Foucault, Genealogy, Emergence: Re-examining the Extra-discursive

Foucault, Genealogy, Emergence: Re-ExaminingThe Extra-Discursive

NICK HARDYjtsb_446 68..91

INTRODUCTION

The objective of this paper is to offer an alternative theorization of a well known,in-depth account of restructured social relations—viz. Michel Foucault’s argu-ment in Discipline and Punish (1975/1991) of the production of “delinquent” popu-lations. The argument draws upon and develops two key concepts found in thework of Foucault and in critical realism: the “extra-discursive” and “emergence”.By developing an account that sees delinquents as neither wholly the outcome of(Foucaultian) “technologies”, “programmes” or “strategies”, nor as (criticalrealist) “ontologically distinct” entities, an analysis can instead be made thatunderstands delinquents as constituted by their imbrication in an immanent andcontingent field of power relations that forms—but does not determine—them asa distinct group.

This paper aims to refine Foucault’s position by pursuing two interlinked linesof inquiry. The first is to deepen Foucault’s conception of power by extending theuse of the extra-discursive as a part of his ontological grounding of power. Thismeans that power “relations” come to be understood as contingent and multi-vectoredand power “effects” as multi-causal. The second is to show that these relations havea constitutive effect upon the subjects on which they exert influence. Seen inthis way, subjects become examples of constitutive emergence—meaning they areneither mere discursive constructs (i.e. “defined”) nor ontologically distinct (i.e.physically/essentially unique). With such an understanding, Foucault’s latergenealogical position gains greater analytical depth and critical realism acquireswider scope to incorporate more complex and historically rich accounts.

The extra-discursive—as an explicitly employed but not fully theorized conceptin Foucault’s work—is much more prominent in his earlier “archaeological” stagethan in his later writing. Although his move to “genealogy” was an overt attemptto create a more dynamic engagement with the extra-discursive, the success of thismove is debatable, and forms the larger subject of this paper. There is already a

Journal for the Theory of Social Behaviour 41:10021-8308

© 2010 The AuthorJournal for the Theory of Social Behaviour © The Executive Management Committee/Blackwell Publishing Ltd. 2010.Published by Blackwell Publishing, 9600 Garsington Road, Oxford, OX4 2DQ UK and 350 Main Street, Malden, MA02148, USA.

body of literature that examines the analysis and the effects of the extra-discursive(and of emergence) in Foucault’s work. Some follow Foucault’s own lines ofthought (Dreyfus and Rabinow, 1982), others inquire into comparisons or pos-sible integrations (especially with regard to other theoretical perspectives such asMarxism) (Balibar, 1988/1992; Marsden, 1999; Read, 2003; Joseph, 2004), whileothers directly critique Foucault’s work (Lecourt, 1972/1975; Dupont andPearce, 2001; Pearce and Woodiwiss, 2001; Datta, 2007; Frauley, 2007; Jessop,2007). Many of the latter object to the manner in which Foucault does notadequately define, integrate or explain the extra-discursive and its interaction withthe discursive (Dupont and Pearce, 2001; Pearce and Woodiwiss, 2001; Datta,2007; Frauley, 2007). As this paper will argue, the implicit use of the extra-discursive as a key, but ill-defined, aspect in explaining the formation of emergententities or concepts and is a sticking point within Foucault’s theory. This results inhim being caught (in Discipline and Punish, 1975/1991, for example) in a largelydescriptive—as opposed to analytic—account of events.

Alongside adapting Nietzsche’s concept of “genealogy”, Foucault also usedNietzsche’s concept of “emergence”. In Nietzsche, Genealogy, History (Foucault,1971/2003) he used the idea of emergence to signify the sudden appearance of anew “domination”; where domination signifies the end result of a Nietzscheanstruggle between protagonists for the position and the ability to define, andtherefore shape, the part of the world they have come to dominate. Foucault usedNietzsche’s “Entstehung” to denote these instances of “emergence”. This allowedhim to chart the disjointed movements of “history”, which he saw as neitherprogressive nor rational but only the “endlessly repeated play of dominations”(1971/2003: 358).

In his later work, however, Foucault seemingly adapts his own genealogicalanalysis to encompass more than just Nietzschean “dominations” as beinginstances of emergence. In his lecture series Territory, Security, Population (2004/2007; see also Foucault, 1976a/1990: 140), he outlines a convincing argumentthat depicts the “emergence” of discursive concepts—such as “population”—aspart of a developmental process (as opposed to a sudden emergence caused by anew domination). However, as this paper will discuss, this understanding posestwo challenges to his own work: first, that the process argument is inconsistentwith his stated position that emergence is only ever produced from “domina-tions”. Second, it adopts as an important aspect of this development a complexinterplay between the discursive and the “extra-discursive”; an interplay that, aswas highlighted above, Foucault does not theorize at a similarly sophisticated levelas he does the discursive.

Foucault is not the only theorist to employ a concept of emergence. It is alsoused in critical realism which treats emergence as the (sometime) product ofcomplex interactions (“events”) between various mechanisms, powers and enti-ties. One of the most detailed accounts of emergence in critical realism is outlinedby Elder-Vass (2005). He defined a concept of “compositional” emergence that

Foucault, Genealogy, Emergence: Re-Examining The Extra-Discursive 69

© 2010 The AuthorJournal for the Theory of Social Behaviour © The Executive Management Committee/Blackwell Publishing Ltd. 2010

expands upon Bhaskar’s (1978/2008) original and Archer’s (1995) subsequentdevelopment of what constitutes the form of the social and the material world.This has also been a theme with other critical realist writers (e.g. Pearce, 2007;Kaidesoja, 2009). As this paper will argue, a critical realist understanding ofemergence offers Foucault a means of developing his own account of emergence,allowing him to incorporate both “sudden” and “process” emergence. Criticalrealism sees both instances as “variations on the same” and not as ontologicallyseparate; the latter arguably being Foucault’s position, for he sees the first(“sudden”) as the effect of a domination and, strictly speaking, should not see thesecond (“process”) as a form of emergence at all.

The paper begins by outlining Foucault’s theorization of the extra-discursive insome of his key earlier archaeological works. It then moves to show the change inemphasis found in his later work after adopting his genealogical methodology andwhy he incorporated Nietzsche’s particular understanding of “emergence” as animportant part in this new inquiry. The paper then examines how Foucault’s useof emergence does not always correspond to his own position and argues that thisinconsistency is linked to his under-theorization of the extra-discursive. Afterdeveloping a brief outline of critical realism and critical realist accounts of com-positional emergence (that offer a different perspective of both “reality” and“emergent” entities), the paper concludes with a re-examination of a key argu-ment in Discipline and Punish—the production of “delinquency”. This is undertakenusing an integrated approach with critical realism bolstering Foucault’s alreadyrich account. It is worth stating explicitly at this point that it is not the intent of thispaper to propose that a critical realist position can be simply “grafted” ontoFoucault’s work and then seamlessly integrated. This paper does, however, seek todemonstrate that there exists a strong basis for a reconceptualization of theplacement and use of emergence and the extra-discursive within Foucault’s oeuvre.

FOUCAULT’S EXTRA-DISCURSIVE

Three questions quickly become apparent when investigating the extra-discursivein Foucault’s oeuvre. The first two: what and where is the extra-discursive? And thethird: how does Foucault integrate it into his theory? None are easy to answer forseveral reasons. First, Foucault (naturally) changed his ontological position overthe course of the 30 years of his writing. Second, his theory developed throughmultiple inquiries (madness, illness, knowledge, punishment, sexuality, etc.) butwas hardly ever articulated in detailed exposition. And finally, as we shall see, hebound himself (albeit inconsistently) to a form of Kantian epistemic fallacy (Datta,2007: 282–3) that hinders his theoretical ability to adequately conceptualize theextra-discursive. This paper focuses on two key texts (supported by others) thatelaborate the extra-discursive in the most depth: The Archaeology of Knowledge (1969/1972; hereafter AK) (along with examples from Foucault’s History of Madness andThe Birth of the Clinic) and The Confession of the Flesh (1977a/1980; hereafter CF).

