Attachments, you say? … How a concept collectively emerges in one research group

Can we interrogate a concept in a manner that is compatible with what the concept itself conveys, while also remaining connected with the specific site in which the concept is being used? This essay undertakes such an exercise with the notion of attachment, as it emerged at the Center for the Sociology of Innovation (CSI) in France, starting in the 1990s. Having played a role in the history I am going to narrate, I will use self-analysis as a resource, even if a somewhat equivocal one. I will explore the virtues that we have collectively lent to the term attachment over the years. To do this, I will contrast three moments when we used it with different orientations at the CSI: to examine economic performativity (Callon); in the critique of emancipation (Latour); and in questioning amateurs’ and caregivers’ activity (myself). I will then briefly address how this relates to sociologists’ necessary commitment in worlds ‘still in process of making’ (W. James).

What keeps us attached?

¹What keeps us attached? Unfortunately, there is nothing like an accident, a separation or the obligation to choose to bring us to an awareness of what carries the real weight among the things, people or situations that we believe we love. We can only be certain of the answer to the question when an experience generates evidence of it. We can not know what keeps us attached without engaging in trials and putting our objects of attachment to the test. No need to find reasons, no need to define criteria: we can feel a loss without knowing precisely what has been lost. No dramatic tension need be invoked either, since these things to which we are attached present themselves in and among more neutral choices too – which wine to drink with this, which music to play when tired, which friend to phone in morose moments … What is more, we can hold on very strongly to something that does not matter to others in the least (‘it’s nothing’, ‘it’s not important’, as we say casually without truly believing it ourselves).

Attachment cannot be registered in terms of causes or intentions, structures or determinations. It does not belong to the vocabulary of action, even if it can be grappled with through this language. Attachment is more like a liability, in the sense the word has taken on in business: an obligation from the past that is brought to bear on the present. Except that – and this is the point – when it comes to attachment no one can do such an accounting. There is no possible equivalence that can be drawn out of this list of incommensurables: between nostalgia for a country; one’s political sensibilities; the way we go about getting into our work or of having breakfast and a devouring passion for rock or opera; the Sunday football game with old buddies or the planet that needs to be saved … The justifications for attachment only come when they are called upon – to defend a taste, a practice or a habit; to share it with others; to oblige ourselves to renounce it; to weigh what it costs to maintain it, ‘live with’ it, or abandon it. Its contingent or occasional appearance, which may seem at first to contradict its imperious nature, is another paradoxical feature of what it means to be attached. Since they must be put to trial and the opportunities for testing are rarely continuous, attachments depend strictly upon circumstances in the strongest sense of the word.² But the fact of being casual, dependent, and undetermined does not prevent them from reappearing insistently in a variety of related manifestations.

An action without actors

What attaches us, and to what are we attached? This question has other merits that follow from the preceding characteristics, but which were less apparent in ‘What keeps us attached?’, the first and somewhat punchier formulation of the question. As we have now grasped step by step, attachments are not only indeterminate, constraining and situational, but they are also symmetrical. Do we hold onto something or does something hold onto us? Beyond such reciprocity, which would remain profoundly dualist, the connections to the diverse natures in which we are entangled, which hold us and hold us together, call for a redistribution of agency³ to be deployed in the interlacing, where every connection does something, but where none is sufficient on its own. As Bruno Latour (1999a) has noted, the idea of attachment immediately implies a challenge to the notion of causality, to the benefit of a much less tidy set of interactions: fumbling, scraping up against, mutual coaching. Instead of clear distinctions between determined and determining things, we pass to the continuity of a less trenchant but infinitely more productive form of distributed action disseminated in these networks.⁴

