 Carly R. Knight |  Isaac A. Reed |  Interviewed by Luis Flores, Perspectives Co-Editor |
In their article “Meaning and Modularity: The Multivalence of ‘Mechanism’ in Sociological Explanation” (2019, Sociological Theory), Knight and Reed disentangle and explore contradictions between “modular” and “meaningful” mechanistic models. The disjuncture, they argue, is grounded in “incompatible causal foundations and entails mechanistic models with distinct and conflicting evidentiary standards.” Below is a conversation with the authors on their provocative article.
Luis Flores (LF)
Invoking Clifford Geertz, your paper opens by suggesting that the concept of “mechanism” in sociology has surpassed a “first flush of celebrity” and is now ripe for an “operation of disentanglement” (234) Could you say a bit about how the idea for this paper emerged, and perhaps, what signs indicated for you that the “mechanism” idea in sociology had arrived at this stage?
Carly R. Knight (CRK)
We first had the idea for the paper a long time ago. My own background with mechanisms came from a causal inference perspective. But I think it was the publication of several theoretical pieces that were trying to pin down a sociological definition of mechanisms that got us thinking about whether these approaches were all compatible. Particularly, I remember reading the Handbook on Analytical Sociology, Neil Gross’s 2009 ASR piece, and Matthew Norton’s Sociological Theory piece all for a class on explanation and feeling that there was a lot of heterogeneity that could be unpacked there. At the same time, it seemed that some researchers were using terms like “mechanism” and “process” interchangeably. When Isaac and I were talking about this one day, we got to wondering if there was anything more we could say about the difference between “mechanism” and “process” and what distinguished these various approaches.
Isaac A. Reed (IAR)
Yes, the paper developed over a long period of time, and I think that was a positive. Whatever clarity it has, was achieved, at least in part, via an extended period of refinement of the original idea. It was really an intense process of writing and rewriting, exchanging drafts over email, and so on—though often, also, of leaving the paper and coming back to it with fresh eyes until we really got it right. For my part, I felt after finishing Interpretation in Social Knowledge in 2010, that the discussion of mechanisms could have gone much further therein. I knew there were missing links between (1) discussions in the philosophy of science around mechanisms and discussions in sociology on the use of concepts in empirical analysis, and (2) the counterfactuals and causal inference perspective in sociology and cultural sociology. I also knew there were flaws in what was going on with mechanisms in sociology, but I wanted to get at them in a way that was truly pluralistic, and which tried to find missing links that were productive for future work in theory and research.
LF
You note the diffusion of the concept but stop short of sketching a genealogy of the drive for mechanistic explanation. We were curious to hear about the historical development of modular and meaningful mechanistic approaches—is there anything to our intuition that the modular form emerges first and meaningful approaches sought, in part, to adopt mechanistic explanation for different purposes? Do you have a sense of the disciplinary or intellectual problems/currents that drove the expansion of the mechanistic idea within sociology—perhaps, as you note, as a placebo for disciplinary ailments?
IAR
This is an interesting question. Certainly, doing such a genealogy would be interesting and worthwhile. I think it would be a bit more fragmented than you suggest, because one would need to include not only the trajectory of thinking about causal inference, but also a more diffuse but powerful intellectual formulation in sociology—including sociology that is quite “humanistic” or “cultural”—which is the sense that some aspects of the social world are quite regularized, but in a way that is not simple or easily grasped by just “looking on the surface” of social life (though I would note that surface/depth is also a metaphor).
What I mean by this is that there has always been a question: is the industrial (and now, algorithmic) social world mechanistic, in the same way as its factories (or AIs) are? On the other hand, the metaphor of mechanism has also been repeatedly used to describe supposedly pre-modern practices, such as ritual or gift exchange. The point is that I think that the metaphor of mechanism is perhaps quite old and widespread in work that calls itself sociology, though perhaps not always by that name, or not always in a codified way. I suspect that it arises when, for varying reasons, we find ourselves reaching for a study of regularity that is also a study of depth.
This is, perhaps, why the genealogical project would be a bit different than what we are doing in the paper, where we take up explicitly the project of “what we talk about when we talk about mechanisms,” now, in this (loosely defined) community of inquiry. Nonetheless, you are right that there is a link: the mechanisms concept, as it emerged into our current language via modular and non-modular accounts, appeared to merge the striving for causal explanation and the more interpretive/historical sensibility of a “mechanistic world” in one fell swoop. Our point is that there is a way to carefully articulate both ambitions, but it is not the way we have chosen so far. And so: there is more work to be done and we hope other folks will take up the larger project implicit in our paper. That is to place under the microscope of intense thought the specimens of sociological thought in our own time, which is to say, our concepts.
