Korth Barbara the Leaps of Faith in Social Science

Barbara Dennis Korth, PhD, Clinical Assistant Professor, Indiana University

School of Education, Bloomington

Abstract: In this paper, the author introduces a post-Meadian concept usefulfor contemporary ethnography. Emergent ideal types capture the imaginary(that which is not bound to space and time coordinates) in the doing ethnog-raphy and making empirical claims (the empirical is considered that whichis limited to space and time coordinates). She uses a heuristic device,grounded narrative fictions, as one way of depicting the imaginary forethnographic audiences. The exploration draws on the ethnography of a kin-dergarten/Grade 1 class. The author does not use many scholarly refer-ences. Instead, she tries to develop two innovations to ethnographic analysis(emergent ideal types and grounded narrative fictions) from within anethnographic example. She tries in this thought piece to explicate the linguis-tic mechanisms through which ethnographers draw on the imaginary to linkempirical patterns with interpretive analyses.

Keywords: Mead, empirical, types, imagined, Habermas, social science the-ory

Citation

Korth, B.D. (2007). The leaps of faith in social science: A study of the imag-ined in the discourse of the real. International Journal of Qualitative Meth-odology, 6(1), Article 5. Retrieved [date] from http://www.ualberta.ca/~iiqm/backissues/6_1/korth.pdf

International Journal of Qualitative Methods 6 (1) March 2007

The Leaps of Faith in Social Science:A Study of the Imaginedin the Discourse of the Real

Barbara Dennis Korth

Authors note

Thanks to Dr. Phil Carspecken forhis dialogue on the topic of thisarticle.

There is a leap of faith in social science(Kierkegaard, 1992) that involves a non-empiri-cal imaginary: that which in any instance is not

accessible through space and time coordinates alone.Let us say that Schutz (1966) has a cube in front of him(the empirical) and that he is imagining a variety ofother possible cubes (the imaginary). He could imag-ine an infinite number of varied cubes. However, as heimagines these various cubes, he is not directly think-ing of the characteristics common to all imaginablecubes, such as rectangularity. Schutz referred to thisset of characteristics, unchanged among all the imag-ined transformations of the concrete thing per-ceivedthe kernel, so to speak, of all possiblyimaginable cubes as the eidos of the cube (p. 114).Schutz was capturing the imagined possibilities in theempirical. Social science is constituted of such leapsfrom the cube before us in space and time to the eidosof the cube and back again to the cube. This is not pecu-liar to social science (I recently started reading a his-tory of algebra the title of which includes theimaginary and real), but in this paper I will refer to thecontext of social science exclusively.

In this paper, the imagined and the empirical willboth be thought of as intersubjectively structured(Carspecken, 1996). Following Schutz (1966) and oth-ers, the empirical will refer to claims that are depend-ent on space and time coordinates and are, in principle,accessible to multiple witnesses (Habermas, 1981). Iwill refer to these empirical claims as the discourse ofthe real. The imagined will refer to paradigmatic struc-tures, mental configurations, concepts, inferences, andso on that are not constructed through a one-to-one de-pendence on space and time coordinates, although theimaginary might make reference to space and time. Inother words, the imaginary is not a mere reflection ofthat we presume to be real. It is something altogetherdifferent yet related. If both the leap of the imaginaryand its contingent, empirical claims are inter-subjectively structured, examining the intersu bjectivestructures that are involved in both would help us betterarticulate and explain social scientific leaps back andforth. This interesting thought project has practical po-tential for qualitative methods.

In this paper, I make a twofold contribution to qual-itative research methodology. First, I specify a sub-stantive concept (emergent ideal type) that linksintersubjective structures of the empirical descriptionswith intersubjective structures involved in explainingthe relation between the empirical descriptions and theimagined. Second, I demonstrate the usefulness of theconcept for ethnography by applying it to a specificempirical example. Here we find the imagined in thediscourse of the real.

In agreement with Habermas (1981, 1987), it is ex-pected that research investigations into actors livedexperiences reconstruct in a radical way what is al-ready internal to participants lived experiences. Thus,it is expected that the imagined in ethnography recon-structs from the imagined in everyday life. Theoreticalconcepts that articulate the imagined would thereforebe substantive and not merely methodological becausethey would simultaneously refer to both a substantiveaspect of understanding as it ensues in everyday lifeand a methodological tool to aid researchers in puttingthe implicit into discourse.

This paper is largely creative, in that it exemplifiesthe creative aspect of doing social science. It is also in-tuitive, in that I have sought to locate substantive con-cepts through ethnographic exploration, radicallyturning inward to understand something that might berelevant beyond that context. There are four major sec-tions: (a) The Discourse of the Real, (b) The Imaginedin Social Science Methodology, (c) Emergent IdealTypes: A Post-Meadian Concept, and (d) Capturingthe Imagined in the Discourse of the Real: AnEthnographic Example. Although the last section ofthe paper is devoted to exemplifying the points througha particular ethnography, I use examples from thatsame ethnography in earlier sections of the paper.

The discourse of the real

I am using the phrase the discourse of the real, whichdemands some definition and explanation. I introducedthis phrase to refer to the empirical: that which is con-tingent on space and time coordinates and is usually ac-cessed through the senses. This point calls for someclarification. Postmetaphysical ideas of empiricism in-volved a discrete subject-object distinction. Objectswere thought to exist outside and separate from sub-jects. Truth was the closest representation in discourseto the natural existence of the object prior to and out-side of the subject who was investigating it. That whichwas assumed to exist external to the subject and wasavailable for study by multiple subjects was consideredreal, but social reality was not as easy to describe be-cause it was difficult to make a definitive line betweensubject with social reality as object. In this section ofthe paper, I review three social science appropriationsof reality. These appropriations reflect contempo-rary debates on the topic of reality. First, I reviewBhaskars (1986) critical realism. Second, I reviewHarrs (1998) pragmatic realism. Last, I reviewHabermass (1981, 1987) presupposed realism.

Bhaskar (1991) claimed that reality canunequivocably (and no longer anthropocentrically orepistemologically) be accorded to things (p. 27)


http://www.ualberta.ca/~ijqm/

2 Korth LEAPS OF FAITH

themselves. According to Kivinen and Piiroinen(2005), Bhaskars critical realism assumes that the ob-jects of study exist independent of the observersclaims about them. Only by means of presuming thatthere are intransitive objects, scientific knowledge canbe seen as progressing toward ever more accurate de-scriptions of the human-independent world, as becom-ing closer to reality as it is (p. 232, italics added).Critical realism makes strong ontological claims. Thestudy of social life, for Bhaskar (1986), involves thestudy of social structures that are real entities, althoughthey are indirectly accessible to social science throughcausal effects. In other words, structures exist in timeand space through their causal effects. Nevertheless,the idea of social structure is meaningful only if it re-fers to some significant feature of reality (Bhaskar,2001, in Kivinen & Piiroinen, 2005, p. 233), the signif-icance of which is empirically assessed through causalrelations. Bhaskars claims about reality directly in-form his criticalism: If we are to act on reality with crit-ical intent, we must be able to identify what is real(Bhaskar, 2002). He might ask, for example, if we can-not say that poverty is real given its effects, then onwhat grounds would we commit ourselves to changingthe conditions of poverty? If we cannot say that genderis real, then on what grounds can we describe things asunequal based on gender? For Bhaskar, there is a sin-gular reality that exceeds and precedes its observers.Reality is not determined by various perspectives;rather, various perspectives can increase overall under-standing of the real world.

Bhaskars (2001, in Kivinen & Piiroinen, 2005) no-tions of reality depend on the idea that for a single phe-nomenon or object of interest there would correspond areality. Putnam (1990) pointed out that the causal crite-rion used by Bhaskar for locating social structures inthe empirical world requires acts of choice on the re-searchers part. Those acts of choice cannot be ex-plained or reduced to the system of objects themselves.To fill in, Kivinen and Piiroinen argued for a morepragmatic rendition of reality, namely that causes canonly be seen from an actors point of view, in respect ofthis or that purpose (p. 233), and therefore realitymust also be mediated by perspective and intention-ality.

