1 Introduction

In this paper we wish to challenge the widely held view that practice-oriented approaches to ethics entail a conservative perspective on certain practices—e.g. that they do not promote any (fundamental) change in our ethics of human dealings with animals, or do not formulate clear principles that help us to achieve this change. In a word, these approaches are accused of endorsing moral conservatism, and typically, at any rate, Wittgensteinian approaches to ethics face that charge, too.Footnote 1 Although this issue has been dealt with in the literature, we believe the defenses of practice-oriented ethics set out so far are insufficient and fail to exploit the full potential of Wittgenstein’s later work on moral philosophy.Footnote 2 We think this can be demonstrated in the context of animal ethics. We will counter the accusation of conservatism by discussing Cora Diamond’s work on animal ethics. Our aim is to show that Diamond’s writings can be read as a critique of some human-animal relationships, and a call for change. We are therefore going to scrutinize Diamond’s conception of the difficulty of reality (henceforth DoR).Footnote 3 Difficulty of this kind can be described as an experience that serves as a basis for changing the status quo and shaking up established practices. This approach complements traditional approaches to animal ethics that are eventually advocating drastic changes.Footnote 4

To understand DoRs and their potential for changing the status quo, we shall look at what bottoms out morals. We will show that the foundation consists of two types of moral certainties: certain propositions and transcendental basic moral certainties. Whereas the former convey content, the latter are rules, like “equals are to be treated equally and unequals unequally.” And particularly the latter are important as they help to get thinking back on track in the experience of DoRs—which are experiences of uncertainty that occur when habituated practices are shaken. This might be the case when we encounter animals in instrumental relationships—like in the laboratory.

This paper is fundamentally influenced by Diamond’s approach to moral philosophy and animal ethics in particular. We read her as follows: if we want to criticize how we treat and conceptualize other animals, we should not reduce reality to as few principles as possible, as is the strategy of traditional animal ethicists. Rather, Diamond challenges us to look at our lived practices, which are variegated and inconsistent, but in themselves offer the possibility of critique. It is not about finding a consensus in morals but about understanding conceptualizations, such as those of other animals, and whether the thinking and conduct that goes along with them is justified. This perspective combined with basic moral certainties and DoRs helps us to reconsider human-animal relationships within the fuzziness of lived morality.

2 Basic Moral Certainties in Wittgensteinian Ethics

Before we explore the ethical scope of DoRs, with particular attention to animals, it will be helpful to clarify the idea of reflection bottoming out in Wittgensteinian ethics. We will argue that practice-oriented approaches to ethics—and especially the Wittgensteinian-Diamondian approach presented here—are not necessarily morally conservative but offer tools for moral criticism and reorientation. Moreover, we are going to argue that this is not despite, but because of their recourse to the practices and language uses which encompass our form of life. DoRs, as we will show below, can shake up established practices and reveal their foundations. But what exactly lies beneath these practices? Following a recently debated understanding of Wittgensteinian approaches to ethics, we can say that a fundamental feature of our form of life is its reliance on basic moral certainties (henceforth BMCs).Footnote 5 They structure moral thought and conduct, they form the “bedrock,” as Wittgenstein puts it, through which language use is made possible.Footnote 6 In what follows, we shall draw on the work of Nigel Pleasants on BMCs as our point of departure.

According to Pleasants, the “badness of death,” the “wrongness of killing” and the “wrongness of unwarranted infliction of pain and other forms of suffering” are BMCs.Footnote 7 These, however, must not be conceived of as absolute moral truths, but rather as “regress stoppers,” i.e. final moral demarcation lines which function as moral foundations for further moral thought and conductFootnote 8:

A basic certainty is something that cannot be sensefully asserted, explained, justified, questioned, or denied first-personally; and indeed no-one would even think of doing so outside a philosophical debate on the phenomenon. That it is very wrong to kill an innocent and non-threatening person, absent special excusing or justifying circumstances, is so fundamental to our human form of life and individual moral consciousness as to be recalcitrant to propositional formulation.”Footnote 9

BMCs are fundamental to morality, and “immune to justification, challenge and doubt, and hence cannot be objects of first-personal knowledge.”Footnote 10They are general and necessary requirements for moral thought and actions. Hence, they are not open to discourse, and instances of making exceptions prove the rule. This is also echoed by Pleasants:

The core idea of what basic moral certainty consists in is that its object is the wrongness of killing, inflicting pain, etc. per se. But this basic certainty coalesces with the socially acquired belief that it is sometimes permissible, and sometimes even required, to kill and inflict pain on innocents, when apparently weighty reasons support or demand doing so.”Footnote 11

Following Pleasants, exceptions to a BMC such as the wrongfulness of killing presuppose this corresponding certainty. The justification of a killing will only prove meaningful if there is, in the background, the certainty that killing is per se wrong. Here, Wittgenstein would say that one has “reached bedrock,” and the “spade is turned,” meaning that justifications are exhausted and one is inclined to say: “‘This is simply what I do.’”Footnote 12 It is certain for us that we do not kill somebody without weighty reasons, and that it is something akin to tragedy if somebody is killed in an act of self-defense—this is simply how we, as a moral community, think and act. A person who thinks killing fellow humans is in any circumstances perfectly fine would justifiably be regarded as morally alien, as they are not participating in the same form of life that is expressed in a shared moral framework whose bedrock consists of BMCs.