70 Nick Hardy


The Archaeology of Knowledge

A post hoc elaboration and consolidation of Foucault’s theoretical and method-ological positions from his earlier works, AK contains key refinements of his earlyformulation of the extra-discursive. Instead of arguing for a clean separation of thediscursive and the extra-discursive, in AK he incorporates the extra-discursive intosome discourses. Key to Foucault’s argument is his concept of discursive formationsand their associated rules of formation. These he describes as:

Whenever one can describe between a number of statements, such a system of dispersion,whenever, between objects, type of statement, concepts, or thematic choices, one can definea regularity (an order, correlations, positions and functionings, transformations), we willsay . . . that we are dealing with a discursive formation. . . . The conditions to which the elements ofthis division (objects, modes of statement, concepts, thematic choices) are subjected to we shallcall the rules of formation. The rules of formation are conditions of existence (but also of coexist-ence, maintenance, modification and disappearance) in a given discursive division (1969/1972:38; emphasis in original).

A discursive formation can be understood as the particular ordering of the fourelements that constitute it: things (objects), things said (types of statement), ideas(concepts) and groupings (thematic choices). Each discursive formation is differentbecause each one has particular rules of formation, dictating what can and cannotbe said, whom or what has the ability to speak, what is determined/identified toexist or not exist, and what can or cannot change within its boundaries (e.g. therules of formation for cardiology as a discursive formation does not allow for the“soul” as an object; Christian discursive formations, however, do). Foucault’sargument here is that not only is there a framework that structures what is and isnot “allowed” to take place “within it”, but that also there are elements of theextra-discursive actively engaged as part of this process; indeed, it appears that thediscursive formation would not be able to operate in the same form without thoseincorporated extra-discursive elements. This becomes much clearer when hediscusses the various components of discursive formations in more detail.

Objects emerge through a combination of “surfaces of emergence” (eg. thefamily, workplace, prisons, etc.), “authorities of delimitation” (experts such asdoctors or priests), and “grids of specification” (where objects are typologized andrelated to one another, e.g. hysteria, melancholia, etc.) (ibid.: 41–42). Enunciativemodalities govern “what can be said” and are structured by “who is speaking”, “theposition of the subject” and “the institutional sites from where the doctor makeshis discourse” (ibid.: 50–52). Concepts are structured through “forms of succession”(the processes by which statements (as part of concepts) follow and are followed byother statements), “forms of coexistence” (how similar statements are placedtogether), and “procedures of intervention” (where some statements are judged tobe “continuous” whilst others are “discontinuous”) (ibid.: 56–59). Strategies, finally,are particularly interesting as they relate to fissures—sometimes intentional,



sometimes not—generated within a discursive formation by the manoeuvring ofdifferent factions.1 Strategies are constrained by “points of diffraction of dis-course” (aspects of contention or similarity), “the economy of the discursiveconstellation” (the particular “rules and customs” of a particular sub-set of knowl-edge), and “the function . . . [the discourse] must carry out in the field of non-discursive practices” (e.g. psychiatry enacted upon “the mad”) (ibid.: 65–68; emphasisadded). Thus the pre-existing extra-discursive becomes co-opted into discursivitiesand, when a discourse becomes powerful enough, the extra-discursive then beginsto be reformed by that discursivity.

It is already clear that discursive formations are not static entities (indeed, theconcept of “strategies” could be seen as something of a forerunner to the Entstehungidea of “egoistic splintering”, as will be discussed below). But added to this,Foucault argues that discursive formations exist in a way that allows them tochange their “position” or “alignment”. Discursive formations can manifest as ageneralized knowledge (savoir) or as one of the multiple “operationalized” specificknowledges (connaissances) that “orbit” around the savoir. Both savoirs and connais-sances continuously operate to (re)structure the “objects” within their authority,often in highly particularized ways. Foucault’s discussion of the savoir of madnessin the 18th century serves as an example: an earlier savoir of madness containedparticular connaissances—such as the psychopathological, jurisprudence, policeregulation, etc.—that are now absent from the contemporary savoir of madness(ibid.: 184–185) (see also Foucault, 1963/1994: 196 on disease moving from “evil”to “nature/death”). The central savoir discursive formation had a series of associ-ated connaissance discursive formations that changed over time, as well as havingthose connaissances incorporate elements of the particular savoir (e.g. the move of“medicine” from a religious to a scientific savoir and the corresponding change inits methods, concepts and objects).

In arguing this position, Foucault is claiming that there is structure to dis-course(s)2 and that it is through the various “rules” of discursive formations thatthe world is understood (and, to some extent, also shaped). The extra-discursive,then, forms not only key elements within discourses (objects, entities, etc.), but alsothe external structures that discourse applies itself through (e.g. the pre-existingsocial institutions that become “surfaces of emergence” for discursive objects)(Dupont and Pearce, 2001: 145; Pearce and Woodiwiss, 2001). Foucault’s twoearliest books History of Madness (1961/2006) and The Birth of the Clinic (1963/1994)both theorize the complex interaction between the discursive and the extra-discursive in this manner.

In Birth of the Clinic, Foucault presents a case that the clinic was formed to aid theteaching of medicine (1963/1994: 61), being a site of specimen collation: “[i]n theclinic . . . one is dealing with diseases that happen to be afflicting this or that patient:what is present is the disease itself, in the body that it appropriate to it, which is notthat of the patient, but that of [the diseases’] truth” (ibid.: 59). Indeed, this appearsto be an instance of the extra-discursive being specifically organized according to a

72 Nick Hardy


particular discursive understanding (ibid.: 58). But this organization was not simplya “one way” discursive ordering: the space was organized according to the(extra-discursive) reality of disease and patients, but the clinic was “probably thefirst attempt to order a science on the exercise and decisions of the [medical] gaze”(ibid.: 89)—i.e. explicitly around the requirements of the physicality of the doctorsthemselves. But the medical gaze itself had changed because now it was:

no longer the gaze of any observer, but that of a doctor supported and justified by an institution,that of a doctor endowed with the power of decision and intervention. Moreover, it was a gaze thatwas not bound by the narrow grid of structure (form, arrangement, number, size), but that couldand should grasp colours, variations, tiny anomalies, always receptive to the deviant (ibid.: 89).

New objects were to present themselves to the medical gaze in the sense that, and at the same time as, the knowingsubject reorganizes himself, changes himself, and begins to function in a new way. It was not, therefore, theconception of disease that changed first and later the way in which it was recognized; nor was itthe signaletic system that was changed first and then the theory; but together, and at a deeper level,the relation between the disease and this gaze to which it offers itself and which at the same time itconstitutes (ibid.: 90; emphasis added).

It is a construction of the “space” that disease could manifest itself within (thediscursive spur for the alteration of extra-discursive place); but that constructionthen reflected back upon the discursive, altering it too. Foucault here outlines aseries of complex interactions between discursive and extra-discursive: “the clinicis not, therefore, that mythical landscape in which diseases appear of their ownaccord, completely revealed; [the clinic] makes possible the integration, in experience, ofthe hospital modification of constant form. . . . By means of the endless play of modifi-cations and repetitions, the hospital clinic makes possible . . . the setting aside ofthe extrinsic” (ibid.: 110; emphasis added). What was problematic to medicine—and what it latently found beneficial after the formation of the clinic—was pre-cisely that the extra-discursive did not conform to the dictates of the discursive,meaning that a space such as the clinic suddenly enabled new “reorganizations”of the “knowing subject”.

Foucault’s magisterial History of Madness (1961/2006) offers similar insights.After the “great confinement” of 1656 where 1 in 100 Parisians were to suddenlyfind themselves deemed “Unreasoned” and thus requiring incarceration, the madbecame “lost” in the prison system. They were only “found” again 135 years later,a few years before the French Revolution:

. . . The consciousness of madness evolved during the course of the eighteenth century. It did notevolve as part of a humanitarian movement that slowly forced the human reality of the mad tobe acknowledged. . . . Nor was it the result of a pressure of scientific need that forced it to bemore attentive to what madness actually had to say about itself. . . . If the mad were progres-sively isolated, and the monotony of insanity was divided into rudimentary species, that was notthanks to medical progress or any humanitarian approach. The phenomenon was born inside confine-ment, and it is inside confinement that the keys to this new consciousness of madness are to be found (ibid: 397;emphasis added).



The “discovery” of madness in the prison and hospital system (ibid.: 417–418) isexplicitly linked, in Foucault’s work, to the extra-discursive reality of confinement.The isolation of the mad—i.e. their continued incarceration while other groups(beggars, vagabonds, petty criminals, etc.), with whom they were originally con-fined, were subsequently moved out or into other spaces such as workhouses—contributed to the alteration and construction of new discursive formations relating tomadness.