For pragmatists, actors meaningfully use socialconditions, structures, and so on both intentionally andunintentionally. Reality is explained in terms of usefor theorists like Harr (1998) and Habermas (1981,1987). Harr, who was described by Kivinen andPiiroinen (2005) as a pragmatic realist and by Krausz(2000) as a constructive realist, claimed that reality ex-ists so long as people believe in it. In other words,Harr contended that there is a world of stuff independ-

ent of interpretive activity (schools, for example) butthat this world is known to humans only through theway in which it is ascribed. Furthermore, all ascribableproperties are interpretation dependent. Thus, therewould be no necessarily singular ontological reality in-dependent of human need. Harr (Harr & Bhaskar,2001) proposed that our commitments to realities werepractical-moral ones. Kivinen and Piiroinen called thisrealism with a small r. Ontologies would be justifiedaccording to the extent to which they help understandphenomena in a useful way. From this perspective,multiple realities (generated through use and need) arethe most plausible conceptualization for reality be-cause different needs and uses will concomitantly gen-erate multiple realities (notice how the theory itselfresorts to a single definition of reality).

Habermass (1981) pragmatic view of reality wouldalso be considered realism with a small r. ForHabermas, reality is presupposed and indicatedthrough ontological claims embedded in truth claims.What we end up with is a more tamed version of multi-ple realities, because Habermas posited three specificforms of truth claims, each of which presumes a partic-ular kind of ontology. The presumed or presupposedreality embedded in the claim becomes part of the wayin which conversing humans determine whether theclaim holds up. Each of the three types is linked to oneof three ontological presuppositions. For objectivity,actors presuppose an external, objective reality (muchlike Bhaskars [2001, in Kivinen & Piiroinen, 2005] re-ality). We make reference to this reality when we talk.We also make reference to a subjective world or real-ity, and we make reference to a social world. Multiplerealities, to this way of thinking, can be organized ac-cording to three formal systems of reference or presup-position. We must assume reality to talk, exchange,improve life, and so on. Each of these three ontologi-cally presupposed realities is implied in establishingthe validity of the claim and in understanding theclaims themselves. I interpret this to mean that we haveways of discussing reality though which a variety ofexperiences and perspectives are dialogically situatedwith a co-arising possibility for multiple realities, all ofwhich are reciprocally interdependent (not free float-ing) on the things said about them.

One important point that all three of these theoristsraised is this: Social reality itself is not something ofdirect experience. It is, rather, socially mediated. Forthis reason, I use the phrase discourse of the real withthe idea that discourse indicates the social mediationnecessary to talk, think, and act on what is real. In linewith Habermas (1981), however, I am trying to suggestnot that reality is an ontology-first concept but, rather,that reality is pragmatic supposition to any truth

International Journal of Qualitative Methods 6 (1) March 2007http://www.ualberta.ca/~ijqm/

Korth LEAPS OF FAITH 3

claims. There can be multiple such suppositions. Em-pirical claims presuppose the very particular reality ofobjects and phenomena in space and time.

The imagined insocial science methodology

The imagined is paradigmatically and intersubject-ively linked to actions that can be described with spaceand time coordinates, or what I will also refer to as thediscourse of the real (reality here is a presupposition tothe claims) (Habermas, 1981; McCarthy,1994;Tugendhat, 1986). Schutzs (1966) ideas help us seethis.

Schutz (1966) distinguished between two processesof imagining: eidetic universalization and empiricaluniversalization. According to Schutz, empirical uni-versalization is an inductive process not only contin-gent in the sense that [its] formation starts from aparticular given contingently in factual experience, butalso in the sense that the conceptualization proceeds onthe ground of comparison with likewise contingentlygiven similarities (p. 80). In statistical analyses, mea-sures of central tendency are examples of empiricaluniversalizations. Eidetic universalization, on theother hand, grasps the invariant content, the innerstructure, without which the object could not be thekind of object it is (p. 81). Continuing on, Schutzclaimed that this invariant element prescribes the lim-its to all possible variations of the same prototype; it isthat element without which an objectivity of this kindcan neither be thought nor intuitively phantasied(p. 81). New statistical methods involving factor analy-sis are more capable of mapping out eideticuniversalizations. Schutzs ideas were not explainedthrough a theory of intersubjectivity; rather, his workemphasized experience as the ground for empiricalperception.

I started this paper by calling what Schutz (1966)was holding a cube, which referred to something thathe might really have been holding. The use of the nameto identify the object requires a leap. Wittgenstein(1969) argued that words and names are not unambigu-ous, even when defined through a set of discrete attrib-utes or essence of the thing named. He proposedinstead that meaning ensues from the way in whichwords are used. Use indicates a series of similarities, orfamily resemblances. These family resemblanceswork off the idea that the link between empirical phe-nomena and meaningful paradigmatic interpretationsis linguistic and is based on a primary dialectic of simi-larity and difference.

Wittgenstein (1969) also argued that these mental,linguistic associations were the product not of a pri-

vate, idiosyncratic process but, rather, of a social one(Tugendhat, 1986), rendering them always, at least inprinciple, understandable and defensible to others. Ac-cordingly, family resemblances, concepts, and otherways of talking about various theories about the imagi-nary, are assumed here to be intersubjetively struc-tured, even when private. This intersubjective qualitymakes Wittgensteins family resemblances accessibleto the interpretive social sciences. We can readWittgenstein to imply that intersubjectivity andpragmatics form the ground for understanding what wecan observe rather than experience (as Schutz [1969]had posited). This is important to keep in mind as weassess the following ideas.

Types, metaphors, models, and concepts are someof the many different means by which social scientistshave leapt to something beyond the empirical. We dothis in everyday life as well. For example, if I report thesimple claim that there are six chairs in the room, thisclaim implicates, among other things, the concepts ofzero, chair, and in. Moreover, the idea of chairness,for example, does not exist in space and time but isnevertheless imagined, or is part of something that isimagined, when I claim that there are six chairs in theroom. The claim itself also implies how it would betested empirically. In this way, the imaginary can as-sume a central position in making sense of qualitativedata and in the construction of ethnographic texts. It isalso tacitly assumed in the analysis and report of quan-titative data (Bhaskar, 1998; Torgeson, 1958). In thissection of the paper, I examine a few of the varied so-cial science conceptualizations of this imaginary. Thepurpose is not to explore all such possibilities but tonote a few of the ways in which the imaginary has beenconceptualized as intersubjective and social so as tomark a path toward the concept that I will be develop-ing in this paper: emergent ideal types. Notice that I amnow talking about concepts of concepts.

Below I describe four examples through which Iseek to articulate the imaginary in social science,namely Webers (1949) ideal types, Dexter andLaMagdeleines (2002) use of metaphors, Lakoff andJohnsons (1980, 1999) ideal prototypes, andHabermass (1981, 1987) concept of the ideal speechsituation. Of the four, Webers ideal types andMagdeleine and Dexters metaphors are most specifi-cally about doing social science. Lakoff and Johnsonsideal prototype postulates a conceptual connection be-tween ideal and typical concepts as metaphors withinhow folk make sense of their everyday experiences, al-though this is not a distinctly social science concept.For each of these three, the imagined is produced as atype of concept which differs from some real, observ-able manifestation. This contrast of the imagined to the



real is fundamentally different from Habermass use ofthe term ideal.

Webers ideal typesin the conduct of social science

Webers (1949) ideal type is a formulation reconstruct-ing implicit ideals in empirical phenomena from whichall specific empirical phenomena will deviate to vari-ous extents. An ideal type, therefore, will not exist em-pirically in the sense that one would be able to findempirical examples that conform to and thereby du-plicate the ideal type. For Weber, ideal types were amethodological tool useful in the analysis of data, es-pecially historical data.

According to Berger (1976), Weber described fourdifferent kinds of ideal types, all of which worked offof a couple of key principles. First, they were derivedinductively from across multiple examples of similarcultural phenomena. Ideal types do not manifest as ag-gregates of characteristics; instead they include char-acteristics that might have manifested regularly acrossthe types if not for some disturbing influences; that is,empirical phenomena were thought to have these char-acteristics in common under ideal situations. Theseideal situations did not exist because of the presence ofactual disturbing influences. Those disturbing influ-ences, whatever they were, disallowed the genuinepureness of a phenomenon to manifest empirically.Berger wrote, The source of the dissimilarity [be-tween the concept and its empirical counterpart] is tobe found in the fact that the situational conditions varyand that the inner states (treated as motivational) arecomplex (p. 125). This idealization represents notwhat the phenomena have in common but, rather, whatthey would have in common when liberated from otherinfluences.

