How do moral theories stand to each other?

Moral theories, such as the variations on virtue ethics, deontological ethics, contractualism, and consequentialism, are expected – inter alia – to explain the basic orientation of morality, give us principles and directives, justify those, and thereby (if all goes well) guide our actions. I examine some functions and characteristics of the extant moral theories from a moral metatheoretical point of view, in order to clarify the generally assumed rivalry between them. By thinking of moral theories in analogy to languages it is argued that different moral theories are neither simply competing nor simply complementary; their respective orientations justify using them, in virtue of the problems they help to solve. But even if considerations about the functionality of a theory and the context in which it is created play an important role, they can neither be sufficient to determine these theories’ relations to one other nor for choosing between them. The challenge is to set criteria for the quality of a moral theory on a moral metatheoretical level and, in particular, to make room for future views on morality.


Introduction
Outside the university campus, people are wondering what they ought to do, but inside academic moral philosophy the question generally is: which theory about what to do is superior?
S. Guinebert ( ) Institute of Philosophy, University of Bremen, Enrique-Schmidt-Str. 7, 330 440, 28359 Bremen, Germany E-Mail: svantje.guinebert@uni-bremen.de A moral theory consists of more or less connected claims arranged to determine what a morally good or right action or stance is, and what it is that makes it either right or good. It thereby conveys, more or less implicitly, a set of assumptions as to what morality is; that view in turn specifies the type of norms or rules to be found and how they can be grounded. The many variations on virtue ethics, deontological ethics, contractualism, and consequentialism, present us with interesting proposals. Moral theories, understood this way, are generally supposed to be mutually exclusive: their multitude and diversity raises the question whether one moral theory should be accepted as the only right or true one.
The aim of this article is to examine some of the functions and characteristics of classic moral theories from a moral metatheoretical point of view, and to consider how far seeing moral theories as analogous to grammars clarifies what is generally assumed to be the rivalry between them. How can we approach the question of what makes a moral theory good or right? What are the specific strengths of different moral theories and how are we to assess their aims and scopes? Can we seek out specific spheres of life that invite the deployment of one kind of moral theory rather than another? I will argue that different moral theories are neither simply competing nor simply complementary; the orientation of each such theory is matched with, and to be justified by, the range of problems it helps to solve. But while the functionality of a theory plays an important role, it is not the whole story; making room for the moral viewpoints and theories of the future means taking on the challenge of settling criteria for the success of a moral theory.
I will proceed as follows. First, I will further explicate the point of departure of our investigation: here, we will be engaged not in moral theory, but in moral metatheory. Then I will briefly discuss an analogy between moral theories and maps, and argue that we will do better to think of moral theories as analogous to languages. Third, I will talk through some of the general features of moral theories, as well as some of the distinguishing features of specific positions. Finally, I will speak to the question of how we should understand the relations in which moral theories stand to each other, and how we might choose between them.

