Human Kinds and Looping Effectsin PsychologyFoucauldian and Hermeneutic Perspectives

Svend BrinkmannUniversity of Aarhus

Abstract. Recent work in the historiography of psychology has suggestedthat the discipline must be seen as involved in the constitution of its ownsubject matter. Two questions arise. First, what does this tell us about thesubject matter of psychology? Second, how should we understand andinvestigate the processes through which such making up people occurs?This article addresses these questions by arguing, first, that psychologicalcategories refer to human rather than natural kinds. In contrast to naturalkinds, human kinds can exert effects on themselves. Then two differentapproaches to the looping effect of human kinds are sketched: aFoucauldian analytics of the techne of psychology, and a hermeneuticunderstanding of humans as self-interpreting beings. It is argued thatFoucaults perspective, while rich in its critical potentials, lacks anadequate understanding of those values and meanings which are a part ofhuman practical life.Key Words: Foucault, hermeneutics, human kinds, looping effects, re-flexivity

The psychological ways of thinking about people have not just served aspassive representations of human subjects, but have in fact deeply influencedhow humans think and feel, and indeed influenced human subjectivity itself.Since the birth of psychology, humans have increasingly come to think aboutthemselves in the light of psychologys concepts and categories, and theirlives have become dependent on psychological technologies such as tests andtherapies. Psychology is in the business of making up people, to use IanHackings phrase (1986). One dilemma arising from this is sometimesreferred to as the reflexive problem: psychology is produced by, produces,and is an instance of, its own subject matter (Richards, 1996, p. 5).

In what follows, I ask what it tells us about the nature of psychologyssubject matter if the discipline can make up people. Following Ian Hackingand Kurt Danziger, I argue that it tells us that psychological objects arehuman kinds (what these are will be explained below). Based on the

Theory & Psychology Copyright 2005 Sage Publications. Vol. 15(6): 769791DOI: 10.1177/0959354305059332 www.sagepublications.com

philosophical natural kind semantics developed by Saul Kripke and HilaryPutnam in the 1970s, I provide an argument to the effect that we have noreason to believe that psychology is working or will be able to work withnatural kinds. Psychological objects (such as intelligence, learning, depres-sion) are not naturally existing objects, for, unlike natural kinds, theirexistence depends on certain descriptions and discursive contexts. Unlikenatural kinds, human kinds interact with their descriptions and categories;there is a looping effect of human kinds (Hacking, 1995b).

Next I discuss two competing ways of understanding the looping effectof human kinds. How do the processes of making up people occur? FirstI give an account of the Foucauldian approach to this matter, particularly asit has been developed by Nikolas Roses analytic of human technologies,stressing what he calls the techne of psychology. Foucault himself thoughtthat psychology was a dubious science (Dreyfus & Rabinow, 1982, p. xx)that will never reach the level of Kuhnian normal science, but which isdeeply and unavoidably enmeshed in political, social and cultural functionsin different ways. From early on, Foucault was aware of the normativedimensions of psychological knowledge, that is, of its ability to constituteand transform its subject matter. In an early interview from 1965 he said: Idont think psychology can ever dissociate itself from a certain normativeprogram. . . . Every psychology is a pedagogy, all decipherment is a ther-apeutics: you cannot know without transforming (Foucault, 1998, p. 255).

I subsequently compare the Foucauldian approach to the hermeneuticunderstanding of looping effects, which frames humans as self-interpretingbeings, and which, unlike Foucault, aims to preserve a space for genuinevalues and meanings in the practices through which psychology makes usup. Instead of beginning the analysis with the techne of psychology and aproblematization of practices (Stern, 2000), the hermeneutic approach repre-sented by Martin Heidegger, Hans-Georg Gadamer and Charles Taylorinsists on beginning with hermeneutic articulation, and it preserves a causalrole of meanings and values in changing and improving practices (Taylor,2000, p. 123). For that reason I argue that the hermeneutic approach ismorally superior to Foucaults aestheticization of human existence.

Natural Kinds and Human Kinds

Let me begin by asking whether psychological categories pick out naturalkinds; and, if not, then what do they refer to? Natural kinds are such thingsas tigers, water or gold. They are groups of naturally occurring phenomenathat inherently resemble each other and differ crucially from other phenom-ena (Danziger, 1999, p. 80). It is one of the goals of natural science todiscover and identify natural kinds, and it is a distinctive mark of the naturalsciences that they are able to identify these. As Karl Marx (1894/1971)

THEORY & PSYCHOLOGY 15(6)770

famously said: All science would be superfluous if the outward appearanceand the essence of things directly coincided (p. 817). The search for naturalkinds is the search for the underlying essences of things. We want to findnatural kinds, because, if we succeed in doing so, we have an overview ofthe common essential properties of a class of things that allow us to explainother, more superficial, properties of this class of things. Take water as anexample. Water has certain observable properties: it is colorless, tasteless,has a specific boiling point and freezing point, and so on. The underlyingessential property that allows us to explain all these superficial properties isits molecular structure (H2O). Scientific categories can be said to refer tonatural kinds if the term for them were to remain applicable even if we wereto discover that they possessed completely different properties than we hadtraditionally thought (Collin, 1990). For example, if humans developed finersenses of taste, and found that water is not really tasteless, we would stillcall it water. The reason for this is that its underlying essence remains thesame, although its superficial phenomenal properties appear differently.

The discussion of natural kinds has been significant in modern linguisticphilosophy, challenging widespread views of linguistic meaning. Mostphilosophers (and psychologists) have thought that meanings are mentalentities of some sort. Knowing the meaning of water would, on thistraditional account, involve being in a certain psychological state; having acertain mental representation. Philosophers Saul Kripke and Hilary Putnamsuccessfully challenged this view in the early 1970s. Putnam (1973), whomI concentrate on here, asserted: Cut the pie any way you like, meaningsjust aint in the head! (p. 704). Putnam defended an externalist account ofmeaning on the grounds that the meaning of such terms as water could notbe accounted for as concepts, or indeed mental entities of any kind. Insteadtheir meanings had to be understood as rigidly fixed to the natural kindpicked out by the term. The illustrative and now-famous science-fictionexample given by Putnam is the twin-earth example. We should imagine thatan expedition from our planet earth finds another planet far away from ours,which in every detail is just like our planet. They name it twin-earth. On thatplanet, there is a liquid, which in all immediate phenomenal respects is justlike our water. People on twin-earth have exactly the same practices withthis liquid as we have with water on our planet. They drink it, boil it andfreeze it, just like we do. But when the expedition analyses the liquidchemically, it turns out that its molecular structure is not H2O, but XYZ. Thequestion then is: how should the compound be classified? Putnam arguedthat we would intuitively say, and rightly so, that the compound is not water,but merely something that looks just like it in all observable respects.Putnam thus argued that natural kind terms (e.g. water) refer to naturalkinds (essences, e.g. H2O), just as proper names refer to particularpeople, without the mediation of mentally represented meanings or defini-tions (Haslam, 1998). These terms are rigidly fixed to their referents.