Invoking Clifford Geertz, your paper opens by suggesting that the concept of “mechanism” in sociology has surpassed a “first flush of celebrity” and is now ripe for an “operation of disentanglement” (234) Could you say a bit about how the idea for this paper emerged, and perhaps, what signs indicated for you that the “mechanism” idea in sociology had arrived at this stage?
Carly R. Knight (CRK)
We first had the idea for the paper a long time ago. My own background with mechanisms came from a causal inference perspective. But I think it was the publication of several theoretical pieces that were trying to pin down a sociological definition of mechanisms that got us thinking about whether these approaches were all compatible. Particularly, I remember reading the Handbook on Analytical Sociology, Neil Gross’s 2009 ASR piece, and Matthew Norton’s Sociological Theory piece all for a class on explanation and feeling that there was a lot of heterogeneity that could be unpacked there. At the same time, it seemed that some researchers were using terms like “mechanism” and “process” interchangeably. When Isaac and I were talking about this one day, we got to wondering if there was anything more we could say about the difference between “mechanism” and “process” and what distinguished these various approaches.
Isaac A. Reed (IAR)
Yes, the paper developed over a long period of time, and I think that was a positive. Whatever clarity it has, was achieved, at least in part, via an extended period of refinement of the original idea. It was really an intense process of writing and rewriting, exchanging drafts over email, and so on—though often, also, of leaving the paper and coming back to it with fresh eyes until we really got it right. For my part, I felt after finishing Interpretation in Social Knowledge in 2010, that the discussion of mechanisms could have gone much further therein. I knew there were missing links between (1) discussions in the philosophy of science around mechanisms and discussions in sociology on the use of concepts in empirical analysis, and (2) the counterfactuals and causal inference perspective in sociology and cultural sociology. I also knew there were flaws in what was going on with mechanisms in sociology, but I wanted to get at them in a way that was truly pluralistic, and which tried to find missing links that were productive for future work in theory and research.
LF
You note the diffusion of the concept but stop short of sketching a genealogy of the drive for mechanistic explanation. We were curious to hear about the historical development of modular and meaningful mechanistic approaches—is there anything to our intuition that the modular form emerges first and meaningful approaches sought, in part, to adopt mechanistic explanation for different purposes? Do you have a sense of the disciplinary or intellectual problems/currents that drove the expansion of the mechanistic idea within sociology—perhaps, as you note, as a placebo for disciplinary ailments?
IAR
This is an interesting question. Certainly, doing such a genealogy would be interesting and worthwhile. I think it would be a bit more fragmented than you suggest, because one would need to include not only the trajectory of thinking about causal inference, but also a more diffuse but powerful intellectual formulation in sociology—including sociology that is quite “humanistic” or “cultural”—which is the sense that some aspects of the social world are quite regularized, but in a way that is not simple or easily grasped by just “looking on the surface” of social life (though I would note that surface/depth is also a metaphor).
What I mean by this is that there has always been a question: is the industrial (and now, algorithmic) social world mechanistic, in the same way as its factories (or AIs) are? On the other hand, the metaphor of mechanism has also been repeatedly used to describe supposedly pre-modern practices, such as ritual or gift exchange. The point is that I think that the metaphor of mechanism is perhaps quite old and widespread in work that calls itself sociology, though perhaps not always by that name, or not always in a codified way. I suspect that it arises when, for varying reasons, we find ourselves reaching for a study of regularity that is also a study of depth.