Moral Metatheory
"So, now just tell me: after all that, which of these is the right one?" It frequently happens that students of philosophy start a class on moral theory with the expectation that they will ultimately learn which moral principles are the right ones and how to assure that their moral decisions are correct. In my view, and I'm probably not alone, the challenge is to reorient those expectations: a teacher aims to present and discuss a number of different theories, and take up their fundamental orientations, their geneses, and their strengths and difficulties. At that point, we are leaving the field of theorizing about morality, to enter into the area of theorizing about moral theories. To think about the merits, the purposes and the status of moral theories isborrowing a locution from Jason Brennan -to engage in moral metatheory (Brennan 2007).
Philosophizing about morality is a different endeavor than philosophizing about moral theory. If we want to take moral theories as objects of philosophical investigation, as we aim to do here, we have to take a step back: "Generally, moral metatheory is not about the semantics, ontology or epistemology of moral judgments and terms. Rather, it concerns the semantics, ontology, epistemology, and methods of moral theories themselves." (Brennan 2007: 28) If we mean to examine different moral theories, we have to address questions like: What is the aim of a moral theory? To which (kind of) questions is it an answer? What makes a moral theory good, helpful, adequate, wrong, or better than another?
Generally speaking, moral theories -as the term is used in this paper -answer the question of how to behave in a morally good or right way. They are theories about morality insofar they are systematizations and theoretical representations of a substantial view of morality: within a moral theory, the claims are anchored in consistent assumptions about what is morally relevant, how we find that out, and what morality consists in. They are normative in the sense that they make claims as to what makes an action, a choice or a stance morally valuable. 1 Thereby, a moral theory takes a stand on what moral norms are: are they human inventions, or a moral law that exists outside of space and time, or rather the outcome of an agreement? It invokes metaethical and anthropological presuppositions, which are expressed in its choice of ethical concepts. That is why, with a view to certain subsidiary aspects of morality, we can assign moral theories to different groups, depending on how a specific moral theory answers questions such as: what concepts are morally especially relevant? -duty or virtue, for example -or the question of what metaphysical assumptions have to be taken on board -are there moral properties or moral facts? -or what makes an action morally right or good. Thus, we assign it to one of the familiar groups of moral theory, that is, consequentialism, deontological ethics, virtue ethics or contractualism. 2 Accordingly, the term "moral theory" will be used in the following for a systematic compilation of claims about morality that unites statements and explanations of what is a morally right or good action or stance, why it is and how we produce or adopt it. 1 It is almost impossible to provide more than these very rough statements about what moral theories are in general: any further specification would be applicable to one theory, but not another. Because of the fundamentally different views on morality that different moral theories convey, we cannot give a definition of moral theory as we could give one of car, expecting to fill in further details as we descend to particular variations on it. 2 It is controversial if even this categorization is appropriate (see, for example, Crisp 2015 for an argument that virtue ethics should not be considered as a standalone theory, but rather as a form of deontology), and we have to be aware that considerations typical of one family of moral theories also figure into moral theories that seem to fit in another group. As Christoph Halbig's differentiation between virtue ethics, ethics of virtue, and theory of virtue (Halbig 2013) demonstrates, virtues, as one of the key ethical concepts, can be found in moral theories that belong to very different groups of moral theories. We differentiate between different groups of moral theories in which the several concepts have different weights; although we know that categorizations often involve simplifications and a sacrifice of precision, we can seek out basic assumptions that can be assigned to different groups of moral theories. Now, if moral theories generally claim to render an account of morality, how can we evaluate moral theories and decide which of them are good, right or better? 3

Thinking about Moral Theories via Analogies
When we ask what to expect from a moral theory and what relations they stand in to each other, it will help to have the right analogy. We will try out two analogies: one between theories and maps, and one between theories and languages.

Moral Theories as Maps
David Schmidtz proposes that we think of moral theories as analogous to maps: a theory is a map insofar as it gives us a representation of the object it depicts. (Schmidtz 2007: 433) Such an analogy is helpful up to a certain point; it helps to clarify what we can expect from moral theories. They are neither truths nor realities, but (at best) helpful representations of a certain kind of reality, built to help us orient ourselves in our world. The analogy does justice to the fact that no theory can portray the reality as it is. And it does justice to the fact that every moral itinerary has to be endorsed by the agent in order to give guidance.
But there are at least two difficulties we encounter if we think about moral theories analogously to maps: firstly, moral theories differ from maps insofar as they have the moral destination already built in; and second, they are not only alternate maps of a given terrain, but rather, owing to their fundamentally different understandings of morality, present different versions of the terrain.
At first glance it might seem that they represent one and the same area, namely morality -but their different understandings of the term complicate that assumption. These complications would not be so important if the moral maps all came to the same conclusion. In On What Matters, Derek Parfit argues that consequentialism, deontological ethics and contractualism can be brought together into what he calls the Triple Theory, and have to be understood as different routes to the same destination. He writes that it "has been widely believed that there are ... deep disagreements between Kantians, Contractualists, and Consequentialists. That, I have argued, is not true. These people are climbing the same mountain on different sides" (Parfit 2011: 419). Such a position could be illustrated by three different maps representing the same area, giving different possible ways to one destination: the mountain peak.
If you ask Google Maps for directions to one and the same mountain peak, once for bicycles, another time for pedestrians, and a third for cars, you will get different route descriptions, depending on what is permitted to each. A bicycle can take some one-way streets the driver of a car is not allowed to, but not, for example, the freeway. But given the very different meanings and functions the disparate theories assign to morality, it seems to me that we should accommodate not only different routes, but also different destinations.
Imagine you are entering an elevator in your administration building to get to your office. Just before the elevator door closes, someone slips in, looks around and says "Where do I have to go?" In order to be helpful, you have to ask him what his destination is, and only then give him the appropriate instructions. However, if the representative of a moral theory is asked where to go, she answers very differently from the helpful occupant of that elevator, by giving both an itinerary and the destination: "What we all seek is Eudaimonia. To achieve it, you have to find the mean between excesses and extremes; just follow the wise person". Alternatively, "if you are exercising your capacity to reason practically, you will necessarily recognize that the most important feature you have to take into account and prioritize is autonomy (or: the best outcome for all concerned). In order to do that, you have to execute the CI-test (or: make choices so as to maximize the balance of pleasure over pain)." The different destinations are built into the different understandings of morality, and are expressed in the different ethical concepts moral theories prioritize. While maps are representations of a terrain, moral theories are depictions of two things at once: they represent fundamentally different terrains 4 , by providing a representation of how the moral world looks, and they give us abstract guidelines, with justifications for those guidelines, which are supposed to escort us through that world.
These differences between maps and moral theories are important enough to tell against the use of this analogy, and to speak for another analogy, namely, thinking about moral theories as like languages.