BRINKMANN: HUMAN KINDS AND LOOPING EFFECTS 771

If we accept this analysis of natural-kind terms, the next question to ask iswhether natural kinds exist in psychology, or indeed in any social science.And here the answer seems to be no (Collin, 1990). As things stand, we haveno reason to believe that any of the terms used in social science refer tonatural kinds. It would not be conceivable, for example, to think that wecould discover that a hospital is really not an institution whose purpose it isto treat illnesses, but something completely different, for example a place fordeveloping means for biological warfare. If it turns out that some hospital defacto functions like this, we will have to conclude, not that we did not knowwhat hospitals were, but that this specific institution was not a hospital. EvenFoucauldian unmasking of the rise of treatment institutions in modernitycannot make things otherwise. If some radical Foucauldian could show thatall hospitals in fact have worked single-mindedly to oppress rather than curepeople, the right conclusion would be that bona fide hospitals have neverexisted, only institutions calling themselves hospitals, but with different,un-hospital-like underlying rationales.

To cite an example perhaps more congenial to psychology: even if wesome day succeed in identifying depression with a specific neural dysfunc-tion, it seems unlikely that we would say that we would thereby havediscovered a natural kind (the essence of depression). This is because, toreiterate the twin-earth example, if it turns out that depressed people ontwin-earth, that is, people with anhedonia and all other phenomenal expres-sions of depression, have a completely different neuronal make-up comparedto depressed people on earth, then we would still not give up our classifica-tion of these people as depressed. They would still be rightly described asdepressed even though their neuronal structure corresponded to the neu-ronal structure of happy and carefree people on earth. Unlike water, there isnot in the case of depression a natural kind, some underlying essence, whichfixes the meaning of the concept depression. Instead, its meaning is tied tothe superficial, observable properties of depressed behavior and its sig-nificance in the human life world. In this regard, it is impossible to separatethe superficial phenomenal properties and some underlying essentialproperties when it comes to concepts in the human and social sciences. Inthese sciences, concepts cannot be about some completely hidden essence,because, as Anthony Giddens (1976) has argued, concepts in the socialsciences are concepts about concepts. That is, they are concepts thatsomehow must hook up with the participants own concepts (this argumentwas also forcefully made by Peter Winch [1958/1963]). Understanding inpsychology and the social sciences always means understanding from withinthe practices and cultures researched.

Psychological categories thus have to relate to the phenomena of the lifeworld and to folk psychologies. Does this mean that psychology itself isa natural kind? Nothing like this follows, and it would indeed go againsthistorical evidence to claim that psychology or the psychological refers to

THEORY & PSYCHOLOGY 15(6)772

a natural kind (Danziger, 1997). The discipline and its subject matter areinfluenced by theories about it, and it must be borne in mind that there aremultiple psychologies around, rather than a single, monolithic body ofknowledge. John Dewey (1917/1980) used to distinguish between twopsychologies, and he argued that in psychology:

. . . all phenomena can be divided into the physiological and the social, andthat when we have relegated elementary sensation and appetite to theformer head, all that is left of our mental life, our beliefs, ideas and desires,falls within the scope of social psychology. (p. 54)

I am willing to grant that some categories that are used in psychologicalresearch may in fact refer to natural kinds. It cannot be excluded that thereare genuine natural kind terms in parts of what Dewey called physiologicalpsychology, but in social psychology, where psychological phenomena aremeaningful and relate to human action, there are none.

But if such psychological categories do not refer to hidden essences ornatural kinds, then what do they refer to? Philosopher of science IanHacking answers that psychological categories refer to human kinds. Whatare these? Initially, we can outline three ways in which human kinds differfrom natural kinds. Natural kinds are characterized by: intelligibility outside discursive contexts; indifference to the descriptions applied to them; and independence of categories and kinds. Human kinds, by contrast, are characterized by: intelligibility only within a discursive context; interaction with the descriptions applied to them; and emergence together of categories and kinds.Hacking (1995b) says that the chief difference between natural and humankinds is that the human kinds often make sense only within a certain socialcontext (p. 362). One can be a captain or a samurai, for example, only incertain social settings, where certain discourses and descriptions are avail-able. One can be a king only in a world where the institution of monarchyexists (water, on the other hand, was H2O before the institution of chem-istry). Human kinds indicate kinds of people, their behaviour, their condi-tion, kinds of action, kinds of temperament or tendency, kinds of emotion,and kinds of experience (pp. 351352). Hacking has analyzed in varyingdetail multiple personality disorder (1995a), fugue (1998), homosexual-ity (1986), suicide, teen-age pregnancy, adolescence, child abuse,autism and Hispanic (1995b) as examples of human kinds.1

An important feature of human kinds is that they can exert effects onthemselves (Martin & Sugarman, 2001). They are affected by their classifi-cations and interact with their classifications, sometimes affecting theclassifications themselves. Human kinds can even intentionally try to change

BRINKMANN: HUMAN KINDS AND LOOPING EFFECTS 773

how they are classified (e.g. when homosexuals objected to being catego-rized as pathological). This is the looping effect of human kinds: Peopleclassified in a certain way tend to conform to or grow into the ways they aredescribed; but they also evolve in their own ways, so that the classificationsand descriptions have to be constantly revised (Hacking, 1995a, p. 21). Andfurther:

Inventing or molding a new kind, a new classification, of people or ofbehavior may create new ways to be a person, new choices to make, forgood or evil. There are new descriptions, and hence new actions under adescription. (p. 239)

Why do new kinds of description, for example those produced bypsychologistsHacking (1995b) refers to Freud as the king of theloopersprovide for new kinds of action? Because in so far as humanactions are intentional, they are actions under a description (Hacking, 1995a,p. 234). We only say that people act intentionally if they are able to providesome kind of description of their action that renders it meaningful bylocating it in a discursive context. I can only have the intention to vote for agiven candidate at a democratic election if I can describe the physicalbehavior commonly associated with voting (going to the voting booth andmarking the piece of paper) as an act of voting. One cannot act intentionallyunder a description unless this description is discursively available. AsHacking (1995a) says:

When new descriptions become available, when they come into circulation,or even when they become the sorts of things that it is all right to say, tothink, then there are new things to choose to do. When new intentionsbecome open to me, because new descriptions, new concepts, becomeavailable to me, I live in a world of new opportunities. (p. 236)

Hacking is clear that this does not amount to any facile social construction-ism. He characterizes himself as a realist (Hacking, 1986). Human kinds areindeed very real, and although we are in the business of potentially makingup new kinds of people when we invent new classifications and new fieldsof description (Hacking, 1995b), this does not make the kinds any less real.Some people do suffer from Multiple Personality Disorder. They (and thecommunities in which they live) are not faking it, although they could nothave acted the way they do, and eo ipso suffered from this condition, beforethe relevant field of description became available some time late in the19th century. Orto be preciselate in the afternoon of 27 July 1885,when Charcots student, Jules Voisin, described the very first MPD-patient,Louis Vivet (Hacking, 1995a, p. 171). In contrast, it can reasonably beargued that people were able to suffer from tuberculosis even before thisillness was named (although they may have been unaware of the cause oftheir disease). In the case of tuberculosis there is a relevant natural kind

THEORY & PSYCHOLOGY 15(6)774

involved that causes the disease, which is not dependent on a discursivecontext, that is, on how it is described and classified.