But there is also another element to the extra-discursive and madness: the veryact of confinement in 1656 was only possible because of the pre-existence of the (thenalmost totally empty) leper houses. Leprosy was dealt with by exclusion, but whenleprosy was seen to be diminishing “the lowly spaces set aside for it, together withthe rituals that had grown up . . . to keep it at a . . . distance, suddenly had nopurpose” (ibid.: 5). But, importantly, “[o]nce leprosy had gone, and the figure ofthe leper was no more than a distant memory, these structures [the leper houses]still remained” (ibid.: 6). No discourse develops against a tabula rasa backdrop; alldiscourses form and develop in situ, influenced, structured and constrained by thecontext in which they exist. The confinement of the mad took place and was onlypossible because of the (contingent) pre-existence of the empty leper houses.

The Confession of the Flesh

The exactitude of the analysis which Foucault developed in AK regarding theextra-discursive is arguably lost as he continued his writing and the developmentof his genealogical accounts post-1971 (Dupont and Pearce, 2001: 133–134). Butwith this loss comes a more dynamic interpretation of the discursive/extra-discursive interaction. Foucault’s concept of dispositif [“apparatus” or “social appa-ratus”] is an example of this, centring around the idea that, upon obtaining adominant position, a particular group constructs and then maintains a particulardiscursive and social position that is to the detriment of others. From this position,they are able to (partially, but continually) influence both the formation of thediscursive and the extra-discursive—and they also gain the ability to “react to”unforeseen circumstances. In the Confession of the Flesh interview conducted in1977, Foucault describes a dispositif as:

. . . [First], a thoroughly heterogeneous ensemble consisting of discourses, institutions, architec-tural forms, regulatory decisions, laws, administrative measures, scientific statements, philo-sophical, moral and philanthropic propositions—in short the said, as much as the unsaid. Suchare the elements of the apparatus. The apparatus itself is the system of relations that can beestablished between these elements [NH]. Secondly, what I am trying to identify in thisapparatus is precisely the nature of the connection that can exist between these heterogeneouselements. . . . In short, between these elements, whether discursive or non-discursive, there is asort of interplay of shifts of position and modifications of function which can also vary verywidely. Thirdly, I understand by the term ‘apparatus’ a sort of . . . formation which has as itsmajor function at a given historical moment that of responding to an urgent need. The apparatus

74 Nick Hardy


thus has a dominant strategic function (Foucault, 1977a/1980: 194–195; emphasis in original,except “[NH]”).

A dispositif is both a “re-wiring” of existing meaning (e.g. what it means to be“Unreasoned” (Foucault, 1961/2006: 102); what it means to be a “worker”(Foucault, 1973/2000: 82)) and the “imposition” of new meanings (i.e. the dis-cursive connections) between a range of heterogeneous elements (e.g. concepts,objects and institutions). The effect of this is twofold: initially, it changes theunderstanding and perception of the world from one perspective to another.Secondly, it gives an ability to “shape the social” by directing and moulding socialconduct according to new parameters established by the dispositifs particular(re)configuration of relations.3 While Foucault remained largely silent on aspectssuch as the generation and formation, the installation and removal, and the meansof maintaining dispositifs, he did elaborate on what a dispositif “does”. Arguing thata dispositif has two important effects, Foucault linked them heavily to his concep-tion of strategy and strategic manoeuvring:

On the one hand, there is a process of functional overdetermination, because each effect—positive ornegative, intentional or unintentional—enters into resonance or contradiction with the othersand therefore calls for a readjustment or a reworking of the heterogeneous elements that surfaceat various points. On the other hand, there is a perpetual process of strategic elaboration (ibid.: 195;emphasis in original).

[T]he apparatus is essentially of a strategic nature, which means assuming that it is a matter of acertain manipulation of relations of forces, either developing them in particular direction,blocking them, stabilising them, utilising them, etc. The apparatus is thus always inscribed in aplay of power, but it is also always linked to certain coordinates of knowledge which issue fromit but, to an equal degree, condition it. This is what the apparatus consists in: strategies ofrelations of forces supporting, and supported by, types of knowledge. . . . [T]he apparatus in itsgeneral form is both discursive and non-discursive, its elements being much more heterogeneous(ibid.: 196–197; emphasis in original).

Strategically, then, a dispositif can be seen as a kind of “management system”operating to enable relations of forces and particular knowledge (i.e. discursive)positions to mutually reinforce one another. This explicit link to strategy is,arguably, because Foucault saw dispositifs as being the mechanism through whicha dominant position could be maintained—a position that includes elevating andenabling particular discourses to “explain” (and therefore define) specific occur-rences, instances or objects. The dispositif operates as the means by which aparticular discursive position is able to “access” relations of forces, allowing it toimpose a particular discursive definition upon the extra-discursive—and, as far aspossible, upon other discursivities as well. But, importantly for the discussion inthis paper, the extra-discursive is not a passive, malleable object waiting to be“defined” by a particular discursive position: it continually “bites back”, as dem-onstrated by the multitude of unexpected events and outcomes that constantly



occur. These unexpected outcomes mean that a dispositif must be continuallyrepaired and/or modified in order to maintain particular relations betweenknowledge(s) and forces (ibid.: 195; see also Deleuze, 1988/1992; Datta, 2008).Indeed, it operates to enable particular programmes and their related technolo-gies (Foucault, 1975/1991; 1979/2003; 1982a/2003; Defert, 1991: 214–215) tobe operationalized, but also to tactically react to unexpected events (Foucault, CF:195–196; 1975/1991: 279–280; 1982b/2003: 142).

But there is also a telling phrase Foucault used regarding the apparatus: “amatter of a certain manipulation of relations of forces” (CF: 196). This is not onlya key component of his understanding of genealogy and emergence (as discussedfurther below), but “relations of forces” are also key in his understanding ofpower. Foucault developed his concept of power over the course of several yearsdefining it as “relations of force” (1976a/1990: 93; 1997/2003: 15). The par-ticular forms that these relations of force take is specific to both their particularmilieu (e.g. family, school, workplace, etc.) and related to the time period inwhich they are produced (e.g. sovereign, disciplinary, bio-political and govern-mental).4 The effect of power is to “act upon [subjects’] actions: an action uponan action, on possible or actual future or present actions” (1982b/2003: 137)—i.e. the present or possible “future” actions of a subject are created and curtailedto varying extents (as well as operating to “shepherd” subjects into new socialcontexts).

Subjects exist in a social and physical world that contains various forms andoperations of power relations, but there are particular sites that operate as con-centrated “blocks” where power relations “constitute regulated and concertedsystems” (1982b/2003: 136). Prisons are examples of such “blocks”—Foucaultterming them “an instrument and vector of power” (1975/1991: 30)—and it is inplaces such as these that subjects are subjected to multiple power relations, allattempting to produce particular outcomes. The resultant subject is (re)shapedand (re)formed out of this “political field”, through these “political technolog[ies]of the body” (1975/1991: 25–26). Dispositifs, in this account, play an importantrole in both creating, consolidating and aiding particular blocks as well as supportingthe particular “relations of force” that enable the reproduction of a particulardominant group at a particular time (CF: 203). A dispositif is as close as Foucaultcomes to articulating the form and presence of a “meta-structure” present insociety (1976a/1990: 93, 94).

But a question remains: how can Foucault have made the jump from under-standing the extra-discursive in the detailed analytic offered in AK to the dynamicdescription offered in CF? The crux of Foucault’s reasoning can be found in hisargument for the adoption of a “genealogical” methodology. As the paper willnow highlight, Foucault argues for a method that draws out the disruptive form ofhistorical change—and one that also includes a prominent position for a conceptof “emergence”. However, Foucault’s ability to maintain an interlinkage betweenthe discursive and the extra-discursive is disputed.

76 Nick Hardy


FOUCAULT, GENEALOGY AND ENTSTEHUNG

Foucault’s adoption of Nietzsche’s genealogical method, as Rajchman (1985: 115,117) terms it, “turns the problem of knowledge into a problem of power. . . . Fou-cault’s turn to genealogy was thus his attempt to analyze the connections betweenbodies of knowledge and techniques of domination.” If archaeology sites subjectsin historical contexts (ibid.: 113) (i.e. there is no “core” or “essence” to subjects,they are formed within the epistemes and discursive formations—God, Reason,labour, etc.—that exist around them), then genealogy is Foucault’s attempt toexplain how these truth constructs change; how dominations “produce” forma-tions. Foucault’s line of attack reconceptualizes the problem in terms of “truth”:“[t]he political question, . . . is not error, illusion, alienated consciousness, orideology; it is truth itself. Hence the importance of Nietzsche” (Foucault, 1976b/2003: 318; see also Weir, 2008).