Pleasants draws on Wittgenstein’s remarks in On Certainty to formulate BMCs. Wittgenstein there argues that pre-epistemic certainty plays a central role in our lives. Basic epistemic certainties include “the fact that physical objects do not just spring into and out of existence; that the ground under our feet will not suddenly give way; that one has never been to the moon; what one’s name is.”Footnote 13 These certainties are—in a way analogous to BMCs— “immune to questioning, doubting and testing, but are also beyond verification, affirmation and appeal to evidence, grounds or reasons.”Footnote 14 So, “[t]hat it is wrong to kill an innocent, non-threatening person” is “just as certain as any logical or analytic truth, or any object of basic empirical certainty.”Footnote 15 Consequently, theories entailing that killing is wrong do not, according to Pleasants, add anything meaningful to the BMC of the wrongness of killing, since that BMC is not in need of any proof—everybody cannot but approve it tacitly. Similarly, there is no need for a scientific verification that the Earth existed before our birth, that it exists now, and that it will exist tomorrow.Footnote 16 We have no standpoint from which we can prove this, from which we would know that these certainties are certain, but by the same token we have no reason to doubt them. As Wittgenstein puts it:

Giving grounds, however, justifying the evidence, comes to an end;—but the end is not certain propositions’ striking us immediately as true, i.e. it is not a kind of seeing on our part; it is our acting, which lies at the bottom of the language-game.”Footnote 17

Jeremy Johnson refines the notion of a BMC. He calls any moral principle that can be neither justified nor supported by evidence “a bedrock moral principle.”Footnote 18 Since bedrock principles are foundational to our language games, Johnson ascribes a form of foundationalism to Wittgenstein which is not to be understood as a substantialist theory, but which “merely describes features of the conceptual scheme we inherit and learn—our foundation. If we did not inherit this scheme, language-games involving truth and justification would not be possible.”Footnote 19 For language games to work we need certainties:

These are not claims that might turn out to be wrong. They are the background against which and the foundation upon which meaningful claims can be made which might turn out to be wrong. They are not grounds for belief, they are the ground for belief.”Footnote 20

BMCs are therefore immune to justification, challenge and doubt because what “is truly foundational is something which nothing imaginable would speak against.”Footnote 21 So, the way we think and act morally rests upon a bedrock that is immovable. In this “heterogenous foundationalism,” as Johnson calls it, beliefs that can be doubted rest on a foundation which itself cannot become the object of plausible doubt because it structures the way we think.Footnote 22

BMCs are by no means relative to culture, according to Pleasants. They are cultural universals which transcend history, culture and moral language games. The only variations possible are deviations from the basic certainties.Footnote 23 Exceptions vary, too, from culture to culture, but the unshakable BMCs are, according to Pleasants, fundamental to human nature. So, BMCs derive from the “reflection of our human social nature, our deepest collective interest in living together cooperatively, productively, harmoniously, and in a way that enables us to flourish both individually and collectively.”Footnote 24 As we are going to argue, Pleasants’ depiction of the non-relativistic nature of BMCs is problematic. In turn, we are going to develop an understanding of BMCs as pseudo-premises which are not cultural universals but rather similar to logical axioms. Therewith, we follow—in part—an argument that Johnson has recently put forward.Footnote 25

Now, we would like to further sharpen the notion of a BMC and show that it functions like a rule, i.e. a form of thought. Consider this remark by Wittgenstein: “Compare the meaning of a word with the ‘function’ of an official. And ‘different meanings’ with ‘different functions.’”Footnote 26 An official is someone involved in the smooth running of a system. A system, apparently, makes some things possible, and others not; and the official is someone who implements a procedure, not someone who chooses it.Footnote 27 Following this metaphor, we believe that BMCs are fundamental forms of thoughts, i.e. rules, and do not contain content—they are, therefore, not propositions. As long as something is uttered within the framework of certainties, within the system, language games work. Basic certainties indicate a path or block it. Uttering something that lies outside the framework of moral certainties is as senseless as claiming that 2 + 2 = 5 (pace logical certainty), or that these hands of mine do not exist (pace epistemic certainty). We think that this interpretation is more in line with Wittgenstein’s understanding of basic certainties as rules:

