Introduction

At the end of his Philosophical Investigations, Wittgenstein writes that if a lion could speak, we would not be able to understand it. The dog and horse trainer and philosopher Vicky Hearne remarks, with the wisdom of rigorous impertinence, that if a lion tamer read Wittgenstein, s/he couldn't help but be astonished: “What does he mean? That if my lion Sudan started talking to me, we would stop being able to understand each other?” (Hearne, 1994, p. 168). It is perhaps here that the major difficulty philosophers have with the animal question is outlined. There is always a “we” that imposes itself: an enrolment, a collective capture in a problem whose terms, which are given from the outset, we have to accept.

For my part I feel much more at ease with the sad observation made to the sociologist Jocelyne Porcher by a worker at an industrial pig farm: “personally, I believe that if animals could speak, we would get shouted at every day” (Porcher in Despret, 2007b). I feel all the more at ease as this suggestion takes on significance in contrast to what breeders say in those farms where humans and animals live in relation to each other, have a mutual knowledge and take seriously what breeding means: “we don't stop talking with our animals”. I'll come back to this.

Of course, one could always envisage a simple explanation, setting out – still with Vicki Hearne – down the path of humour rather than that of denunciation: philosophers have an enormous problem of rivalry when it comes to language. Thus, she remarks, “human philosophers tend to talk strangely when the topic of parrots comes up, as if they believe that their stature depends on the diminished status of parrots” (Hearne, 1994, p. 3). “Because”, she continues, “the problem of parrots is that they need to control the exchanges. You want to engage them in conversation, a conversation that rapidly takes on the appearance of an IQ test, and you find yourself led astray by a habit common to all talking parrots: their refusal to let another individual choose the topic of conversation.

You go up to a parrot, and he's probably in a cage and you're not, so you feel pretty superior, maybe you even think you can feel sorry for the parrot, and you ask the parrot how he is, and he says something gnomic like, ‘so's your old man’ or ‘how fine and purple are the swallows of the late summer’. Then the parrot looks at you in a really interested, expectant way, to see if you're going to keep your end up (…). You start trying to figure out what the parrot means by it, and there you are. You haven't a hope of reintroducing whatever topic you had in mind. That's why philosophers keep denying that parrots can talk, of course, because a philosopher really likes to keep control of a conversation” (ibid., p. 5).

The same thing could be said of behaviourists, although conversation would not be at the centre of their problem of control, even less at the centre of what really interests them. The fact that no parrot has ever spoken to a behaviourist testifies to this. The reason for the silence of parrots in this kind of apparatus is simple enough and makes Wittgenstein right, in part – providing one takes into account not the point of view of the human but the hypotheses the parrots seem to forge. In view of the situation suggested to them, they deduce that if they spoke, nobody in that situation would hear them. Because – as later research, having succeeded in the miracle of having a conversation with them, has shown – parrots have a pragmatic rather than a referential conception of language. They cannot speak if they don't feel they are speaking to someone. And this someone is cruelly lacking in the behaviourist apparatus. Under the banner of objectivity, everything is constructed in such a fashion as to render the researcher as impersonal as possible, to make of him a being replaceable by whosoever, which is precisely the contrary of what defines a person.

In the light of the work of Irene Pepperberg, the psychologist who succeeded in making parrots speak and be understood (Pepperberg, 1995, pp. 11–15), another reason can be suggested. Her practice shows us that to teach a being to speak presupposes not only a tolerance of but also a profound interest in misunderstandings. All parents know that the teaching and learning of language requires, at the outset, that meanings be proposed for things that don't necessarily have any. When Alex, the first grey parrot from Gabon to collaborate in Pepperberg's research, inadvertently produced a new signifying sound, the researchers would act as if this sound was intentional and respond to this new act of language as if Alex had wanted to demand something or comment intentionally. The effect of the misunderstanding, of the “as if”, is that a sound produced accidentally can thus become a word that signifies something for the parrot because it has signified something for the researcher. Meanings are constructed in a constant movement of attunement, which makes them emerge. This strategy, which tunes up meanings, which gives them and adjusts them, is inscribed more broadly in Pepperberg's work in an apparatus that redistributes control. The fact that the parrot may or may not have had the intention of producing the new combination is not important, Pepperberg explains, because we simply want to show the parrot that phrases can have meaning and that they can be used to control, or at least influence, one's environment and the actions of those who take care of him or her. Besides, we know that for birds, learning a language cannot be summed up in learning how to sing, but also in learning how the song must be used. Language must be learned/taught in its pragmatic function: it is an effective means of acting and of making others act.