I am aware that that point has been denied, most forcefully by BrunoLatour (2000), who has recently argued that although we have found tracesof tuberculosis in the mummy of Pharaoh Rameses II, he could not havedied of tuberculosis for the simple reason that the bacillus was not identifieduntil more than 3000 years after his death. Latour believes not only that allour categories are socially constructed (which is obviously true), but alsothat all kinds referred to by our categories, including natural ones, areequally so constructed. All kinds are, to use Hackings terminology, humanones. I cannot here defend in great detail why I believe a distinction betweenhuman and natural kinds is still needed. But against Latours monism ofkinds, it can be said that it becomes difficult to explain how natural thingsresist our dealings with them in a way that human things do not necessarilydo. Latour (1988) has defined the real as that which resists us: Whateverresists trial is real. . . . The real is not one thing among others but rather agradient of resistance (pp. 158159). He is thus clear that we cannotsocially construct nature as we would like it to be, but how can this bemaintained if one is unwilling to grant that parts of reality exist in somedefinite form external to social practice? If something resists our practices ofclassification it must be because there are structures (natural kinds) prior toour categories that limit how we can fruitfully classify it. Furthermore,unlike water and other natural kinds, human kinds are intrinsically part ofsocial practice and take an interest in how they are classified, becausehuman kinds are laden with values. This in fact is Hackings argumentagainst Latours insistence that all natural kinds are at bottom social (seeHacking, 1995b, pp. 365368). It makes no difference to water what we callit, but it does make a difference to human kinds. After all, as AlasdairMacIntyre (1985) once remarked, molecules dont read chemistry textbooks,whereas humans do read psychology books that affect their self-understandings.

In this context it is important to bear in mind that human kinds are just asreal as natural kinds. There is no ontological difference between them in thatsense. Human kinds do not belong to a different, non-natural world. In asense they are just as natural as natural kinds, but they are natural in adifferent way, perhaps akin to what Hegel and modern Hegelians callsecond nature (Hegel, 1821/1991, p. 35; McDowell, 1994), that is, natureas it expresses itself in and through social practice. Human kinds as second-nature entities are no less natural than natural kinds (first-nature entities), norare they necessarily easier to change. If our category white rhinocerosrefers to a natural kind, then we can wipe out all instances of this naturalkind quite easily, unfortunately all too easily. We cannot similarly eliminatedepression, although depression is a human-kind term. And the fact thatdepression is not a natural kind does not make the condition any less serious

BRINKMANN: HUMAN KINDS AND LOOPING EFFECTS 775

for those who suffer from it. The distinction between natural and humankinds tells us absolutely nothing about the existential persistency of any ofthese kinds, or about the seriousness of their existence.

An Analytic of Human Technologies

I shall now consider two different frameworks that purport to help usunderstand practices of classification and looping effects in psychology, theFoucauldian and the hermeneutic. The goal is not to provide detailed analysesof actual looping effects, but rather to ask: What is the proper approach andattitude towards human kinds and their looping effects in psychology? AsHacking said, human kinds indicate kinds of people, behavior, action,temperament, emotion and experience. They thus indicate something abouthuman subjectivity. Human-kind terms are always applied to, or by, humansubjects. For that reason, the following discussion will center on the notion ofsubjectivity, and I begin with Foucaults conception of the subject, beforemoving on to Roses analytic of human technologies.

Foucault on the SubjectFoucaults historical and philosophical argument involves the view that thehuman subject as studied by the social sciences did not exist as an object oftruth until early in the 19th century (Smith, 1997, p. 857), when the socialsciences created a subjugated subject by rendering it calculable, manageableand governable. Foucaults overarching goal had been to create a history ofthe different modes by which, in our culture, human beings are madesubjects (1994b, p. 326). This assertion provides us with a fruitful way ofreading his oeuvre.

Foucault is probably most widely known for his critique of modernity aspresented in Discipline and Punish (1977). He was here intent to examinehow modernitys social scientific disciplines, psychology among them,historically emerged as attempts to govern and discipline individuals andpopulations. In modernity, new disciplines, institutions, techniques anddiscourses made possible new forms of power exertion. In contrast to earlier,pre-modern forms of power, such as public transportation of prisoners inchains and public executions, modern forms of power work in subtler andmore hidden ways, in closed prisons and bureaucratic structures supportedby scientific expertise. Foucaults point is not that earlier kinds of visiblebrutality were morally better than modernitys veiled brutality, but that wemust develop new ways of analyzing power in order to account for itsmodern forms. We can no longer work with a conception of power ascentered with a subject, a single sovereign, for example a king or a leader,since modernitys power relations are much more diffused and distributed in

THEORY & PSYCHOLOGY 15(6)776

heterogeneous practices. Modern power is capillary, as Foucault said, nolonger connected to obvious physical superiority, but to truth, knowledgeand expertise, dispersed across social life.

A radical point in Discipline and Punish is that the human subject is aneffect of power. It is not primarily subjects, who, from a position outsidepower relations, intentionally exercise power in order to promote theirspecific interests, for being a subject with interests in the first place is onlypossible because of power relations. Power is thus not merely oppressive,according to Foucault, but also productive: it produces subjectivities. Howshould we understand this strange claim?

Foucaults illustrative example is the panopticon, a prison structuredeveloped by Jeremy Bentham, the father of utilitarian moral philosophy.This structure would enable a single guard to monitor the behaviors of alarge number of prisoners. Foucault is interested in this architectonicstructure because it represents the ideal type of modernitys rational-bureaucratic form of power exertion, and it nicely illustrates the point thatpower should always be understood, not in abstractum as belonging toatomistic subjects or different classes, but as concretely embedded intechnologies, practices and physical objects. The panopticon should renderindividuals visible at all times, but it should also make the individualsmonitor themselves. Prisoners unavoidably turn the guards gaze towardsthemselves, and thereby become constituted as self-monitoring or self-examining subjects. Here are reminiscences of a theory of self-consciousness as found in G.H. Mead, where human beings always relate tothemselves through the gazes and reactions of others. In Foucault, however,this self-relation is portrayed as a relation of power, where subjects areconstituted as self-examining individuals. A self-examining individual hasthe soul, psyche or subjectivity as a correlate (Coles, 1992). In Foucaultseyes, the soul is the prison of the body. The soul (or mind) is that which isproduced when the body is worked upon and disciplined in specific ways.This can be called a physicalistic aspect of Foucaults argument.

Foucaults point is that the self-monitoring subject was constituted inmodernity through a number of institutional practices: in schools, factories,prisons, hospitals, courts of law, and in the whole system of treatment ofindividuals at large. The soul, the inner psychological world, became thepanopticon that we all carry within ourselves (Coles, 1992). Psychologybecame the science of the inner world, and therapies and other similar powertechniques became normalizing, subjectivity-constituting social technolo-gies, especially when Western men and women became what Foucault calledconfessing animals, constituted as subjects through confessional technolo-gies (Foucault, 1980). We are not simply subjects. Rather, we are madesubjects through processes of subjectification.

In the final books and articles written by Foucault before his death in1984, his interests shifted from external power techniques to what he called

BRINKMANN: HUMAN KINDS AND LOOPING EFFECTS 777

technologies for individual domination (Foucault, 1988a). He becameinterested in the history of how an individual acts upon himself, in thetechnology of self (p. 19). Technologies of the self are tools with which anindividual acts on herself to create, re-create and cultivate herself as asubject. Through his historical investigations, Foucault succeeded inlocating different kinds of technologies of the self: the Stoic practice ofletter writing, Augustines confessions, examinations, asceticism and inter-pretations of dreams.

