【按】如何精读与细剖一篇理论文本？如何不像只是summary地读？如何触摸文本的“肌理”？如何既constructive又critical？

——这可能在多数做社会学理论研究的常见困扰。

——为此，赖得吕炳强老师赐稿，下面贴出，他是如何解剖《Sociological理论大缸》推送过的一篇AJS发表的关于民族志中机制解释的论文。

——第56期：民族志也能【做出】机制式因果解释？实用主义回答.（下面评论的那篇AJS）

——第27期：开学季！课程大纲《社会学实践的哲学规章》（这是吕炳强老师在16fall于CUHK开设的一门课程。）

______________________

Commentson A Pragmatist Approach to Causality in Ethnography, by Iddo Tavory and StefanTimmermans（AJS,2013,119 (3): 682–714）

 以下P.+num是 这篇AJS论文的页码。Lui即吕炳强老师的剖析文字。

P. 683:

Whatever their provenance, all approaches to causality share common challenges: how to move from themessiness and abundance of empirical observations to a simplified causalexplanation and how to demonstrate the influence of indirect, temporally and spatially removed, processes. [...]

Lui: 

This can be regarded as the statement of the problem.

This article draws on pragmatist writings and processual approaches to causality to develop an alternative account of identifying and constructing causal explanations with ethnographic evidence. We propose three interrelated activities for identifying a causalexplanation. The first activity entails identifyinga causal sequence based on meaning-making structures. As a method of inquirythat rests on participation and close, detailed observation of people movingthrough their lives, ethnography has a unique ability to investigate unfolding moments of action. Building upon the semiotics of Charles Sanders Peirce, we argue that ethnographers should trace processesof meaning-making-in-action. The sequential process inwhich current meanings build further on previous meanings not only opens an analytical vantage point for the ethnographic study of causality but also allows ethnographers to systematically simplify messy, abundant empiricalmaterials. The second activity is to iterativelyrework the proposed explanation through an examination of variation. A semiotic chain of action in a single observed instancedoes not necessarily constitute a generalizable causal explanation. To firm upthe causal explanation, ethnographers are able to take advantage of thestructure and temporal dimension of meaning making to generalize acrossobserved variation and to recursively rework their proposed causal explanation. We suggest that ethnographers systematically examine three formsof observed variation in order to buttress their causal claims. These forms of variation include differences among prototheoretically similar situations,or data set variation, variationin meaning making over time, and intersituational variation, where different kinds of actions acrosslocales and interaction patterns are related to a common causal explanation.Checking for variations provides insight into the common resources and structural conditions. The result is a causal account thatmay include temporally and spatially removed processes without resorting to“invisible” social forces.

Lui: 

(1) The “semiotic chain of action” is recorded sequentially on the physical clock, that is, the physical time.  (2) “Temporally and spatial removed” means necessarily “invisible”. The real distinction is between “processes” and“social forces”. Watch out for the distinction as we read on.

P. 684:

The third interlinked activity consists of engaging the proposed causal explanation within a broader intellectual community. Following pragmatist insights, we add that the usefulness of these forms of variation as evidence for causal explanations is always tied toan attempt to convince a “community of inquiry”that a particular account fits observations and that the proposed causal explanation is better than plausible alternatives. The power of a causal explanation is relative to alternative explanations that the ethnographer can expect various audiences to raise. The added value of a theory resides in the way it is taken up by others and makes a difference within communities of inquiry.

Lui: As you have said, the introduction of“community of inquiry” is a contribution of this paper. Note that these “alternative explanations” are expected to be raised by various audiences, not there searcher himself. A simple way to say the same is: Situate one’s research inthe existing research literature.  Notethat Tavory and Timmermans made a distinction between “data set” and “evidence.

P. 684:

The intellectual contribution of our pragmatist approach to causality in ethnography is threefold. First, weground causality in the construction of a temporal generalization anchored in actors’ observed meaning-making processes, capitalizingon both the methodological strength of ethnography and the strength ofpragmatist semiotics. This approach does not imply a “first-person” approach toexplanation, where we privilege actors’ reasoning. Instead, their actions (whether verbal, cognitive, or otherwise),as ethnographically observed, form the bedrock ofanalysis. [...]

Lui: 

It is an all-researcher’s-view. What isstudied is actually pieces of actor’s speech, not their actions. If thisapproach is followed through, the researcher will be led eventually to the notion of Saussure-Bourdieuen network of speech—“the bedrock of analysis”. I suspect that what is called the “pragmatist approach” is simply a semioticapproach, the Peircean perhaps. The term comes from Neil Gross, see References.Watch out for this as we read on.   