Weber (1949) proposed that social science re-searchers could develop these ideal types as a methodof capturing, in an exaggerated and one-sided way, theessence of some cultural phenomena. Once con-structed in discourse as an ideal type, the researcherscould bring these ideal types to bear on the empiricalworld. Through this way of thinking, the empiricalworld, of course, would not measure up because of thepresence of distorting features within the culture. ForWeber, the essence of ideal types was radically imagi-nary and external to the real.

An example of one of Webers (1949) famous idealtypes is the bureaucracy, where no single empirical ex-ample of a perfect bureaucracy exists in the empiricalworld, but, rather, each approximates a perfect idealbureaucracy. Weber did not connect this heuristic so-cial scientific tool with the interpretive processes in

which people intuitively engaged while making senseof everyday social phenomena, for example in their ev-eryday interactions with bureaucracies.

Dexter and LaMagdeleines metaphorsin qualitative research

Dexter and LaMagdeleine (2002) explored the use ofmetaphors in qualitative research. They argued that aresearchers attempts to both understand the essence ofcultural phenomena and bring out that essence for oth-ers involves the use of metaphors. According to Dexterand LaMagdeleine, metaphors can be both generativeand reflexive. Metaphors exaggerate or foregroundsets of characteristics or components for particular pur-poses. Thus, researchers can talk about points at whicha metaphor fails to work in contrast with the more suc-cessful aspects of a metaphor-choice. Multiple meta-phors can be applied to the same set of data. Thosemultiple metaphors indicate varying positions throughwhich particular interpretations can be rendered. Suchmetaphors are developed through a mapping processthat involves (a) locating best fit, (b) heighteningawareness of saliency and distribution, and (c) grasp-ing form and content relations. This process results inthe articulation of an image or picture that is meant toconvey a whole set of internally meaningful relationsby developing an external construct that would simul-taneously deliver the essence of those relations. Ac-cordingly, metaphors are derived by researchers inorder to deliver complicated relations of findings. Themetaphor is an articulation of the imagined.

What the metaphor articulates is the imagined es-sence of cultural phenomena, which is neither what it isempirically nor what it is metaphorically. The imag-ined exists in the space between these. In this way, met-aphors can both generate understandings of data andreflect understandings of data. They summon meaningfor the reader.

It is interesting to compare Dexter and La-Magdeleines (2002) use of metaphors with Webers(1949) ideal types with respect to ethnography. Theprimary similarity is that metaphors and ideal types inthese specific contexts are developed inductively, sug-gesting that the relationship between the imagined andthe empirical is inductive. Furthermore, both meta-phors and ideal types are heuristic for social science.There are important differences across these concepts.The metaphor does not attain the status of measuringpost; it merely meaningfully delivers aspects of theimagined. Although metaphors fit various empiricalphenomena to certain extents, they fit on the level ofinvoking an imagined set of internal relations similar tothose discovered empirically; whereas ideal types are




articulations of ideal, imagined versions of the culturalphenomena and can be used directly to locate where, inwhat way, and to what extent empirical phenomenavary. Metaphors, on the other hand, help to hand overthe essence of the phenomenon in its complexity with-out emphasizing the variability of empirical phenom-ena. Webers ideal types are imagined, but metaphors(as talked about by Dexter and LaMagdeleine, 2002)are expressions of the imagined. For Dexter andLaMagdeleine, metaphors are explicitly described asintersubjectively structured, and this is why they canwork as expressions of the imagined.

Let us compare the two using Schutzs (1966) dis-tinction between eidetic and empirical universal-izations. Barber (1988) interpreted the difference be-tween eidetic and empirical universalizations as one ofdegrees. This interpretation would say we move to-ward the eidetic universalization by first inducing em-pirical universalizations. Both Weber (1949) andDexter and LaMagdeleine (2002) would conform tothis interpretation of Schutz: The use of induction leadseventually to grasping the essence, or eidos, of a phe-nomenon, which can be expressed through a metaphor(in Dexter and LaMagdeleines case) or can produce anideal type (in Webers case).

Lakoff and Johnsonsmetaphors and prototypes

Dexter and LaMagdeleine (2002) wrote about meta-phors as heuristic tools for qualitative researchers, butLakoff and Johnson (1980, 1999) talked about meta-phors as necessary for the ways in which people makesense of their own experiences. These metaphors aremore organic than the ones Dexter and LaMagdeleinedescribed. Actually, Lakoff and Johnson developed astrong theory of metaphors, which would include thekind that Dexter and LaMagdeleine wrote about, withmore organic metaphors as primary. An example of anorganic metaphor is force, whereby the body experi-ences force physically and comes to conceptualize as-pects of the nonphysical world metaphorically in termsof force (Lakoff & Johnson, 1980). Lakoff and John-son called these embodied metaphors. The imagined isthe form the physical experience takes as somethingelse. These organic metaphors form crucial systems ofmeaning that are invoked through actors everyday so-cial intercourse. They are intersubjective with an expe-riential basis.

Lakoff and Johnson (1999) also wrote about proto-types, contrasting ideal prototypes with typical proto-types by using the examples of the typical husband (atypical prototype) and the ideal husband (an ideal pro-

totype). Lakoff and Johnson asked readers to conjureup the difference in imagery between these two types.

Neither prototype is meant to be taken as an actualempirical possibility (although for different reasons).The typical prototype (Lakoff & Johnson, 1999) is ageneralization of aggregates across empirical exam-ples. This typical prototype is similar to Schutzs em-pirical universalization, whereas the ideal prototype ismore like Webers (1949) ideal types (what the phe-nomena would be like without the interference of dis-torting features). Imagining the ideal prototypeinvolves thinking of this as a flawless or perfect ver-sion of the prototype, whereas imagining the typicalprototype involves imagining characteristics of theempirical types deduced with their distorting influ-ences.

The ideal in Habermassideal speech situation

Thus far, we have seen the word ideal connote some-thing like perfect, pure, or undisturbed, which missesthe interpretive generativity that we find inHabermass (1981, 1987) use of the word ideal in ref-erence to the ideal speech situation. Habermas pro-vided a more directly intersubjective, presuppositionaluse of the term ideal. For Habermas, the ideal speechsituation is both presupposed and counterfactual in allattempts toward reaching understanding. The idealspeech situation is the description of conditions underwhich communicatively achieved understanding ispursued, even fallibly; namely, conditions of reciproc-ity and symmetry: that all participants or participantsperspectives are given equal opportunity for expres-sion and challenge in any communicative effort aimedat understanding. The ideal speech situation is a condi-tion presupposed in understanding and in reachingconsensus.

Often enough Habermass (1981, 1987) idealspeech situation is interpreted as if it is an idealistic be-lief in what might be possible within actual speech situ-ations. This misinterpretation results in peoplethinking that Habermass theory is very unrealistic andrenders the ideal speech situation as something akin toeither (a) Schutzs (1966) empirical universalization asan empirical universalization of speech situations or(b) Webers (1949) ideal types, where the ideal speechsituation is the situation that would obtain in the empir-ical world if all distorting elements were eliminated.This last interpretation is inaccurate because it is insuf-ficient. Habermas has argued that the ideal speech situ-ation is usually counterfactual (McCarthy 1994). Thisconception, when taken alone, leads people to limittheir understanding of a Habermasian critique.



Through this limitation, Habermass speech situationis no more than Webers ideal type applied to commu-nicative contexts. The misinterpretation and limitationare the effect of a similar theoretical mistake in under-standing how the ideal speech situation is naturalisti-cally and philosophically connected to the empiricalworld. Given the more common use of the term ideal,these mistakes are understandable; however, they arealso worth clearing up.

According to Habermas (1981, 1987), the idealspeech situation is presupposed by any communicativeact precisely because it is the way in whichintersubjective sensemaking can be constituted. Theideal speech situation is a force in the holistic impetusto even act meaningfully (Carspecken, 1999).Habermas has not only described a utopia or a pureform of communication; he has reconstructed thatwhich makes possible any attempt to make sense ofoneself and others. The principles of democracy andegalitarianism implicit in the ideal speech situation canunfold as empirical phenomena (for example, throughvoting), putting the ideal speech situation in the statusof limit case, but the ideal speech situation itself is fun-damentally a constitutive force presupposed in anycommunicative act, even nondemocratic, non-egalitar-ian ones. Therefore, the ideal speech situation (a) iscarried through acts, not merely referenced by them;(b) works through intersubjective (position-taking)structures; and (c) cannot be limited to empiricalclaims, although it will always be connected to them.