The rule, once stamped with a particular meaning, traces the lines along which it is to be followed through the whole of space. […] When I obey the rule, I do not choose. I obey the rule blindly.”Footnote 28

As long as one follows the rule, one is in the realm of certainty. “Blind” means that one does not have to reflect on it—to “obey the rule blindly” expresses a habituated practice.Footnote 29 The utterance that “one ought to eat stones as they are nutritious” is senseless because the proposition “stones are nutritious” is, given the notion of stone and its subsequent rules, incommensurate. Hence, BMCs delineate permissible moves in language, like the move from 2 + 2 to 4 in the eqaution 2 + 2 = 4; BMC does not lie in the result, but in the moves towards the result. Doubting them would be senseless, i.e. outside the language game. Moreover, making basic certainties explicit is like adding a pseudo-premise to an argument.Footnote 30 For instance, take the example of equal consideration being given to the infliction of pain on someone:

P1: The intentional infliction of pain on a human individual capable of experiencing pain is morally significant.

P2: Some non-human animals are similarly capable of experiencing pain.

C: The intentional infliction of pain on such non-human animals is morally significant.

This sets out a convincing line of thought. Consider how strange it would be to claim that this is not warranted, as a form of practical thinking, since premise 3 is missing —which may be formulated as:

P3: Equals are to be treated equally and unequals unequally.

We cannot but think with this rule: being convinced means that one has followed a rule. To make P3 explicit is redundant in everyday language games: It adds nothing. So, P3 would be like a pseudo-premise and can justifiably be excluded from the argument.Footnote 31 Such premises are not in need of explications in everyday language uses since they are no plausible objects of doubt. This example shows, by the same token, that a BMC becomes palpable only if followed inside a context. “That’s why ‘following a rule’ is a practice.”Footnote 32 Applying this idea to Pleasants’ argument shows that the proposition that killing is wrong results from an argumentation that follows a BMC, while this proposition itself is not a BMC, as we will argue.

So, basic certainties are not, in any sense, related to objects; they enable and delineate the forms of thought that relate to the object in question. This gives the notion of a BMC a major twist that, so far, has not been identified in the literature: It is not a moral subject matter that is certain, but instead the certainty experienced by a person as a consequence of following the rule. In other words, it is the lived following of the rule that creates certainty, not a particular belief or proposition, because basic certainties lie within the rule rather than the belief or proposition.

As BMCs are forms of thought, i.e. rules, and do not contain propositional content, we think the following is a proper BMC: Equals are to be treated equally and unequals unequally. However, without a habituated practice of following a BMC, it becomes futile. Here, we diverge from Pleasants, as we do not consider the badness of death and wrongness of killing to be BMCs. These are propositions conveying content, but they do not describe forms of thought and are built on the wrongness of unwarranted infliction of pain and other forms of suffering. Further, their subject matters are plausible objects of doubt.

First, in the human case it is crystal clear that the wrongness of killing can be doubted. The wrongness of killing (and analogously the badness of death) is a certain proposition, it refers to states of affairs which are normatively evaluated. Normally, killing is wrong, but there can be warranted exceptions as Pleasants notes himself.Footnote 33 These exceptions may encompass self-defense or euthanasia. The warranted exceptions of killing are seldomly clear cut, they may vary in their plausibility from case to case and are thus a constant issue of practical and theoretical discourse, as is the case in medical ethics. The Stoics, like Marcus Aurelius, even valorized suicide, which is basically killing oneself, when existence becomes undesirable. In fact, for the Stoics “[d]eath itself was unimportant,”Footnote 34 what mattered was the way of dying. Accordingly, the wrongness of killing and the badness of death are open to discussion and not immune to justification, challenge and doubt; hence, these propositions do not fulfill the requirements for BMCs since they are proper objects of doubt.

Second, the fact that moral subject matter, such as the conviction that killing is wrong, can be doubted becomes even more obvious when considering animals. Most people are critical of intensive animal husbandry but many of them also support so-called humane farming and slaughter. Setting aside the question whether humane killing is practically possible, we do not have certainty that no one would question that killing animals is always wrong, even for food. Moreover, there may even be good reasons to kill and eat humanely raised animals, for example “there may be some societies where people need to eat meat because they do not have access to enough food or finding adequate vegetarian alternatives would be prohibitively expensive.”Footnote 35 Again: we routinely kill animals for food, and this practice is far from being unproblematic, thus still a matter of debate.Footnote 36 However, the belief that there is a good way of eating animals has permeated cultures for centuries and is still widespread today.Footnote 37 This proves the point we wish to make here that the proposition that killing is wrong is not beyond doubt and therefore no BMC. Further, what about the badness of death in the animal case? It is imaginable that the premature death of a cow is not bad for her, if she lacks (biologically or individually) any capacity to conceptualize her future or her own death.Footnote 38 So, we reject the BMC of the badness of death for animals: It is open to debate, deniably bad, but not certainly bad. As we saw in the example of the Stoics, the same is true for human beings.