P. 685

PRAGMATIST SEMIOTICS AS AMECHANISM-BASED APPROACH TO CAUSALITY

The philosopher of scienceJulian Reiss (2009)  noted that variousworking definitions of “causality” exist in thesocial sciences, each compatible with particular formsof evidence and disciplinary aims. Thus,for example, econometricians tend to treat the notion of causality assynonymous with prediction, while for many “large-N” researchers causality is a matter of establishing a regularity over time. Because of ethnography’s strength in capturing unfolding meaning-making processes (Emerson, Fretz, and Shaw 2011, p. 1), causal explanationsbased on ethnographic evidence are most amenable to mechanism-basedaccounts, which explain how a social phenomenon came into being or acts (seealso Hedström and Ylikoski 2010). In an influential essay on mechanistic causality, Peter Machamer, Lindley Darden, and Carl F. Craver (2000, p. 3)define mechanisms as “entities and activities organized such that they areproductive of regular changes from start or set-up to finish or termination conditions.” Mechanism-based accounts assume that explanations can bedecomposed into parts and thus specify generalizable processual links byshowing how, continuously, an explanandum leads to anexplanans. Importantly, mechanisms gain explanatory power when processesoccurring on a lower order of aggregation can be shown to explain how something was produced on a higher level of aggregation or abstraction (see also Machameret al. 2000, p. 13; Stinchcombe 1998, p. 267). Indeed, Gross (2009, p. 363)observes that “all work on social mechanisms assumes that mechanisms are thegears in some social machinery and thus stand in a relationship of lesser to greatervis-à-vis the causal effect they bring about.”

Lui:

(1) A mechanism has a beginning and an end, usually with more than one path between them and sometimes with loops in-between as well. An unfolding meaning-making process recorded by the ethnographer is the realized (andobserved) path but there are very likely many realizable paths in a mechanism.Check whether the newborn screening test cited by Tavory and Timmersman as the example has a well-defined end or not, though it definitely has a well-defined beginning (the announcement of a positive result to the parents). 

(2) The Mechamer-Darden-Craver paper is well-known in the theory of mechanism.

(3) A mechanism needs not to involve layers.

 (4) Tracing a causal explanation is always from an explanandum bacward to its explanans. Watch out for the final explanadum as we read on.

P. 686:

Drawing on pragmatism, Gross has made the case for moments of social action—problem solving in social practice—asalternative foundational processes for mechanism-based explanation in thesocial sciences. Gross rightly assumes that the kinds of explanations socialscientists look for are necessarily grounded in human action and (often)reflexivity. Gross’s pragmatist account of socialaction leads him to define social mechanisms as “composed of chains oraggregations of actors confronting problem situations and mobilizing more orless habitual responses” (Gross 2009, p. 368).Centering attention on habitual problem solving appropriately shifts the unitof analysis in the mechanisms literature to the momentof action, along with its situational and socially emergent features,without making untenable assumptions regarding human nature and rationality. The notion of “habitual problem solving,” however,has its limitations as the “bottoming out” level for a sociological, andespecially ethnographic, approach to mechanisms. Whereas mechanisms need toprovide an intelligible language through which causal processes take place, thefocus on problem solving leaves quite a few questions unanswered: How exactly do people solve problems? What are the differences between habitual and unreflexive versuscreative and novel problem solving? Gross emphasizes that meaning makingis central to any pragmatist theory of action; in fact, he says that a“mechanism is interpretive all the way down” (Gross 2009, p. 369; see also Reed 2011). And yet, Gross’s focus on subjects’ habits, collective habits, and bundled repertoires of “habit sets” (Gross 2009, pp. 370–71) as forms of social and epistemic culture and resources for action bypasses the pragmatist-semiotictheory of meaning precisely where it might be of most use to describe howmeaning is made. Because ethnography has a first-row perspective onmeaning-making-in-

action and because of the deeply reflexive nature of themethod (Hammersley and Atkinson 2007; Burawoy 2009), we argue that a pragmatist semiotic account of meaning making may offer a more fruitful foundational building block of social mechanisms than habitual problem solving. We therefore suggest that the pragmatist attempt to look for social building blocks from which toconstruct an intelligible processual account can be aided by what Charles Sanders Peirce, the founder of pragmatism, termed his “semeiotics.” Rather than beginning from individuals’ choices, or even with habitual problem solving, weargue that it is useful to start with the process of meaning making—whether in its most creative or its most habitual form. By focusing on the structure of meaning making we argue that ethnographers can provide precisely the kind of intelligible and continuous pragmatic account.

