ever used in many ways in sociology in order to enhance sociological explanation.

The author makes a case for a “historically infused sociology”, because when history is taken more seriously, time is also taken more seriously. History provides the social context of the sociological research matter: sociologists should thoroughly historicize their basic categories of analysis such as class, gender, and race, and they should fully exploit the explanatory and interpretive potential of narratives. Sociologists tend to see the analytical categories (e.g., gender, class, race): 1) in isolation from other concepts and categories, 2) fixed in meaning, and 3) positionally defined from social structure. This permits quantification and replication, but also it reduces sociologists’ “ability to discern the temporal fluidity and mutability of analytical categories”. In contrast, historians tend to see the analytical categories as historical products and processes which are elastic, ‘shifting’ and ‘unstable’ in content and definition because their meanings adapts through time. While for sociologists the analytical categories have fixed meaning and conceptual independence, for historians, the analytical categories have no fixed meaning and are conceptually tightly intertwined. Historians’ conceptualization is inconsistent with sociologists’ methodological conventions and analytical goals.

Historically infused sociology adopts narrative: First, the data “should consist of, and closely track, social actions through time”. Second, these data should be basically sequential in their “very definition and construction because social action is itself temporally sequential.” Sociologists should think about variables as actions, motives for action, and as consequences of actions. Third, sociologists should take advantage of information which conveys and carries action, such as stories, biographies, and other narratives. For Griffin, most sociological explanations do not incorporate time into the logic of explanation. Incorporating time into sociological explanations implies the use of an explanatory mode that is intrinsically temporal. Thus, narratives can be used to interpret and explain sociologically in order to distill the “logical from the chronological” and discerns the “theoretically general in the historically particular.” In sum, to take history seriously is to take time seriously and to use time as context and as narrative in sociological explanations.


Social Theory, Modernity, and the Three Waves of Historical Sociology, by Julia Adams, Elisabeth Clemens, and Ann Orloff. Russell Sage Working Paper, 2003.

From Historical Sociology to Theoretical History, by Gareth Stedman Jones. British Journal of Sociology, Vol. 27, No. 3 (1976), pp. 295-305.

The Uses of Theory, Concepts and Comparison in Historical Sociology, by Victoria E. Bonnell. Comparative Studies in Society and History Vol. 22, No. 2 (1980), pp. 156-173.

Historical Sociology in the History of American Sociology, by Mildred A. Schwartz. Social Science History, Vol. 11, No. 1 (1987), pp. 1-16.

Social History and Historical Sociology: Contrasts and Complementarities, by Theda Skocpol. Social Science History, Vol. 11, No. 1 (1987), pp. 17-30.

The Role of General Theory in Comparative-Historical Sociology, by Edgar Kiser; Michael Hechter. American Journal of Sociology Vol. 97, No. 1 (1991), pp. 1-30.

Odious Comparisons: Incommensurability, the Case Study, and "Small N's" in Sociology, by George Steinmetz. Sociological Theory Vol. 22, No. 3 (2004), pp. 371-400.

"We're No Angels": Realism, Rational Choice, and Relationality in Social Science, by Margaret R. Somers. American Journal of Sociology Vol. 104, No. 3 (1998), pp. 722-784.

Initial Conditions, General Laws, Path Dependence, and Explanation in Historical Sociology, by Jack A. Goldstone. American Journal of Sociology Vol. 104, No. 3 (1998), pp. 829-845.

Path Dependence in Historical Sociology, by James Mahoney. Theory and Society Vol. 29, No. 4 (2000), pp. 507-548.

Revisiting General Theory in Historical Sociology, by James Mahoney. Social Forces 83(2):459-489, 2004.

Narrative, Event-Structure Analysis, and Causal Interpretation in Historical Sociology, by Larry J. Griffin. American Journal of Sociology Vol. 98, No. 5 (1993), pp. 1094-1133.

How Is Sociology Informed by History?, by Larry J. Griffin. Social Forces Vol. 73, No. 4 (1995), pp. 1245-1254.

Historical Methodology: Comparative and Macro, by Jennifer Platt. Timelines, Newsletter of the ASA History of Sociology section, April 2007.

See related writings on comparative-historical sociology.