Third, what counts as unwarranted in the wrongness of unwarranted infliction of pain and other forms of suffering can also be doubted. There is no certainty of what counts as unwarranted or warranted. Think of the administering of a vaccine (to a person or an animal) or the castration of stray dogs and cats for population control. Even though these actions are obviously intended to benefit the beings in question, they are creating harm in order to generate the benefit. These actions may be viewed as warranted if the positive effects outweigh the negative effects. This proposition has such a strong plausibility, however, that one is inclined to frame it as a BMC—especially because the wrongfulness of killing and the badness of death seem to rest on it. But it is a very strong certain proposition for which deviations are thinkable.

To this point, we hope to have shown that the wrongness of killing, the badness of death and the wrongness of unwarranted infliction of pain and other forms of suffering are neither fully applicable to the human and to the animal sphere nor BMCs. Here, we wish to make an important distinction between certain propositions and transcendental BMCs. The former reflect content and not a rule—they seem to be moral certainties due to their intuitive, seemingly unquestionable plausibility but they are not. The latter convey a rule but have no content. Proper BMCs—such as: equals are to be treated equally and unequals unequally—are thus transcendental; they are like pseudo-premises which do not need to be mentioned.Footnote 39 They are transcendental in the sense that they organize our thinking. We think along with them without thinking about them. A BMC must fulfill the criterion that what is morally certain, what is “truly foundational is something which nothing imaginable,” according to Johnson, “would speak against.”Footnote 40 Pleasants’ suggestions of BMCs are thus certain propositions and not transcendental BMCs.

Further, we have identified the transcendental BMC that equals are to be treated equally and unequals unequally. The main characteristic of a BMC is that it is a rule and has no content. It is a form of thought; not following it comes at the cost of producing nonsense; doubting it equals going astray and being off trails. But BMCs alone—this is due to their rule character—do not press to consider our dealings with animals as problematic. The function of rules can only be seen in practice, when dealing with reality. But sometimes, reality becomes resistant to habitual ways of thinking about it, which is the experience of a DoR. To fully understand Diamond’s thoughts on DoRs, we shall first discuss her approach to animal ethics and her critique of traditional approaches in more detail.

3 Cora Diamond’s Approach to Animal Ethics

Diamond’s approach to animal ethics, in a brief outline, is as follows. Practices and their correlating concepts are expressed in language games, including variegated and inconsistent notions like human or animal, which structure our imagination and ways of thinking. Diamond ascribes special moral importance to merely being human, in a sense going beyond the biological understanding of humans.Footnote 41 This is at odds with traditional moral individualism, which is the view that it is individual characteristics that underwrite ascriptions of moral status, not membership of a group, community or species.Footnote 42 In clear contrast to the idea that merely being human is irrelevant to moral consideration, Diamond emphasizes the importance of being human. To explain this, she offers a remarkably simple observation: We do not eat our dead, after they have died, for example, in an accident, even if their flesh may be first class.Footnote 43

This, at first seemingly descriptive, observation refers to what we would like to denote as our shared fabric of meaning: Diamond tries to direct her reader’s attention to the fact that we can only grasp our moral relations to other human beings, and to other animals, within the context that forms our shared practice. Being human involves being situated in concrete practices. And our practice is fundamentally connected with not eating deceased human beings. Eating our dead is perceived as a sacrilege or may be justified solely as an act of sheer misery; claiming that this act has no moral significance would signal a rejection of our entire form of life (as situated beings within a shared life practice) as morally irrelevant. Crucially, for Diamond, we cannot evaluate moral problems abstractly, i.e. without taking such shared and lived practice into account. Oppositely, Katarzyna de Lazari-Radek and Peter Singer, for example, consider such abstraction the ideal starting point for ethical deliberationFootnote 44, but for Diamond moral consideration rests upon a shared context that is already saturated with meaning. So, what Diamond criticizes is not moral individualists’ perception but their argumentation:

It is their arguments I have been attacking, though, and not their perceptions, not the sense that comes through their writings of the awful and unshakeable callousness and unrelentingness with which we most often confront the non-human world. The mistake is to think that the callousness cannot be condemned without reasons which are reasons for anyone, no matter how devoid of all human imagination or sympathy.”Footnote 45

According to Diamond, it is neither necessary nor possible to escape the situated human perspective when granting animals moral consideration. She shows that we can perceive animals as fellow creatures, i.e. as “fellows in mortality” who may be “sought as companyFootnote 46 and to whom we extend moral concepts that are crucial in human life. The application of concepts like justice, pity or charity to animals only proves meaningful when those concepts are extended from human life to animal life, and that requires a situated human perspective. Without understanding the significance of, say, pity in human life, Diamond suggests, we cannot grasp what it means to empathize with an animal. Human life is the inescapable point of reference. Thus, what we owe to animals depends on an extension of our conception of human life. Does her account remain anthropocentric or even speciesist?