From this perspective and to keep the parrot interested, the researchers gave it control of its rewards. A correctly named or appropriately described object would be offered but sometimes Alex wouldn't want it, and would prefer something else: a titbit or to go for a walk. The researchers would comply with good grace. The reward, then, translates for Alex as the right to “want” and to take a position in relation to what is offered to him. For the scientists, this redistribution of control and this more equitable apportionment of “wanting” translates the success of their apparatus: Alex testifies to his capacity to accomplish tasks that were hitherto considered as exceeding the capacities of non-humans. Not only would he speak, describe, count, classify objects in abstract categories, and use concepts like “same” and “different”, but he was also able to use speech so as to influence the behaviour of others: “come here”, “I want to go to that place”, “no”, “I want this”. By restating and inverting the question of control – no longer a solution requiring purification but a problem to be negotiated – the laboratory authorized a superb exchange of properties between the researchers and their subject.

Consequently, one can understand how a conception of objectivity, defined as the refusal to suggest the meanings perceived momentarily by oneself alone (this corresponds to one of the possible figures of subjectivity), has rendered the learning of speech by parrots impossible. An infant doubtless wouldn't have managed any better. More generally, one should consider that the manner in which behaviourists consider language – and the regime of proofs that testify to one's competence to use it – is lacking one condition essential to permitting linguistic exchange: the possibility of misunderstandings. Without this possibility of continually adjusting and negotiating what is understood, it is the understanding itself that is compromised,Footnote 1 because exchange, pragmatically speaking, can only be achieved when there is a continuous reprisal of translations and betrayals of meaning. If Sudan the lion could speak, would this “we” of “we understand [each other]” still be possible? Constructed with her tamer through the invention of a dialect,Footnote 2 a constant attention towards the other, through doubts and vital risks engaging them both and above all through the possibility of not understanding everything, they would both be referred to another mode of existence.

By identifying language use with modes of existence, one might rediscover the celebrated response that Daniel Dennett opposed to Wittgenstein, arguing that if a lion could speak it would no longer be lion enough to teach us anything on the subject of “lioness”. A priori, Dennett's response might have no interest other than that of demonstrating once again the lack of seriousness with which philosophers treat the question of the “we”. A lion that was able to speak would not be able to do so in the name of the “we” of lions. This is a singular decision. Certainly it rests on a burdensome conception of the naturalness of animals, which explains the decision without being able to justify it. The mode of existence of lions is subordinated to that of an essence “lioness”, guaranteed by the identity of the species and the stability of its repertoire of behaviour. But let's leave that problem, which in the last instance only translates some tired old reflexes faced with the menace of the blurring of the frontier between humans and animals.