In the essay Nietzsche, Genealogy, History (1971/2003, hereafter NGH) Foucaultoutlines his adoption of genealogy5—and his use of “emergence”—in the mostdetail. Foucault introduces three terms that he takes from Nietzsche: Ursprung[origin], Herkunft [descent] and Entstehung, which “designates emergence, themoment of arising” (NGH: 357; emphasis in original). Foucault argues that “tra-ditional” accounts of history are either forms of teleological essentialism (ibid.:363) or are metaphysical ideals of continuity (ibid.: 361). What is needed instead,Foucault argues, is a “wirkliche Historie” (ibid.: 259) that:

leaves nothing around the self, deprives the self of the reassuring stability of life and nature, andit will not permit itself to be transported by a voiceless obstinacy towards a millennial ending. Itwill uproot its traditional foundations and relentlessly disrupt its pretend continuity. This isbecause knowledge is not made for understanding; it is made for cutting (ibid.: 361).

With the Nietzschean dramaticism put to one side, Foucault uses Ursprung, Herkunftand Entstehung to destabilize key elements of the “traditional historical” sense. Afterproblematizing the notion of “origins” in history or historical groups and thencriticizing the idea of “descent” or lineage in thought, population or society,Foucault uses “emergence” to interrupt the ideal of continuity. Foucault’s concep-tion and use of Entstehung is “the moment of arising,” meaning “[e]mergence is thusthe entry of forces; it is their eruption, the leap from the wings to centre stage” andthat they “result from substitutions, displacements, disguised conquests, and sys-temic reversals” (ibid.: 357, 358, 359). Entstehung is the unseen “happening” (i.e. notjust the “presence”) of opposing forces conflicting against one another; it is:

a place of confrontation but not as a closed field offering the spectacle of struggle amongequals. . . . [I]t is a “non-place”, a pure distance, which indicates that the adversaries do notbelong to a common space. Consequently no one is responsible for an emergence; . . . since it alwaysoccurs in the interstice. In a sense, only a single drama is ever staged in this “non-place”, theendlessly repeated play of dominations. The domination of certain men over others . . . (ibid.:358; emphasis added).



For Foucault, then, emergence explicitly results in the imposition of a new domi-nation (although not necessarily a domination over a defeated opponent) and isalways “produced in a particular state of forces” (ibid.: 357; cf. Foucault, 1969/1972: 45; Bhaskar, 1978/2008: 51). One of the tasks of genealogy, therefore, is todetermine what these forces were at a given time. There are three particular“states of forces” that Foucault sees as promoting Entstehung: the first is where agroup (or “species” as Foucault terms it) is victorious in fighting against outsidersor to keep elements within itself suppressed; the second is where, with no outsideforces, there is a struggle of “egoisms” within a group so that it (in a sense) devoursitself through a “splintering of forces”; the third is, again with no outside forces,where the strongest element(s) within a group splits away from the weaker (Fou-cault, 1971/2003: 357).6 Here Foucault is specifically adopting and then bindinghimself to a Nietzschean conception of emergence. It is a position that highlightsthe struggle between different protagonists in order to dominate—and thereforeto define, shape and control—their particular part of the world. Genealogy maybe the investigation of the circumstances leading up to the emergent moment ofa new domination, but the moment itself always occurs in the “pure space of thenon-place”.

But is this really the genealogy that is “gray [sic], meticulous and patientlydocumentary” (ibid.: 351)? Why does Foucault limit himself to only calling “domi-nations” emergent? Asking this question poses two problems for Foucault’s posi-tion: the first is with the “object” of the inquiry itself and the second is with the“parameters” of the inquiry. In arguing that instances of emergence can onlyoccur as a “domination”, Foucault is effectively restricting the scope of emergenceonly to the effect(s) of human contestations—i.e. to those “things” which are capableof imposing Nietzschean “dominations”. This immediately rejects other instancesthat may, in fact, be emergent but are not a domination. The second problem isthat genealogy seemingly restricts avenues of inquiry into determining only the“state of forces” (ibid.: 357) that (dis)enable or (dis)allow a particular dominationto take place. This effectively excludes other “simple” social relations that maynecessarily be present, but which may not be determined as an active “force” inthe sense that Foucault uses it to explain the forces involved in a moment ofEntstehung.

An interesting development during the course of Foucault’s writing was that heseemingly did adopt a somewhat “softer” conception of genealogy and emergencein some of his writings after NGH. In Territory, Security, Population (2004/2007),Foucault develops an account of the “emergence” of the concept of “population”(ibid.: 21–22, 70–74) and argues for its centrality to the production of other newconcepts within existing “domains of knowledge” [savoirs] (ibid.: 76). Politicaleconomy, for instance, is an off-shoot “domain of knowledge” from an originalseventeenth century analysis of finances (i.e. the simple analysis of wealth): it tookthe notion of “population” to create a new “subject-object” of study and then anew domain of knowledge to study it (ibid.: 76–77). But this did not happen by

78 Nick Hardy


some accidental turn of thought or slip of the quill pen. In an earlier discussion ofan eighteenth century issue that was linked to “population” (the issue of grainscarcity), Foucault argues that:

You could read the principle of the free circulation of grain as the consequence of a theoreticalfield and also as an episode in the mutation of technologies of power and an episode in theinstalment of the technique of apparatuses of security [i.e. ‘governmentality’ NH] . . . (ibid.: 34;emphasis added).

Indeed, Foucault links this process to genealogy explicitly:

Instead of considering [scarcity] in terms of an archaeology of knowledge, I would like toconsider it from the perspective of a genealogy of technologies of power. I think we couldreconstruct the function of the text[s concerning scarcity], not according to the rules of forma-tion of its concepts, but according to its objectives, the strategies that govern it, and the programof political action it proposes (ibid.: 36).

The emergence of “population” as a concept does not, seemingly, correspond tohis earlier notion of Entstehung as the “interstice”, the “non-place” or a “leap fromthe wings to centre stage” (NGH: 358). Instead, “population” appears to be aconcept that developed through the various reflections and reverberationsbetween the discursive and extra-discursive: i.e. it was created from the “constantinterplay between the techniques of power and their object [which] graduallycarve[d] out in reality, as a field of reality, population and its specific phenomena”(Foucault, 2004/2007: 79).

Foucault is seemingly adopting a form of emergence, then, that can be firmlylinked to process. But it is here that a contradiction in Foucault’s position becomesapparent: if emergence is meant only to be the dominations of Entstehung, thenhow can he correlate this with his own investigative work that charts instances of“gradual” emergence? Furthermore, if Foucault himself finds it necessary todepart from the methodology he outlined in NGH, does this indicate a tacitrecognition that an integrative approach is required for a “process emergence”argument to work? “Process emergence” may be part of an overdetermined7

process—in that there is something akin to an influencing, directive element (e.g.the strategies that drive the “techniques of power”) affecting the others—but thereis also certainly a series of interconnections between the discursive and theextra-discursive, creating various promotions or hindrances to the establishmentof a concept.

CRITIQUING AND DEVELOPING FOUCAULT’S QUIET RELIANCEON THE EXTRA-DISCURSIVE

The crux of the problem facing Foucault is his inability to give the same analyticweight to the extra-discursive in the genealogical accounts as he did in his earlier



archaeologies. However, this is not just a methodological oversight: it stems froma much larger ontological problem. The problem originates in what Lecourt(1972/1975) argues is the tension (beginning in Foucault’s archaeological work)between knowledge (of the extra-discursive) and the production of the intangible,meta-level “archive” (and the savoirs that the archive encompasses). Knowledge ofthe extra-discursive is structured, argues Lecourt, by this already-existing archive,which, in turn, gives rise to the discursive formations that then structure evenfurther the understanding of the subjects within them: “the task of the “archae-ology” is in fact to constitute the theory of the “discursive” instance insofar asit is structured by relations invested in institutions and historically determinateregulations” (ibid.: 198). Another example is Foucault’s argument that dis-cursive connections are, in fact, subordinate to “what we might call “primary”relations, . . . which independently of all discourse or all object[s] of discourse, maybe described between institutions, techniques, social forms, etc.” (Foucault, 1969/1972: 45; also in Lecourt, 1972/1975: 199). Lecourt argues that it is regularity whichFoucault highlights as important in discourse—a “regularity” which is an effect ofconfining and systematizing structures (Lecourt, 1972/1975: 201–202). Theimportance of Foucault’s work is that it offers a very strong explanation forthe pre-formation of knowledge, thereby reducing the primacy of human subjects asthe unfettered agents of social change. Lecourt’s conclusion is that Foucault doesnot take his argument far enough in acknowledging how much he (Foucault) reliesupon extra-discursive elements in order to maintain discursive formations.

