Introductory remarks

Ash Watson had already mentioned Andrew Abbott’s essay “Against Narratives: A Preface to Lyrical Sociology” (2007) in her interview with Marc Ortmann.¹ It appears to have become a key reference text for those who incorporate fictional writing into their research practice, as well as for those in the field of qualitative social research who are seeking new, creative approaches.

Abbott’s plea for lyrical writing begins with a critique of narrative sociology and the conception of temporality expressed within it. Narratives that attempt to explain the emergence of a social situation, he argues, cannot avoid turning back to the past and recounting the branching sequence of events in such a way that they culminate in the situation under consideration. This, however, has the consequence that social situations and actions ultimately appear as determined; decisions are no longer recognized in terms of their potential to have been otherwise. Everything becomes a necessary part of a “story”.

Abbott sees in lyricism – which turns to the moment and makes its emotional significance a central concern – a way of escaping the pitfalls of narrative. Alongside a particular stance toward both the object and the reader – namely engagement and emotionality – which is characteristic of lyrical texts and corresponds to the experience of the social world, Abbott also describes in his essay the literary devices through which this effect can be achieved: evocative imagery intended to elicit concrete feelings; vividness created by illuminating the same situation from multiple perspectives; and the use of personification and figurative language.

SJM: Dear Professor Abbott, I am very pleased that you agreed to answer a few questions. As you know, the interview will primarily focus on your article on lyrical sociology, a central paper in the context of sociological fiction.

AA: It should be understood that by writing the original article on “lyrical sociology” I did not mean to propose lyricism as a specific method. I meant to suggest that lyricism should be a stance that is possible and respected within the discipline of sociology. I meant this position very specifically as a counteractant to an overemphasis on narrative, whether of individual cases or of variables, an emphasis to which my own development of sequence analysis had of course contributed.

SJM: I used to read your text on “lyrical sociology” with students in seminars, and their immediate reaction was: “That sounds really exciting. But if I would write like that, my supervisors would throw it back to me!” What would you answer if someone tells you that it is only particularly privileged sociologists who are allowed to write lyrically?

AA: I don’t know whether all supervisors would throw such work back or not. It is my general impression that sociology used to be much more lyrical than at present. The Chicago School of the 1920s tended to write with a strongly lyrical stance. For good or ill, the scientism of the midcentury suppressed this lyrical stance, outside of ethnography. The essentially narrative property of “causality” became central in the discipline’s self-image. As politicization took off after 1968, the focus changed again, but remained narrative; but now the narrative was a moral one as well as a causal one. So yes, it is probably the case that work was more lyrical in the past. As for the matter of privilege and rank, I don’t think that at present the suppression of the lyrical approach is somehow dependent on rank. I had a lyrical piece on theory rejected very recently, by a journal that had asked for a paper from me. And that experience seems to counteract the claim that privilege grants the right to be lyrical, since I imagine that people who worry about such things think of me as a “particularly privileged sociologist”. The obvious way to deal with this problem would be to create a journal for lyrical sociology. I could have done this myself had I not had another journal to worry about. But this avenue is still open for anybody to try. The question is whether enough younger people genuinely have the urge to do truly lyrical work and, even more important, to read it. And we must face the fact that it is not easy to write great lyrical work, any more than it is easy to write great lyric poetry. There is no reason to assume that people who find themselves in sociology as a field of work are any better at writing effective lyrical texts than anyone else. It is one thing to have emotions; it is quite another to write about them in a way that grasps readers effectively.

SJM: Did you teach students how to do “lyrical sociology”? And if so, how did you do it? How, for example, do you get someone to construct impressive, concise linguistic images?

AA: No, I’ve never taught a course about lyrical sociology. But looking at the articles that have cited the article suggests that I should teach such a course, organizing that course not so much around my piece as on how other people have reacted to it. The reactions seem interesting. It is not always the case that reactions to my work seem interesting. I once wrote an article about articles citing my book on professions, and made the depressing discovery that about half of those articles did not understand the main argument of the book. But a glance at the web reveals something quite different about the lyrical sociology piece. One finds a quite interesting collection of articles on lyrical sociology, some of which seem to be moving in directions I didn’t myself imagine. Oddly, my current theoretical work has forced me far from lyricism. I have had to remove all metaphors from my theoretical system. They turn out to be limiting, because theoretical work must be very, very clear, and metaphors always carry dangerous baggage. Writing deductive theory is very far from lyricism. But what one needs in a general sociological methodology is something that embraces both ends of this continuum at a single time. I may get to that eventually, but at present I’m located in the rigorous world of deductivism. To be sure, my notebooks are full of lyrical things.