Habermas (1987) developed these points by draw-ing on Meads (1934) theories about the generalizedother, intersubjectivity, and the self. The most basiccritical potential associated with the ideal speech situa-tion lies originally in the juxtaposition of communica-tions meaningful dependency on the ideal speechsituation and its deferment of the ideal speech situationrather than on its secondary status as a limit casethrough which empirical acts might be measured.Therefore, the ideal of an ideal speech situation refersto presupposed principles through which communica-tive activity is actually possible but which is not likelyto be fully empirically embodied in that same commu-nicative activity. This dual character of Habermassideal speaks both to its generativity and to its pragmaticinterpretability in everyday communication, and it isthereby also distinguished from both Schutzs (1966)eidetic and the empirical universalizations in the do-main of the imaginary.

The imaginary, in this sense, is a more radically in-ternal concept than those that we have explored thusfar. It deepens our understanding of internalinterpretability without stepping outside of it

(Habermas, 1981). The term presupposed is used toindicate the leap to the imagined. In this case, the pre-supposed is that which can be reflectively put into dis-course and used to describe empirical phenomena. Inaddition, there is another important difference here:The empirical world is conceptualized not solely as aworld of existing things but as a world to whichinteractants can refer when communicating andachieving understanding.

Emergent ideal types:An introduction

Emergent ideal types, as I have conceived of them, aresocial attitudes regarding identities, roles, dispositions,and other self-related structures that are organized intoa coherent system of normative expectations. Theyfunction as a limit case presupposed when social actorsinterpret themselves and others as particular types ofpeople. The types emerge, in the first place, from theeveryday interactions of social actors meaningfully en-gaged with one another. Their value for social scienceis linked to this. Emergent ideal types are ideal forself-other relations in the same double-edged way thatthe ideal speech situation is ideal for reaching under-standing: They are generative and interpretable. Emer-gent ideal types are the imaginings through whichones interactions give one a sense of identity, self, andotherness. They are not ideal representations of a per-fect image of something but, rather, of dialogicallyimagined possibilities [hence the generativity] for sub-ject-specific interpretations given a shared set of nor-mative relations (hence the interpretability). They aretypes because they have some substantive stability,that is, a boundedness that renders them recognizableas such-and-such. In other words, interpretations co-alesce in way that can be referred to with some unity.

Emergent ideal types:A post-Meadian concept

Having established some sense of the social sciencecontext for thinking about the imaginary in relation tothe real, in this next section of the paper I develop theconcept emergent ideal types as part of the imaginarydomain with leaps to the empirical: the discourse of thereal. Emergent ideal types take us into a narrower dis-cussion of the imaginary because we begin focusingspecifically on the imaginary as it relates toethnographic discourse about real subjects. The pointswill be exemplified using ethnographic data and find-ings from a previous study (Korth, 2001).



Meads generalized otherand emergent ideal types

Mead (1934) developed a dialogic theory of the self ca-pable of explaining how individual selves and groupnorms are inherently mutually constitutive. Meadsconcept of generalized other is crucial to this theory.According to Mead,

We get an other which is an organization ofthe attitudes of those involved in the same socialgroup. The organized community or socialgroup which gives to the individual his unity ofself may be called the generalized other. Theattitude of the generalized other is the attitude ofthe whole community. (p. 154)

Mead used the example of a game to illustrate this.A player must grasp what each possible play meansfrom the perspective of all the players positions; theorganization of these positions is what he refers to asthe generalized other. In other words, it is the unitaryorganization of these perspectives that constitutes thegeneralized other for particular social groups; it is notthe mere accumulation of different positions towardones actions but, rather, the qualitative unity throughwhich all particular positions might grasp their rela-tionships to one another for the entirety of the event. Tokeep with Meads game example, players might simi-larly grasp the game as a whole and the possible playsas parts of that whole. This involves grasping thenorms implicit in each of the possible plays as well asthe system of norms that serves to unify the variouspossibilities.

The unitary aspect can be spotted in how ones ac-tions would be evaluated. According to Carspecken(2002), this evaluation involves taking a third-personposition on ones own actions by position taking withthe generalized other. I might therefore evaluate myplay as batter using this unified grasp of the system ofpossibilities and perspectives that comprise the gener-alized other of the baseball game: the set of expecta-tions regarding what batters are supposed to do inrelation to the rest of the players and the goals of thegame. According to Meads theory, all interpretationsinvolve this kind of position taking with a relevant gen-eralized other.

With respect to the interpretive process,Carspecken (2002) wrote,

We do not simply take the position of anotherperson, we do so by simultaneously taking theposition of a third person who could commenton the interactions of all participants and we as-

sume that the other participants in the interac-tion can also occupy this position of the general-ized other. (p. 17, italics in the original)

Therefore, the third-person position is one that couldcomment on the expectations for play regarding anyone of the players positions: one unified third personposition through which all relevant positions are un-derstood.

If the generalized other for a particular interactionis made up of the shared set of norms implicit in actingappropriately, then the attitudes that coalesce (let ussay as player positions) assume the form of what Iam calling emergent ideal types. These emergent idealtypes are often taken for granted within a system ofnormative relations constituting the generalized other.To use more of Meads (1934) formulations, we couldsay that emergent ideal types are nodes of social atti-tudes that are organized into a whole as the generalizedother. They are the nodes of attitudes that one wouldrecognize as relevant to understanding ones self fromthe generalized other. Here we can see both the genera-tive and interpretive aspects. Emergent ideal types aregenerative, in that they are the attitudes that constitutethe possibility of acting in particular sorts of ways andin that they open up the possibility for socially interpre-tive relations and recognition.

Returning to the game example, remember that thegeneralized other is the system of rules and norms thatunite all possible plays from all possible positions inthe game. The positions of pitcher, catcher, first base-man, and batter reflect coherent nodes of social atti-tudes and indicate possible emergent ideal types.Emergent ideal types are ideal imaginings in the sensethat (a) they are presupposed in interactions both struc-turally (in terms of normative relations implicit to thegeneralized other) and substantively (in terms of thecontent boundaries), (b) they do not exist in time andspace, and (c) they are both generative and interpretive.This proposal is post-Meadian because it extends thesketch that Mead laid out to include the dialectic of dif-ference in the constitution of unity. So far, I have intro-duced the concept of emergent ideal types from withinthe domain of the imagined.

Theoretically speaking, a person constructs socialidentities (or mes, to use Meads concept of the self)by position taking with the generalized other(Carspecken, 2002). Whatever me is constructed byposition taking with the generalized other is done inpart by identifying myself within the normative struc-ture constitutive of the generalized other (Carpsecken,2002). Emergent ideal types work off the principle ofdifference. Much of what is written about the general-ized other captures the sense of singularity in terms of



position taking with the generalized other, but few au-thors have tried to develop the dialectic with differencepresupposed by any particular generalized other. Dif-ference is both presupposed and produced the minuteone position-takes with a generalized other. It is thisposition taking and the part-whole dialectic betweenthe generalized other and emergent ideal types thatseem to generate narratives that actors use to displaythe sense of unity of self that Mead considered an effectof the generalized other. This whole process (positiontaking with a generalized other, grasping thepart/whole relations presupposed by the generalizedother, comprehending specific activities through thegeneralized other, recognizing particular identities inreference to the generalized other, and producing nar-ratives which display the unity of self) involves differ-entiating emergent ideal types.

It would also be useful to locate emergent idealtypes in relation to social positioning theory, a con-trasting postmodern theory of identity. Davies andHarr (1999) advanced Foucaults work to argue thatpeople identify themselves by positing an identity thatholds a relation to social discourse. When people inter-act meaningfully, they do so by establishing positionswith those in the interaction, histories of meaning, andrelevant social contexts. These positions co-createidentities. In making utterances, individuals seek toestablish positions for themselves, both in an immedi-ate relation with an other, but also in relation to themany other exchanges that go to make up patterns ofdiscursive exchange (Winslade, 2005, p. 353). More-over, according to Harr (1998), humans take up twoidentity projects through their activities. They seek toestablish their honorability within a social group, andthey also seek to establish their uniqueness as one veryparticular, individual being. These projects are motiva-tional, and they require that the person be positionedsocially. There is no singular positionality, as onesprojections can be multiple. They are most certainlysituational and ongoing. One does not accomplish theprojects of being socially honored and individually ap-preciated once and for all. Neither does one accom-plish these projects in the absence of social interaction.With Harrs (1998) conceptualization we see the uni-tary aspect of social identity involved in the project ofbeing honored by ones social group and the differenti-ating aspect of identity involved in the project of seek-ing individual recognition as unique being.