Moral Theories as Languages
We expect moral theories to tell us (inter alia) what is right or good. That is something that we also want from a language textbook -not concerning the morally right or good, of course, but concerning the linguistically correct. Up to a certain point, to endorse a specific moral theory is like speaking a particular language.
A protagonist in Muriel Barbery's novel The Elegance of the Hedgehog explains what grammar is for in the following way: "Personally, I think that grammar is a way to attain beauty. When you speak, or read, or write, you can tell if you've said or read or written a fine sentence. You can recognize a well-turned phrase or an elegant style. But when you apply the rules of grammar skillfully, you ascend to another level of the beauty of language. When you use grammar you peel back the layers, to see how it is all put together, see it quite naked, in a way. And that's where it becomes wonderful, because you say to yourself, 'Look how well-made this is, how wellconstructed it is! How solid and ingenious, rich and subtle!' ... I find there is nothing more beautiful, for example, than the very basic components of language, nouns and verbs." (Barbery 2008) In the analogy as developed here, the grammar she talks about corresponds to the ethical concepts we find at the core of our moral theories. To study moral theory may lead one to appreciate another instance of beauty: one can be thrilled and impressed by the eloquence and comprehensive thinking of moral philosophers, as when we are inspired by the Kantian celebration of the human capacity to react not only to needs, natural necessities, and heteronomy, but to think in laws. Some readers will react with consternation to his moral thinking; others may therewith be reconciled with humanity. In any case, when we look at the widely held, more or less "solid and ingenious, rich and subtle" moral theories, we can learn something about the views of morality they are expressions of, and reflect on their respective scopes, functions, and strengths. Just as it helps to compare different languages, to learn more about both one's native language and others, it helps to examine different moral theories, seeing them as the systematization of different understandings of morality.
Similitudes on different levels invite us to explore the reach and potency of an analogy between moral theories and languages. Accordingly, it is not surprising that various attempts of this kind have already been made. 5 The point I am after is that moral theories have basic concepts and structures that express their respective views regarding morality (as grammars are underlying structures of languages and allow for the expression of a certain worldview), and that these structures and basic concepts necessarily affect and shape how we think about moral relevance and goodness. 6 In particular, we will examine three enlightening parallels: moral 5 As far as I can see, similarities and differences between morality and language have been so far examined with other objectives and priorities. See for example the discussions of the Linguistic Analogy (LA) in Dwyer, Huebner, Hauser 2010 andin Enoch 2013; and for an application of it Schwind 2018. It is noteworthy that the discussion of LA is located mostly within or at the intersection of moral psychology and descriptive morality. See also the discussion between Sripada and Harman, concerning the innate bases of certain aspects of morality (Harman 2008;Sripada 2008). Harman notes various indications that speak in favor of developing an analogy between languages and moralities. 6 Benjamin Lee Whorf formulated a very similar thought about language: "Languages shapes the way we think, and determine what we can think about." (See, for his investigations on the relation between language and thinking, for example Whorf 2012). The Whorfian hypotheses about the relationship between human language and human thinking are controversial, especially those that might suggest that some natural languages are better for talking or thinking about some kinds of things, and others for others. It has to be noted that in the context of moral theories it would make no sense to engage in a discussion of a natural or innate superiority of any group or person behind a moral theory. theories and languages have room for normativity, they change over time, and they are expressions of a certain view.