SJM: Your text is regarded by many art-based approaches as a kind of founding text or theoretical legitimization for creative methodological approaches. What do you think about that?

AA: I’m completely surprised – mainly because I don’t know what it is that these art-based approaches are approaching. This is just another example of the fact that one has little idea what becomes of one’s work, as I discovered while researching the article about papers citing my professions book. A book or article is like a note in a bottle cast into the sea. You have little idea who will read it and how. In tight scholarly communities, this is less true: one can assume that most readers will get the major points of an intervention. But today’s scholarly communities are far too large for that kind of loose consensus. Perhaps this was one of the reasons lyricism flourished in the 1920s, when academic sociology in the United States contained considerably less than a thousand people.

SJM: Could you put in a nutshell why it is that affect is so important for doing sociology?

AA: It’s a long topic, not a nutshell matter. Affect is important because human social life moves forward through the participation of human beings, and affect is an essential portion of the way action unfolds in the human organism. We can think about this in two different ways, which are reflected […] above. The first way relates to thinking about lyrical sociology as a stance. There, affect is important both because of the affect of the observer/sociologist when writing in the lyrical mode, and also because of the general rule just noted, that affect is part of how action unfolds in the human organism. The second way relates to general social ontology. In considering social ontology, affect matters because it is potentially a part of all social entities, not only individuals, but also networks, groups, ideologies, symbols, etc. Such an assertion draws less on the emotion side of lyricism and more on its temporal quality as purely local in time, as momentaneous. These two senses of affect must both be recognized in a general social theory. Put another way, the notion of emotion as we commonly understand it describes the momentary reactions of a human organism to a situation, which inevitably involve the bodily states of affairs that we commonly call emotions (and which the James/Lange theory makes the defining quality of emotion). But there are momentary reactions of other kinds of social entities to social affairs, and these can be considered “emotions” as well. But they don’t have a simple relation to the bodily states of affairs of individuals.

SJM: In your approach to lyrical sociology, you propose the use of poetic principles of representation to gain insight. This seems to me to be very close to the approach of Siegfried Kracauer or Walter Benjamin, who spoke of figurative sociology. Neither author was considered a sociologist (but philosopher, essayist, or journalist). Do you believe that lyrical or figurative approaches are only good for a life in the shadows?

AA: As it happens, Kracauer and Benjamin are among the many important intellectuals that I don’t happen to have read. So I can’t say anything about the relation of my work to theirs.

As for whether lyrical or figurative approaches are only good for a life in the shadows, I am not quite sure what is meant by “life in the shadows”. I will assume that what is being asked here is whether lyrical/figurative approaches can become the mainstream (or part of the mainstream) of the social sciences. I don’t know about this possibility. The social sciences are at present in a kind of crisis. There are many reasons for this, which I considered at length in my Marc Bloch lecture of 2015² and my discussion of the future of knowledge at the WZB in Berlin in 2017.³ But the most important one is really the spectacular success of social science from 1890 to about 1990, in which four scholarly generations created the astonishing richness that fills our libraries. To respond to that amazing success, we have to create a whole new approach to social science. In my view, that will require a new theory of social ontology and a corresponding new theory of social morality. Luckily, the chaotic history of the last fifty years has raised a whole new generation of problems for consideration. So not only are the general topics known, so also is the fundamental body of examples. Once we free ourselves from the paralysis induced by neoliberal management of intellectual life, and from the dated politics we have inherited from modernity, the way ahead is clear. To be sure, it is not clear who will pay for the new social science, whether it can be made a “job” with sufficient remuneration, and what audience will read it. But the future general agenda of social science seems quite plain, and if lyricism takes no place in that agenda, it will have been our fault. Luckily, the immense threat of AI makes the crisis even clearer than before. Dealing with the threat of AI will require minds that are broad, eclectic, but extremely rigorous – people who can write algorithms, do ethnography, and write poetic texts, and do so “all at once,” not at different times of day, as Marx thought. They must be people who can put all those skills in the service of a new vision of social life. Some of us have started into that world. The question is who will follow.