Contra historical materialists, Foucault claimed that technologies of theself are relatively independent of their socio-economic and political condi-tions. For that reason, humans are much freer than they think (Foucault,1988b). The consequence of the fact that the self is not given us to discoverin any universal way is that we must create the self as a work of art(Foucault, 1984). The later Foucault still conceived of subjectivity as aneffect of power relations produced in concrete practices, but individualswere now understood as being able to determine, to some extent, whichpractices they will allow themselves to be constituted by. We can use theproductive nature of power to shape our selves, which is what Foucaultcalled freedom.

Technologies of the self thus need not be oppressive. They can be linkedwith ethics as practices of freedom. By ethics, Foucault meant not theabstract philosophical discipline, but a practice, namely the practice ofsubjective self-formation (Foucault, 1984). Ethics is the practical relation-ship that a self ought to have to itself, and, in this sense, it can be studiedindependently of moral codes, and also independently of scientific truth.There is no connection at all, according to Foucault, between ethicalproblems and scientific knowledge. We can just as little demand that ethicsshould be true scientifically as we can demand that a work of art should bebased on scientific criteria.

That is why we must create ourselves as a work of art. There is no deep,inner truth about human nature, or about individual subjectivity, to beunearthed by science, so the goal of practicing technologies of the selfshould not be to decipher such an illusory truth. If this happens, thentechnologies of the self become oppressive. They come to belong in thehands of scientific expertsmasters of truthand then such technologiesdegenerate into techniques of domination. That is why Foucault wants toreplace the classical Western ideal know thyself with what is for him a moreprimary demand: care of the self (Foucault, 1988a, 1990).

The Techne of PsychologyThe central point in Foucaults view of the human subject is that individualhuman beings are constituted as subjects (subjects to be known andcontrolled, and subjects who know and control) by practical technologies,

THEORY & PSYCHOLOGY 15(6)778

infused in power relations. So we see that this is Foucaults version of howpsychology and other human sciences are able to constitute human subjectsand make human kinds loop: by providing society with technologies forsubjectivity constitution. In his early writings, one part of the loop isemphasized at the expense of the other, namely the loop from technologiesof power to human subjects. In later writings, Foucault turned his attentiontowards the other side of the loop: to the ways humans themselves caninfluence their subjectivities.

Nikolas Rose is arguably the most significant Foucauldian to haveaddressed psychology. Inspired by Foucaults notion of governmentality(Foucault, 1994a), he has analyzed how governmentality or simply govern-ment have become psychologized. The exercise of modern forms ofpolitical power has become intrinsically linked to a knowledge of humansubjectivity (Rose, 1996b, p. 117). Government refers to the strategies forthe conduct of conduct (p. 116) that have proliferated over the past twocenturies, significantly supported by the psy-knowledges, the disciplinaryregime of psychology (Rose, 1985, 1996a, 1999). Foucault and Rose intendthe notion of governmentality to connect broader forms of power andpolitical governmental control with finer micro-processes of power (Hook,2003), for example in psychological practices such as therapy or testing.

Roses analyses deal with the techne of psychology (Rose, 1996a, 1996b).In Aristotles classical usage, techne is the kind of knowledge needed toproduce things, and the techne of psychology is thus the kind of knowledgeneeded to produce subjectivities. Rose (1996b) wants to consider psycho-logy not merely as a body of thought but as a certain form of life, a modeof practicing or acting upon the world (p. 116). Living a life, Rose argues,has become a psychological and even a therapeutic business. We are todayobliged to make our lives meaningful through the search for happiness andself-realization in our individual biographies, and psychological techniqueshave become imperative in this regard. The self has become a commodity ora raw material to be worked upon, and psychological expertise has beencentral in devising ways of working upon the self. The goal is freedom, themodern liberal self is obliged to be free (Rose, 1996a, p. 100), andpsychological expertise and power have become ethicalized (p. 92), linkedto authority concerning the good, meaningful life.

In Roses development of Foucaults ideas, the subject becomes anassemblage: the localization and connecting together of routines, habits, andtechniques within specific domains of action and value (Rose, 1996a, p. 38).The subject is assembled through practical technologies, and we shouldbegin with a minimal, weak, or thin conception of the human material onwhich history writes (p. 37). The key to understand psychology and itshistory lies in understanding how it has served to assemble human subjectsthrough its technologies. Rose thus develops what he calls an analytic ofhuman technologies (p. 27) where looping effects, the way psychology

BRINKMANN: HUMAN KINDS AND LOOPING EFFECTS 779

affects its subject matter, are conceived as a technological affair. On thispoint, Rose is critical of what he sees as social constructionists emphasis onmeaning, narrative and linguistic discourse:2

Subjectification is not to be understood by locating it in a universe ofmeaning or an interactional context of narratives, but in a complex ofapparatuses, practices, machinations, and assemblages within which humanbeing has been fabricated, and which presuppose and enjoin particularrelations with ourselves. (p. 10)

Technology refers to any assembly structured by a practical rationalitygoverned by a more or less conscious goal. Human technologies are hybridassemblages of knowledges, instruments, persons, systems of judgment,buildings and spaces (Rose, 1996a, p. 26). The normative aim in Rose andFoucault is not to liberate humans from the thralldom of technology, for, intheir eyes, subjectivity is necessarily and always a technological assemblage.Rather, it is to question the more or less conscious goal that informs thetechnologies through which humans are currently assembled as psycho-logical subjects (particularly the liberalist goal of governing autonomousselves), and the ways our identities and moralities consequently have beenbound to an inner psychological realm. However, I believe that an analyticof human technologies and the view of ethics and subjectivity developed byFoucault have no answer to give us concerning an ethic of existence moreadequate than the one currently informing the psy-sciences. For we are givenno resources for evaluating whether psychological technologies affect us ingood or in bad ways other than a turn to the individuals subjective self-formation. As we shall see in the following, the Foucauldian analytic ofhuman technologies cannot account for the role that values and meanings playin social practices, however rich it may otherwise be in its critical potentials,and it cannot give us grounds for assessing psychologys products.

Hermeneutic Processes of Self-interpretation

Foucault and Roses technological approach to human subjectivity and thelooping effect of human kinds can be contrasted with an alternativehermeneutic approach, which understands humans as self-interpreting be-ings. In introducing the notion of looping effects of human kinds, Hacking(1995b) originally did not want to enter the VerstehenErklaren debate inpsychology and social science: The Verstehen dispute has partly to do withmethodology, a subject that I abhor . . . I believe there are some deepinsights on the Verstehen side of the argument, but here they are irrelevant(p. 364). The Verstehen problematic is irrelevant because Hackings under-standing of human kinds demands causal analysis rather than Verstehen ormeanings (p. 364). I believe, however, that Hackings view of things could

THEORY & PSYCHOLOGY 15(6)780

become (even) more interesting to psychologists if it were stated withreference to hermeneutic philosophy, notably ontological hermeneutics. Ashe says in one place, the essential difference between the looping effect ofhuman kinds and other looping effects is that people are aware of what theyare called, adapt accordingly, and so change, leading to revisions in facts andthen knowledge about them (Hacking, 1995b, p. 388). This being aware,which humans are capable of, is exactly what characterizes our existence asDasein, according to Martin Heideggers (1927/1962) ontological herme-neutics (cf. Brinkmann, 2004; Stenner, 1998). This has also been sensed byMartin and Sugarman (2001), who use Hackings discussion of human kindsas a springboard to develop a hermeneutic approach in psychology. Theyclaim that human kinds demand a hermeneutic understanding, because,while human kinds are definitely real, it is a reality in which theythemselves are deeply involved (p. 194). The idea of humans beinginvolved in their own reality is the focus of ontological hermeneutics as ithas been developed by Heidegger, Gadamer and Taylor, among others.While Foucault was primarily engaged in the problematization of practices,and would leave the task of assessing the values of different practices to theindividual in the process of creating herself as a work of art, hermeneuticwriters tend to de-emphasize individual choice and claim that the values ofpractices cannot be accounted for as something subjective. Instead theytypically point to communities and traditions as the source of values.