Some Thoughts on Normativity
In everyday life, we know how to speak and we are able to differentiate grammatically incorrect from correct sentences, even if often we cannot state the underlying rule and cannot explain why the utterance is correct or incorrect. Matters might be seen as similar in the case of morality: We know that you do not need to work in moral philosophy and to argue about moral theory in order to develop a view as to what is morally right or wrong.
Textbooks for languages and grammars as well as moral theories provide systematizations of normative statements: they tell you what is right or the best, and why. But the role of normativity seems to be completely different when it comes to languages and morals, because -at least at first glance -the normative force of their rules and principles are drawn from very different sources. Due to the role of normativity, in both cases we lack the tools available in other fields for corroborating theories at our disposal. Like language textbooks, moral theories do not provide predictions on the basis of previous observations; but unlike these textbooks, there is no field investigation that would be decisive for the normative force of their guidelines. Linguists indeed take an inductive approach, but what they draw on and transform into a concept are not (only and primarily) predictions about how people will speak, but which use of the language is correct. Thus they are formulating normative statements, converting a statistical norm to an instruction; whoever wants to speak correctly has to do it this way. Moral theories are even farther away from such an anchor: recourse to statistical data plays no role. We cannot rely on experiments or use special telescopes to verify our assumptions about moral norms or values. 7 Moral theories express and justify a way the world could be, and a way of doing and being that is normatively conceived -the question of whether we ever observe a morally good or right act has no conclusive force concerning the correctness of a moral theory. 8 7 However, researchers in other disciplines cannot simply rely on just any empirical data to falsify a theory. Even in physics, not every observation that does not fit a theory is simply a reason to reject the theory. If Adams and Leverrier had not looked for a still undiscovered planet that exerts gravitational force on Uranus, but had abandoned Newton's gravitational theory instead, they would not have calculated the mass and position of Neptune (Okasha 2002: 15). As Chomsky says, in physics "you're trying to understand how it works, not just describe a lot of phenomena", and that means that you have to be careful about the informative value of empirical observations: The "Galilean style [...] is the recognition that it is the abstract systems that you are constructing that are really the truth [...]. And so, it often makes good sense to disregard phenomena and search for principles that really seem to give some deep insight into why some of them are that way, recognizing that there are others that you can't pay attention to." (Chomsky 2002: 98 f.).
As far as normativity goes, moral theories are even more disconnected from empirical observation than grammars. On the one hand, there is the widely shared assumption that a norm, in order to count as a moral norm, has to be justified and valid independently of contingent circumstances. On the other hand, moral theories specify more or less abstract and general norms to provide orientation and guidance -and to ensure this, in addition to the dissociation from the real world I have just mentioned, there must be a certain connection to it. To give helpful answers, in compliance with questions and problems that arise in the real world, they have to keep up with the times, so to speak.
That there is at once dependence and independence of the worth of moral theories from the current reality can be rendered comprehensible if we differentiate between a theory being effective in an absolute sense (that it gives us what we need to know about being and acting morally, for all time and in all morally relevant situations) and its being effective in specific kinds of situations and circumstances (it gives us what we need to know for specific morally relevant situations). This is not to be confounded with the kind of relativity that is often taken as the counterpart to absoluteness, as if we had to choose between the frying pan (absolutism) and the fire (relativism). We are dealing here with a very particular form of relativity: the connection between moral guidelines and the specific kind of situation for which they are appropriate. The importance of a connection between norms and problems they help to solve can be seen on a variety of levels, if we focus on their change and continued development.