Harr and van Langenhove (1991) argued that so-cial positioning theory provides details necessary toexplain how discourse is involved in the production ofsocial relationships and the claiming of personal iden-tities. Indeed, it is possible to advance a notion of

identity as a product of the clustering of repeated iden-tity positions accepted and taken up in a multitude ofconversations, rather than resorting to essentialist ex-planations of personality (Winslade, 2005, p. 355).Dominant social discourses set the parameters for whatis appropriate to talk about, what seems most impor-tant, what constitutes a successful argument, and so on.Social positioning theory offers an explanation analo-gous to Meads (1934): one positions ones self (analo-gous to the ME in Meads self) in relation to dominantdiscourses (analogous to Meads generalized other)and thereby simultaneously positions others. As posi-tions are co-produced, they create relational attributes,for example who among the interactants have the rightto talk or respond in particular ways. The positions arereal to the extent they are visable and viable throughthe discourse such that they are believable to others.The positions, in effect, become the manifestations ofparticular realities. The two identity projects (estab-lishing ones self as honorable through the view of par-ticular social groups and establishing ones self asindividually unique) both require social positioning.The interpretive milieu is implicit in these projects andin the social positioning theory itself. Metaphysical de-scriptions of identity, with their links to humanism, arenot credible for Harr (1997), yet the expressivism thatemerges as an effect of these two identity projects is in-teresting to him.

Following this line of thinking, our identities arereal because they are real for the actors, and discoursesare real for the same reason, but reality is not a unitarything existing distinct from the subject, and the imagi-nary does not stand in contrast to this reality. ForDavies and Harr (1999), social positioning is imma-nent and is reproduced moment by moment in linguis-tic action, not from an abstraction. Abstractions arepart of creating theories (Aronson, Harr, & Way1995). Davies and Harr would likely interpret emer-gent ideal types as a form of discourse, real in their ef-fects and not properly explained through as apresupposition in acting. According to Davies andHarr, the extent to which these ideal types manifest inaction, they are real. They would use a real history ofexperiences and thoughts as well as manifest a contem-porary, situated identity interdependent on the realforces of the conversations to explain similar phenom-ena. The project, from their perspective, would not beabout capturing the imaginary in the real, because inthe context of emergent ideal types there would only bethe real. This problematizes my uses of the terms idealand imaginary, and I will return this problem in theconclusion.




Capturing the imaginaryin the empirical: An example

In this section of the paper, the empirical articulation ofemergent ideal types is illustrated. There are two sub-sections and a conclusion, all of which are intended toillustrate how one captures and then displays this imag-inary.

Methodology and context

I am using an ethnography of a U.S. kindergar-ten-Grade 1 (K-1) class to provide a concrete exampleof the philosophical points. (Korth, 2002). The class Istudied had high level of student activity; a colorfuland stimulating environment; class pets; student move-ment and talk; learning centers; computers; huge col-lections of puzzles, games, books, blocks, and toys;and the noticeable absence of desks (including ateachers desk). In general, the classroom was a veryrich, safe, orderly, well-planned environment withinwhich children were able to pursue their learning inter-ests. The social as well as material and academic mi-lieu was intended by the teachers to facilitate groupwork and interaction.

This ethnography began when the school opened inJanuary 1997. I observed the schools only class at thetime (a K-1 class of 18 students with 2 teachers). Ispent 8 hours a week from January through May ob-serving the class. I became a regular part of the class-room life. Consequently, I was able to talk frequentlyand informally with the students as part of the everydaylife of their classroom experiences. In addition, twoother field researchers were making weekly observa-tions. I conducted hour-long interviews with parentsand several hour-long interviews with teachers and theprincipal. Classroom observations were audiotapedand videotaped. Interviews were audiotaped. All datawere transcribed to create thick records. These were re-viewed by the three of us involved in data collection.This produced hundreds of pages of data. These datawere analyzed by me using peers and members asdebriefers. These validity techniques were coupledwith negative strip analysis and consistency checks(Carspecken, 1996).

To understand the classroom life, as ethnographer Itried to learn the insider perspectives, in the lan-guage of Mead (1934). This meant coming to under-stand the generalized others, which were most sharedfor participants of the classroom. As someone who wasa relative outsider, this generalized other was alwaysonly partially shared by me, because although I learnedto appreciate myself through position taking with thesevarious generalized others, I was always also position

taking through a generalized other that involved myethnographic attitude more directly.

I spent a great deal of time trying to understand andreconstruct the activities in the classroom across time,persons, and interactions so that I could begin to inter-nalize the generalized others that were presupposed byteachers and students. Doing this resulted in the loca-tion of tacit, commonly referenced, unified sets ofnorms. The norms presupposed certain ways of actingin the class that were referenced as a type of person.When I understood the interactions, it was partly be-cause I understood the pragmatically structured waysin which class members recognized appropriate waysof acting and what that meant from various perspec-tives. Thus, there was a reciprocal and dialogic rela-tionship between my intuiting the types and myreconstructive analysis of classroom activities. Thesetypes were not an accumulation of traits or a deductionof qualities. Rather, the word type refers to recognizingand participating in a particular bounded way of under-standing self and others. The emergent ideal types re-flect parts in dialectic with the generalized other aswhole. These emergent ideal types can be differenti-ated as possible cultural limit cases in actual interac-tions.

I will discuss two of many emergent ideal types re-garding classroom success: the Emergent Ideal Type:Nice Girl and the Emergent Ideal Type: Smart, Capa-ble One[BOY]. To illustrate the methodological process,I describe some of the ways in which classroom actorspointed me toward an understanding of success in theculture so that I can describe the emergent nature of theanalysis and pose a hermeneutic definition of success,through which I came to recognize two relevant emer-gent ideal types. Following that, I present the two spe-cific emergent ideal types through a strategy that I callgrounded narrative fictions. All of the names used inthis paper are pseudonyms.

Internal references to successand emergent ideal types:

Capturing the imaginary in the real

There were many references to success in the class-room; sometimes specific children were referenced assuccessful, and sometimes democratically establishedrules marked ideas of success, for example. Sometimesthe references were constructed through language,sometimes through nonverbal acts, and sometimesthrough logical relations. By paying attention to these,I learned to recognize what it meant to be thought of assuccessful in this particular K-1 class. This is not thesame thing as saying that I learned to recognize thegeneral characteristics of success (although this fol-




lows), nor is it the same as saying that I captured theeidos of success. What I learned was the way in whichthe interpretation success was generated andbounded through a unified system of norms and a dif-ferentiated set of types.

One morning I was in the classroom sitting at a ta-ble with four children who were working on a puzzletogether. A couple of boys, including Mark, were sit-ting on the classroom floor nearby, also collaborativelyputting together the pieces of a puzzle. The childrensitting with me were talking about their puzzle. Thenone of them turned to me, pointed to one of the boys sit-ting on the classroom floor nearby, and said, You seehow many pieces he can work? Thats because whenhe talks nobody understands him. The children werepointing out to me a particular student, Mark, whomthey took to be successful, and they were also articulat-ing some of the ways in which they noted success: inthis case, verbal vocabulary and the complexity of thepuzzle he was able to solve (Korth, 2002). The studenttelling me about Mark and his friend was indicatinghow, through a generalized other, he might posi-tion-take in relation to Mark and boys like him. Bytalking to me about this, he was pointing toward a wayof making sense of himself and others, a way that heexpected was shared with those he was talking aboutand others in the class. The student was not merely pro-viding me with a set of characteristics to list out. Hewas helping me interpret himself and others by sharingwith me this particular way of getting Mark.

During center time one day, a group of childrenwere shaping play dough. Allyssa pressed her shirt tothe play dough and produced textures in interestingpatterns. Another child approached the table and toldthe children that they were supposed to be making di-nosaurs. Janie countered, The other teacher said wedidnt. Then Janie went on to say, We copy fromwhatever Alyssa does. Kun adds, Yeah Allyssaknows what to make. Another child adds, She al-ways helps us, then Janie concludes the interaction bysaying, We do what Allyssa does. This example, likethe earlier one, illustrates how children referred to aparticular student while simultaneously foregroundingcharacteristics that they associated with success intheir classroom environment.