Ontological Hermeneutics

Originally, hermeneutics was developed as a methodology for interpretingtexts, notably biblical texts (see chapter 9 in Richardson, Fowers, &Guignon, 1999, for an illuminating history of hermeneutics). With WilhelmDilthey in the late 19th century, hermeneutics was extended to human lifeitself, conceived as an ongoing process of interpretation. Heideggers Beingand Time (1927/1962) is often cited as the work that inaugurated the shift toontological hermeneutics proper. The question of methodological herme-neutics had been: How can we correctly understand the meaning of texts?Epistemological hermeneutics had asked: How can we understand our livesand other people? But ontological hermeneuticsor fundamental ontologyas Heidegger also called it (1927/1962, p. 34)prioritizes the question:What is the mode of being of the entity who understands? (Richardson etal., 1999, p. 207). Being and Time aims to answer this question and can thusbe said to be an interpretation of interpreting (p. 208).

Heideggers name for the entity that understands is Dasein, and the beingof Dasein is unlike the being of other entities in the universe. Physicalentities such as molecules, tables and chairs are things that have categorical

BRINKMANN: HUMAN KINDS AND LOOPING EFFECTS 781

ontological characteristics, whereas human beings, or Dasein, are historiesor events and have existentials as their ontological characteristics(Polkinghorne, 2004, pp. 7374). The existentials make up the basicstructure of Dasein and are affectedness (Befindtlichkeitwe always findourselves thrown into situations where things already matter and affect us),understanding (Verstehenwe can use the things we encounter) and articu-lation or telling (Redewe can to some extent articulate the meanings thingshave) (Dreyfus, 1991). Of course, humans also have living physical bodieswith categorical characteristics, we are alive, but in addition we have lives thatunfold in time, and evolve around what is deemed meaningful and valuable.

Dasein primarily exists as involved in a world of meanings, relations andpurposes, and only derivatively in a world of physical objects. In oureveryday lives we live absorbed in pre-established structures of significance,or what Heidegger called equipment, which serve as a background thatenables specific things to show up as immediately meaningful and value-laden, given our participation in different social practices. We do notexperience meanings and values as something we subjectively project untothe world, for the world in which we live meets us as always already imbuedwith meaning and value. Only when our everyday, unreflective being-in-the-world breaks down, when our practices of coping with the equipmentsomehow become disturbed, do entities appear with objective character-istics distinguishable from human subjects.

For Heidegger and later hermeneuticists like Gadamer and Taylor, under-standing is not something we occasionally do, for example by followingcertain procedures or rules. Rather, understanding is the very condition ofbeing human (Schwandt, 2000, p. 194). We always see things as something,human behavior as meaningful acts, letters in a book as conveying somemeaningful narrative. In a sense, this is something we do, and hermeneuticwriters argue that all such understanding is to be thought of as interpretation.This, however, should not be understood as implying that we normally makesome sort of mental act in interpreting the world. Interpretation here is notlike the mental act of interpreting a poem, for example. It is not an explicit,reflective process, but rather something based on skilled, everyday modes ofcomportment (Polkinghorne, 2000). In the ontological hermeneutical per-spective, interpretation is not so much something we consciously do, assomething that is pre-reflectively lived and depends on a tacit meaning-giving background. Interpretation depends on certain pre-judices, asGadamer (1960/2000) famously argued, without which no understandingwould be possible. Knowledge of what others are doing, and also of whatmy own activities mean, always depend upon some background or contextof other meanings, beliefs, values, practices, and so forth (Schwandt, 2000,p. 201). There are no fundamental givens, for all understanding depends ona larger horizon of non-thematized meanings.

THEORY & PSYCHOLOGY 15(6)782

Self-interpretationAccording to Taylors reading of Heideggers fundamental ontology, we areself-interpreting animals (Taylor, 1985b). This means that humans are whatthey interpret themselves to be. This does not imply that any individualperson can make himself Napoleon by trying to interpret himself that way,but rather that human communities of interpreters and their historicaltraditions constitute the meanings of our self-interpretations. Gadamer(1960/2000) has said:

In fact history does not belong to us; we belong to it. Long before weunderstand ourselves through the process of self-examination, we under-stand ourselves in a self-evident way in the family, society, and state inwhich we live. The focus of subjectivity is a distorting mirror. The self-awareness of the individual life is only a flickering in the closed circuits ofhistorical life. That is why the prejudices of the individual, far more thanhis judgments, constitute the historical reality of his being. (pp. 276277)

As Hacking pointed out, there must be access to specific fields ofdescriptions if we are to be able to act intentionally in specific ways. Andthese fields of description that constitute our self-interpretations only existin historically evolved traditions. Gadamer (1960/2000) argues that thismakes the condition of human and social science quite different from theone we find in the natural sciences where research penetrates more andmore deeply into nature (p. 284). In the human and social sciences, therecan be no object in itself to be known (no natural kinds) (p. 285), forinterpretation is an ongoing and open-ended process that continuouslyreconstitutes its object. The interpretations of social life offered byresearchers in the human and social sciences are an important addition tothe repertoire of human self-interpretation, and powerful fields of descrip-tion offered by human science, such as psychoanalysis, can even affect theway whole cultures interpret themselves. As Richardson and colleagues(1999) spell out Gadamers view, this means that social theories do notsimply mirror a reality independent of them; they define and form thatreality and therefore can transform it by leading agents to articulate theirpractices in different ways (p. 227). On the face of it, this seems not tocontradict the Foucauldian notion of how different practical technologiesinscribe themselves on embodied subjects, thus disciplining or subjectify-ing them, in specific ways. So what is the difference?

Values and PracticesThe difference is that ontological hermeneuticists such as Taylor believethat we can to some extent articulate the background practices thatconstitute our self-interpretations, and, in doing so, get in touch with thevalues and goods that define them as practices. This is in contrast toFoucault, who portrayed background practices (without using this term) as

BRINKMANN: HUMAN KINDS AND LOOPING EFFECTS 783

a field of power, mediated by technologies. Foucault did not want toarticulate the meanings and values of practices, but rather to problematizethem (Stern, 2000). His account of the rise of disciplinary power, forexample, does away with all references to human values and intentions,and portrays disciplinary practices as just growing in modernity. But wecan, Taylor (2000) contends, articulate the reasons why these practicesgrew, and thereby refer to the intentions and values that people havewanted to realize. There are no doubt good reasons to criticize theseintentions and values, but the point is that if we want to understand thesepractices, we have to refer to the values that guided them. Social practicespresuppose both materiality and value-laden human self-interpretations:[W]hat we see in human history is ranges of human practices that are bothat once, that is, material practices carried out by human beings in space andtime, and very often coercively maintained, and at the same time, self-conceptions, modes of understanding (Taylor, 2004, p. 31).