This is, perhaps, why the genealogical project would be a bit different than what we are doing in the paper, where we take up explicitly the project of “what we talk about when we talk about mechanisms,” now, in this (loosely defined) community of inquiry. Nonetheless, you are right that there is a link: the mechanisms concept, as it emerged into our current language via modular and non-modular accounts, appeared to merge the striving for causal explanation and the more interpretive/historical sensibility of a “mechanistic world” in one fell swoop. Our point is that there is a way to carefully articulate both ambitions, but it is not the way we have chosen so far. And so: there is more work to be done and we hope other folks will take up the larger project implicit in our paper. That is to place under the microscope of intense thought the specimens of sociological thought in our own time, which is to say, our concepts.
| CRK I agree with Isaac that a genealogical project would be fascinating but difficult—and a bit different from what we were up to. At a very basic level, it seems clear that mechanisms, at least as they have been most recently conceptualized, were meant to be a response to the idea that social science was about discovering generalizable “social laws” or “covering laws.” That’s at least how Hedstrom, Swedberg, and Elster, among others, framed the promise of mechanisms: we are not looking for laws, we are looking instead for empirical generalities. | “…the metaphor of mechanism is perhaps quite old and widespread in work that calls itself sociology, though perhaps not always by that name...” |
But my guess is that different conceptualizations of mechanisms will reveal very different genealogies. Apart from what Isaac mentioned with respect to mechanisms in causal inference versus cultural approaches, there’s a big difference between conceiving of mechanisms as forms of social explanation—the way Hedstrom and Swedberg do or the way Analytic Sociology does— and conceiving of mechanisms as the basis for causation—the way critical realists do. These approaches pull from very different theoretical traditions such that it is sometimes not really clear how much these approaches share, beyond similar terminology. Because of this, a genealogy of “mechanism” would be both incredibly useful but also likely very unwieldy.
IAR
Right! Note how both Swedberg, Hedstrom et. al. were offering “depth” in their response to the “surface” causality of covering laws and certain habits of statistical analysis; Bhaskar and the critical realists also offered “depth” in response to positivism. But when you dig down, is the use of the metaphor the same, in terms of the logic of theory and research that is implied? Perhaps not… this is the kind of thinking that led to our writing the paper together.
LF
In pointing to heterogeneity within a seemingly unitary term, your paper is reminiscent of Gabriel Abend’s (2008) “The Meaning of Theory,” also in Sociological Theory. However, instead of arriving at a “semantic predicament,” you end by identifying a metaphorical one, laden with the politics of epistemic authority. Recognizing the “metaphorical” uses of mechanism, you argue, means abandoning a degree of “epistemic capital” (248). This loss, as you note, may be particular to proponents of meaningful mechanisms. If unveiling the metaphorical character of “mechanistic explanation” exposes asymmetries of epistemic capital, what might be the next challenges for sociological theorists hoping to support something like epistemic pluralism, or at the very least, a clearer division of labor in social inquiry?
CRK
It is really nice to have our paper compared with Abend’s “The Meaning of Theory” since I see our paper as engaged in a very similar style of project. And while we do not explicitly tackle the normative question Abend refers to as a “semantic predicament”—in our case, how ought sociologists use the term “mechanism”— I think we do end up in a similar place to Abend. By that, I mean our ultimate goal was to argue for a pluralistic approach and to acknowledge, much like Abend does with theory, that the multiple ways that mechanism models were being used were already generative. So even though we believe attempts to nail down a single definition of “mechanism” would be unlikely to resolve the inherent contradictions among different approaches, the actual practice of using a plurality of approaches is useful.
IAR
Right! Note how both Swedberg, Hedstrom et. al. were offering “depth” in their response to the “surface” causality of covering laws and certain habits of statistical analysis; Bhaskar and the critical realists also offered “depth” in response to positivism. But when you dig down, is the use of the metaphor the same, in terms of the logic of theory and research that is implied? Perhaps not… this is the kind of thinking that led to our writing the paper together.
LF
In pointing to heterogeneity within a seemingly unitary term, your paper is reminiscent of Gabriel Abend’s (2008) “The Meaning of Theory,” also in Sociological Theory. However, instead of arriving at a “semantic predicament,” you end by identifying a metaphorical one, laden with the politics of epistemic authority. Recognizing the “metaphorical” uses of mechanism, you argue, means abandoning a degree of “epistemic capital” (248). This loss, as you note, may be particular to proponents of meaningful mechanisms. If unveiling the metaphorical character of “mechanistic explanation” exposes asymmetries of epistemic capital, what might be the next challenges for sociological theorists hoping to support something like epistemic pluralism, or at the very least, a clearer division of labor in social inquiry?