Diamond’s notion of human life has been criticized. Cary Wolfe discusses the problems of assuming “that this human being is an essentially homogeneous and undifferentiated creature that is capable of a more or less transparent relationship to its own nature.”Footnote 47 He questions the idea that we always have a clear concept of what it means to be human that precedes our moral relationships to animals. In fact, communities have never been just human, humans and animals have always been forging relationships in “mixed communities.”Footnote 48 Accordingly, extending concepts of what is morally important to animals should depart from a conception of human life that is not (and never was) exclusively human:

Rather, ‘we’ are always radically other, already in- or ahuman in our very being—not just in the evolutionary, biological, and zoological fact of our physical vulnerability and mortality, our mammalian existence, of course, but also in our subjection to and constitution in the materiality and technicity of language that is always on the scene before we are, as a precondition of our subjectivity.”Footnote 49

According to Wolfe, being human is unstable, oscillating between being human and being animal. We are fundamentally incapable of conceptualizing being human in Diamond’s sense; this inability, then, is characteristic of human life, not transparency.Footnote 50 Wolfe understands this oscillation as an unbridgeable foundation for human-animal relationships.

Nevertheless, Diamond’s approach is not fundamentally threatened by this criticism. Diamond emphasizes the importance of shared concepts that humans engage in as language users which they cannot share linguistically with other animals. And Diamond is aware of the difficulty of grasping human life in relation to animal life: “We are mysteriously like them, mysteriously unlike them.”Footnote 51 Wolfe’s critique is valid to the extent that Diamond overestimates the importance of particularly being human, but Diamond is correct in emphasizing that the significance of moral concepts is tethered to practice and can only be extended to other animals from there—this is right, even if a self-transparent relation to one’s own human nature is not possible.

And further, being human is not a rigid, dogmatic concept that is not open to criticism whatsoever. Rather, Diamond understands morality as fundamentally polyphonic.Footnote 52 However, she also highlights the importance of concepts that have proven useful for moral reasoning.Footnote 53 In the context discussed in this paper, this amounts to the following: in order to do justice to our relationships with non-human beings, we cannot neglect the significane of the concept of justice in human-human interactions—this is what it means to extend concepts from the human to the animal sphere. To get rid of such conceptions structuring our moral lives (often pre-reflectively) is the signature strategy of traditional moral individualists. For them, the world is fundamentally devoid of morally saturated concepts that may structure thinking in a good way—such as the concept of being human and concepts like justice which are fundamentally tethered to it.

We wish to emphasize that extending concepts to other animals, in Diamond’s sense, always refers to specific human-animal relationships in which morally relevant, and tangible, interactions take place, and not to abstract matters. These relationships, though, are characterized by the way we introduce morally relevant differences to our dealings with animals using colliding concepts. Two examples may help to illustrate this: Most people would find it repugnant, beyond being uncivilized, if someone were to slaughter their dog, which usually lies beside or on their couch while they are watching TV, for a barbecue party. Pets, a category of domestic animals with which many people in industrial societies interact more often than they do with wild animals, are not something we eat. If we eat an animal, it is not a pet. In fact, a pet is given a name, and thereby “it is given some part of the character of a person.”Footnote 54 By the same token, we use the concepts livestock and vermin to exclude animals from the class of fellow creatures.Footnote 55 Thus, we already hold morally inflected concepts of what non-human animals are, and we do not approach these animals in a morally neutral way. These examples illustrate that we engage with a variety of complex and conflicting concepts of the animal while not knowing what the animal is.

Diamond shows, with these examples of conceptual conflict, that humans think within conceptual compartments, as it were, of what animals are, and that these compartments correlate with ways in which an animal should be treated.Footnote 56 Rather than extending the circle of moral status, which typifies moral individualists’ approach to animal ethicsFootnote 57, Diamond considers how moral concepts can be justifiably demarcated in their application. Let us explain this in more detail.