In the hermeneutic perspective, the background that constitutes oursubjectivities is not just technological but also meaningful, infused withvalues and affected by human intentions. Practices involve what Taylor(2004) has recently talked about as social imaginaries. Practices are onlypossible if they make sense in some way, and this they do only byincorporating values, self-conceptions and modes of understanding (i.e. ifthey are imagined in certain ways). Taylor argues that we in fact canchange these background practices by articulating the values that informthem. This is the task for social theory as a form of practice (Taylor, 1985c).For example, the modern atomistic view of society is indeed, in Taylorsview, guided by a genuine human value, namely freedom. In articulatingpractices of freedom in modern societies (or practices of subjectification, asFoucault saw them) we must refer to the value of freedom if we are tounderstand them. The way to criticize them is not to do away with allreferences to values, but rather to see whether they realize the values onwhich they are purportedly built. And then we might find that a truly atomistsociety in fact needs a maximum of bureaucratic surveillance and enforce-ment to function (Taylor, 1985c, p. 110). This then defeats rather thanrealizes the value of freedom. Thus, ontological hermeneuticists reject theatomist view of society by way of immanent critique (background articula-tion), rather than Foucauldian problematization.

Taylor is worried that Foucaults problematization of practices collapsesinto moral subjectivism, the view that all moral orders are equally arbitrary(Taylor, 1989, p. 99), and he sees Foucaults late turn towards an aestheticsof existence and the self as a work of art as a way to restate the idea thatthe good is all in the register of individual autonomy (Taylor, 2000, p.133). According to Taylor (1989), Foucault offers charters for subjectivismand the celebration of our own creative power at the cost of occluding whatis spiritually arresting in . . . contemporary culture (p. 490). With Foucault

THEORY & PSYCHOLOGY 15(6)784

and also Rose we are offered no way of assessing what is valuable and whatis dangerous in the ways psychology works as a dubious science toconstitute its own subject matter, other than the rather empty one of lettingindividuals decide what should contribute to creating their selves as works ofart (see also Taylor, 1984). There is no recognition of the values on whichpractices necessarily operate, and this is why the pure analytic of humantechnologies is inadequate from the hermeneutic viewpoint.

If Foucaults aestheticization of existence can be characterized as sub-jectivist and morally problematic, what resources for evaluating practices,including the ones informed by psychologys operations, are then availableto the hermeneuticists? Traditionally, three non-subjectivist answers havebeen given. First, there is the classical Aristotelian answer that points to anessential, ahistorical human nature that can be either nourished or thwartedby the theories we make about it. Second, there is the rationalist or Kantiananswer that points to rational and universal procedures for securing justice.A significant theory here is Jurgen Habermass discourse ethics (e.g.Habermas, 1993). Although Aristotles conception of practical wisdom hasinspired many hermeneutic philosophers, including Heidegger and Gadamer,and even though Habermas can be counted among hermeneutic theorists,most contemporary hermeneuticists argue that neither human nature (at leastin the essentialist, Aristotelian sense) nor universal procedures of rationaldiscourse can give us the moral criteria required to evaluate the loopingeffects of human kinds. Human nature (as first nature) is too thin to give usguidance, and the Habermasian procedures of discourse ethics are oftenperceived as too rigid, too objectifying and too ethnocentric.

The third hermeneutic answer points to the historical development ofpractices as the source of moral guidance, but in this context, I can merelyprovide the gist of what this answer looks like. We can view Taylors entirephilosophy as a way to articulate this third alternative. For Taylor, goodinterpretations in social science are those that enrich our practices byenabling us to lead better lives. According to which criteria? As Schwandt(2000) says, For Taylor, what counts as better interpretation is understoodas the justified movement from one interpretation to another (p. 202).Taylor would reject Habermass view that we can begin from nowhere everytime we evaluate some interpretation or judgment. Instead, practical reason-ing is always, Taylor (1995a) argues, comparative, or reasoning in transition.We can demonstrate that a given position is superior to a rival, but wecannot show it to be true simpliciter (Brinkmann, 2004, p. 71).

When we articulate the background that shapes our ideas about ourselves,we are engaged in an exercise in practical reason, but one wholly unlikemodernitys instrumental reason. Articulating the background may open theway to detaching ourselves from or altering part of what has constituted itmay, indeed, make such alteration irresistible; but only through our un-questioning reliance on the rest [of the background] (Taylor, 1995b, p. 12).

BRINKMANN: HUMAN KINDS AND LOOPING EFFECTS 785

Pilar Palacios

Pilar Palacios

Pilar Palacios

Pilar Palacios

Pilar Palacios

Pilar Palacios

Pilar Palacios

Pilar Palacios

Pilar Palacios

We inevitably criticize our self-interpretations from within certain other self-interpretations that exist as parts of historically evolved traditions. WhenFoucault recommends that we create ourselves as a work of art, he forgetsthat for something to count as a work of art, a society and a tradition of artappreciation are required with certain normative standards. Obligationstowards this society and its social relations precede the individuals ownfreedom and self-formation since such relations are what make situatedindividual freedom possible in the first place (Taylor, 1985a, p. 209). Inaddition to pointing to the necessary reflection on our cultural embedded-ness, Taylor has offered a number of other ways of establishing ways ofevaluating interpretations, for example when we try to understand andevaluate the self-interpretations of another society. In this case, we shouldwork out a language of perspicuous contrast in which we can formulateboth their way of life and ours as alternative possibilities in relation to somehuman constants at work in both (Taylor, 1981, p. 205). This is close toGadamers notion of a fusion of horizons, and Taylor believes that this willpermit us to rightfully criticize the practices of other cultures, as well asthose of our own.3

Conclusions

I have attempted to answer two questions. First, what does it tell us about thenature of the subject matter of psychology if the discipline is involved inmaking up people? Second, how should we understand the processesthrough which such making up occurs? In response to the first question itwas argued that it tells us that psychologys objects are human kinds, whichinteract with their classifications and are able to exert influence on them-selves. Unlike natural kinds, human kinds demand social discursive contextsand relevant fields of description in order to exist. In response to the secondquestion, I investigated two different ways of answering it, the Foucauldianand the hermeneutic. Below are listed the main features of their answers, andhow they frame the workings of psychology.

Foucauldian perspective: Psychology has come to constitute our sub-jectivities through a complex historical power play, which has tied us to theproject of our own identities in the hands of psychological masters of truth.Rather than articulating what we are, we should problematize and refusewhat we are (Foucault, 1994b, p. 336), after having realized that the self isnothing else than the historical correlation of the technology built in ourhistory. Maybe the problem is to change those technologies (Foucault,1993, pp. 222223). We should face the fact that psychology is a techne thatproduces human subjects, and, in trying to change the technologies thatshape us, we should embrace an aesthetics of existence and createourselves as a work of art.