CRK
It is really nice to have our paper compared with Abend’s “The Meaning of Theory” since I see our paper as engaged in a very similar style of project. And while we do not explicitly tackle the normative question Abend refers to as a “semantic predicament”—in our case, how ought sociologists use the term “mechanism”— I think we do end up in a similar place to Abend. By that, I mean our ultimate goal was to argue for a pluralistic approach and to acknowledge, much like Abend does with theory, that the multiple ways that mechanism models were being used were already generative. So even though we believe attempts to nail down a single definition of “mechanism” would be unlikely to resolve the inherent contradictions among different approaches, the actual practice of using a plurality of approaches is useful.
| “...attempts to nail down a single definition of “mechanism” would be unlikely to resolve the inherent contradictions among different approaches, the actual practice of using a plurality of approaches is useful” | In describing mechanisms as metaphorical models, our goal was to point to the fact that (1) mechanistic explanation isn’t some grand unifying endeavor and (2) mechanistic explanation in and of itself does not solve the causality question. If this is the case, then, we hope abandoning mechanisms’ epistemic capital is a way to open the door for other similar kinds of meaningful models—like narrative, for instance. I also want to say that I don’t think our argument is particularly deflationary for meaningful mechanistic approaches. If anything, we had hoped to point to just how few of the relationships in our social world evince the kind of independence modular models require. |
As for supporting epistemic pluralism, I am pretty optimistic on this front. I think social theorists should continue the projects they are already engaged in—that is, specifying how our explanations work. For example, in the realm of comparative/historical sociology, I was part of a group trying to tackle the question of what case-study comparison was doing if it wasn’t principally aiding in causal explanations. If, for example, the Mills method doesn’t successfully establish causation, then why do we compare? These are the kinds of questions social theorists are already working on and, at least from where I stand, epistemic pluralism seems alive and well.
IAR
I think sociological theorists who wish to pursue this sort of epistemic pluralism should examine concepts that are used with high frequency in sociology, and really try to arrive at careful and robust understandings of them. Note that I write “understandings”, not “definitions.” This is not so much a definitional issue as one of meaning in communities of inquiry, and that means pursuing something like the method of working that we did in this paper—combining philosophical reflection with analysis of sociological practice, and both of those with some argument about what we should mean by these concepts. I think one might call the project “theory as concept development,” and the method “working epistemics.” Or something like that.
LF
A central task in mechanistic explanation, you highlight, is slicing through social complexity, “abstracting away a messy social reality into tractable and generalizable models” (238). The distinction between the two approaches revolves around the modular versus relationally interdependent character of these incisions. This core criteria of mechanistic explanation, however, seem to be shared by macro/grand theoretical traditions: approaches to field theory and Marxist class analysis, for example. We’re curious about where you see distinctions and similarities in this “carving of complexity” between mechanistic explanation and what is often called “grand theory”? The question seems prescient, since an appeal of mechanistic explanation has, to a degree, been its “meso-level” departure from the styles of abstraction in “grand” theories. It some ways, grand traditions prescribe the carving up social complexity into more or less standard pieces, but as you argue, approaches to “carving up” in mechanistic explanation are not fully deductive, but follow from ontological presuppositions.
CRK
This is a great question. Mechanistic models are certainly not alone in this endeavor at abstraction. Any form of social scientific explanation is necessarily engaged in abstraction since all models are idealized representations of complex phenomena. That’s true not just of mechanism models but also economic models, scientific formulas, maps, etc. We argued that what distinguished the meaningful from the modular mechanistic models was the relationship between these models’ components parts. That is, modular mechanisms models require a kind of independence among their components that just doesn’t hold for a lot of the social world that we might wish to model, for instance, the kinds of interactions or semiotic relationships that characterized meaningful mechanism models.
You raise a really interesting question about the difference between how grand theory versus middle-range theories go about modeling complexity. Generally, the distinction between grand theory and mechanisms is made in reference to levels- “macro” versus “meso” explanation, as you say. I think another fruitful way of characterizing that difference comes from the philosopher Philip Kitcher’s work on explanatory unification. He doesn’t discuss social scientific grand theory, exactly, but the general idea is that powerful, deep explanations are those whose abstracted models are able to unify seemingly distinct events and regularities with a minimal theoretical apparatus. Laura Franklin-Hall, who we cite in our paper, makes a similar point about how high-level explanation work by offering causal economy—high-level explanations give a lot of “bang for their buck.”
So grand theory isn’t only giving us an ontological architecture with which to describe the social world-- it is also makes otherwise confusing social reality coherent by showing how disparate events are manifestations of similar underlying social forces. That is part of what makes them so attractive. Apart from that, any kind of grand theory is obviously extremely committed to the content of their unified causal explanation—whether its economic determinism for Marx or technologies of power for Foucault. So, both in terms of the form and content of their explanations, grand theories aspire to a kind of unification that I don’t think mechanistic approaches aim to, or really can, achieve.
