1. Congratulations on the publication of your book, Analytical Sociology: Actions and Networks, which demonstrates the power of the theoretical framework of analytical sociology in explaining a large array of social phenomena. How did the writing process begin?

A group of European scholars interested in analytical sociology and related research agenda started to meet annually in 2009 to discuss how mechanism-based explanations can be built and empirically tested. This group received the name European Network of Analytical Sociologists. In 2011, I had the great privilege to organize the annual meeting of this group in Paris at the University of Paris-Sorbonne. This was the first meeting attended by a fair proportion of scholars from outside Europe, which led us to change the name of the group to International Network of Analytical Sociologists. In addition to meta-theoretical papers, several empirical-oriented papers inspired by the analytical sociology principles were also presented at the Paris conference. I thought that these two elements, i.e. larger participation across the world and more concrete analyses, required being valued and developed through a book. This was my original motivation in making a proposal to Wiley: I wanted to show that analytical sociology was entering a new phase of development.

2. Who should read the book and why?

Social scientists interested in sociological theory and general sociology will find in the book a detailed discussion of what analytical sociology is as well as an illustration of the kind of empirical research this approach is inspiring. Quantitatively-oriented social scientists will find in the book original methodological procedures to track the mechanisms of social life through statistics, mathematical models, simulation, or lab and in field experiments. Philosophers of science interested in causality, explanation, and mechanisms will find in the book materials that illustrate how the concept of mechanism is used within a specific approach to sociological theory, i.e. analytical sociology. Thus they will be in a position to compare this approach to other forms of mechanism-based theorizing in social sciences, like critical realism, or in other disciplines like biology. Although highly specialized, the book is also meant to be a source of inspiration, both substantive and methodological, for doctoral students in sociology.

3. One of the principles of the book is that leading scholars apply the theoretical principles of analytical sociology to understand how puzzling social and historical phenomena including crime, lynching, witch-hunts, tax behaviours, Web-based social movement and communication, restaurant reputation, job search and careers, social network homophily and instability, cooperation and trust are brought about by complex, multi-layered social mechanisms. Could you please tell us more about how this can be applied?

All empirical-oriented contributions share the same starting point: they clearly formulate a why-question and, in order to answer this question, they attempt to reply a how-question. To reply a how-question means to make it explicit what entities, properties, activities, and connections brought about the observed events, patterns, or trends. This can be done in different ways. Some contributors collect empirical evidence directly to demonstrate that this or that part of the postulated mechanism was really operating in the real-world. Other authors look for statistical signatures at the aggregate level of the postulated mechanisms.

Still other authors design a formal model of the postulated mechanism and used analytical or computational tools to determine the consequences one should expect if this mechanism was at work. Several contributions combine in creative ways these different strategies. The book makes a strong case for simulation methods, in particular agent-based computation models, as a pillar of mechanism-based theorizing. Since social mechanisms are usually not observable in their entirety, several contributors argue that detailed models are needed to make the analysis possible and, in the absence of fine-grained empirical data, simulation is a powerful tool to study these models.

4. The analyses presented in this book rely on a wide range of methods which include qualitative observations, advanced statistical techniques, complex network tools, refined simulation methods and creative experimental protocols. Are any of these a result of your teachings at GEMASS, Centre National de la Recherche Scientifique (CNRS) and University of Paris–Sorbonne, which have thus informed your writings?

GEMASS is a small research unit at the CNRS founded in the 1970s by the French sociologist Raymond Boudon. At that time, Boudon was one of the rare scholars in sociology to argue in favour of the use of numerical simulations deductively to study theoretical models of social mechanisms. Thus, in this sense, the stress that I put in the book on a specific modern form of computer simulation, namely agent-based modelling, is without a doubt related to some of the intellectual orientations of the research unit to which I belong, i.e. GEMASS. I have also regularly taught classes on this method over the last seven years in the Department of Sociology at the University of Paris-Sorbonne (as well as at other universities in the US, Germany, and Italy). Thus, teaching and students’ (sometimes violent) reactions to the idea that one can write computer codes to represent social mechanisms have certainly shaped my way of thinking about analytical sociology and its potential contribution to our understanding of the social world.