About Changes through Time
Conceptions of moral theories and morality, like languages and grammars, change over time. We are all familiar with the continued development of Utilitarianism, introduced as a systematic account by Bentham and refined by Mill and his heirs 9 , as well as with the evolution of other moral theories. The development of languages and grammars can for example be observed in the emergence of French to Latin: not only words change over time, but also very basic grammatical structures. 10 There are interesting commonalities and differences that will support the proposal I have just made: we should not determine the goodness of a moral theory by referring it to any absolute standard of rightness, but focus on its connection to specific situations and circumstances and examine its response potential.
Languages change depending on social and personal needs of expression, i.e., for pragmatic reasons. They change when new entities appear that have to be given names; sometimes new terms are created, sometimes terms are borrowed from other expedient, depending on the factual state of abundance and actual nature of the agent), but they are never sufficient to develop and validate a moral theory. 9 Although "proto-utilitarian positions can be discerned throughout the history of ethical theory", with early precursors such as "Cumberland, Shaftesbury, Hutcheson, Gay, and Hume." (Driver 2014). 10 Meisenburg and Gabriel note for example the loss of morphologically marked cases as one of the most typological features that vary from Romanic languages to Latin. In Old French we have a system with two cases, in other Romanic languages the system of cases has widely been abandoned (Gabriel/Meisenburg 2007: 140). languages (as "PC" found its way into German, while French invented the "ordinateur"; maybe one day we will need a new kind of category in between subject and object, together with a new kind of verb, for the behavior of robots). In contrast to the rules of a grammar, moral principles and guidelines are usually supposed to be grounded and justified in an ahistorical way: moral theories and moral philosophers typically purport to have seized the moral once and for all. 11 We are used to thinking of moral theories as universal and eternal. However, we should think not only of languages but also of moral theories as children of their time, 12 and examine how and why they undergo change.
There are two main ways for moral theories to change that correspond to two sources of criticism: moral theories can be modified because deficiencies and gaps are revealed either by their application (whether it be in real life or in thought experiments) or on an abstract level, in theoretical debates. Often these two reasons for development and change appear together and are argumentatively combined. 13 Generally speaking, moral theories change by and by as single aspects are picked out and discussed in theoretical debates. Such revisions yield the variations that together form the familiar groups of moral theories. Larger and more extensive changes occur if a traditional theory is judged fundamentally insufficient; then a moral theory can be built that grounds a new group of theories.
As we need new terms and maybe even new grammatical structure, so we need new moral rules and principles, as the world in which we live changes: for example, the biblical commandments were helpful in earlier times (e.g. levirate marriage); some norms are helpful for the local environment, but need to be reworked when the scope of our activity changes (see for example Hans Jonas' attempt to develop an Ethics for the Technological Age, 2003); in present times, we need guidelines and orientation for handling hyperspecialization (a problem that was not even imaginable some hundred years ago; cf. Millgram 2015). It is not the rightness of a moral theory that changes over time, but the frequency of possible applications, i.e. the connection between different moral theories and the questions and problems agents actually face. 14 11 Note that an Existentialist has no difficulties here if he assumes that we create the values in relation to the actual world we are living in, as Sartre describes it in Being and Nothingness. 12 This is a bitter pill to swallow, especially if we believe that a philosopher has to "overcome his time", as Nietzsche writes at the beginning of The Case of Wagner, 1888. 13 In Rawls' Theory of Justice (Rawls 1971), we find a way of dealing with moral theories that unites both. In a nutshell, he claims that morality has to be about welfare; but that is not enough because we also need justice; the method of reflective equilibrium (cf. Nelson Goodmans introduction to this method in Goodman 1955:67) allows us to design a theory that collects thoughts both on justice and on welfare, accounting in addition for commonsense intuitions concerning morality, and thereby combines more abstract concepts like justice with pragmatic, historical views about questions of implementation. In this regard, he follows Rousseau who in his Social Contract presented an attempt to bring together justice and utility (Rousseau 1762, esp. I .1 and passim). 14 Consequently, a particular theory can become moot, in that it supplies answers for which there is no longer any need. Since extant moral theories were developed in answer to historically localizable and context-related questions and problems, we can recommend thinking about what questions we might ask in ethics today and in the future. Note that we are thereby saying nothing about absoluteness (or the sense or possibility of it) in ethics and morality. Concerning the difference between propositions about absolute and relative goodness, Wittgenstein ends his Lecture on Ethics thus: "Ethics so far as it springs from the