IAR
Furthermore, there is no particular reason why so-called grand theory cannot admit one or another kind of mechanistic explanation. It depends upon the question it is used to answer, and the overall intellectual project—that’s the pragmatist nature of the human sciences for you. I would also point out that what is sometimes referred to as grand theory—though I’m not sure we should constantly use a pejorative term if we want to really have a discussion about this—tended to have a third component beyond the two mentioned by Carly: an ambition to explain/interpret something that was in some way a historically specific, but extraordinarily widespread, phenomenon (“transition to capitalism in Europe”). Dan Hirschman has written about this a bit I believe. I consider Greta Krippner’s work in this light—it articulates concepts and makes unifications, but it is also explains one really important thing that affects millions of people (e.g. financialization, etc.)
IAR
I think sociological theorists who wish to pursue this sort of epistemic pluralism should examine concepts that are used with high frequency in sociology, and really try to arrive at careful and robust understandings of them. Note that I write “understandings”, not “definitions.” This is not so much a definitional issue as one of meaning in communities of inquiry, and that means pursuing something like the method of working that we did in this paper—combining philosophical reflection with analysis of sociological practice, and both of those with some argument about what we should mean by these concepts. I think one might call the project “theory as concept development,” and the method “working epistemics.” Or something like that.
LF
A central task in mechanistic explanation, you highlight, is slicing through social complexity, “abstracting away a messy social reality into tractable and generalizable models” (238). The distinction between the two approaches revolves around the modular versus relationally interdependent character of these incisions. This core criteria of mechanistic explanation, however, seem to be shared by macro/grand theoretical traditions: approaches to field theory and Marxist class analysis, for example. We’re curious about where you see distinctions and similarities in this “carving of complexity” between mechanistic explanation and what is often called “grand theory”? The question seems prescient, since an appeal of mechanistic explanation has, to a degree, been its “meso-level” departure from the styles of abstraction in “grand” theories. It some ways, grand traditions prescribe the carving up social complexity into more or less standard pieces, but as you argue, approaches to “carving up” in mechanistic explanation are not fully deductive, but follow from ontological presuppositions.
CRK
This is a great question. Mechanistic models are certainly not alone in this endeavor at abstraction. Any form of social scientific explanation is necessarily engaged in abstraction since all models are idealized representations of complex phenomena. That’s true not just of mechanism models but also economic models, scientific formulas, maps, etc. We argued that what distinguished the meaningful from the modular mechanistic models was the relationship between these models’ components parts. That is, modular mechanisms models require a kind of independence among their components that just doesn’t hold for a lot of the social world that we might wish to model, for instance, the kinds of interactions or semiotic relationships that characterized meaningful mechanism models.
You raise a really interesting question about the difference between how grand theory versus middle-range theories go about modeling complexity. Generally, the distinction between grand theory and mechanisms is made in reference to levels- “macro” versus “meso” explanation, as you say. I think another fruitful way of characterizing that difference comes from the philosopher Philip Kitcher’s work on explanatory unification. He doesn’t discuss social scientific grand theory, exactly, but the general idea is that powerful, deep explanations are those whose abstracted models are able to unify seemingly distinct events and regularities with a minimal theoretical apparatus. Laura Franklin-Hall, who we cite in our paper, makes a similar point about how high-level explanation work by offering causal economy—high-level explanations give a lot of “bang for their buck.”
So grand theory isn’t only giving us an ontological architecture with which to describe the social world-- it is also makes otherwise confusing social reality coherent by showing how disparate events are manifestations of similar underlying social forces. That is part of what makes them so attractive. Apart from that, any kind of grand theory is obviously extremely committed to the content of their unified causal explanation—whether its economic determinism for Marx or technologies of power for Foucault. So, both in terms of the form and content of their explanations, grand theories aspire to a kind of unification that I don’t think mechanistic approaches aim to, or really can, achieve.