However, the question could become interesting if it were posed concretely from the point of view of the lions, not in an abstract and general manner. Would a lion that speaks still be recognized as a lion by its conspecifics? What matters, from the point of view of a lion, to make it say to another lion “you are still one of us”? This is the problem that researchers attempting to reintroduce captive animals into their natural milieu encounter every day, as they are confronted with unbelievable difficulties in reintegrating them into social groups, as we see with chimpanzees. But if this is indeed the way in which certain ethologists or primatologists can pose the question – and I'm thinking here of the manner in which Shirley Strum proposed to interrogate her baboons by subordinating her questions to what matters for the baboons – one will notice that Dennett approaches it differently: we must ourselves be the judges of what it is that ensures that a lion has something to teach us on the subject of what it is to be a lion. That is to say, we have slid surreptitiously from the question of representing to the question of the representative.Footnote 3 As a consequence of this, the “we” that designates us as “teaching us something on the subject of lioness” becomes totally abstract, in the same way that the obvious fact that lions can teach us nothing if they are not well represented by the persons that the lions make speak disappears.Footnote 4 We will know nothing about lions if we follow Wittgenstein and Dennett because the latter are evidently not pertinent spokesmen for lions, of what makes a world for them, what makes them active, what frightens them, what makes a “we” for them when they hunt together, even when they participate in a circus act. The question is not what is a lion, but “how does one become a lion”, not only in the lion community, not only in the lion's species, but also “how does one become a lion” in the work of scientists constructing what it is to be a lion. This is a question of becoming: of that of which the animal is rendered capable by the apparatuses that interrogate it, by the narratives that guide these apparatuses, by the hours of work spent observing them. It is sufficient simply to compare what we knew about baboons 30 years ago and what we can say about them today.Footnote 5 Scientists have found it necessary to properly learn from them, to pass from the question of the representative animal (which, for many reasons and in many senses, was the baboon of the initial research) to that of the good representer: how can what I say about lions or baboons be authorized by them?

To return to those animals whose entry into language makes philosophers so uncomfortable, the appropriate response to this question is the one that best testifies to the success of Pepperberg's project with Alex. It is the question that the researcher poses at the end of her work with the parrot: what does Alex really authorize me to say about parrots? I cannot, she says, affirm that all parrots talk, nor that all the grey parrots of Gabon talk. Alex is not representative of parrots; no parrot could be. The givens appear to us instead as a means of sketching out the competences that can, with the appropriate environmental support, figure in the list of capacities of the species. Here then is not what parrots are but what they might be rendered capable of.

This rendering capable at the same time indicates what is at work here: Alex talks because Pepperberg desires it and demands it of him, and because she was able to subordinate her desire to what makes sense for Alex in the matter of speaking. She was able to negotiate with Alex over what in speech could interest him.Footnote 6 Alex talks because for diverse reasons his desire overlaps with that of Pepperberg. In other words, Alex doesn't talk in the name of a “we” of parrots successfully imposed by scientists, but in the name of a “we” constituted by the assemblage of a parrot and human beings equipped with an apparatus aimed at making the parrot talk well.

Is this to say that one might have to renounce all generalization, compromising one of the requirements for doing science? Generalization is always possible, but it is constructed in another way: it is constructed bit by bit. In 1990, Kyaro, a young male aged three and a half months, and Alo, a seven-month-old female, came to join the team. Interrogated under the same conditions, they could testify to the possibility that the competences of Alex are present in other grey parrots. Consequently, this generalization bit by bit, from success to success, is no longer expressed in terms of what parrots are, but in terms of the possibilities that the apparatus could actualize. Generalization has changed sides: it is no longer on the side of the parrot neither on the side of the researcher, who ought to represent an anyone guaranteeing objectivity: now it qualifies the appropriate apparatuses.

In this way, by resisting the possibility of letting itself be captured in its alternatives, Pepperberg's practice escapes the two fatal and perfectly foreseeable traps of subjectivity. By taking an interest in what constitutes the appropriateness of a material apparatus that transforms those it interrogates, and by fully agreeing to situate herself in a regime of transformations and accomplishments that mingle with and give form to desires, the question of the subjectivity of the researcher “influencing” the competences of her subject no longer has any reason to cause concern. The question is now about the effectiveness of the apparatus, the researcher's desire no longer being anything but one of the modes of this efficacity. The apparatus reveals nothing, it testifies instead to the power of the transformations themselves. As a corollary of this resistance, the question of the subjectivity of the parrot no longer has much sense, short of restricting it and rendering it very concrete. If the parrot can talk, we do not know what it is, nor what parrotness is, nor anything about the point of view of parrots on the world. But we do learn in a viable manner about its point of view on the apparatus. We learn something about its point of view on the new materials with which it will make a world: colour boxes, numbers, words, a grammar, forms, humans and abstractions. In the same manner that the refusal to talk, in other apparatuses, constitutes an expression of the parrot's opinion in relation to the relevance of what it is asked, the fact that it engages with, accepts and actively transforms what becomes a part of its world, translates an extension of this world and therefore an extension of its subjectivity as “parrot-with-human”.