The signature strategy for moral individualism is to detect morally relevant, that is, empirically detectable characteristics that can be referred to in order to make the ascription of moral status to animals plausible. Predominantly, the characteristic invoked is sentience, or the ability to experience pleasure and pain and therefore to have interests. Hence, since sentience matters morally in humans, it also matters morally in animals (if it can be detected). Human persons are sentient and have strong future-oriented interests which are frustrated by being killed. Thus, their capacities put moral claims on moral agents. If other beings have similar characteristics and similar interests, the circle of beings that deserve moral consideration has to be expanded. This is due to the principle of equal consideration of interests.Footnote 58 In this respect, apes and humans have a similar status, as the former are also sentient, even if they hold less strong future-oriented interests than the latter. According to this logic, bees have a different moral status than human beings and apes because (and so long as) they differ in respect of sentience and future-oriented interests. According to this expansion model, it is worse to kill a human than an ape, and worse to kill an ape than a bee. Singer’s approach can thus be described as a “hierarchical pathocentrism.”Footnote 59

In opposition to moral individualists, Diamond’s animal ethics addresses the question of justified moral demarcation lines by referring to established practices. Within these practices pre-reflective attitudes structure our perceptions and dealings with animals. Pets and farm animals (or livestock) are morally inflected concepts that go along with an understanding of seemingly unproblematic conduct. But here an important question arises, namely: how is the demarcation between caring tenderly for the former throughout their lives and killing the latter justified? So, following Diamond, the task of animal ethics is to examine legitimate or illegitimate forms of moral demarcation and not to extend the moral community to previously morally neutral beings, as moral individualism suggests.

This understanding of how to approach human-animal relationships morally quite fundamentally deviates from traditional moral individualists who desire to derive their moral claims from empirical facts, i.e. facts which are revealed by natural sciences like ethology or comparative cognition. So, correlating with their specific set of characteristics—like sentience, intelligence, self-awareness—an individual (human or non-human) merits a particular treatment. But this results in a major problem: traditional moral individualists advocate a reductionistic understanding of what counts as morally important in human-animal relationships, neglecting lived practices and particular relationships, and aim to create a unified picture of morality by reducing morality to as few principles as possible.Footnote 60 Against this picture of an alleged consensus about what morally matters, Diamond’s philosophy is in strong opposition.Footnote 61

4 Difficulties of Reality in Human-Animal Interactions

We have shown that it is reasonable to assume that the following is a transcendental BMC: equals are to be treated equally and unequals unequally. Then, we have argued that concepts of animals, i.e. what animals supposedly are, are divided into different practices as compartments with particular moral infrastructures, that can be expressed in language games. Within such compartments, certain kinds of moral behavior appear to be unproblematic and others do not, depending on the respective compartment. This brings us back to our core question and the accusation that practice-oriented accounts lack the resources for moral change and remain conservative.

The concepts of what animals—supposedly—are can be mutually exclusive, as is the case with pets and livestock. If animals are treated with respect for their individual lives and are given names etc., we refer to them as pets, separating them from other practices like being bred and killed for food, which is the livestock compartment. However, these compartments usually do not collide in practice since our world is organized around them: In everyday life, there seems to be nothing wrong with our conviction that we have special duties towards companion animals, who we may treat as family members, while simultaneously eating farm animals without difficulties. Divergent norms, or patterns of conduct, may co-exist and provide order in our everyday life. However, sometimes we may find that these patterns of moral behavior collide and trigger uncertainty. We suggest that this uncertainty arises not merely because something in reality is “resistant to our thinking it”Footnote 62 but also because we have the BMC that equals are to be treated equally and unequals unequally.

In the following, we would like to examine the notion of a DoR, in Diamond’s sense, and refine it with the Wittgensteinian idea of BMC. Phenomenologically speaking, DoRs arise when language games run aground, and uncertainty emerges. According to Diamond, DoRs are:

experiences in which we take something in reality to be resistant to our thinking it, or possibly to be painful in its explicability, difficult in that way, or perhaps awesome and astonishing in its explicability. We take things so. And the things we take so may simply not, to others, present the kind of difficulty, of being hard or impossible or agonizing to get one’s mind around.”Footnote 63

Diamond illustrates the incomprehensibility of a DoR which is “painful in its inexplicability” and makes one aware of the limits of thinking by referring to Ted Hughes’ poem Six Young Men in which the narrator looks at a photograph of six young men who are all dead by that time:

What interests me there is the experience of the mind’s not being able to encompass something which it encounters. It is capable of making one go mad to try, to bring together in thought what cannot be thought: the impossibility of anyone’s being more alive than these smiling men, nothing’s being more dead.”Footnote 64

Our mind cannot grasp this, just as we cannot grasp that 5 + 3 = 7 or that we may treat equals unequally; the DoR imposes something outside the language game that is not making sense: non-sense. So, what happens here is this: “Language is shouldered out from the game, as the body from its instant and heat.”Footnote 65 Here, language no longer seems appropriate to express reality. As in the example of a non-moral DoR above: It explores the perceived impossibility of someone being alive and dead at the same time.

