Cultural Sociology After the Credibility Revolution
How should we do cultural sociology after the credibility revolution?
A few years ago, this would not have been an alarming question. Work in cultural sociology was bifurcated between what we can call “cultural” and “cognitive” traditions (Kaidesoja, Hyyryläinen, and Puustinen 2022), one focusing on the classic questions around meaning and cultural practices, while the other focusing on cognitive processes underlying human thinking and behavior.
However, despite clear differences in theory and question formulation, both of these traditions shared a relatively common attitude toward quantitative empirical work. Objects that we deem culturally relevant were rarely framed in terms of treatments, counterfactuals, and causal estimands. Instead, culture was mainly invoked as an organizing force in social life: something that structures perception, facilitates certain practices over others, or stabilizes social action across time. In practice, this involved a tradition of empirical practice that prioritizes model-based thinking: strategies aiming to map cultural structures via clusters, dimensions, latent spaces or formal models.1
1 Some of these models (Goldberg and Stein 2018) were, I’d say, much more generative than certain other projects (e.g., Mohr and Duquenne 1997), due to our dispositional tendency to get excited by pretty images.
This strategy makes sense, because it is typically dubious (or at least odd) to think that “culture” is something that operates as a manipulable treatment. We instead define culture via its role in a larger system—either as a force that, say, motivates people (Vaisey 2009) or something that changes in response to some social changes (Fishman and Lizardo 2013). It was rarely the case that these causal relations were really articulated as contrasts between well-defined counterfactual states.
The situation changed, however, with what is now commonly referred to as the credibility revolution, though the implications of this transformation are yet to be reckoned with in cultural sociology.
Across sociology and the social sciences more broadly, the empirical emphasis shifted toward counterfactual reasoning, in one form or another, at the expense of model-based thinking. While this shift in emphasis still makes sense only in the context of large structural models (such that a given causal estimate is scientifically relevant only when understood within the context of a larger scientific model2), it is nonetheless pervasive, and in many respects, achieved a clarifying function when set against the existing scholarly practices. At the same time, however, the implications of these changes are obviously unlikely to be similar. I am thinking particularly of fields whose core objects of study do not naturally align with policy-based or treatment-based reasoning in, say, economics.
2 A point rarely made explicit.
Given the kinds of questions cultural sociologists tend to ask or the kinds of objects they tend to study, I think it is fair to think that this turn may obstruct—or have already obstructed—theoretically well-motivated empirical work, by narrowing what counts as an “acceptable” evidence.
That may very well be true, but I believe that once this new standard is taken seriously, a more serious question emerges for the study of culture. This question asks whether cultural processes can be meaningfully expressed in the language of potential outcomes. This can be a contested move, because many cultural theories involve forms of causation that are diffuse, cumulative, and (perhaps because of this very reason) open to being relatively vague. What does it mean to postulate a world where an actor did not acquire a particular schema? What would it mean to imagine the absence of “habitus,” “fields,” and “capital”? These questions immediately point to a strong tension between how culture is typically theorized and how causal claims are currently expected to be formalized.
 |
|---|
| The “Effects” of Habitus |
Let me return to the initial question I posed: how should we do cultural sociology after the credibility revolution? I think we need important changes in our paradigm to greet this moment.
One suspicion I have heard repeatedly from people is that the tension is practically unresolvable because, well, we cannot manipulate culture. Seen this way, the pressure coming from the credibility revolution will either invalidate classic questions in cultural sociology because it is not really possible to answer them, or completely reshape what cultural sociology can do. But I think the issue here is that the potential-outcomes framework does not really demand that cultural objects be manipulable in any literal sense. Instead, it demands that our questions involve intelligible comparisons between well-defined units. Put differently, it clarifies what, exactly, a claim about cultural change commits us to observe in the real world, and which contrasts must hold for that claim to be meaningful.
A useful illustration of this clarification role recently came from Kratz and Brüderl (2025), who re-defined the effects of aging as what would be observed if the same individuals aged under otherwise comparable conditions. What makes this paper particularly examplary is that it shows, in quite a strong fashion I believe, how counterfactual thinking might have a clarifying function for many classic questions in cultural sociology. That being said, even the authors themselves acknowledge that there is a bit of an “uneasiness” that stems from this strategy, acknowledging that
viewing research on the age–happiness relationship through a causal inference lens is subject to debate [… and that] age is not “manipulable” and inferring whether aging affects happiness relies on strong assumptions […] which are likely violated [11].
Nonetheless, as Kratz and Brüderl (2025) says, this is precisely the point. Only when we put “the effect of aging” as a problem like this may we appreciate that our seemingly straightforward research questions actually hide many conceptual problems. The payoff is thus a clarification of what the claim would mean, which comparisons would justify it, and which assumptions are doing the real explanatory work. In that sense, cultural sociology can benefit from the credibility revolution less as a tool for statistical estimation than as a device for conceptual clarification.