It is not simply by chance that I have summonsed some very particular situations in order to resist this “we” imposed by certain philosophical practices (which returns so easily when it is a matter of animals, as if on this subject we could only be in agreement). They are situations in which humans and animals work together. They are, above all, situations in which humans and animals accomplish things together. I could say it again in other terms, by taking up anew those by which I designated Alex's accomplishment: they are situations of the extension of subjectivity.Footnote 7 What makes us “one of us” for beings of one species will, like a proposition, overlap with what will become “one of us” for beings of another species.Footnote 8

The same situations can be encountered in certain practices of domestication characteristic of the work of breeders. They place animals one nevertheless has the habit of expecting little from (besides the fact of feeding us), such as cows and pigs, on centre stage.Footnote 9 Although the question of “working together” seems less evident to breeders than to animal tamers or a scientist whose practice is very close to the latter, their testimony seems to me to describe a very similar adventure, an adventure in the course of which subjectivities overlap, are transformed, actualized and extended to the subjectivity of the other. In tackling the question of subjectivity and its possible confiscation today, I have all the more reason to appeal to these breeders. One of my motives for accepting the sociologist Jocelyne Porcher's offer to work with her on the relations involved in breedingFootnote 10 was principally linked to this desire to resist the “we” that philosophers impose on us when they undertake to represent us in order to decree the difference between animals and us.

As it was a matter of resisting the “we” that academic knowledge imposes in order to constitute the difference between animals and humans, it was first of all necessary to learn to present ourselves properly and address others properly. On the one hand, it was necessary to learn to address breeders not as properly representative beings but as good representers. For instance, this was what guided the choice of those we encountered: we sought breeders who we thought would be both interesting and interested by our questions.Footnote 11 On the other hand, it was a matter of not seeking to impose “our” problem as common a priori.

In fact, “our” problem was constituted by the convergence of three problems whose solutions we wished to experiment with. The first was that of Jocelyne Porcher. In the course of her previous enquiries she had arrived at the conclusion that an interesting and reliable manner of talking about the relations between breeders and their animals was to consider that they were working together. To consider that humans and animals work together permits the multiple aspects of their relationship to be thought in a way that other concepts cannot grasp: the judgement that animals make about humans, in the form of the “judgement of relation”, the contrast between exploitation and collaboration, the question of the sentiment of accomplishment, that of gift and of exchange, etc. Now, if in her previous work Jocelyne Porcher had heard multiple anecdotes that showed that animals actively collaborate in the work of their breeders, take the initiative to make it possible, have an autonomous part in this work, the question could not have been posed. When she asked a breeder, “don't you feel that your animals are working with you?” the breeders generally refused this proposition, by stating, “no, the animals don't work. It is us who work”. And yet, the anecdotes testified to the contrary.

The second and the third problems were mine. I was feeling more and more acutely irritated in hearing philosophers, psychologists, sociologists and anthropologists define, with crude generalizations, the specificity of humans and their difference from animals. There was clearly a double “we” in this story, which posed a problem: first, a “we know full well” and then, a “we are different”. How could this “we” impose itself so self-evidently if not as the effect of an academic strong-arm tactic? I wanted to break with this “we” and find out how people who live with animals, who make animals live and who live by them could themselves envisage this difference. I was going to ask them to create this breach in the “we”.