Lui: 

(1) “Moments of social action” is regarded as“problem solving in social practice”. But “problem solving” is very close towhat I mean by the actor’s interpretation (that is, narrative and strategy) inthe hermeneutics of my theoretical sociology. 

(2) Tavory and Timmermans distinguishbetween “habitual and unreflexive” and “creative and novel” problem solving.But I do not think the actor can solve any problem in his “mundane” state, andhence he can solve problems only when he is in his “reflexive” state. 

(3) Is“problem solving” the same as “meaning making”? Probably not.

P. 687:

Although Peirce’s theory of signs is notoriously technical, thebasic logic of his semiotics is straightforward (see Liszka 1996; Short 2007).As a way to ground the logic of scientific inquiry, Peirce broke down different aspects of meaning-making-in-action. In contrast to the later division of the sign into the signified and the signifier (Saussure [1916] 1986), Peircedevised a threefold partition. He wrote: “I define a sign as anything which is so determined by something else, called its Object, and so determines an effect upon a person,which effect I call its interpretant, that thelatter is thereby mediately determined by the former” (Peirce 1992, 2:478). Meaningmaking thus consists of three interlinked parts: a sign, an object, and aninterpretant. The first of these elements is the sign,which we can think of as the signifier in thesame way that smoke signifies a fire or that a word signifies a concept orobject. The sign does not exist on its own but is always in relationship to anobject. It is the utterance, pointing finger, picture, or whatever vehicleactors use to represent an object in a certain way. The second related elementis then the object, any entity aboutwhich a sign signifies, an actual thing “out there”or an idea in our head. Peirce’s key insight,however, was that meaning making is not an abstract but a practical achievement. To capture this point, Peirce argued thatevery act of meaning making includes an interpretant—theeffect of the sign-object through which any act of meaning making receives its practicaldefinition. The interpretant is a reaction that theinterpreter undergoes while making sense of a sign. Simply put, a sign isnot a part of an act of signification unless it has some kind of effect—an understanding, emotion, or action. Temporal movement is inherent in the action of meaning making, and signifying is thus always thrust into the future.

Lui: 

I draw the connection between the Saussureanand the Piercean semiotic systems:

The term “sign” takes different meanings in the two semiotic systems. It makes good reason to call the Piercean system “pragmatism” becausethe actor is brought in.

P. 688:

Peirce then made a second crucial move: the interpretant is always potentially the sign for anotheriteration of meaning making. Whether we are alone or whether we areenmeshed in interaction with others, each understanding or action throws usinto another round of meaning making. This is obviously true in a conversation with others, when the speaking turn of one party becomes the sign the otherparty acts upon, but it is even true in a soliloquy,when each thoughts gives rise to the next. Thus, a useful way to conceptualizemeaning making is to draw out the semiotic chains through which it isconstituted. Peircean semiotics can be imagined as a spiral of meaning making, where the interpretant of one iteration of meaning making maybecome the sign for another. This account of meaning-making-in-action is onethat ethnographic mechanism-based explanations could find extremely useful. Itis a processual approach, which works on a lowlevel of aggregation and which compels the researcher to construct a continuous, and intelligible, causal account.

Additionally, focusing on meaning making, we move from theproblematic attempt to ground mechanisms in the characteristicsof agents (e.g., rationally choosing ones) to the characteristics ofmeaning making in action; we provide the “how” of both explicit problem-solvingand habitually embodied action. And, perhaps most important, we present anaccount that ethnographers follow meaning-making-in-action, they are able to provide compelling semiotic

accounts of iterations of meaning making (Short 2007).

Lui:

The terms “sequence of action” and “semiotic chain ofaction” appear in the following paragraph.

（明天推送下半部。微信有字数限制，对英文算字数貌似不是按单词数。）

————————

（Sociological理论大缸第61期）

超链接：

第60期 百年《美国社会学学报》,只有54篇“历史”论文？！清单。

第57期 “我们需要的不仅是机制！”—阐释论的不满与反攻

第56期 民族志也能【做出】机制式因果解释？实用主义回答

第55期  别太抽象理论、别太微观行动，快告诉我【机制】到底要怎么找！

第51期 活着的美国社会学理论家有三代？关注哪些议题？《ST》主编Emirbayer透露