Before drawing to a close this wholesale unwrapping of traits there is one other somewhat forgotten dimension of this expression, attachment. Attachment signifies a connection, restriction, restraint and dependence that reminds us that we are prisoners, confined on all sides by our history and our environment. But the word has been unburdened of all negative connotation. These meanings have by no means been able to detract from its positive value, be it emotional or moral. In its most common meanings, attachment first invokes the relationship of a mother to her young, in the most corporeal dimension of bodily contact, dependence and protection, and love for another. This is the attachment that psychologists have endlessly theorized (e.g. Bowlby 1969, 1973, 1980). The other more frequent usage of attachment relates to profound convictions and strong engagements. It is the attachment to values like freedom, invoked in the footnote above. Indeed, even if in a much less exclusive fashion, less precise words such as bond or tie can also, without taking on any false meanings, teeter from a negative idea like constraint or restriction to an idea that is not just positive but is cherished and revered, as if it were charged with all the richness of the things that are associated with it: bonds to our nearest and dearest, to places, to memberships, to origins – however in ‘attachment’, the ‘-ment’ says that it has to be made, even if it makes us in return.

Inversely, attachment has no reason to limit itself to things that are agreeable, grand or sublime. It is not in the content of what happens to us that it takes on its positive value but in the way it is experienced. I would go so far as to say that, to the contrary, it is with regard to difficult problems – such as destructive behaviours or painful situations, states of weakness or vulnerability, disability, chronic illness, alcoholism, drug addiction, Alzheimer’s or simply the process of aging and decline – that speaking in terms of attachment generates the most vigorous alternative forms of understanding (Hennion & Vidal-Naquet 2015). This is precisely because in turning existence into a long series of tests in the more technical and less affective sense that I just gave to this word, these problems, difficulties or weaknesses impose and oblige a turn towards pragmatism. These problems impose an abandonment of the paralyzing oscillation between medical definitions and social definitions in order to concentrate attention on the constitutive dimension of what it is, in fact, to be in these conditions. We carry these conditions with us, but in spite of our own bodies we must also, during every new ordeal or trial that tests us, get ourselves used to our own state while reciprocally rearranging, taming and ceaselessly and regularly remaking this state into our own.

The body endures the test

The attachment is what gets experienced.⁵ The experience requires a body. It is fortunate that bodies, having returned to the forefront of the social sciences, are in vogue. Far from being a distraction that steals the analytic scene, this presence raises other questions. What is this body that is tested, feels, works and is worked, endures and responds, and forges itself through experiment as in the body of the athlete, the patient, the amateur, the handicap, the pianist? At once the body that each one is and the body that each one has, as Merleau-Ponty strikingly reminds us,⁶ is the body that we cannot escape, but that at the same time is a support and a medium that reacts. It is not any more of a given than objects or states of being. It is also defined, to the contrary, by this strange already–evoked mix of indeterminacy and resistance. The body is decidedly made of exactly the same material as the attachments that weigh on it.⁷ No balance sheet can stop the body’s deployment. It is entirely tributary, flowing forth from the contacts it makes. Bodies are stubborn and persistent, but their suppleness and reformation is continuous.

Let’s take a little step back. In my work, insisting on the corporeal engagement of the amateur was an endeavour to render patience to the passionate (Hennion 2007), following in the footsteps of Science & Technology Studies (STS) and Actor-network Theory (ANT), which also give their objects agency. The point was not to satisfy some kind of lacuna in these analyses by opening onto the ‘subject’ in the way that they had already opened onto the object. No, the subjects are already there in ANT, taken into the networks that they perform, distributed in the devices that render them active, linked by the associations of which there are at once traces and tracers. What all of this vocabulary demonstrates is that the term attachment is far from being a novelty to the Centre for the Sociology of Innovation. It has arisen progressively, starting from parallel but distinct considerations, and out of research done by several of us. I would go so far as to say that it very aptly collects the CSI’s history, notably with regards to the limits of ANT’s expression (Law & Hassard 1999). In sticking terms together to illustrate paradoxes, ANT has managed to take up the very terminology of the sociology it combats. This is true not only on the side of the actor, returned to the saddle once his excessive qualities were removed. This is equally true for the network. For everyone except in ANT, where networks refer to an open fabric of heterogeneous connections, the word network means an assembly of closed and homogeneous pipes, whether it be for gas or social networks.⁸