IAR
Furthermore, there is no particular reason why so-called grand theory cannot admit one or another kind of mechanistic explanation. It depends upon the question it is used to answer, and the overall intellectual project—that’s the pragmatist nature of the human sciences for you. I would also point out that what is sometimes referred to as grand theory—though I’m not sure we should constantly use a pejorative term if we want to really have a discussion about this—tended to have a third component beyond the two mentioned by Carly: an ambition to explain/interpret something that was in some way a historically specific, but extraordinarily widespread, phenomenon (“transition to capitalism in Europe”). Dan Hirschman has written about this a bit I believe. I consider Greta Krippner’s work in this light—it articulates concepts and makes unifications, but it is also explains one really important thing that affects millions of people (e.g. financialization, etc.)
| There is, however, a further issue here. I think it is correct to criticize some aspects of the tradition of grand theory (whatever that is) as sometimes veering towards armchair philosophy (though as Andrew Abbott always pointed out, the so-called classical theorists loved data, historical evidence, etc.). But I would draw a distinction between the pejorative armchair and the necessity of reading relevant philosophical debates (philosophy of science, political philosophy, etc.) as part of one’s attempt to develop claims in the human sciences. One of the reasons I’m confident we really got to something important in this paper is that we forced ourselves to think with both the philosophers and the sociologists, while keep our own questions and concerns clearly in mind. | “...both in terms of the form and content of their explanations, grand theories aspire to a kind of unification that I don’t think mechanistic approaches aim to, or really can, achieve” |
LF
To conclude, we’re curious to hear if/how your approach to sociological research has been informed by the arguments arrived at in your paper? How do you think the metaphorical and multivalent character of “mechanism” will impact your own orientation toward sociological research?
IAR
In my own work, causal pluralism (and its relationship to the interpretation of cultures) was necessary from the beginning, because my first research project required a lot of thinking about an event that was both real and a frequently-used metaphor in American culture (the Salem Witch Trials, used perhaps most famously, by Arthur Miller to describe McCarthyism). I was always concerned with what the witch trials revealed about certain cultural-political dynamics. (I finally got to the heart of the matter of Salem, I think, in my forthcoming book, though there remains one more paper to be written). But this work also required an analysis and use of concept-as-metaphor: in this case “performance” as a way to think about power. In my own work, when this way forward succeeds, it produces a very satisfying return for intellectual efforts made: aspects of empirical materials that one cannot get at quite as well with other metaphors become illuminated. And that’s all you can ask for in this vocation, is it not?
CRK
Well, I’m a lot more thoughtful about when I use the word “mechanism”! But I also think it has made me a lot more thoughtful about the kinds of explanations that I’m working on generating. A lot of my recent work, for instance, has been in computational sociology where I’m attempting to characterize long-run changes in culture, discourse, and rhetoric. These abstract characterizations are transposable and generalizable, but are not really casual and are not mechanisms. They are more like cultural genealogies, and it has been interesting to think about what these kinds of genealogies do: what kind of unique explanations we get from this kind of historical “carving” of the social world. But perhaps that’s another paper.
To conclude, we’re curious to hear if/how your approach to sociological research has been informed by the arguments arrived at in your paper? How do you think the metaphorical and multivalent character of “mechanism” will impact your own orientation toward sociological research?
IAR
In my own work, causal pluralism (and its relationship to the interpretation of cultures) was necessary from the beginning, because my first research project required a lot of thinking about an event that was both real and a frequently-used metaphor in American culture (the Salem Witch Trials, used perhaps most famously, by Arthur Miller to describe McCarthyism). I was always concerned with what the witch trials revealed about certain cultural-political dynamics. (I finally got to the heart of the matter of Salem, I think, in my forthcoming book, though there remains one more paper to be written). But this work also required an analysis and use of concept-as-metaphor: in this case “performance” as a way to think about power. In my own work, when this way forward succeeds, it produces a very satisfying return for intellectual efforts made: aspects of empirical materials that one cannot get at quite as well with other metaphors become illuminated. And that’s all you can ask for in this vocation, is it not?
CRK
Well, I’m a lot more thoughtful about when I use the word “mechanism”! But I also think it has made me a lot more thoughtful about the kinds of explanations that I’m working on generating. A lot of my recent work, for instance, has been in computational sociology where I’m attempting to characterize long-run changes in culture, discourse, and rhetoric. These abstract characterizations are transposable and generalizable, but are not really casual and are not mechanisms. They are more like cultural genealogies, and it has been interesting to think about what these kinds of genealogies do: what kind of unique explanations we get from this kind of historical “carving” of the social world. But perhaps that’s another paper.
RSS Feed