On the other hand – this was my second problem – I was more and more driven to call into question the manner in which research investigations are carried out and had for some years been looking for other apparatuses, ones which would substitute a research for pertinence and interest for the search for truth. The necessity for me to modify my research practice had already imposed itself on me during an investigation I had carried out at the start of the 1990s in the refugee camps of the former Yugoslavia, when I had realized that the effects of the investigation risked stigmatizing the people I had been questioning, and would thus contribute to a deterioration in their condition. To address people as refugees, an identity in which they could not recognize themselves, for example, only repeated the process of exclusion [by which they had become refugees in the first place].Footnote 12 Anonymity prolonged what one could call a regime of insults – “you refugees” – a regime of insults that was all the more violent because of the fact that being called or considered a refugee (the undeserving, nobodies, people of the third zone, someone from the other side of the world, to cite just some of the insults) was experienced by those people as extremely disabling. A first move then was to break with the practice of anonymity to the extent that it only reinforced the feeling of “being nobody” and deepened the contrast between “them” and “us”.

Anonymity, this unquestioned condition of most research investigations, is not simply a characteristic of the research process. From the outset it translates a certain type of relationship and a certain manner of defining those whom one addresses. In other words, it suggests a very specific kind of subjectivation.

To break with anonymity is, consequently, to break both with certain manners of presenting oneself and with the way in which one suggests that others present themselves, or, indeed, get represented. Consequently, I had every reason to want to prolong this rupture in less dramatic contexts: it is by erasing one's name that one becomes the “subject” of an investigation. I understand “subject” here in the way that psychologists designate “whosoever” comes to occupy the position of the uninitiated in research. Anonymity plays an essential role in the research apparatus, in the form of an induction, which largely exceeds that of a “feel free to say what you want”, and even “don't be afraid to talk” which is given as motive: feel free, because actually what you say will have no consequences. It only takes a couple of seconds, though, to realize how people will perceive themselves and translate the situation when they are told that what they say will have no consequences.

The result, the stake, even, of anonymity, is to produce a radical asymmetry of expertises: on the one hand there is the “researcher–author” – the author of questions, of interpretations of hypotheses, of constructions of problems; on the other, there is a social actor: witness, informant, someone having opinions, beliefs and representations for which the researcher will take charge of the analysis. For example, scientists cite their colleagues by name, whereas the witnesses they talk to are all anonymous. This is only one of the many ways in which the asymmetry of expertises is recalled and the possibilities of being authorized to think and of being recognized as putting oneself to the test of thought, indeed of being engaged by what one says, is maintained. Authors are cited, given names, while one lists the interchangeable holders of opinions. The researcher is engaged by what he or she affirms, the witness knows in advance that what he or she will say engages him or her in nothing, because what he or she says are only representations, consigned from the outset by the stamp of his or her distance from the truth. Such practices bring back the difference between representers and representatives incessantly.

In fact, this difference is only the symptom of a more general situation of which certain research habits are characteristic. They rest on a procedure that demands submission from those who are questioned: submit to questions, submit to the inevitable play of interpretations that will judge one's testimony, one's beliefs, indeed one's unconscious motives, submit to the theories that guide research, submit to the problem that is imposed on them and to the manner in which the researcher constructs and defines it. The subject is summonsed by a problem that he or she often has nothing to do with, or in any case has nothing to do with the manner in which the problem is defined, just as the researcher isn't usually preoccupied by the manner in which his problem may or may not be a problem for whoever it summons. And most of the time the subject mobilized in this way will agree to respond to questions without calling into question their interest, their appropriateness or even their politeness,Footnote 13 as, evidently, the scientist “knows better”. In this regard parrots are less aware of the prestige of science and are not worried about resisting and expressing their opinions.

We therefore sought the form of an apparatus that would be best placed to break with what always risks producing submission in the research process. The solution amounted to exploring the place where those who we would be questioning would be best placed to “object”, to addressing them where they would be getting on with their affairs and therefore where they would be both interested and interesting. Now, the fact that our questions posed a real problem was going to lead us to the proposition of a solution: Jocelyne Porcher's question could not be posed; as for mine, if I took seriously my own reticence with regard to the relevance of the response, I ought to have envisaged that the question itself of the difference between humans and animals was perhaps not relevant for the animal breeders. We therefore asked them to help us construct our questions.