Attachment/detachment

The word attachment made its first appearance at the CSI as a theoretical expression in 1992, in a document authored by Callon (1992, p. 62). The paper dealt with theories of markets and techno-economic networks, criticizing the misleading image of producers and buyers always meeting face-to-face. Callon took up Chamberlin’s radical thesis on monopolistic competition (1933) in full force – a thesis that has frequently been weakened when it has been assimilated with imperfect competition, where in fact it means the exact opposite. Chamberlin inversed the ordinary vision of the market, a move that was very much within the spirit of pragmatism.⁹

Pragmatism enacted an analogous reversal between relations and their terms: each transaction is unique, it has to be reproduced each time, in its irreducible singularity.¹⁰ It is not the equilibrium of supply and demand that constitutes the law, while the reality of markets is to be ‘imperfect’. To the contrary, what can never be but instable compromises is the possibility of partial substitution, of adaptations that permit repetition, of standardizing products and needs, of duplicating purchase situations. And nothing but those compromises, difficult to achieve, ceaselessly revisited, permits the achievement of transactions which are in every instance singular.

Through this vision of markets we can better understand the procession of intermediaries that is constantly being extended – designers, marketers, pollsters, publicists, merchandisers, etc. – while their increasingly sophisticated tasks remain unintelligible under the law of ‘the’ market which only needs to be applied.¹¹ We can also understand that the cumulative efforts of these various intermediaries are aimed at enhancing the particularities of transactions, and not towards rendering them all similar and transparent, as though all that has to happen is to spill the preferences of the buyer contained in a vase on one side over a connected list of disposable goods on the other.¹² No mechanism of the sort exists. It is to describe the permanent work of economy that Callon referred to attachment and detachment: we perpetually depart from a tangled ensemble of links, within which preferences and products are being mutually defined from moment to moment. The best sellers can do is to lightly displace these configurations in their own favour, after having tried to make configurations appear to clearly suggest certain choices, that is, by unknotting certain links and tying others together. This analysis led straight to the more specific notion of framing-overflowing, a sharp criticism of embeddedness, and the theorization of the economic discipline as being performative of the real economy (Callon 1998, 2007).¹³

Sensing and valuing

It is true that attachment has lacked the body, not in principle but due to the terrain and the objects of inquiry privileged by Callon, namely science, techniques and markets. For the same reasons, the term attachments has lacked morality and politics, which are the norms, wills, intentions, values that economy is so good at keeping at bay.¹⁴ As we have seen, Bruno Latour also took up the word attachment, in 1999. He contrasted it to the detachment and emancipation that he assimilated to the cult of autonomy. He used it to denounce the injunction to obey only one’s own law: autonomy has become the only law of the moderns. This time the proposition was quite polemical, saturated with morality and politics. It envisioned a new programme in which the essential was not to liberate oneself but to sort the good attachments from the bad, by leaning not on grand overarching principals, but ‘the justice immanent in things’ (Latour 1999b, p. 25). But how can we do this? Trapped in Latour’s long-term battle against modern dualism, the text’s foremost aim was to detach itself from the previous divisions, from the philosophy of liberty and sociological determinism. How can we decide upon the quality of attachment? And how do we nonetheless detach ourselves from an attachment, if there are some that are harmful or less good than others?

This reference to attachment finds its provenance in a seminar at the CSI in 1999, on a thesis in progress. The session on Émilie Gomart’s particularly engaging research at the Blue Clinic (1999) had a very unusual effect of collective revelation. The mission of the clinic, located in a Parisian suburb, was to treat drug addicts with methadone. Obliged to find the dosage that was suitable to each situation, the personnel navigated in plain view on a case-by-case basis with regards to effects and reactions of the substitute. They formatted their decisions girdled by practical regulations, which were immediately revised when faced with the urgency of any particular case. We saw a similarity between this very dark and worrying case and that of amateurs whose activities are typically valorized; both of which could be continuously described with a related vocabulary. Hence the startling effect of the comparison made in the seminar, later elaborated in a joint article (Gomart & Hennion 1999). The description of work being accomplished under maximal constraint by clinic personnel clarified the grammar of attachment, while also giving it more weight.