THEORY & PSYCHOLOGY 15(6)786

Pilar Palacios

Pilar Palacios

Pilar Palacios

Pilar Palacios

Pilar Palacios

Pilar Palacios

Pilar Palacios

Pilar Palacios

Pilar Palacios

Pilar Palacios

Hermeneutic perspective: Psychology has come to shape our backgroundunderstandings (of what is meaningful, of what is worthwhile, of what isvaluable, of what we are) and thereby come to influence our self-interpretations. Today we therefore pre-reflectively relate to ourselves andothers in psychologically informed ways. We imagine life in psychologicalterms. The space of possible intentions and actions has been significantlyaffected by the new fields of interpretation opened up by psychology. Byarticulating the ways in which psychology has informed our backgroundunderstandings (and thereby formed us as agents), we are engaged in aprocess of practical reasoning, namely in investigating if the values thatguide us are worth being guided by.

From the hermeneutic critique of Foucaults physicalismthat it doesnot take into account that human actions and practices are guided by realnotions of valueI have argued that although the Foucauldian analytic isrich in its critical potentials, it is possibly defective as a basis for aconstructive approach to develop psychology as a morally responsiblescience. In this light, Foucaults approach is morally subverting, and, in aparadoxical way, comes to support some of those aspects of modern socialscience that it sets out to criticize: namely its commitment to a kind of value-neutrality. The Foucauldian problematization of social practices leads to thenotion, at least as perceived by Taylor, that values are nothing more thansubjective projections ad modum Nietzsche. Foucault is beyond both struc-turalism (objectivism) and hermeneutics, as indicated by the well-knownbook by Dreyfus and Rabinow (1982), for he was deeply suspicious of anyperspective that would posit the human being as a detached, meaning-givingsubject (such as phenomenology). However, as Dreyfus and Rabinow spellout, ontological hermeneutics gives up the phenomenologists attempt tounderstand man as a meaning-giving subject, but attempts to preservemeaning by locating it in the social practices and literary texts which manproduces (p. xv). This is in contrast to the Foucauldian strategy ofproblematization that does away with meaning and value.

If we return to the discussion of Hackings contribution to the philosophyof human kinds, we should notice what he says about the very purpose ofstudying the looping effects. In response to a question about whether ourbeing aware of looping effects can provide a condition for making the questa moral one for the betterment of humankind, Hacking (1995b) answered,Yes, this is the expression of a noble hope, and he added: [T]he loopingeffect is not necessarily a bad thing. Liberation movements of which manyof us approve are part of the game (p. 394). Human kinds are politically andmorally laden: Human kinds are kinds that people may want to be or not tobe, not in order to attain some end but because the human kinds haveintrinsic moral value (p. 367). I have argued that psychologists couldbecome equipped to understand the moral values of their human kinds if

BRINKMANN: HUMAN KINDS AND LOOPING EFFECTS 787

Pilar Palacios

Pilar Palacios

Pilar Palaciosmrito

Pilar Palacios

Pilar Palacios

Pilar Palacios

Pilar Palacios

Pilar Palacios

they would turn to hermeneutic theory, perhaps after having appropriated thevalid critical perspectives of Foucault and his followers.

Notes1. In psychology, Kurt Danziger (1997) has explicitly referred to Hackings use of

human kinds to analyze such kinds as intelligence, behavior, learning,motivation, personality and attitudes. Danziger (2003) has also referred topsychologically developed human kinds as psychological objects, whose biog-raphies he is interested in tracking. Psychological objects first come intoexistence, Danziger argues, when they are described and classified. The cate-gories and the kinds referred to by the categories grow together. Like Hacking,Danziger is clear that human kinds are distinguished from natural kinds in thatthey are self-referring and intrinsically part of social practice.

2. Foucault (1984) himself argued that subjectivity is constituted in real practiceshistorically analyzable practices. There is a technology of the constitution of theself which cuts across symbolic systems while using them (p. 369). One is notmade a subject through language and discourse alone, as some constructionists,but also deconstructivists, tend to claim. Foucault dismissed Jacques Derridastextualization of practices and the Derridean claim that there is nothing outsidethe text. In a response to Derrida, he calls such textualization a little pedagogythat gives its practitioners limitless sovereignty by allowing them to restatethe text indefinitely (quoted from Flyvbjerg, 2001, p. 115).

3. In Explanation and Practical Reason (Taylor, 1995a) a number of other waysare delineated of evaluating competing interpretations as a kind of reasoning intransition. When interpretations or theories are in disagreement, the superiortheory will be the one that makes better sense of inner difficulties than its rival, orpresents a development which cannot be explained on the terms of the rival, orshows that it itself has come about through a move that can be intrinsicallydescribed as error-reducing compared to its rival (p. 54).

ReferencesBrinkmann, S. (2004). The topography of moral ecology. Theory & Psychology, 14,

5780.Coles, R. (1992). Self/power/other: Political theory and dialogical ethics. Ithaca,

NY: Cornell University Press.Collin, F. (1990). Naturlige klasser, semantik og metode i samfundsvidenskaberne

[Natural kinds, semantics, and method in the social sciences]. Filosofiske Studier,11, 724.

Danziger, K. (1997). Naming the mind: How psychology found its language.London: Sage.

Danziger, K. (1999). Natural kinds, human kinds, and historicity. In W. Maiers,B. Bayer, B. Duarte Esgalhado, R. Jorna, & E. Schraube (Eds.), Challenges totheoretical psychology (pp. 7883). North York, Ont.: Captus.

Danziger, K. (2003). Where history, theory, and philosophy meet: The biography ofpsychological objects. In D.B. Hill & M.J. Kral (Eds.), About psychology: Essays

THEORY & PSYCHOLOGY 15(6)788

Pilar Palacios

at the crossroads of history, theory and philosophy (pp. 1933). Albany: StateUniversity of New York Press.

Dewey, J. (1980). The need for social psychology. In J.A. Boydston (Ed.), Themiddle works: Vol. 10. 19161917 (pp. 5363). Carbondale: Southern IllinoisUniversity Press. (Original work published 1917.)

Dreyfus, H. (1991). Being-in-the-world: A commentary on Heideggers Being andtime, division I. Cambridge, MA: MIT Press.

Dreyfus, H., & Rabinow, P. (1982). Michel Foucault: Beyond structuralism andhermeneutics. Hemel Hempstead: Harvester.

Flyvbjerg, B. (2001). Making social science matter: Why social inquiry fails andhow it can succeed again. Cambridge: Cambridge University Press

Foucault, M. (1977). Discipline and punish: The birth of the prison. New York:Pantheon.

Foucault, M. (1980). The history of sexuality: Vol. 1. An introduction. New York:Random House.

Foucault, M. (1984). On the genealogy of ethics: An overview of work in progress.In P. Rabinow (Ed.), The Foucault reader (pp. 340372). London: Penguin

Foucault, M. (1988a). Technologies of the self. In L.H. Martin, H. Gutman, &P.H. Hutton (Eds.), Technologies of the self (pp. 1649). London: Tavistock.

Foucault, M. (1988b). Truth, power, self. In L.H. Martin, H. Gutman, & P.H. Hutton(Eds.), Technologies of the self (pp. 915). London: Tavistock.

Foucault, M. (1990). The history of sexuality: Vol. 3. The care of the self. London:Penguin.

Foucault, M. (1993). About the beginning of the hermeneutics of the self: Twolectures at Dartmouth. Political Theory, 21, 198227.