To explore the moral implications of DoRs for animal ethics in detail, we use the example of a laboratory mouse. This is one compartment describing a mouse, but there are other compartments of what this animal can also be for us: scientific tool (laboratory mouse), vermin, pet, liminal animal, fellow creature etc. A DoR can be experienced as “the difficulty of human life in its relation to that of animals, of the horror of what we do, and the horror of our blotting it out of consciousness.”Footnote 66 In our example this would be the case when the treatment and conception of a mouse simply as a scientific tool becomes resistant to our thinking it, i.e. when the use of the word mouse as a laboratory mouse becomes difficult. DoRs show that a single compartment is not exhaustive and exclusive in explaining what the mouse is.

DoRs are, in a word, starting points for reflection, prompts for the reconsideration of language games. As language users we have the urge “to establish an order in our knowledge of the use of language: an order for a particular purpose.”Footnote 67 But each order, i.e. each compartment, is only “one out of many possible orders; not the order.”Footnote 68 This already indicates why DoRs emerge in the first place: there is no abstract point of view from which we can determine what the mouse is. We only have different language games to approach it: the language games of utility, gratitude, biology etc. Experiencing a DoR shows that a language game fails in its claim to describe what the mouse is.

But, as long as nothing triggers a DoR, orthodox ways of thinking about the mouse will run smoothly. Equally, while orthodox ways of thinking seem sufficient, DoRs will not emerge. The mouse might be seen as a delicate instrument used to gain knowledge in the practice of animal testing.Footnote 69 However, smooth thinking about the mouse as research instrument can be interrupted. Hence, some sort of trigger is needed to cause an experience of a DoR. Triggering interruptions can take various forms: the reading of literature on mouse behavior or a mouse’s ability to experience pain, fiction dwelling on the human-mouse relationship or the sudden emergence of doubt. What these triggers have in common is that they suddenly reveal a blind spot in one’s thinking about the mouse. What one has not seen before suddenly becomes apparent as an interruption of one’s formerly smooth conceptualization in a particular compartment.

Whichever trigger is involved, it is followed by the personal, individual experience of being thrown off track; and a realization that one’s former ways of thinking the mouse are no longer sufficient. In the case of laboratory mice, a DoR may be experienced by one person as astonishment or outrage—while others may not experience triggering interruption at all. One person may be startled that mice are used in huge numbers for medical experiments while alternatives may be at hand. Another may feel astonished that a mouse can be perceived as a fellow creature, much like the way people are perceived. Attempting to share this experience, imperatives such as “Look at the mouse—in a different way!” may emerge. However, what it is that shakes up one’s thinking is not brought forward in explicit terms. It is experienced as an uneasiness with the traditional way of thinking, with the use of the term mouse in the particular context.

This shows that the language game being played is not fit to offer a solution to the problem experienced. Seemingly rigid uses of terms like laboratory animal or delicate scientific tool (thought of as extreme points) collide with the concept of a mouse as a fellow creature or companion. While the former concepts correlate with the moral language game of utility, the latter are used in moral language games of gratitude, friendship and compassion. However, as a result of the difficulty of comprehending the experienced phenomenon, no language game is readily available to resolve the experience of the colliding concepts. What happens is more like a struggling and exploring of different language games that can accommodate the felt trouble and difficulty. What is the being in front of me now? What “it”—or rather: this individual—is supposed to be no longer runs along familiar tracks.

It becomes apparent that the seemingly certain conceptual compartments describing what animals are need to be reassessed, and that this reassessment leads to their rearrangement. This is where BMCs come in. They help us to get thinking back on track. In other words: BMCs help us to regain order. BMCs are implicit, lived rules, shared in moral thought and action; they are transcendental to any moral language game and function as regress stoppers. So far, we have identified the following as a transcendental BMC: equals are to be treated equally and unequals unequally. But as we have shown further above, BMCs have no content but are forms of thought and function as regress stoppers. Accordingly, we cannot infer from our identified BMC in which respect mice are equal or unequal to humans (or to anything else).

At this stage of the experience of a DoR, one needs to examine the aspects in which mice are equal or unequal to humans. Whatever the aspect, it results from what one values as equally morally important, it is not something given or imposed by biology. So, it is an engaged act of valuing, not an abstract act of considering, that determines in which aspect mice are equals or unequals. Mice and men are equals in the aspect of shared embodiment, i.e. vulnerability, as Diamond notes herself.Footnote 70 (Note that vulnerability reflects one communality in which mice are equals to humans. Other framings of similarity are imaginable such as being an interdependent member of a social group.) Valuing the mice’s vulnerability and ours equally, and then perceiving reasons to inflict suffering on mice for science or the greater good as insufficient, runs contrary to the moral language game of utility that now no longer seems appropriate. In fact, one has identified a case of unequal treatment of equals, i.e. practice had previously made them unequals so that we could use them for these purposes. As a result, using mice for human ends, even if these are medical ends, is not immune to doubt anymore—in fact, to be more precise, it never was immune to doubt because using mice or animals for human ends is not a BMC. This means that agreement in action is challenged, and that the pre-reflexive structures of our language games become conscious. The conceptual compartments are not basic forms of thought but claims about what things in the world supposedly are. They may be misconceived as part of one’s Weltbild, but they do not belong to the bedrock; they rest on it. They are, consequently, open to rearrangement.