This is a strange grammar indeed, perpendicular to the active/passive axis, difficult to put into words. My work with Geneviève Teil on taste explores the ways in which the amateur ‘has himself love’ things (Teil & Hennion 2004; Hennion 2015b). This awkwardly attempted to find an expression of such a grammar of taste: so, in the thread of « faire faire », ¹⁵ we used therefore « se faire aimer », or « se mettre à aimer » ¹⁶ (‘to have oneself love something’, and reciprocally ‘to assist things to express themselves’). In fact, we had to resort to many other laboured turns of phrase, like ‘to put oneself in a condition to be actively affected by’, or ‘to let oneself go into the grip of the object while continuing to value (in the lovely double sense of the word) it’. How could we signify such a paradox: the fine articulation of attentiveness and fastidious control (in things, even when these go unrestrained) with abandon and zeal (which is itself immediately re-restrained by evaluation, narratives of experience and appropriation into stories)? … Attachments, like these formulations, are heavy and circular. They play outside of the linguistic dualism between active and the passive and between subject and object, which is best expressed by words such as pleasure or passion.¹⁷ Nobody understands these to be about passivity; they are instead kinds of actions to the power of two that target their opposite: overflowing boundaries and possession by an object.

Substitutes and displacements

The richness of the amateur’s practice, which encompasses all of this, suddenly held us as in a mirror the image of the terrible dependence of the drug addict. What keeps us attached? … Far from ceding to an irresponsible complacency, by emphasizing the traits that are common in the opposing pleasures of drugs and music, we draw from the clarity that the sequence connection-hold-detachment-substitution-reattachment provides. We forbid the recourse, altogether too easy, to a posture of choice and mastery over situations by autonomous subjects in the case of the amateur, and of mechanical dependence within the empire of things in the case of the drug user. Humans do not seek to untie themselves from their connections. They seek to live through them – or to die by them. Neither the amateur nor the drug addict seeks to ‘liberate’ themselves or to be their ‘true’ self.¹⁸ The descriptive weakness of an analysis of these behaviours made through the lens of action was strikingly obvious to us.

Likewise, we grasped the obligatory pragmatism of the addicts’ chaperones. They were forced to depart from the present state of things. They operated by trial and error to facilitate displacement and to permit not a liberation, but partial substitutions. They were able to judge their own actions only from the effects that were produced. This description resembles that of a trainer or a professor. There was a double lesson: of not postulating liberation but substitution, and of a moral programme linked to a series of unforeseeable tests whose results were altogether unknowable to anyone not committed to the project. With much greater acuity we also pointed to the necessity of drawing distinctions between various cases, even if most of them were pressing. Nevertheless, answers could only be determined in the situation, through trial and error. Decisions were made in the heat of the moment as to what intervention would be. This was a unique manner of proceeding, departing from bonds, testing them to detach from some while reinforcing others. The drug addict was living in negative what amateurs live in the positive: not a series of mastered choices flowing out of a world of stable objects, but a trajectory made up of withdrawals and retractions of the self, crafting little by little, experience after experience, bodies that were different.

It is easy to see retrospectively that the delicate situation Émilie was describing, for the addicts, the personnel and herself, provided us with a laboratory case for raising questions that ANT was leaving in the dark, but which it gave all the resources to take up in again in a fresh way. This contrasted with the amoral tunic that both philosophical adherents of imperative categories and critical sociologists, each wielding a dualism attacked by Latour, were zealously attempting to throw over ANT: questions about the body, about what it experiences and what it reveals, about the moral value of practices, the political sense of an intervention like that of the social workers at the clinic, about the place of sociology in this uproar, and about an activity which is revealed to be permanently in a state of testing and constant adaptation. And such a state makes its moral dimensions ever more present, or more pressing.¹⁹ These problems cannot be dealt with by creating ethical committees or by regurgitating a charter, but rather by inventing little procedures, rules and controls, by creating time discipline and scheduling doses, by learning in the moment to feel where the limits lie … 

An obligation towards commitment was also an obligation towards pragmatism, and reciprocally an obligation towards pragmatism was an obligation towards commitment! With attachment, the Blue Clinic re-bestowed morality unto ANT – not an overbearing morality that insists on being applied by cutting itself out of the very situations in which it was needed, but an attached morality of examining trials in which action and the criteria for action were found within one another from within the fibre of existing ties, an attachment that risks the production of a discrepancy (Stengers 2006) and attempts an uncertain progression.