Datta (2007: 282–283) takes this critique further, drawing a parallel betweenBhaskar’s argument for Kant’s epistemic fallacy8 and a similar position thatFoucault adopts. Foucault, Datta argues, is unable to accurately theorize theextra-discursive precisely because he is trapped by the logic of his own position:the discursive formations presented in AK, or “networks of power” (e.g. Foucault,1976a/1990; 1977b/1980; 1982b/2003), all produce “knowledge of the world”that is necessarily constructed by pre-existing discursive structures. This effectivelyexcludes Foucault from theorizing the extra-discursive because, without a sepa-rate theoretical space from which to access the extra-discursive, Foucault himselfcannot claim to have knowledge “outside of” his own discursive position. It alsoundermines his ability to conceptualize discursive formations (Dreyfus andRabinow, 1982: 99), dispositifs or even the “simplest” of power relations. Thechallenge that Datta outlines for Foucaultian theory is that it necessarily requiressome conception of the extra-discursive with which to work.

The outcome of these critiques, then, positions Foucault’s work at an impasse:effectively he is relying on an important—but untheorized—relation between thediscursive and the extra-discursive where, in his argument for the genealogicaloperation of discursive relations, each continually exerts an influence upon theother. The dynamism of the genealogical accounts is precisely the “creativetension” between the discursive and the extra-discursive. But an effect of this isthat emergence, for Foucault, cannot be analyzed as an extra-discursive element

80 Nick Hardy


precisely because it is, in effect, “unrealizable” in Foucaultian theory as an“object” of inquiry. This means that he is restricted not only on what can countas “emergent”, but also what he can articulate as taking part in that emergent“event”—and, possibly, why it was understood to always happen in the “non-place” (NGH: 357).

Critical Realism, Compositional Emergence and Causation

One route to reconcile the difficulties faced by Foucault can be found through thearguments presented by critical realism. Initially developed by Roy Bhaskar(1978/2008), critical realism argues that the natural world is stratified into threedistinct ontological levels: the real, the actual, and the empirical. The real containsthe multitude of intransitive, natural, generative mechanisms, which are, accordingto Bhaskar, “nothing other than the ways of acting of things” (1978/2008: 14).The actual is where various conjunctive events occur; viz. when mechanisms comeinto conflict or congruence with one another and so produce varied effects. Theempirical consists of the experiential world (and manifests only a fraction of themechanisms that are in operation in each particular event). It is in the empiricalthat scientific experimentation is undertaken that can, under the appropriateconditions, produce fallible knowledge (i.e. non-permanent knowledge that is open toboth revision and refutation) of experiences, events and mechanisms. This avoid-ance of the relativist “sceptical abyss” is achieved by acknowledging the ability ofhumans to gain some understanding of certain aspects of the external world,whilst still recognizing the potential fallibility of that knowledge—a clear enoughposition because the ontologically distinct external world does not rely on any formof human understanding in order to continue to exist and operate (Frauley andPearce, 2007: 4–5).

However, scientific knowledge is a product of what Bhaskar terms “closedsystems”; a concept he elaborates when circumscribing the applicability of KarlPopper’s “falsification” thesis. Popper’s position only works, argues Bhaskar(1978/2008: 128), within the context of closed systems—i.e. where only thosemechanisms contained within a system have the potential to act, and are to theexclusion of others (ibid.: 53). By contrast, open systems, such as social relations, arethe exact opposite of closed systems as they contain a huge number of causaltendencies that are intermittently and unpredictably active (and therefore ineffect), creating an unknown number of events that are unable to be either isolatedor excluded in order to produce reliable predictions. This is an explicit rejectionof the Empiricist approach which states that only what can be demonstratedthrough experimentation is “knowledge”—i.e. realism argues there are mecha-nisms that can be postulated as being in effect, even if they cannot (at a particulargiven point in time) be empirically proven. This acknowledgment of the multi-level complexity of the natural world by critical realism explicitly “brings theoryback in” as a necessary component part of any complex explanation.9



Emergence: Formation, Maintenance and Causal Effects

The concept of emergence used in critical realism is different from Foucault’s.While Bhaskar himself did not use emergence in an entirely consistent way(Kaidesoja, 2009),10 a more rigorous version of emergence is outlined and devel-oped by Elder-Vass (2005), one that he terms “compositional emergence”. UsingCunningham’s (2001) different conceptions of emergence found in philosophicaland scientific work, it is clear that all conceptions of emergence share a commonposition: that to be “emergent” an entity must have at least two elements that existin some form of relation to one another in order to produce the emergent entity.Distinctions and disagreements begin to develop when questions of whetherspecific structured relations between elements are necessary for emergent propertiesto exist, and/or whether an entity requires already complex internal elements to formemergent properties (Cunningham, 2001: S68-S70). Elder-Vass’ work distin-guishes compositional (or “synchronic”) emergence that structures relations betweenan entity’s constitutive elements. This particular structuring gives (“diachronic”)rise to properties that are emergent because they are causal powers not already held bythe entity’s constituent elements.11

Two important corollaries are produced by this conception of emergence: (1)the morphogenesis and morphostasis of the emergent entity; and (2) the irreduc-ibility of emergent properties to be explained by their constituent parts. The twoprinciples of morphogenesis and morphostasis were developed initially by Buckley(1967: 58–66) and then greatly expanded along critical realist lines by Archer(1995). Morphogenesis is the principle that certain forces are necessarily requiredfor an entity to form. It is not enough to have the constitutive elements of an entitysimply “together”, it is also necessary to have a particular set of forces present thatform the elements into a particular relationship. Morphostasis, by contrast, details thefactors necessary for an entity’s constituent elements to maintain their morphoge-netic (and therefore synchronic) relations between each other (Elder-Vass, 2005:324). These principles require any explanatory account to detail both the formationof an emergent entity as well as how and why the entity endured over a given periodof time (long or short).

Emergent entities, by definition, have properties that are not “explainable” bytheir constituent elements. However, two mistakes can occur when analyzing anemergent entity: “explanatory” and “eliminative” reductions. Explanatory reductionreduces an analysis to a discussion of the various elements found within the entityitself. Whilst it is a valid argument to state that each constitutive element of anentity is necessary for its formation or that these elements must be structured in aparticular way (for if the elements were not there in that particular configurationof relations, how could the entity form?), the emergent property is not explainableby a simple appeal to the various elements that individually constitute the emer-gent property holding entity. As Elder-Vass states (2005: 322; 2007b: 163–164),“eliminative reduction” or “level abstracted” arguments are unhelpful when

82 Nick Hardy


trying to explain complex entities and their causal power(s). Eliminative reduction,however, takes the explanatory reduction argument one step further: it reduces anentity to its smallest part, explaining higher-level operations by an appeal to the“foundational” element(s) of that entity. This completely ignores the (various)levels of synchronic relations within an entity that (each) give rise to diachronic,and therefore “emergent”, properties. The opposite, but corollary, mistake is levelabstraction: here an entity is understood as a simple whole, its complex structuringof constitutive elements either overlooked or ignored.12 In order to provide anaccount of an event, therefore, it requires a downwardly inclusive view detailing the“internal stratification” (Elder-Vass, 2007b: 164) and relationships of an entity’scomponent elements (but not reducing an entity to these elements).

The concept of compositional emergence is important because it offers anaccount of how complex relations between elements can give rise to entities withproperties that are not originally held by their constitutive elements. Furthermore,a key aspect of this is that emergent entities are not determined by their constitutingelements, instead they are only constrained by them.