5. What is it about the area of analytical sociology that fascinates you?

Two features of analytical sociology are especially interesting for me. On the one hand, analytical sociology is a theoretically-guided empirical-oriented approach. Although this may seem trivial in other disciplines, much sociology is still concerned either with pure, speculative theory construction, or with pure, although often refined, empirical descriptions. By contrast, analytical sociology has a strong explanatory ambition which makes theory at the service of the explanation of clearly defined empirical puzzles. On the other hand, since analytical sociology has the ambition to realize this explanatory requirement by describing the underlying mechanisms of social life, it raises complex methodological issues on what methods can be mobilized to track activities and events that are intrinsically difficult to observe empirically. Thus, analytical sociology is challenging on a methodological plan, which I find especially stimulating.

The possibility to combine substantive-oriented simulations with more classical statistical techniques is, to my modest opinion, one of the frontiers of quantitative methodology in social sciences for the coming years.

6. Why is this book of particular interest now?

As testified for instance by the symposia that several journals devoted to analytical sociology, the international visibility of this perspective increased quickly over the last year. Many scholars are still concerned however with the real distinctiveness of this approach. A book clarifying the theoretical principles of analytical sociology and proposing several empirical illustrations of these principles thus seems of particular interest now.

7. What were your main objectives during the writing process? What did you set out to achieve in reaching your readers?

I would say that I had two main goals in writing this book. On the one hand, I wanted to explain in what sense analytical sociology can be considered as a distinctive theoretically-guided approach to empirical research in today’s sociology. The introductory chapter attempts clearly to expose the main theoretical and methodological principles of this approach and, at the same time, admits its internal heterogeneity. My idea with this chapter was to avoid/solve current misunderstandings surrounding analytical sociology.

On the other hand, the book clearly wants to make a case for advancing analytical sociology in a more empirical-oriented direction. Indeed, one of the objections that this approach received was that it is too much concerned with meta-theorizing about what a “good” explanation is and much less with proving empirically what one can gain by providing “good” explanations.

8. Were there areas of the book that you found more challenging to write, and if so, why?

The introductory chapter was without a doubt the most difficult part to write. Indeed, in this chapter, I wanted to simultaneously reconstruct, in a rigorous way, all debates about analytical sociology and propose a specific understanding of this perspective. To create coherence between my own introductory chapter and the rest of the book was also challenging because some of the book’s contributors have a different view on what makes analytical sociology specific. To solve this issue, I conceived the theme/variation solution according to which each contribution is presented as a “variation” on a common “theme”, the common theme being the principles discussed in my own chapter. Finally, the writing of sixteen short introductions (one for each chapter) to create coherence and connections among contributors was without a doubt a demanding and challenging exercise.

9. What will be your next book-length undertaking?

This is a difficult question. I have been recently approached by Polity for a book about the role of social networks in the creation of social inequalities. This is a tempting proposal, which is very in line with my current research interests. However, I must confess, that, at the moment, I am still not sure whether I really want to undertake the writing of a new book. I have several papers in progress and it is probably more reasonable to focus on completing these projects before starting a new book project.

10. Please could you tell us more about your educational background and what was it that brought you to recognise statistics as a discipline in the first place?

I was trained in sociology at the undergraduate level at the University of Trento (in 1998-2001). I discovered statistics there as the main tool that sociologists can use to describe complex patterns of association at the aggregate level. When I moved to Paris-Sorbonne for my PhD, I started to endorse a more critical view on statistics and progressively realised that, to reach truly explanatory goals, statistics is not enough. This is the reason why I started to explore other formal tools and, in the end, started to get deeply interested in computer simulations and in agent-based computational modelling in particular. The flexibility of this method to design theoretically-informed models of social mechanisms is a feature that sociologists still fail to appreciate. The possibility to combine substantive-oriented simulations with more classical statistical techniques is, in my modest opinion, one of the frontiers of quantitative methodology in social sciences for the coming years.