“In the course of the investigations I have carried out with animal breeders, I (Jocelyne Porcher) have often heard anecdotes, stories, sometimes even ways of talking which suggested that animals, in a certain manner, collaborated in the breeders' work. Now, when I have tried to raise this question directly and in more depth with the breeders, I have encountered resistance or incomprehension. Evidently it is not a good question to ask. But the recurrence of testimonies on this subject encourages me to persevere. So, in your opinion, as a breeder, how do you think I should construct my question so that it has a chance of being understood and of being interesting?” Similarly, the question that was my responsibility, that of difference, was formulated in this way: “we have observed that the difference between humans and animals is a question that a large number of philosophers, psychologists, anthropologists and sociologists have looked into. Evidently this question interests them. In other respects, obviously, the answers that they have given are different from those that would be obtained if one asked other people, who have other relations with animals, and this contrast interests us. Now, if this question agitates academics we are not sure that they are particularly interested in breeders nor that the question is appropriate. So, for you as breeders, how do you think we should construct and ask the question for it to have a chance both of interesting those we ask and of receiving interesting answers?”

Formulated in this way, these two questions show what we have tried to construct and how it modifies the way in which research investigations are traditionally undertaken. We know that problems are only interesting if they interest: we therefore had to ask the breeders themselves to construct the interest they might take in our investigations, even if – this was the risk – we were told that our problems are not interesting, relevant or shareable. Thus, for example, when Maria Celeste Guimaraes, an elderly breeder from North Portugal, mischievously interrupts her younger friend by stating “the difference between human [l'homme] and animal? Her, she doesn't know, she doesn't have a man [d'homme]”. We understand that she is translating, through humour, the fact that the ambit of our question rests on a rather simplistic presupposition: that we all know very well what a man is.

On the whole, and apparently paradoxically, we were demanding the maximum of kindness in relation to our research, while creating the conditions for maximum “recalcitrance”Footnote 14: “your question is not appropriate”. That is also why for each one of our demands we insisted on the expression “you, as a breeder”, which, it seemed to us, would enable those we were questioning to raise objections and at the same time to help us. One process recurred throughout the majority of our interviews: the breeders incessantly uprooted our question, displaced it, modified its ambit, and when they found the right way of formulating it, they answered. In other words, they responded to the question that interested them. “Difference” here takes on a totally unexpected aspect.

Thus, we often heard this proposition: “the animal understands us better than we understand animals”. More specifically, Manuel Calado Varela states, “the animals know what we want but, we, we don't know what they want”. This way of setting up the contrast is anecdotally supported: “when I open the doors, the cows know I want them to go out, but I don't know if they really want to go out”, “in fact, they know us better”. Several very similar testimonies to these evoke the fact that animals know us in a manner that is sometimes incomprehensible for us.

The transformation of the question on which they worked often led the breeders to privilege the type of question that allowed the issue of the competences of their animals to be tackled. But behind this choice, another difference takes shape, a difference which matters: that between farms where humans and animals talk to each other, make each other propositions, get on and present modes of subjectivity to each other; and farms where such relations are rendered impossible. For all our interlocutors, this fact emerges: animals “pay attention” to their breeder and turn out to be passably good translators of intentions.

The breeding apparatus is an apparatus that creates subjectivities. We found testimonies to this incessantly.

“It's a bit like a child who you give a present to every time you visit. And one day you visit without a present … For example, when we go to see them (the cows) without changing their field, they don't take it very well. They will carp. But me, I say to them ‘I've got the names of everyone who is crying!’ How better to draw the contrast between the sad observation of the industrial farm worker – if the animals could talk we would get shouted at every day – and what the breeder Philippe Roucan tells us. In some farms, animals talk, respond and can even answer back – if only to judge the work of the breeder or indeed to shout at them. The proof of this is that Bruno Greindl's cows know that there are ‘limits not to be crossed. For them in relation to us and us in relation to them; there is a moment when they will explain to us that that's enough'. The humour with which Philippe Roucan replies to his cows’ manifesting their disagreement marks and accompanies a passage: that of animals into the world of ‘speaking for speaking's sake’. This humour sanctions and celebrates the active inversion of relations, the transgression of categories, the joyous impertinence of the exchanges proposed. It is there, doubtless, that the offerings of subjectivity find the opportunity for their best actualization: in the best farms, talk is incessant. And because there is talk, there is talking back.