Real and Actual Causation

With emergent entities possessing causal powers, an important distinction withincritical realism regarding “causation” itself needs to be highlighted, as emergententities can potentially blur this distinction. Bhaskar’s original position was thatemergence occurred when “the operations of the higher level [entity] cannot beaccounted for solely by the laws governing the lower-order level in which . . . thehigher-order level is ‘rooted’ and from which . . . it was ‘emergent’,” and thismeant “[i]n short, emergence is an irreducible feature of our world, i.e. it has anontologically irreducible character” (Bhaskar, 1978/2008: 113). However, thiscomes into conflict with the distinction made within critical realism between“real” and “actual” causation (Elder-Vass, 2007b: 168–174), as it raises theimportant issue of what “type” of causal properties an entity possesses.

Real causation refers to the effects of “mechanisms” (sited in the “real”), whichare the “ways of acting of things,” leading Bhaskar to claim that the “worldconsists of mechanisms not events” (1978/2008: 14, 47). Bhaskar (ibid: 50) termsthese mechanisms “tendencies”, not “powers”, for mechanisms do not automaticallycause an effect because “[f]or whereas powers are potentialities which may beexercised, tendencies are potentialities which may [already] be exercised . . . or“in play” without being realized or manifest in any particular outcome” (ibid.).Actual causation, on the other hand, refers to the specific circumstances—thecontingent event, sited in the “actual”—under which the various causal tenden-cies of mechanisms interacted with one another to produce an outcome.13

The problem, then, for critical realism is that Bhaskar’s ontological siting ofmechanisms becomes “quite inappropriate for the discussion of what is happening



over multiple levels when we turn to causation at the level of individual instances”(Elder-Vass, 2007b: 168; emphasis added; see also Pearce, 2007: 50–53). Indi-vidual instances (“events”) are important because they are an interplay of differentmechanisms—but mechanisms can also be the product of complex, multi-levelentities that are reliant upon the morphostatic continuation of their particularsynchronic levels that produce diachronic properties (and are therefore emergentmechanisms). The level abstracted view used to explain real causation cannot takeinto account the structured complexity of actual causation precisely because, atleast in the case of entities, it is unable to conceptualize the particular levels ofsynchronic relations within each entity necessary to produce an event. It may not alwaysbe necessary to analyze using such a downwardly inclusive view, but an entityexists as a morphostatic pyramid of previous events all necessarily maintainingtheir particular synchronic composite relations in order to produce emergentproperties that allow for the entity to exist (Elder-Vass, 2007b: 170).

In this sense, Foucault’s studies offer a huge benefit to critical realism preciselybecause they investigate the complexity of compounded and/or focused socialrelations—i.e. the continuous “layering” of discourses and the extra-discursive.Entities can be seen to exist as a consequence of events in the actual and canthemselves possess causal powers. Understanding actual causation, therefore,means attributing causal powers to the entities that possess them and also account-ing for the morphogenetic and morphostatic forces that formed and maintainthese entities and the functioning of their causal powers.

REVISITING FOUCAULTIAN EMERGENCE: THE CASE OF PRISONSAND DELINQUENCY

A very interesting, but under-theorized, account of the complex interactions ofpower, the discursive, and the extra-discursive can be found in Discipline and Punish(1975/1991; hereafter DP). Here Foucault, tangentially at least, attempts toexplain the stubbornness of the extra-discursive to conform to discourse. Thecentral argument made by Foucault in DP is that punishment changed its formaway from a model based upon the retribution of a wronged sovereign to anattempt at an all encompassing system of oversight, direction and control.However, this attempt failed: while there was a move away from the norm ofretribution, the attempt at creating a norm of discipline did not manifest asenvisaged. “What replaced the public execution was not a massive enclosure, itwas a carefully articulated disciplinary mechanism—at least in principle” (ibid.: 264;emphasis added). Despite the strong standing of the savoir of discipline (i.e. thegeneral, or “meta-”, knowledge of disciplining), it ended up producing unexpectedeffects—and one of these was the creation, via the newly reconstituted prisonsystem, of a “delinquent” population.

Foucault argues that the prison should not be seen as developing through thecyclical stages of “prison, failure, reform”, but instead as constituted by:

84 Nick Hardy


A simultaneous system . . . superimposed on the juridical deprivation of liberty; a fourfold systemcomprising: the additional, disciplinary element of the prison . . . ; the production of an objectiv-ity, a technique, a penitentiary ‘rationality’ . . . ; the de facto reintroduction, if not actual increase,of a criminality that the prison ought to destroy . . . ; [and] the repetition of a ‘reform’ that isisomorphic, despite its ‘idealism’, with the disciplinary functioning of the prison . . . (ibid.: 271).

Prison has a form and function quite peculiar to itself, meaning delinquency wasthe direct result of the relations internal to the “disciplinary” prison:

[P]rison, apparently ‘failing’, does not miss its target; . . . it reaches it, insofar as it gives rise toone particular form of illegality in the midst of others, which it is able to isolate, to place in fulllight and to organize as a relatively enclosed, but penetrable, milieu. This form is, strictlyspeaking, delinquency (ibid.: 276–277).

And:

An entirely unforeseen effect which had nothing to do with any kind of strategic ruse on the partof some meta- or trans-historic subject conceiving or willing it. This effect was the constitutionof a delinquent milieu very different from the kind of seedbed illegalist practices and individualsfound in eighteenth-century society. What happened? The prison operated as a process offiltering, concentrating, professionalizing and circumscribing a criminal milieu (CF: 195–196).

Foucault’s argument revolves around the idea that prison inadvertently produceddelinquency because when disciplinary discourses were applied (in the form of“programmes”, operating through “technologies”) to the extra-discursive (i.e. thealready existing) world of institutions and individuals, they had unexpected effects.This, then, is not a case of delinquency being “created” through discourse alone(it was not the “aim” of that discourse; if it was, then it would imply that alldiscourses simply “produce” what they discuss), nor was it a simple redefinition orredesignation of an existing population as “delinquent”. Foucault’s position is thatthe particular constitution of the “disciplinary” programmes and technologiesemployed in prisons acted with the extra-discursive to produce an entirely new (andunforeseen) population.

Foucault’s stated position is that delinquency is “fabricated” by the prison system(DP: 278). But Foucault faces a problem with the fabrication explanation: it is verysimilar to a “level abstracted view” because it collapses multiple and complexprocess into one. This position cannot conceptually account for the ontologicaldepth of structures such as prisons because it does not engage with a key element:the extra-discursive. Foucault should be bound by his own theoretical positionthat the very social structures which enable a subject to become an “object ofknowledge” therefore also create a new form of that object (Datta, 2007: 280).However, it is clear that the social structures involved in the prison system createddelinquency independent from knowledge.

Potentially a much more beneficial position for Foucault is to engage with theconcept of “constitutively emergent” entities. One possible route to developing



this could be through Foucault’s own conception of “power” alongside an under-standing of the complex composition of emergent entities. On its own, powerseemingly relies too much on explaining its effects through its productive capaci-ties as a positive force, “creating” at least as much as it (negatively) “constrains”.However, if the concept of constitutive emergence is used alongside an under-standing of the extra-discursive, then a more convincing argument can be made.Unless one is willing to argue that the delinquents and criminals produced byprisons are ontologically distinct (an argument this paper is not making), anaccount must instead explain the difference between the “disciplinary” prison andthe previous form of “retributive” prison.

Understanding subjects not as formed but as continuously “imbricated” inpower relations and political fields, the perspective shifts from the subject as a“product” of power relations to becoming a “manifestation” of power relations.There is enough stability to the strongest of these (society wide) power relations(see Foucault on “domination”, 1982b/2003: 143–144) so that key aspects have acontinuous (constitutive) presence. In this sense, subjects can be said to never quiteobtain a fully morphostatic status: they continually require the presence of the(strongest) social and political relations that constitute them. However, the effectof these relations is not determinative—and this is where the “emergentist” explana-tion comes fully into operation: the subject is constituted by the particular powerrelations in the particular political and social fields in which they are present. But,crucially, subjects do not find themselves determined by these relations (restricted,most certainly, but not determined). From an emergentist perspective this isbecause subjects are not wholly bound by the “form” of their constitutive powerrelations precisely because they have capacities that are irreducible to their constitutiverelations—i.e. subjects are socially emergent entities.

The extra-discursive, here, intercedes and interrupts the operation of pro-grammes and technologies. It does this through both the “physical” (e.g. archi-tecture and buildings, natural processes, etc.) and the “inter-discursive” (i.e. otherold or alternative discourses) (Foucault, 1968/1991: 58): the extra-discursiveimpacts upon “current” discourse and upon the physical elements that this dis-course utilizes. The extra-discursive forms an important element that continuallyintroduces destabilizing effects that interfere with the political strategies thatdominant groups attempt to implement. The extra-discursive, therefore, operatesboth as an explicitly theorized factor “within” technologies and practices, but also“externally” to technologies and practices, an aspect which Foucault implicitlyadopts.