The immanent quality of things

My own work on attachment has begun with the experience of amateurs and pursues this interrogation on the immanent quality of things, which has been examined in the raw by following the drug addicts’ urgent and sombre case. There is a double necessity of thinking about the relationships between bodies and things that precedes the subject–object dualism, and of thinking about situated evaluation that is internal and not external to experience and occurs over the course of testing. The first step is returning the body of the taster, treated as an accomplishment, to the scene. In comparing wine and music but also in comparing more far–flung objects like the climber’s canyon, the footballer’s field, or even the singer’s voice, we effectively capture all the delicate work that must be deployed, on the self, on one’s body, and one’s spirit (Hennion 2016a). That work has to be done both in the moment and over the long term in order to become sensitive to the differences that matter. This motion has a double meaning. As much as it is a form of bodily work, taste is about working up objects to make differences in them emerge, which reciprocally can only come across to those who sense these differences. Attachments through and through – but from the point of view of the amateur who knows this better than anyone else, the paradox is inversed: what strikes her is that things are not any more given than bodies.²⁰ To those who are not interested, the trajectory of a balloon, the asperities in a rockface, the swing of a racket, the pitch of a voice, are nothing more than chance occurrences in an inert world, totally uninteresting. Symphonies, matches or wines, it’s all the same thing. Whether it is hip-hop or French lute sonatas, repertoires are simply boring catalogues. Adorno once said that to love music, one must already love it. But there are two ways to interpret the indifference of things that an indifference to things provokes.

This first is to take this necessity of having experiences as a problem of knowing or of being ignorant of objects which exist nonetheless, for which we ‘have’ a preference or not. Then, the moment of contact, the indecisions that occur due to scoring a goal, being surprised by a wine, or touched by the sound of music are not but secondary effects for which the principle must be sought elsewhere, such as in education, milieu or social games of belonging. Again, taste is seen as a simple act of adherence to a culturally coded practice. We return to this interface: on one side there are objects and their properties, on the other, subjects and their preferences, and the question boils down to a four square table: tell me who you are and I’ll tell you what you appreciate. This assures us in passing of the distribution of the disciplines – those that should occupy themselves with objects, and others that are adapted to tastes. The matter is much simpler. The problem has slid from a taste to be produced and that produces in the same gesture a repertory of objects and its amateurs, to a system of identities that correspond to a set of things. This is the reappearance of the dualism between needs and things that Callon was criticizing in the case of economy.

In a second, opposite way, we take the co-construction in forming tastes seriously. The word ‘taste’ itself says as much. It indicates the taste of things just as well as having a taste for things. Neither of the two is a given, but they discover each other in the act of tasting/testing (Hennion 2007). The process of making oneself appreciate is indissociable from the process of making things arrive. Making things happen, deploying their pertinent differences at the same time as we deploy the capacity and the pleasure of feeling these differences. To think about the activity of the amateur as an attachment is to insist on the fact that it is work done out of a tightly woven tissue of individual and collective past experiences which have built up the sensibility of the amateur, little by little. These experiences have defined a domain filled with objects and nuances, of know-how and repertoires, of criteria and techniques, of common histories and contested evolutions. Thinking in this way therefore means that the activity-sensibility of the amateur is caught in tight linkages, without which it would have no consistency. This activity-sensibility does not remotely resemble the free choice in the great supermarkets of disposable goods.