“We had left for the day, we came back, it was winter, we came back at 7–7.30 in the evening. Someone calls us saying ‘you have a cow in difficulty, along by the fence, and she's having trouble calving’. It was drizzling, there was fog, the works. I set out with the torch. The cows were spread out over thirteen hectares, but I knew more or less which area to search in. I go there with the torch. Suddenly I hear the noise of hooves. The herd had detected me, the entire herd was on top of me. I say ‘shoo, it's me’. They recognize me, everyone stops. For two minutes they stop like that, I talk so they recognize me. I see them turn around and go off. And I say to them, joking ‘don't you want to help me look for Semba?’ And they've taken me to the cow. They've all made a semi-circle around us along the fence. They've waited there and they've taken me to the cow, who had, effectively, given birth. She couldn't get up anymore. They knew why I had come and I think maybe that they must have known I could do something for her”.

The talking back and forth, exchanging judgements about intentions is incessant. Because that is what it's all about in the constant flood of exchanges, it seems to me: it's a matter of incessantly adjusting the intentionalities between animals and humans – I know that you know what I intend to do. Language is not limited to creating an overlapping awareness between two speakers (Hearne, 1994, p. 136); it “populates” each of the beings present with perspectival propositions, which are so many propositions of intentionalities. One gives words like one ascribes intentions. This practice, which inscribes the animal in the world of “speaking” and which contributes to “populating” it, works to blur the frontiers between humans and animals. One makes say, one makes ask, one puts oneself in the place of, so as to “populate with”. One doesn't interpret, one experiments. This does not arise from what is usually referred back to empathy, but instead to a non-immediate form of knowledge, which allows the construction of the perspective of those one knows. One does not put oneself at the place one populates the place with. One doesn't substitute one point of view for another; on the contrary, everything is done by the addition of points of view. Breeders are perspectivists: each perspective is made up of a translation of intentions.

Moreover, we find this trait again in nearly all the breeders questioned: they attribute to animals the capacity to attribute intentions to their breeder. It has been said – we heard it many times – “animals know what we want better than we know what they want”. This possibility of thinking one's own intentions from the point of view of the animal that perceives them – what I am calling the perspectivism of the breeders – largely exceeds the situations in which it is a matter of reconciling wills even if they are articulated in a certain manner. For example, Bruno Greindl recounts the dilemma of a charming, chubby little heifer whom he had attempted to tame. She was part of the herd but we felt we were tempting her: “we had the impression that she didn't want to betray her friends, she wanted to stay … yes she wanted to stay with her friends. She knew perfectly well that we weren't bad, but I have the impression that it's a bit like the wolf that mocks the dog because the dog has become a friend of humans. I have the impression that, in the same way, she didn't want to become too friendly because otherwise she would …”.

Or when André Louvigny associates the question of intelligence with the capacity to understand the intentions of humans: “that's what I believe the intelligence of an animal is. It is being able to enter into situations of which one isn't the master but which one enters all the same by having confidence in the [person] who induces this situation”.

It seems to me that the situations that we have encountered can be brought together under the same sign: they are situations of the exchange, and hence extension, of subjectivities. Breeders adopt perspectives in which they authorize their animals to think and judge their intentionsFootnote 15 and to respond with their own intentions. They are all the more perspectivist for recognizing the animal's ability, sometimes much better than their own, for adopting the other's point of view. These competences are invested in all the more because they doubtless result from the process that guided the fact of breeding animals – for humans – and of being bred by humans – for the animals (on this subject see Haraway, 2003, p. 3).

This extension of subjectivities, which breeding situations activate quite faithfully, resembles what can be called intersubjectivity: becoming what the other suggests to you, accepting a proposal of subjectivity, acting in the manner in which the other addresses you, actualizing and verifying this proposal, in the sense of rendering it true. This is doubtless one of the reasons why in this type of farm, the question of the difference between animals and humans takes such unexpected routes and results in shared perspectives, intelligences and intentions, resemblances, inversions and exchanges of properties between humans and animals.