But the extra-discursive should not be understood as always being a disruptiveforce. Foucault directly links (DP: 279–280; CF: 195–196) aspects of the extra-discursive that were available to the bourgeoisie as the means by which theyabsorbed delinquency back into existing social relations. This was achieved by theirstrategic implementation of new and revised tactics—a key factor being theircontrol of the dominant dispositif. “Delinquency, solidified by a penal system

86 Nick Hardy


centred upon the prison, thus represents a diversion of illegality for the illicit circuits ofprofit and power of the dominant class” (DP: 280; emphasis added). Further:

one finds an immediate re-utilization of this unintended, negative effect within a new strategywhich came in some sense to occupy this empty space, or transform the negative into a positive.The delinquent milieu came to be re-utilized for diverse political and economic ends, . . . . Thisis what I call the strategic completion [remplissement] of the apparatus (CF: 196).

Foucault argues that controlling the dominant dispositif when delinquency was“created” gave the dominant class direct advantage precisely because it allowedthem to re-capture the unintended effects of the disciplinary prison (via, amongother things, a change in the role of the police to become an informationgathering—and therefore a “disciplinary”—system (DP: 213–215, 278, 280)). Thecreation of this “new type” of criminal population was potentially disastrous; notonly did they pose a direct challenge to the validity/applicability of disciplinarysavoirs, but they also challenged the social relations that relied upon those disci-plinary relations (i.e. the formation and utilization of a (largely) compliant work-force). Control of the dispositif, however, enabled the use of delinquents as strikebreakers, as informants, agents provocateurs, smugglers—and as a means of intimi-dating and controlling workers (DP: 279–280)—when otherwise they might havechallenged the dominant position held by the bourgeoisie.

CONCLUSION

This paper has sought to demonstrate that there is a productive relationshipbetween Foucaultian and critical realist theory. Critical realism can provide ameans of conceptualizing the external world that allows an escape from Foucault’sversion of the “epistemic fallacy”, enabling him to successfully theorize the extra-discursive. Introducing an ontologically deeper conceptualization of the extra-discursive would allow Foucault to generate a much stronger account of how theextra-discursive interacts with the discursive and of how they might resonate witheach other. However, Foucault’s present (in)ability to theorize the extra-discursiveis an obstacle to integrating his work but, if achieved, would enable deeperinvestigations that incorporate the sophisticated complexity of AK with the dyna-mism of DP.

A corollary of this position, however, is that critical realists must similarly payheed to the depth of the accounts presented in Foucault’s work. This papercautiously agrees with Day (2007: 117, 138) that “the way in which “postmodern-ist” theory is handled by most realists leaves much to be desired,” and this meansrejecting the idea that the insights “generated cannot be of use to those whoseinterests incline more toward what are commonly known as the natural and socialsciences.” Foucault’s oeuvre contains a wealth of potential routes for realist investi-gation—his post-structuralist accounts of madness, health, sexuality, discipline and



government are, arguably, a catalogue of what realists would term “events”. Theyshow the intimate interactions and connections that exist between the differentsocial collectivities that together constitute the totality of “society”.

The argument developed in this paper is an attempt to generate such a dia-logue. Foucault’s seeming de facto reliance upon a concept of compositional emer-gence was critiqued along two lines: first, that Foucault’s theoretical position inwhich knowledge is structured by pre-existing relations meant that he becametrapped within a self-referential loop (i.e. knowledge of a object is not of that“object” per se, but is instead only knowledge of a particular discourse that“explains” that object); second, that Foucault’s Nietzschean concept ofemergence-as-domination reduced its explanatory scope too far and was unableto account for more nuanced and gradual processes (that were not solely explain-able by his other accounts). If Foucaultian theory were to utilize something akinto a critical realist position that acknowledges the fallibility of knowledge and thecomplexity of causes, entities and open systems, it could begin moving to over-come some of these problems. The extra-discursive plays a fundamentally impor-tant role in Foucault’s theory—with Foucault himself refusing to make the choicebetween discourse or structure; instead he wove the two together, albeit with theextra-discursive forming only an intermittently visible thread. Breaking the abso-lutist notion that “knowledge of” a thing equates “construction of” a thing wouldexplain the unintended production of “delinquency”—but this requires Foucault-ian theory to fully acknowledge, and theorize, the extra-discursive.

Nick HardyDepartment of SociologyQueen’s UniversityMackintosh-Corry HallKingston, Ontario, K7L 3N6, [email protected]@googlemail.com

Acknowledgements. I would like to express my deep thanks to Frank Pearce, DaveElder-Vass, Paul Datta and Mira Bachvarova, without whose time, constructivecriticism and guidance this paper could not have been written. I would also like tothank the three anonymous reviewers for their time and constructive insights. Anearlier version of this paper was presented at the 2009 British Sociological Asso-ciation conference in Cardiff, Wales.

NOTES

1 It should be noted, however, that Foucault explicitly states that these strategies takeplace within a discourse and are “not rooted, anterior to discourse, in the silent depths of achoice that is both preliminary and fundamental” (Foucault, 1969/1972: 69).

88 Nick Hardy


2 See Politics and the Study of Discourse (Foucault, 1968/1991) for a more detailed discus-sion of the structure of discourse.

3 Foucault gives an example of a seventeenth century dispositif. French Physiocratsattempted to construct a new set of relations to govern the production and distribution ofgrain (Foucault, 2004/2007: 37). A contemporary example might be the rise of theChicago school “neo-liberal” economics: a discursive formation of “neo-liberal” economicswas constructed, successfully installed, and became a dispositif when it formed economic“orthodoxy” in Anglo-Saxon countries since the early 1980’s (Foucault, 2004/2008: 317–324).

4 Foucault should not be mistaken as arguing that there were different “power-epochs”that enforce corresponding changes in local-milieu power relations. Power relations alwayshave their own particular forms as power is in “local systems” because it “circulates”(Foucault, 1997/2003: 29, 34; 1976a/1990: 99–100).

5 Although the space available in this paper does not allow for a deep discussion withregard to “genealogy”, another source where Foucault gives a good indication of what heunderstands by genealogy is in Society Must Be Defended (Foucault, 1997/2003). Genealogy isabout “the insurrection of knowledges” and “an insurrection against the centralizingpower-effects” (in our society) of organized, unitary discourses such as the savoirs of the“sciences”; genealogy, therefore, “fight[s] the power-effects characteristic of any discoursethat is regarded as scientific” (ibid.: 9). To combat the sheer-edifice of unitary discourses (“aplace for everything and everything in its place”) they have to be disrupted by posing tothem not just counter-claims, but the alternative knowledges within that very “unitary”discourse that have been suppressed during its development and continued operation—“subjugated knowledges . . . : the knowledge of the psychiatrized, the patient, the nurse, thedoctor, that is parallel to, marginal to, medical knowledge” (ibid.: 7).

6 There is an interesting similarity between the “strategies” present in AK and discursiveformations and the “state of forces” promoting “splintering” in NGH.

7 See Althusser, 1965/2005: 100–104.8 Bhaskar articulates Kant’s epistemic fallacy as that “statements of being can always be

transposed into statements about our knowledge of being” (1978/2008: 16, 37)—i.e.knowledge of an object is assumed to accurately reflect that object. Datta applies avariation of this to criticize Foucault.

9 Important critical realist conceptions are the two terms “retroduction” (the process ofidentifying different mechanisms and entities involved in an event) and “retrodiction” (thedevelopment of an account detailing the specific interaction(s) of the identified elements)(Lawson, 1998: 164–165; Elder-Vass, 2007a: 472). Open systems make prediction impos-sible (hence Bhaskar’s use of the alternative term “explanation”, 1978/2008: 135–137),meaning the closest statements that could be made are instead “practical predictions ofcategorical form” (Bhaskar, 1978/2008: 135)—i.e. developing of retrodictive accounts ofprevious events and then (cautiously) applying these to possible future events. Retrodiction,therefore, is a method that investigates “empirically apprehendable phenomena” as“effects that are related in a complex way to underlying and sometimes obfuscated con-stitutive and sustaining processes” (Frauley and Pearce, 2007: 20). As such, retrodiction canbe used to develop accounts of relations within open systems—including where relationsmay operate as closed systems within broader open systems—and to determine the formthat events may take, e.g. cyclical, contingent, synchronic, diachronic, etc. Retrodictiveanalysis is important because it accounts for both entities and events, in order to under-stand the constituent elements in the development and maintenance of social entities aswell as the particular elements of an event.