Once during a math lesson, the class was playingGuess My Number. To play this game, the studentleader sets boundaries within which the secret numberresides, and fellow classmates attempt to efficientlyfigure out what the secret number is using the bound-aries and others guesses. It was Boyds turn to lead thegame because he had correctly guessed the teachersnumber. One of the students said, If ever Boyd is play-ing a game, Im not playing. Other students chimed in

and said the same thing. Their actions within the groupreferred me to a particular child who might be thoughtof by students as less successful. At this point, though,I had little go on in terms of understanding why Boydwould be considered less successful. I could begin tounderstand who students thought of as leaders (andwho they did not want as leaders), but I could not justequate leadership with success.

Students also referenced their peers through the useof eye contact, peer selection, and so on. For example, Iobserved that Jennifer frequently exchanged smileswith other children, often patted other children on theback, and regularly spoke positively to other children.Girls liked to play with Jennifer, and many students ap-proached her for a variety of reasons. She interactedwith others a lot.

In contrast, there were some children who were notgiven attention by their peers. Moreover, once in awhile teachers would make reports to researchers likethis, I am concerned about Dave because he doesntinteract with the other children. Here is a sad but typi-cal interaction for Dave. Dave got up from his chairduring center time and began walking. He walked (andthis does not look purposeful) by the center whereMark was sitting.

Dave: Speaking to Mark, a male classmate whois sitting at a Center in a nearby chair. Couldyou tie my shoes?

Mark: No. Doesnt look up from what he isdoing.

Dave: I dont know how.Mark: Looks away. Gets up and walks off.Dave: Walks over to Cathy, a female classmate,

at the play dough center. To Cathy: Couldyou tie my shoes?

Cathy: Looks at Dave. Shakes her head toindicate no.

John walks passed Dave.Dave: To John, Would you tie my shoes?John: Shakes his head to indicate no.Dave frowns. He leaves the play dough center

and begins walking, somewhat aimlessly,around the room.

Dave: To Allyssa, Would you tie my shoes?Allyssa: Nods her head to indicate that she will.

Stoops down and ties his shoes.Dave: To Allyssa, Thanks. Wanders around

the room as if unsure which center to go tonow.

The teachers encouraged students to help eachother in these kinds of ways, but Dave had a hard timesecuring such help and interacting with his peers. He



spent much of his time in the class alone. These kindsof observations and comments provided an under-standing of what and/or who would not be interpretedas successful by class actors.

Through all of these observations/interactions overtime, I grasped a presupposed set of relations that wereinvolved in how class actors evaluated and interpretedthemselves with respect to success. These various atti-tudes regarding success were organized into a general-ized other whose unitary nature had to do with locatingones self in terms of success. This success was largelyvisible through the popularity of the children amongother children, but it was explained in broader terms:Thats because when he talks, no one can understandhim. Popularity was an internal attribute to success inthis class. The empirical scene made sense through thecultural structures that were implicit in how classroomactors were interpreting each other. Further work usingreconstructive horizon analysis facilitated the articula-tion of the notion of successfulness within the class.The analytic process and the data collection processtook on a dialogic relation to one another, as did thevarious levels of inference involved in understandinghow success was both achieved and pointed toward, in-cluding getting at the boundaries of success and its si-lenced oppositions. This was not a solely inductiveprocess and involved the imaginary as interpretationrather than as the accumulation of a set of characteris-tics. In other words, the eidos of success the sense al-ready presupposed through the empirical character-istics but not equivalent to them was captured.1

Emergent ideal types were (re)constructed in theform of what I call grounded narrative fictions. Thesegrounded narratives were the best display for corestructures that presupposed the emergent ideal types.Grounded narrative fictions were a product of my so-cial science, but they drew off of an impetus to narrateidentity.

Grounded narrative fictions:Writing the imaginary

into the discourse of the real

I played with several forms, including metaphors, in aneffort to display emergent ideal types as they were pre-supposed through position taking with a shared gener-alized other. Grounded narrative fictions were the formthat seemed most intuitively connected to emergentideal types: They were capable of manifesting the in-terpretive fluidity and generativity associated withemergent ideal types and their generalized other.Grounded narrative fictions comprised actual data thatwere reconfigured (not decontextualized but relo-cated). These narratives facilitated the contextualized,

implicit understanding of the subtleties involved ingrasping emergent ideal types as presupposed by ageneralized other. The narratives also brought out theinterpretive boundaries of those types. On the riskyside, grounded narrative fictions read as if the emer-gent ideal types were embodied by particular studentsrather than that they were the radically imaginary pre-suppositions organized through a unified generalizedother.

Methodologically, (a) the emergent ideal typeswere thoroughly reconstructed, then (b) the data werereexamined in light of these reconstructions. This in-volved specifically looking for examples andcounterexamples of interpretive, implicit references tothe proposed types. Returning to the data providedboth an opportunity for a validity check and an oppor-tunity to begin (c) telling the stories that condensed theintuitive experience into a single narrative. The pointof the stories was to set the context through which her-meneutic interpretations might emerge for readers, al-lowing them to grasp the generalized other andemergent ideal types presupposed through the herme-neutics of classroom life.

I will help: A nice girl

Alison (a fictional student who is constructed of actualdata from across time, space, and person to capture thepresuppositions of success which characterize theEmergent Ideal Type: Nice Girl) entered the classroomsmiling. She looked around the room and seemedaware of various things going on. Regardless ofwhether it was center time or group time, she fre-quently acknowledged others in the class. For exam-ple, when she walked by children working in centers,she was likely to smile at them. When Alison joined agroup of students already busily engaged in their activ-ities, she did so by matching what they were doing: Shesmoothly integrated herself into the ongoing flow. To-day, like many other days, she walked past a center,and one of the children said to her, We saved you aspot over here, Alison. Remember we made patterns inthe clay yesterday. She helped others either whenasked or by noticing that someone needed help andthen initiating an offer. For example, while the studentswere writing in their journals, Alison noticed that oneof her classmates was having trouble figuring out howto write a particular word. Alison gently tapped herfriend on the shoulder and pointed to a spot in the envi-ronment where the problem word was written. Alisonhelped others clean up, make things, read, and so on.She seemed highly aware of the goings-on of her fel-low classmates.




Alison was rarely in conflicts. She asserted herselfbut seldom ended up in a confrontational dispute.When she wanted a turn at something but had thus farbeen denied, she said, I want a turn, please, and usu-ally she got one. Here is an example illustrating howshe handled potential disputes without the events esca-lating into a conflict over access to markers. Alisonwas dealing with the problem at the same time that Su-san, a less successful student, was trying to deal withthe same problem.

There was a small, rectangular table with one longside pushed up against a wall. There were four chairsaround the table, one on each end and two on the longopen side of the table facing the wall. Margaret was sit-ting at one end of the table with the bucket of markerson the table directly in front of her. Cathy was sittingnext to Margaret. Both of them were using the markersto work on their journals. Susan was seated in the chairnext to Cathy. Alison joined the three of them by sittingin the chair at the end of the table opposite Margaret.Neither Susan nor Alison could reach the markerswhile sitting in their chairs.

This was what happened next.

Susan: Arms folded across her chest, talking ina loud volume to Margaret. Hey that isntfair. Youre taking all of them [the markers]. Ifyou wouldnt have sat there this wouldnt havehappened.

Alison: Speaking in a calm voice and tone toMargaret. We need to get some too.

Susan: Sounding angry. You cant just dothat. You cant just take all the markers youwant to so nobody else can take them. Standswith a firm-looking appearance at the end ofthe table that is too far to reach the markercontainer. It seems that she wants Margaret toput the marker container in the center of thetable so that everyone can reach it.

Observer Note: Margaret continues to reachinto the container and use markers as she needsthem. She neither looks at Susan nor speaks toSusan, though it is obvious that Susan is upset.

Alison: Walks over to a spot behind whereMargaret is sitting. She smiles at Margaret.Margaret smiles back. She reaches overMargarets shoulder into the container ofmarkers, grabbing a couple of particularcolors. She takes these markers with her andreturns to her chair at the other end of the table.She completes her journal entry and then putsthe markers back into the container. She leavesthe group and has refrained from alienatinganybody or engaging in a dispute.