At the same time, it means that far from dampening the experience of tasting or rendering it repetitive, the weight of these interconnected linkages enlivens taste. These linkages give the performance of taste a chiselled precision that permits the most subtle criteria that define its quality for this one or another to be examined in an open and polemical fashion. This demands that tastes be re-enacted every time, as the word performance, common to sports and to art, aptly indicates. By the same token, this makes it possible for taste to constantly produce something unexpected. Outside of this understanding we would be hard-pressed to explain why football fans return to see matches, why music-lovers go to concerts or why wine enthusiasts try wines other than the ones they already know they love. To enter into the experiences to which amateurs submit themselves each time is demanding: it requires them to return to the condition, sharpen their expectations, mobilize their attention and reinvest in the object. Without this, the object itself would have no chance of arriving in the strongest sense of the term. It unlocks a presence that must be conquered each time. This is the condition under which the performance unfolds with its uncertain rhythm. And sometimes bountiful pleasures get unleashed.

Conclusion: a necessary involvement in worlds ‘still in process of making’ (James)

Finally, using the word attachment amounts to advocating for social inquiries made on sensitive matters and things that count for people. This requires making choices, fighting against other choices and engaging in favour of actions made in order to make things better. Attachments do not belong to people nor define them: depending on situations, forging their existence and history through debates and confrontations, they have to be continuously done and re-done. They both appear as a constraint and a resource to people confronted by destabilizing problems. I developed the notion of taste, amateurs, and then care in order to make sociology recover a capacity to better speak of matters of concern, be they objects of passion or fragile beings. Amateurs and caregivers have a lot in common. In care as in art, every case is different, every situation is particular. The smaller detail may at any time engage things in a bad way: it is all the truer as people are dependent, vulnerable and their tense, often painful state makes all issues delicate.²¹ What interests me here it that so doing, through inquiry, attention, gestures and reflexive returns to what just happened, both amateurs and carers develop an art of making their objects of care exist more − be they things or beings. At the same time, in spite of the many delusions encountered, they are themselves enriched, enhanced by such moments.

This had led me to criticize the ‘sociologism’ of my own discipline. Taste, music, the love for things, the care for people – these are not only a matter of social differentiation or a game of power. They support and enrich both the objects and subjects of the elaborated, debated and instrumented attention they require. Sociology systematically understates actors’ objects and concerns. Its inclination to conceive of ‘the social’ as an autonomous order – and thus take social science for a science of ‘the social’ – amounts to a misunderstanding and misleading scientism (Latour 2005). Either objects are seen as natural and irrelevant to social science, or they are reinterpreted as undisputable values, cultural signs, social markers or even ‘illusions’.²² Positivism, structuralism and critical thought, beyond deep oppositions, share the same request to take distance. This results in an insistent reluctance to acknowledge people’s competences and abilities in producing their own worlds, instead of only being ruled by them without knowing it. The main difference is that problems and proposals arise from working together on unpredictable situations, not from a predefined frame of analysis.

In contrast to the words taste and preference in the case of amateurs, or to norms and dedication in the case of caregivers, the word attachment better expresses our ways of both making and being made by the relationships and the objects that hold us together. This may be the core of what attachments help address: how do we build a social world with and within our objects of concern, of care, of passion– and not in spite of them?

Disclosure statement

No potential conflict of interest was reported by the author.

Notes on contributor

Antoine Hennion is a Professor at Mines ParisTech and the former Director of the Centre for the Sociology of Innovation (CSI, 1994–2002). His research in the sociology of music and culture focuses on the cultural industries, advertising and design, mediators, services and users. At the CSI, he promoted a problematization of mediation crossing cultural sociology and STS. He is now developing a pragmatist approach to diverse forms of attachment, from taste and amateurs’ practices to issues about care, aging and disability, and migrants.

Notes

1. The starting point of this text is Hennion (2010). A special thank you goes to M. Poon and D. Schleifer for their precise and precious remarks on a first English version.

2. Fortunately, our attachment to freedom only gains substance when it is being threatened.

3. A long-term debate, to which ANT scholars grimly contributed (Callon & Law 1995; Law 2003 [1992]; Latour 2005).

4. French uses rich ways of getting away from the binary oscillation between active/passive, such as the reflexive form (« il se produit ceci ») or the double infinitive, like « faire faire »: « il a fait jouer l’orchestre ». This implies no idea of an authority (as in English, ‘he conducted it’), but rather a distributed agency; see other intersting examples below.