The urge to rearrange the conceptual compartments is accompanied by an underlying feeling that something is making a moral claim on one. What is the source of this claim? What one may eventually find difficult is the mistreatment and misconception of the mouse as simply a tool barely (if at all) worthy of moral consideration. The familiar way of dealing with the mouse does not snap into the seemingly rigid compartments anymore. It is precisely what Diamond called further above an “experience of the mind’s not being able to encompass something which it encounters.” DoRs press one into “minding what already matters”Footnote 71—in this case mice as creatures whose wellbeing makes moral claims on us, which becomes palpable by valuing their vulnerability and ours equally because the use of vulnerability on mice makes sense. The mice’s vulnerability is in the world, it imposes itself on our seemingly rigid compartments.

What we see here is the experience of a DoR dissolving the seemingly rigid order of the conceptual compartments. Some mice are no longer conceived as means to ends, although they are not yet fellow creatures either. At this point, new compartments can be established, or the term mouse can be put into a different one. This can be described as an aspect shift, and as the entrenchment of the term or concept of a mouse with new meaningFootnote 72: the use of the term mouse has changed.

Of course, one can fall back into the former language game of utility. However, if one’s awareness of a particular DoR is shared and conceptualized, this might impact upon the form of life and change practices and legitimate ways of thinking about mice (although we must stress that this would represent a new agreement, not in opinion, but in form of lifeFootnote 73). So, when we find a way to redirect our thinking, the world and our experiences of DoRs serve as an important source. Animal ethics is in large part, one could claim, based precisely on the idea of using this resource to change human-animal relationships by thinking of animals in a new way. It is remarkable that an empirically lived uncertainty reveals the foundation of moral thought—it is the BMC that equals are to be treated equally and unequals unequally.

This moral certainty underlies the experienced unruliness of pressing mice into certain compartments. Humans and mice are equals regarding their vulnerability; they are not unequals in this respect and must be considered equally. This certainty rests on the moral sphere and not on the world. But at the same time, the mouse as part of the world counts for its own sake—and the mistreatment and misconception of mice as unequals in their vulnerability is what may provoke an experience of a DoR. In this sense, BMCs enable one to see what already matters, but they must be made visible by something external, i.e. an established practice. As soon as one experiences a DoR, one is pressed to justify why one treats something (for example the mouse) unequally, and the BMC that equals are to be treated equally and unequals unequally helps one’s thinking to get back on track. One has to explain why the exclusion of the mouse as a vulnerable fellow creature is justified. If this justification process fails, one needs to rearrange one’s understanding of what a mouse is.

5 The Wittgensteinian-Diamondian Approach to Ethics and Moral Conservatism

In the previous sections we have tried to show that a Wittgensteinian-Diamondian approach to animal ethics is far from being morally conservative. Instead, it offers a profound practice-oriented approach to animal ethics which actually promotes change in human dealings with animals. In other words, it is compatible with radical critique:

Hence engagement in radical moral and political critique is ‘accepting’ the givenness of our forms of life, for these disputational practices enter into the constitution of the forms. But there is no compelling reason to count the peculiarly modern institutionalised practices of animal exploitation as constitutive of our forms of life. On the contrary […] they are contingent practices within those forms, and could be abolished without perverting the form or even any significant diminishment to human wellbeing.”Footnote 74

Although these remarks are Pleasants’, they apply to Diamond’s moral philosophy as well. Diamond, and hence Wittgenstein, remind us that we are situated human beings who participate in a variety of complex language games.

But we also see some conciliatory notes directed to moral individualism as it spells out the transcendental BMC that equals are to be treated equally and unequals unequally. Famously, Singer once argued that all animals—human and non-human—are equal, i.e. merit equal consideration of interests.Footnote 75 Although his theory was embedded in a—supposedly—morally impartial world, Singer highlights a contradiction that we would all agree exists, and he demands consistency in moral claims: We agree that all human animals are to be considered equally—and if there is no knock-out criterion showing that animals are not equals, we can see that we are treating equals unequally. From a Wittgensteinian-Diamondian point of view, moral individualism’s arguments are not some sort of super language game, superior to all others. Rather, they function as templates that one can put on human-animal relations to remind us of all the features we share with animals. However, the BMC does the moral job and is neither in need of justification nor an object of proper doubt: The rule is followed blindly and therefore BMCs help us as moral agents to find out what already matters.