10 Kaidesoja (2009) provides a very interesting discussion of the ambiguity with whichBhaskar develops “emergence”.



11 A good example of this is water. In the binding of the hydrogen and oxygen atoms,the resultant water molecule has emergent properties that are not contained by eitherhydrogen or oxygen. “One cannot, for example, put out a fire with oxygen and hydrogen”(Elder-Vass, 2005: 318).

12 Analyzing an “everyday object” like a car, for instance, as a complete object with theproperty of auto-motion would be a level abstracted view. In fact, a car is a series of complexrelations between a multitude of specific parts—e.g. the engine, axle, gears, chassis, etc.—that on their own, or assembled differently, do not have the same emergent properties.

13 It does not automatically imply that an effect was produced, for it could be that theinteracting forces nullified one another and so produced “no” outcome—rememberingthat even “no” outcome is still an outcome per se.

REFERENCES

A!"#$%%&', L. (1965/2005). For Marx. Verso.A'(#&', M. (1995). Realist Social Theory: the Morphogenetic Approach. Cambridge University

Press.B)!*+)', E. (1988/1992). Foucault and Marx: The Question of Nominalism. In Timothy

Armstrong (trans) Michel Foucault: Philosopher. Routledge, pp. 38–57.B#)%,)', R. (1978/2008). A Realist Theory of Science. Verso.B$(,!&-, W. (1967). Sociology and Modern Systems Theory, Prentice-Hall.C$..*./#)0, B. (2001). The Reemergence of Emergence. Philosophy of Science, 68(3),

Supplement, Part 1, S62–S75.D)""), R. P. (2007). From Foucault’s Genealogy to Aleatory Materialism: Realism, Nomi-

nalism and Politics. In J. Frauley and F. Pearce (eds) Critical Realism and the Social Sciences:Heterodox Elaborations, Toronto University Press, pp. 273–295.

D)""), R. P. (2008). Politics and Existence: Totems, Dispositifs and Some Striking ParallelsBetween Durkheim and Foucault. Journal of Classical Sociology, 8(2), 283–305.

D)-, R. (2007). More Than Straw Figures in Straw Houses: Toward a Revaluation ofCritical Realism’s Conception of Poststructuralist Theory. In J. Frauley and F. Pearce(eds.) Critical Realism and the Social Sciences: Heterodox Elaborations. Toronto University Press,pp. 117–141.

D&1&'", D. (1991). “Popular” Life and Insurance Technology. In G. Burchill et al. (eds.)The Foucault Effect. University of Chicago Press, pp. 211–233.

D&!&$2&, G. (1988/1992). What is a dispositif? In Timothy Armstrong (trans) Michel Foucault:Philosopher. Routledge, pp. 159–168.

D'&-1$%, H. and R)+*.34, P. (1982). Michel Foucault: Beyond Structuralism and Hermeneutics.University of Chicago Press.

D$53.", D. and P&)'(&, F. (2001). Foucault Contra Foucault. Theoretical Criminology, 5(2),123–158.

E!6&'-V)%%, D. (2005). Emergence and the Realist Account of Cause. Journal of CriticalRealism, 4(2), 315–338.

E!6&'-V)%%, D. (2007a). Social Structure and Social Relations. Journal for the Theory of SocialBehaviour, 37(4).

E!6&'-V)%%, D. (2007b). Re-examining Bhaskar’s Three Ontological Domains. InC. Lawson, J. Latsis and N. Martins (eds.) Contributions to Social Ontology. Routledge,pp. 160–176.

F3$()$!", M. (1961/2006). History of Madness, Routledge.F3$()$!", M. (1963/1994). The Birth of the Clinic, Vintage.

90 Nick Hardy


F3$()$!", M. (1968/1991). Politics and the Study of Discourse. In G. Burchill (ed) TheFoucault Effect, pp. 53–72.

F3$()$!", M. (1969/1972). The Archaeology of Knowledge, Pantheon.F3$()$!", M. (1971/2003). Nietzsche, Genealogy, History. In P. Rabinow and N. Rose

(eds) The Essential Foucault. The New Press, pp. 351–369.F3$()$!", M. (1973/2000). Truth and Juridical Forms. In J. Faubion (ed.) Power, The New

Press, pp. 1–89.F3$()$!", M. (1975/1991). Discipline and Punish, Penguin.F3$()$!", M. (1976a/1990). History of Sexuality Vol. 1, Vintage.F3$()$!", M. (1976b/2003). Truth and Power. In P. Rabinow and N. Rose (eds) The

Essential Foucault. The New Press, pp. 300–318.F3$()$!", M. (1977a/1980). Confession of the Flesh. In C. Gordon (ed) Power/Knowledge.

Pantheon, pp. 194–228.F3$()$!", M. (1977b/1980). History of Sexuality. in Power/Knowledge: Selected Interviews and

Other Writings 1972–1977, ed. Colin Gordon, Pantheon, 183–193.F3$()$!", M. (1979/2003). “Omnes et Singulatim”: Toward a Critique of Political Reason.

In P. Rabinow and N. Rose (eds.) The Essential Foucault. The New Press, pp. 180–201.F3$()$!", M. (1982a/2003). Technologies of the Self. In P. Rabinow and N. Rose (eds)

The Essential Foucault. The New Press, pp. 145–169.F3$()$!", M. (1982b/2003). Subject and Power. In P. Rabinow and N. Rose (eds) The

Essential Foucault. The New Press, pp. 126–145.F3$()$!", M. (1997/2003). Society Must Be Defended. New York, Picador Press.F3$()$!", M. (2004/2007). Security, Territory, Population. Palgrave Macmillan.F3$()$!", M. (2004/2008). The Birth of Biopolitics. Palgrave Macmillan.F')$!&-, J. (2007). The Expulsion of Foucault From Governmentality Studies: Toward an

Archaeological-Realist Retrieval. In J. Frauley and F. Pearce (eds) Critical Realism and theSocial Sciences: Heterodox Elaborations. Toronto University Press, pp. 258–272.

F')$!&-, J. and P&)'(&, F. (2007). Methodological and Epistemological Preliminaries. InJ. Frauley and F. Pearce (eds) Critical Realism and the Social Sciences: Heterodox Elaborations.Toronto University Press, pp. 3–29.

J&%%35, B. (2007). From Micro-powers to Governmentality: Foucault’s Work on Statehood,State Formation, Statecraft and State Power. Political Geography, 26, 34–40.

J3%&5#, J. (2004). “Foucault and Reality” in Capital and Class, #82, 143–165.K)*6&%37), T. (2009). Bhaskar and Bunge on Social Emergence. Journal for the Theory of

Social Behaviour, 39(2), 1–23.L)4%3., T. (1998). Economic Science Without Experimentation/Abstraction. In

M. Archer et al. (eds) Critical Realism: Essential Readings. Routledge, pp. 144–169.!&(3$'", D. (1972/1975). On Archaeology and Knowledge (Michel Foucault). In

B. Brewster (trans) Marxism and Epistemology. New Left Books, pp. 187–213.P&)'(&, F. (2007). Bhaskar’s Critical Realism: An Appreciative Introduction and a

Friendly Critique. In J. Frauley and F. Pearce (eds.) Critical Realism and the Social Sciences:Heterodox Elaborations. Toronto University Press, pp. 30–63.

P&)'(&, F. and W336*4*%%, A. (2001). Reading Foucault as a Realist. In G. Potter andJ. Lopez (eds) After Postmodernism: An Introduction to Critical Realism. Athlone Press, pp. 51–62.

M)'%6&., R. (1999). The Nature of Capital: Marx After Foucault, Routledge.R)7(#0)., J. (1985). Michel Foucault: The Freedom of Philosophy, Columbia University Press.R&)6, J. (2003). The Micro-politics of Capital: Marx and the Prehistory of the Present. State

University of New York Press.W&*', L. (2008). The Concept of Truth Regime. In Canadian Journal of Sociology, 33(2),

pp. 367–389.



Foucault, Genealogy, Emergence: Re-examining the Extra-discursive

Documents

Transcript of Foucault, Genealogy, Emergence: Re-examining the Extra-discursive