Theories as Expressions of Specific Views
It has become proverbial that learning a new language is getting to know a new world. As different grammars underlie and adjust languages, each reflecting a specific worldview and cultural heritage, so different moral theories support and express different interpretations of what counts as morally relevant, what as moral value (s) and what as principle(s). We can learn a new language and thereby acquire a new interpretation of the world. And we can learn to see through the lenses of new moral theories and thereby acquire new conceptions of morality, and their accompanying interpretations of given situations. A language enables us to express our perceptions of our surroundings and of the world, but it also sets limits: Every language conveys different associations, and is particularly good at allowing the expression of one thing, context, and connection, rather than another. 15 In the same way, we have to understand each moral theory as suitable for the expression and structural reinforcement of a specific understanding of what morality is. This is not meant in the same sense as would be the expressivist pronouncement that a moral judgment is the expression of a personal commitment -it is rather a take on what morality is. Thus, one theory highlights autonomy, another pleasure and pain, others equality, and still others stances and virtues. The different aspects that are taken to be morally relevant go hand in hand with the structure and concepts a moral theory provides. The underlying structures of morality, that is, the different ways to understand it, are determinant: Seeing the world through deontological eyes is a different kind of moral understanding than seeing it through contractualist or consequentialist eyes. 16 I want to suggest that these differences are due, inter alia, to the questions moral theories are supposed to answer. I want to see how to draw characteristics out from some well-known versions of moral theories, to show how a moral view is determined by a specific "diction". Can we find questions or even spheres of life that suit specific moral theories (or groups of moral theories) better than others? If so: What are the contexts in which a specific theory provides the best expressive possibilities, and offers the best guidance? desire to say something about the ultimate meaning of life, the absolute good, the absolute valuable, can be no science. [...] But it is a document of a tendency in the human mind which I personally cannot help respecting deeply and I would not for my life ridicule it." (Wittgenstein 1965: 12). 15 Every language reflects a distinctive way of living. This becomes obvious, for example, if we look at the various ways of closing a letter; anyone who has ever tried to find the appropriate words to finish a letter written in a foreign language has encountered these stumbling blocks. Those who have no ears for French standards of written politeness will be astonished to find as a translation of 'Sincerely': «Je vous prie d'agréer, Madame, Monsieur, mes salutations distinguées». 16 So far it is not justified to say that some languages incline one to think ethically in one way rather than another (and that is a descriptive matter we won't cover here), and it is probably also saying too much to claim that moral theories are interpreted in different ways, depending on the language in which they are expressed; we discuss the one and same theory, for example Utilitarianism, transnationally and we manage to identify the same problems. But there are differences between discussions of the same theory within different language communities -some strands of discussion are confined by language of publication or by a vision of philosophy, for example, in parallel discussions within continental and analytic philosophy. And sufficiently disparate languages with different grammatically basic categories can certainly make the understanding of a theory at least very difficult.

Different Moral Theories and Their Respective Strengths
Each in its own way, the extant moral theories tell us what we ought to do and how to find that out. The differences in their answers are due to, among other things, the different personal, social, and philosophical backgrounds against which these theories have been built up and the questions that emerged under those circumstances. It is not by accident that ancient virtue ethics, focused on excellence in one's character, has flourished in times when people believed they lived in a well-ordered cosmos. It is not by coincidence that Hobbes, Rousseau and other contractualists emphasized equality, in times when the social and political background prepared people to think about equal rights. Kant wrote his Critiques when people still could hope that the use and development of reason would solve the various difficulties humanity faces. And the Utilitarianism of Bentham and Mill goes with an aspiration to concretely and effectively change society as a whole for the better, paying attention to human needs and to real-world conditions. Depending on what exactly is being sought and on how the search is performed, 17 different implications are built into the questions and the answers. That goes with a thought Wittgenstein noted in his Philosophical Remarks: "I should like to say: for any question there is always a corresponding method of finding. Or you might say, a question denotes a method of searching" (Wittgenstein 1998: 77). According to David Schmidtz, a "context is a question that motivates us to theorize. 'What are my children due?' is one context. 'What are my employees due?' is another." (Schmidtz 2007: 448) If we want to know which contexts a specific theory is appropriate for, we have to look at the questions to which they give answers and find spheres in which we can take advantage of their respective particular strengths.
In order to get there, I will take a bird's-eye view; this means that I won't do justice to any one theory. Instead I will be briefly pointing out some characteristics of particular variants, in order to move towards an answer to the question of how to find what different moral theories are good for. Some classical, only roughly outlined variants of the extant groups of moral theories will serve to illustrate the following point: In fact, the variants of the different groups of moral theories are supposed to give answers to different questions, and these questions can help us to locate their strengths.