Observer Note: All this while, Susan continuesto express her anger towards Margaret and adispute ensues that eventually requires teacherassistance. In the end, Susan moves to adifferent table to gain access to markers andcomplete her journal entry.

Alison was not vocal in her resolution of this prob-lem. Rather, she was interpersonally smooth, refrain-ing from insisting that Margaret do something aboutthe marker problem, and instead, silently and with apositive approach (smiling), Alison just got a handfulof markers and returned to her seat to quietly completeher work. She left the group having stirred no negativefeelings. She found a way to accomplish her goal whilesimultaneously maintaining positive interpersonal re-lationships with everyone at the table. Alisons successhere did not facilitate the resolution of conflict for Su-san.

Alison talked favorably about her workone-on-one. As she participated in the learning centers,she said things to her nearby classmates like this:Look at this. See what I did. She made similar posi-tive comments about her classmates work, Thatsnice. Alison was a good reader, and she was recog-nized as such by her classmates.

In summary, Alison tended to display happiness,personal assertiveness in one-to-one situations, goodreading skills, engagement with classroom activities,support to fellow students, and positive conflict resolu-tion. She was like a magnet to whom conversation, eyecontact, and references keep returning.

The Emergent Ideal Type; Nice Girl was a non-em-pirical imaginary involved in how class members madesense of one another. References to the Nice Girl in-voked an interpretation pattern that emphasized femalegender and morality.

NICE GIRLS DO THINGS WHICH ARENICE FOR OTHERS

NICE GIRLS HELP OTHERSNICE GIRLS WORK HARD AND PRODUCE

GOOD RESULTSNICE GIRLS ACT FRIENDLY AND GET

ALONG WITH OTHERSNICE GIRLS ACT CONGENIALLYNICE GIRLS DO NOT ACT COMBATIVELY

OR ARGUMENTATIVELY

Therefore, when classmates were position takingwith a shared generalized other, there was an EmergentIdeal Type: Nice Girl, which was one of the presup-posed attitudes organized into the set of norms and val-ues invoked through this generalized other. When




classmates would position-take with this generalizedother, they would recognize themselves and evaluatethemselves by grasping the unity of the generalizedother as well as in relation to the parts, like this Emer-gent Ideal Type: Nice Girl.

Look at me. My idea is bestThe smart, capable one [BOY]

Craig, the name for the fictional character of this emer-gent ideal type, walked confidently into the room as ifhe had a purpose. He faced the world with his headup. His steps were quick, and he got right into an activ-ity where there were other boys. He seemed focused onhis activities, even when they last for only short peri-ods. He was highly verbal and able to articulate theo-ries, ideas, and opinions as well as engage inlogical-verbal disputes with apparent ease. He talkedmore than other children in the room, and he used a no-ticeably more advanced vocabulary. He was consid-ered smart by other children in the class, who usedindicators such as the complexity of puzzles and theway he talked to distinguish him as smart.

Craig was frequently in disputes with fellow class-mates. These disputes did not seem to produce anykind of unhappiness for Craig. In other words, disputesdid not seem to be counted as any kind of lack of suc-cess on his part, as anything to avoid (whereas for girls,a dispute is interpreted as a lack of success, a kind offailure). He seemed, on the other hand, to feel proud ofhis disputes and proud of winning them, as often as hedid. Usually these disputes were verbal but once in awhile became physical. In addition to disputes, Craigwas able to offer counterpronouncements that won theassent of many in the class without requiring much per-suasive effort.

Craig acted with confidence and self-assurance. Hemade frequent bids for leadership during center andgroup activities. It was common for him to interjectsome interesting piece of information or some interest-ing theory or explanation into the ongoing group timediscussions. For example, one day the teacher wasreading a book to the class. She had just read a para-graph that contained the word glacier. Craig raised hishand, and this is what happened:

Craig: Raising arm and wagging his hand.Ummmmmm. Me.

Teacher 1: Looks up. Speaking to Craig,Yes.

Craig: To the teacher, I can tell them [hisclassmates] what glaciers are.

Teacher 1: Looks at Craig. Can you wait?Craig: Can I tell them now?

Right then, he went on to tell his classmates aboutglaciers. His frequent bids for leadership during grouptime resulted in his receiving a lot of attention from theteachers and peers. He spoke and answered and posedquestions, spontaneously moving to the front of the cir-cle to share his ideas with the group. He commanded alot of attention throughout the course of the day andspoke very positively about his own work and ideas.He did this both through small groups and for thewhole class. He was verbally assertive, but in a waythat was much more public than Alisons mode of ver-bal assertiveness.

Craig interacted most often with other successfulboys, although his interactions were not limited to thisgroup. He seemed confident and happy as he under-took classroom activities. He frequently sought and re-ceived class attention and was considered smart by hispeers. Unlike Alison, it was rare for Craig to offer helpto his classmates or to compliment others on theirwork.

The emergent ideal type of successful boy wasnamed The Smart, Capable One[BOY].

SMART BOYS EXPRESS GOOD IDEASSMART BOYS PERFORM WELLSMART BOYS EXPRESS THEMSELVES

WELLSMART BOYS PERSUADE OTHERSSMART BOYS ACT RATIONALLY AND

REASONABLYSMART BOYS EXPLORE THEIR CURIOUS-

ITIESSMART BOYS DO SMART THINGS

The reason I called this emergent ideal type SmartCapable One with the [BOY] as subscript makes sensebecause these characteristics of success were the onesarticulated by the teachers and parents as what wouldbe considered successful in the classroom in a presum-ably ungendered way, but in practice this form of suc-cess was really only actively interpreted as such forboys (Korth, 2002). I wrote the boy as subscript toindicate how gender was hidden in the discourse withinthe claim that this form of success was not gendered.

The imaginary in the discourseof the real: A summary

In the above descriptions, I wove together empiricalexamples to illustrate emergent ideal types in narrativeform. These activities did actually occur but not as theactions of any one particular child. The point of using anarrative like this is to grasp the discursive nature ofthe emergent ideal type and to illustrate how such types



would be recognized implicitly through roles and otherpragmatic structures.

Now it is time to return to the original promise:demonstrating how emergent ideal types capture theimaginary in the empirical. Remember that the empiri-cal has been referred to as claims that are dependent onspace and time coordinates, which are produced in aform accessible by multiple folks. Empirical claimsmanifest one kind of interpretive position from whichto act meaningfully. It is not that the empirical isnon-interpretive; rather, empirical interpretations spe-cifically report on activities that can be seen, heard, andso on by multiple others (at least in principle). Allclaims immediately depend on a set of inter-subjectively shared structures (Brandom, 1994) and con-tent, such as the concept of success, which might betaken for granted in the process of understanding. If itis thought of in this way, then the empirical seems to bea particular kind of differentiated interpretation fromthe more holistic imaginary.

The imaginary is part of the intersubjective web ofpossibilities that might be both impending and refer-enced. Capturing the imaginary involves puttingintersubjective structures and content into discourse.Emergent ideal types are presupposed attitudes com-prising a shared generalized other. The main point ofthe paper was to introduce the concept emergent idealtype, which was theorized as part of the process of cap-turing the imaginary in the empirical. This can beturned around by acknowledging the dialectic betweenthese.

Conclusion

I began this paper with the idea that our ethnographicreports necessarily involve something more than theempirical. The point is fairly well accepted in the quali-tative social sciences, but it is not well conceptualizeditself. In this article, I have contributed to the qualita-tive methodological theory by defining the imaginaryas the intersubjective structures that could be inferredfrom the empirical and on which the empirical waslikewise dependent for its own saliency. This set up theargument that this imaginary is not something peculiarto the conduct of social science. Ethnographic partici-pants, in making sense of their lives, will employ thesame leaps of faith hermeneutically that ethnographersemploy methodologically.

I introduced the imaginary concept emergent idealtype, and then I illustrated its relevance toethnographic data analysis. The emergent ideal type isa substantive concept, in that it captures something thatactors refer to in the process of coming to understandone another. The empirical and imaginary seem, in the

end, to constitute a horizon relation from which actorsreciprocally engage in coming to understand one an-other. This involves differentiating the empirical fromthe imaginary through time- and space-dependentclaims. It is through this iteration in time and space thatthe imaginary is constituted, and yet it is also onlythrough the generativity of the imaginary that the em-pirical can become differentiated. The underpinningexplanation for this dialogic reciprocity is a theory ofintersubjectivity (Carspecken, 2002).