5. « Éprouver » in French means both to try and to feel, like ‘to experience’ in English, but even more strongly: to put to the test and to be afflicted.

6. Merleau-Ponty (2002 [1945]), Part One: ‘The Phenomenal Body’.

7. « De la même étoffe », says James (in French; maybe we should say ‘of the same fabric’!), in « La notion de conscience », 1905 (ed. in James 1976). About James’s pragmatism and attachments, see Hennion (2015c).

8. As Latour (1999b, p. 21) has humorously remarked, in ANT even the hyphen is questionable. On Latour and the CSI in relation to mediation and pragmatism, see Hennion (2016b).

9. He did this staying strictly within the framing of economic reason and within the template of supply and demand, which only adds to the beauty of the thesis.

10. Chamberlin would say that when we buy something, we also buy the moustache of the salesman … !

11. In French one would say « la loi s’applique », not « la loi est appliquée ». Such a use of the reflective is doubtless the contemporary trace of the ancient Greek middle form, which precedes the cleavage of the active from the passive (Latour 1999a, p. 21; Hennion 2007, p. 106). Here the pronominal form clearly conveys the idea that the law of the market is valid by itself: in an economic understanding of economics, things only happen because they have to; nobody has to ‘apply’ anything. 80 years ago, Chamberlin said exactly the opposite.

12. Or furthermore, as Catherine Grandclément’s thesis (2008) shows, supermarkets and self-service, which should in theory be emptied of all intermediaries between the buyer and the product, are actually brimming with posters, indicators, stands, promotional aids, gondola heads, point-of-sale ads and a thousand other sales devices that proliferate every day.

13. In line with heated debates in anthropology and economic sociology, Callon also speaks of entanglement/disentanglement. The aim is to make a sociology of economics able to account for the reality which economics performs, by contrast with a critical sociology denouncing the irrealism of economic models, but also with the theory of embeddedness, having society and culture taking the cradle of autonomous economy in their protective arms. The concepts of framing and overflowing and of performativity put to rest the perpetual dualism between the cold calculating and universal character of the market that severs ties, and a warm, local and humanist social reality (Callon & Latour 1997).

14. Is this why his next research project, with V. Rabeharisoa, was on the AFM [the French association against muscular dystrophy], which permitted him to ask questions about the morality of Gino, a father who refused to take care of his child (Callon & Rabeharisoa 1999a, 1999b) or to speak again of detachment and attachment in a more political sense to deal with the sociologist’s commitment (Callon 1999)?

15. Latour borrowed this formula from Greimas (1976), with his actants (‘what does the doing’) which permitted the detachment of the question of action from that of the subject.

16. Again here this curious ‘middle form’ pronominal in French, pointing at a state neither active nor passive.

17. Nobody sees ‘passion’ as being ‘passive’. However, from another viewpoint the word can be somewhat misleading by adding the idea that attachment is an extreme format: this is not obligatory. Passion is but one format among others. Attachment refers just as much to the tastes that are most familiar.

18. Cf. ‘I love Bach’ (Hennion 2015a, p. 6).

19. The fact that local communities were financing the clinic to be rid once and for all of an insoluble problem, while the militant engagement and the past experience of the personnel favoured an attitude of collective understanding towards addicts, put in place a curious situation that might be seen as a risky social experiment, but which was also denounced as an irresponsible utopia – the creation of a pay-as-you-please 1970s California commune – nay, as legalized drug trafficking. As one supervisor bluntly asked, at what moment does providing a drug substitute change into complicity with the user, pure and simple?

20. This is the definition of pragmata given by W. James (1976 [1912]). One can think also of the place G. H. Mead (1932) gives to the ‘physical things’ in the consciousness we get of the world, through their ‘manipulation’.

21. This double character of dependency to the situation and of high sensitivity, particularly salient in the case of care, made us speak of a ‘situational ethics’ (Hennion and Vidal-Naquet 2015).

22. Indeed, Bourdieu points at etymology (‘in ludo’ means ‘en-jeu’): it still takes concerns for just social ‘stakes’.