But equally one could ask oneself if our particular apparatus didn't favour the choice of route taken by the breeders. This hypothesis in fact suggests the need to envisage the question of apparatuses that activate subjectivity in a broader manner, by asking if the apparatus of questions that we suggested, with its insistent demand to construct the question “as a breeder”, did not activate some modes of existence rather than others.

This then would amount to affirming that in the descriptions given to us by the breeders, cows and pigs became the object of a proposal of subjectivity all the more rich and extended for our questioning being addressed to breeders “as breeders”. It is true that at the outset this “as breeders” which marked each one of our demands to put our questions to work had another aim: it was a matter of underlining the fact that we were addressing ourselves to experts. “You as a breeder” translated the fact that we were addressing ourselves to a specialist with expert knowledge of his or her network of practices, occupying the privileged “situated point of view” for talking about what breeders know and think. In this way we started, perhaps a little hastily, from a presupposition: we assumed that we were all interested in the exchange of what they know and think.

Now, re-reading these interviews and putting them to the test of the question of subjectivity and of the creation of the “we”, it seems that certain breeders translated this “you as a breeder” as an invitation to something completely different to what we proposed: this “you as breeder” is the perspective to which their manner of conceiving their occupation obliges them. For them, breeding is not an empty word. It is to help grow (up) and make happy, even if death, which sometimes renders things paradoxical and difficult, is at the end of the story. This constant reminder is translated, admitted by the breeders as a constant reminder of what a particular proposal of subjectivation sanctions: being the subject of specific practical achievements [pratiques d'accomplissements].

The term “breeder” is a term that they say designates a particular practice, a practice that responds to obligations, a practice that engages, a practice that at the same time creates differences. The people we questioned were often precise about this difference, by insisting on the fact that the difference between ways of breeding was, according to them, a more important question than the question of the comparison of humans and animals. For example, André Louvigny contests the generality of the “as a breeder”: “Even among breeders, there are those who are gentle, competent in their relations with animals and there are brutes, imbeciles. So there already you have a world. There are people who feel nothing for their animals. There are people who look after them properly and there are those who don't really click”. Or Patrick André, who suggests subordinating the question of difference to that of expectations: “Personally, I think that one way of tackling it is simply to ask the question: ‘what do we expect from animals?’ Is the window sufficiently large to try to examine the difference between humans and animals?” He resumes [the matter] with regard to the question of work: “I'm not sure that an industrial farmer expects the same thing from animals as a small holdings farmer, who has a different approach in the sense that he considers that he must be in harmony with nature, whilst the former has instead to adapt nature to his requirements”.

Philippe Roucan arrives at a similar observation, but the manner in which he formulates it should make us pay attention: “Personally, I call breeder (he is talking to Jocelyne, who he knows well) you agree entirely, us, we are breeders, the type of farm that we run, call it, say, a family, not run in an extensive way, like us, with a limited number of animals, well, I think that a breeder is not the same when he has 100 animals as when he has 10,000 because to my mind he won't have the same perception of the animal”. You agree entirely, us, we are breeders he says to Jocelyne Porcher. There is a “we” that translates an agreement about what “being a breeder” means.

I thought I was offering the posture of an expert in “knowledge” of animals. That was the “we” that I wished to construct with the breeders, the locus of a convergence on what could define us, what could interest us. However, despite my good will, this “we” turned out to have been assumed a little too quickly in this story. The important thing for the breeders is not to have good knowledge, but it is to be good breeders.

That is what the animals stood as witnesses for. That is why it is intentions, expectations, perspectives and exchanges of properties that were privileged in the way that the breeders responded to us and worked on our questions. They translate not what cows and pigs are, nor what breeders know of their stock. They indicate what animals become capable of in the practices through which the breeders proudly define themselves: bringing into existence animals that nourish humans in many ways. This includes the nourishing of diverse modes of existence and becomings of subjectivities, the very thing that governed the choice of our enquiry.

Translated from the French by Andrew Goffey.