Ethnography adds a layer of complexity becausenow the empirical is make up of the ethnographersempirical claims about the lifeworld (actors empiricalclaims plus the imaginary through which their empiri-cal claims are sensible), plus the ethnographers ownimaginary. To put forward empirical claims about thelifeworld, one needs to intuit the lifeworld actorsimaginary. I labored over this as a methodologicalproblem and decided that articulating the emergentideal types that one captures when capturing the imagi-nary in the empirical is one way to do this.

Emergent ideal types might be put into discourse inseveral ways. I illustrated one of these: through thecomposition of grounded narrative fictions. I embed-ded the use of cultural reconstructions and the recon-struction of identity claims into the grounded narrativefictions because the narratives recreated an intuitive,naturalistic opportunity for readers to appreciate thehermeneutic grasp of the emergent ideal types. Such anintuitive display will implicitly involve the more basicparadigmatic structures of validity claims(Carspecken, 1996). Moreover, each of these ways ofbringing the emergent ideal types into discourse mir-rors ways these emergent ideal types might be broughtinto discourse in the lifeworld.

The concept of emergent ideal types was linked pri-marily to Meads (1934) generalized other and second-arily to Harr and van Langenhoves (1991) idea ofsocial positioning. Linking the imaginary (emergentideal types) to the imaginary (generalized other, socialpositioning) requires philosophy, but it is relevant toempirical work. I have argued that our qualitativemethods rely both on empirical observations and imag-inary structures. Moreover, what we produce as re-search, this discourse of the real, is never purelyobservation. Qualitative methods will benefit frommore exploration into the dialectic between the imagi-nary and the real in both our understanding of socialphenomena and in our understanding of methodology.

At last, it is possible to problematize the imaginaryas I have used it to describe emergent ideal types withHarrs (1998) pragmatic constructivist realism. Socialpositioning theory provided another plausible way ofdescribing what I refer to as emergent ideal types. Con-



trasting my post-Meadian account with social posi-tioning theory promises interesting insights.However, because this paper has methodological in-terests at heart, it is the challenge Harr poses relatedto the use of a category the imaginary that must beaddressed. For Harr (1998), Davies and Harr(1999), Harr and Langenhove (1991), the discourseof gender identity would be real but indirectly ac-cessed through metaphors, for example. This indirectaccess suggests for me the leap to which I have beenreferring. For Harr (Harr & Bhaskar, 2001) andothers, indirect access is more of a procedural than asubstantive feature of realities. The idea of multiplerealities owing to the various situated aspects of iden-tity and the various discourses prevailed on when se-curing ones identity are compatible with howemergent ideal types are proposed here, but the idealand presuppositional description of those types is an-tithetical to the explanation that would be offered byHarr. Such divergent explanations in social scienceare worth dialogue, and it is my purpose in this paperto continue (not resolve) such conversations.

Notes

1. These interpretations were checked out through interviews,peer debriefing, and negative strip analysis (Carspecken,1996).

References

Aronson, J., Harr, R. & Way, E. (1995). Realism rescued: Howscientific progress is possible (new ed.). Peru, IL: Open Court.

Barber M. (1988). Social typifications and the elusive other: Theplace of sociology of knowledge in Alfred Schutzs phenomen-ology. London: Lewisburg Bucknell University Press.

Berger, T. (1976). Max Webers theory of concept formation:History, laws, and ideal types. Durham, NC: Duke UniversityPress.

Bhaskar, R. (1986). Scientific realism & human emancipation.London: Verso.

Bhaskar, R. (1991). Philosophy and the idea of freedom. Cam-bridge, MA: Basil Blackwell.

Bhaskar, R. (1998). The possibility of naturalism: A philosophi-cal critique of the contemporary human sciences (3rd ed.).London: Routledge.

Bhaskar, R. (2002). Reflections on meta-reality: Transcendence,emancipation, and everyday life. Thousand Oaks, CA: Sage.

Brandom, R. (1994). Reasoning, representing, and discursivecommitment. Cambridge, MA and London, England: HarvardUniversity Press.

Carspecken, P. (1996). Critical ethnography in educational re-search: A theoretical and practical guide. New York & Lon-don: Routledge.

Carspecken, P. (1999). Four scenes for posing the question ofmeaning and other essays in critical philosophy and criticalmethodology: Studies in postmodern theories of education(Vol. 79). New York: Peter Lang.

Carspecken, P. (2002, September). Critical inferentialism andreconstructive data analysis: Applications of inferential se-mantics and pragmatics to critical ethnography. Paper pre-

sented at the Oxford Ethnography in Education Conference,Oxford, United Kingdom.

Davies, B., & Harr, R. (1999). Positioning and personhood. InR. Harr & L. van Langenhove (Eds.), Positioning theory(pp. 32-52). Oxford, UK: Blackwell.

Dexter, S., & LaMagdeleine, D. (2002). Dominance theater,slam-a-thon, and cargo cults: Three illustrations of how usingconceptual metaphors in qualitative research works. Qualita-tive Inquiry, 8(3), 362-380.

Habermas, J. (1981). Reason and the rationalization of society.Boston: Beacon.

Habermas, J. (1987). Life-world and system: A critique of func-tionalist reason. Boston: Beacon.

Harr, R. (1998). When the knower is also the known. In T. May& M. Williams (Eds.), Knowing the social world (p. 37-49).Buckingham, UK: Open University Press.

Harr, R., & Bhaskar, R. (2001). How to change reality: Story vsstructureA debate between Rom Harr and Roy Bhaskar. InJ. Lopez & G. Potter (Eds.), After postmodernism: An intro-duction to critical realism (pp. 22-39). New York: Athlone.

Harr, R., & van Langenhove, L. (1991). Varieties of position-ing. Journal for the Theory of Social Behavior, 21(4),393-407.

Kierkegaard, S. (1992). Concluding unscientific postscript tophilosophical fragments (Vol. 1; H. V. Hong & E. H. Hong,Trans.). Princeton, NJ: Princeton University Press.

Kivinen, O., & Piiroinen, T. (2005). The relevance of ontologi-cal positions in social sciences: Realist and pragmatist view-points. Journal for the Theory of Social Behavior, 32(3),231-248.

Korth, B. (2001). Reforming educational practice against theboundaries of (re)iteration: A critical ethnography of the hid-den curriculum of a constructivist charter school. In G.Walford & P. F. Carspecken (Eds.), Critical ethnographies inEducation (pp. 153-198). London: JAI.

Korth, B. (2002). Critical qualitative research as conscious-ness-raising: The dialogic texts of researcher/researchee inter-actions. Qualitative Inquiry, 8(3), 381-403.

Krausz, M. (2000). Interpretation and its metaphysical entan-glements. Metaphilosophy, 31(1/2), 125-147.

Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chi-cago: University of Chicago Press.

Lakoff, G., & Johnson, M. (1999). Philosophy in the flesh: Theembodied mind and its challenge to western thought. NewYork: Basic Books.

McCarthy, T. (1994). The critical theory of Jrgen Habermas.Cambridge, MA: MIT Press.

Mead, G. H. (1934). Mind, self, & society: From the standpointof a social behaviorist (C. Morris, Ed.). Chicago: Universityof Chicago Press.

Putnam, H. (1990). Realism with a human face. Cambridge MA:Harvard University Press.

Schutz, A. (1966). Studies in phenomenological philosophy.The Hague, the Netherlands: Martinus Nijhof.

Torgeson, W. S. (1958). Theory and models of scaling. NewYork: John Wiley.

Tugendhat, E. (1986). Self-consciousness and Self-determina-tion (P. Stern, Trans.). Cambridge, MA: MIT Press.

Weber, M. (1949). Max Weber on the methodology of the socialsciences (E. A. Shils & H. A. Finch, Trans. & Ed.). Glencoe,IL: Free Press.

Winslade, J. (2005). Utilising discursive positioning in counsel-ling. British Journal of Guidance and Counselling, 33(3),351-364.

Wittgenstein, L. (1969). The blue and brown books. New York:Barnes and Noble.




Documents

Korth Barbara the Leaps of Faith in Social Science