Foucault, M. (1994a). Governmentality. In J.D. Faubion (Ed.) Essential works ofMichel Foucault: Vol. 3. Power (pp. 201222). London: Penguin.

Foucault, M. (1994b). The subject and power. In J.D. Faubion (Ed.), Essential worksof Michel Foucault: Vol. 3. Power (pp. 326348). London: Penguin.

Foucault, M. (1998). Philosophy and psychology. In J.D. Faubion (Ed.) Essentialworks of Foucault: Vol. 2. Aesthetics, method, and epistemology (pp. 249259).New York: New Press.

Gadamer, H.G. (2000). Truth and method (2nd rev. ed.). New York: Continuum.(Original work published 1960.)

Giddens, A. (1976). New rules of sociological method. London: Hutchinson.Habermas, J. (1993). Justification and application: Remarks on discourse ethics.

Cambridge: Polity.Hacking, I. (1986). Making up people. In T.C. Heller, M. Sosna, & D. Wellbery

(Eds.), Reconstructing individualism (pp. 222236). Stanford, CA: StanfordUniversity Press.

Hacking, I. (1995a). Rewriting the soul. Princeton, NJ: Princeton University Press.Hacking, I. (1995b). The looping effect of human kinds. In D. Sperber, D. Premack,

& A.J. Premack (Eds.), Causal cognition: A multidisciplinary debate (pp.351383 and pp. 384394 [Discussion]). Oxford: Clarendon.

Hacking, I. (1998). Mad travelers: Reflections on the reality of transient mentaldisease. Charlottesville: University Press of Virginia

Haslam, N.O. (1998). Natural kinds, human kinds, and essentialism. Social Re-search, 65, 291315.

BRINKMANN: HUMAN KINDS AND LOOPING EFFECTS 789

Hegel, G.W.F. (1991). Elements of the philosophy of right. Cambridge: CambridgeUniversity Press. (Original work published 1821.)

Heidegger, M. (1962). Being and time. New York: HarperCollins. (Original workpublished 1927.)

Hook, D. (2003). Analogues of power: Reading psychotherapy through thesovereigntydisciplinegovernment complex. Theory & Psychology, 13,605628.

Latour, B. (1988). The Pasteurization of France. Cambridge, MA: Harvard Uni-versity Press.

Latour, B. (2000). On the partial existence of existing and nonexisting objects. InL. Daston (Ed.), Biographies of scientific objects (pp. 247269). Chicago, IL:University of Chicago Press.

MacIntyre, A. (1985). How psychology makes itself trueor false. In S. Koch &D.E. Leary (Eds.), A century of psychology as science (pp. 897903). New York:McGraw-Hill.

Martin, J., & Sugarman, J. (2001). Interpreting human kinds: Beginnings of ahermeneutic psychology. Theory & Psychology, 11, 193207.

Marx, K. (1971). Capital (Vol. III). London: Lawrence & Wishart. (Original workpublished 1894.)

McDowell, J. (1994). Mind and world. Cambridge, MA: Harvard University Press.Polkinghorne, D. (2000). Psychological inquiry and the pragmatic and hermeneutic

traditions. Theory & Psychology, 10, 453479.Polkinghorne, D. (2004). Practice and the human sciences. Albany: State University

of New York Press.Putnam, H. (1973). Meaning and reference. Journal of Philosophy, 70, 699711.Richards, G. (1996). Putting psychology in its place: An introduction from a critical

historical perspective. London: Routledge.Richardson, F.C., Fowers, B.J., & Guignon, C.B. (1999). Re-envisioning psychology:

Moral dimensions of theory and practice. San Francisco, CA: Jossey-Bass.Rose, N. (1985). The psychological complex: Psychology, politics and society in

England 18691939. London: Routledge & Kegan Paul.Rose, N. (1996a). Inventing our selves: Psychology, power, and personhood.

Cambridge: Cambridge University Press.Rose, N. (1996b). Power and subjectivity: Critical history and psychology. In

C.F. Graumann & K.J. Gergen (Eds.), Historical dimensions of psychologicaldiscourse (pp. 103124). Cambridge: Cambridge University Press.

Rose, N. (1999). Governing the soul: The shaping of the private self (2nd ed.).London: Free Association Books.

Schwandt, T. (2000). Three epistemological stances for qualitative inquiry: Inter-pretivism, hermeneutics, and social constructionism. In N.K. Denzin &Y.S. Lincoln (Eds.), Handbook of qualitative research (pp. 189213). London:Sage.

Smith, R. (1997). The Norton history of the human sciences. New York: Norton.Stenner, P. (1998). Heidegger and the subject: Questioning concerning psychology.

Theory & Psychology, 8, 5977.Stern, D. (2000). Practices, practical holism, and background practices. In

M. Wrathall & J. Malpas (Eds.), Essays in honor of Hubert L. Dreyfus: Vol. 2.

THEORY & PSYCHOLOGY 15(6)790

Heidegger, coping, and cognitive science (pp. 5369). Cambridge, MA: MITPress.

Taylor, C. (1981). Understanding and explanation in the Geisteswissenschaften. InS.M. Holtzman & C.M. Leich (Eds.), Wittgenstein: To follow a rule (pp. 191210).London: Routledge & Kegan Paul.

Taylor, C. (1984). Foucault on freedom and truth. Political Theory, 12, 152183.Taylor, C. (1985a). Atomism. In Philosophical papers: Vol. 2. Philosophy and the

human sciences (pp. 187210). Cambridge: Cambridge University Press.Taylor, C. (1985b). Self-interpreting animals. In Philosophical papers: Vol. 1.

Human agency and language (pp. 4576). Cambridge: Cambridge UniversityPress.

Taylor, C. (1985c). Social theory as practice. In Philosophical papers: Vol. 2.Philosophy and the human sciences (pp. 91115). Cambridge: Cambridge Uni-versity Press.

Taylor, C. (1989). Sources of the self. Cambridge: Cambridge University Press.Taylor, C. (1995a). Explanation and practical reason. In Philosophical arguments

(pp. 3460). Cambridge, MA: Harvard University Press.Taylor, C. (1995b). Overcoming epistemology. In Philosophical arguments (pp.

119). Cambridge, MA: Harvard University Press.Taylor, C. (2000). Whats wrong with foundationalism? Knowledge, agency, and

world. In M. Wrathall & J. Malpas (Eds.), Essays in honor of Hubert L. Dreyfus:Vol. 2. Heidegger, coping, and cognitive science (pp. 115134). Cambridge, MA:MIT Press.

Taylor, C. (2004). Modern social imaginaries. Durham, NC: Duke UniversityPress.

Winch, P. (1963). The idea of a social science and its relation to philosophy.London: Routledge & Kegan Paul. (Original work published 1958.)

Svend Brinkmann is a PhD scholar in the Department of Psychology atthe University of Aarhus, Denmark. He is working on a research projectentitled Psychology as a Moral Science, and his research interestsinclude the ethical and political presuppositions and implications ofpsychological knowledge, moral psychology, and the pragmatist andhermeneutic traditions of inquiry. Address: Department of Psychology,University of Aarhus, Jens Chr. Skous Vej 4, 8000 Aarhus C, Denmark.[email: svendb@psy.au.dk]

BRINKMANN: HUMAN KINDS AND LOOPING EFFECTS 791