Virtue Ethics
The ancient variant of virtue ethics 18 is commonly taken to answer the question of what sort of person we want to be and how we should live. Someone who accepts virtue ethics as one of the major approaches of normative ethics, that is, as a moral theory discussed in the sense here, might state that it "emphasizes the virtues, or moral character, in contrast to the approach that emphasizes duties or rules (deontology) or that emphasizes the consequences of actions (consequentialism)" (Hursthouse/Pettigrove 2016). While other moral theories may, and indeed often do, also speak to the virtues, for virtue-ethical theories, virtues and vices are foundational, and other normative notions can be grounded in them. We should understand those virtues as excellent traits of character.
Aristotle holds that there is a characteristic function humans have, namely the "activity of the soul in accordance with virtue, and if there are several virtues, in accordance with the best and most complete" (Crisp 2000(Crisp : 1098a. One underlying idea of his theory is the existence of a well-ordered totality in which the human being is embedded. By precise observation and exploration of the human constitution and its position within this arrangement, we can arrive at an impression of the telos of human beings and of how they have to conduct themselves, to be in accord with their nature and the cosmos as a whole. Accordingly, as Odo Marquard observes, the Greek conceptual world and thus also the roots of virtue ethics are elicited in a world shaped by "the teleological, purposive and functional constitution of the world and of course of the human." (Marquard 2013: 58, my translation).
That means that Aristotelian virtue ethics was outlined in a significantly more orderly world than the one in which we find ourselves today. Can we nonetheless assign to virtue ethics a sphere in which its response potentials can benefit us the most? In a nutshell: Virtues are not insulated attitudes, but pertain to a network of connected preferences, hopes, beliefs, and inclinations. The adoption of virtues leads to a kind of personal order that works to orient and guide a good life 19 , and thereby is still of great value at least in the personal and intimate sphere. Within the personal cosmos, virtues we develop and consolidate can play the same orienting role they do for a person living in a well-ordered cosmos; up to a certain point, we can order our life cosmos by developing views of our own, making ourselves into the kind of person who tries to become excellent in some of her activities and relations.

Contractualism
In a broad sense, contractualism can be construed as the view that moral norms derive their normative force from a contract or mutual agreement. Here is how Rousseau formulates the challenge: "Find a form of association that will bring the whole common force to bear on defending and protecting each associate's person and goods, doing this in such a way that each of them, while uniting himself with all, still obeys only himself and remains as free as before" (Rousseau 1762: VI). At the time, the notion that every individual is equal and counts -that every person takes an active part in determining the content of the laws and that doing so is required for their legitimacy -was a pioneering proposal. Important aspects of contractualism are (among others) the more or less implicit idea of equality, as well as the demand to acknowledge this equality by integrating everyone into the decision-making that determines the joint rules of play. The ideal basis of common rules would be -if we follow Rousseau -that every person is integrated in a way that makes her amenable to the law, and makes her follow it, all the while remaining obedient only to the law she has given to herself.
The difficulties are enormous and the approach continues to be controversialstill, some of these problems do not arise from Rousseau's conception, but rather from our contemporary conception of democracy. There is a difference between joint decision-making, where everyone puts his own personal needs and desires on the table, the formulation of norms then consisting in the finding of compromises, and a form of decision-making in which everybody puts on the table what he thinks to be the best norms and obligations for the group and the common undertaking as a whole. We live democracy construed in the first of these ways; but that is not a sufficient reason to simply drop from the agenda the Rousseauvian so-called common will. Collective decision-making need not be endorsing the wishes of the party with the strongest lobby or the summing up of preferences; rather it would be most expedient if the functioning of the decision-making community had priority.
Contractualist thinking, understood this way, should be expedient and helpful in spheres that aim at joint action and constellations of collective life that are delimited in scope, which can however be dissolved at will. Here I'm thinking for example of residential communities, or political parties that have a small, limited membership, but also of personal relationships. These common undertakings are small-scale in the sense that everyone has to be able to influence the procedure and the course of the joint project. Interpreted this way, to apply contractualism would mean to engage in group undertakings in which equality of rights and equal access to decision-making are realized.

Deontological Ethics
The idea of a law one gives oneself found uptake in deontological ethics, and in particular in its Kantian version. Here our capacity to reason is the subject of criticism and is, in Kant's view, proved to be much more important for morality than any individual interest or need could ever be. It is our reasoning that helps us to discover and formulate what "morally to ought to do something" means. In his view, something we ought to do must be something that can be expressed in the form of a law; a law in turn is something that is valid for everyone, independent of personal needs and inclinations. Every personal guiding principle -every maxim -that complies with what is morally right must be consistently thinkable as a law (Kant 1785). 20 What is the strength of an approach in morality that takes freedom from contradiction and valuing practical reasoning as its central features? Freedom from contradiction goes too far when it bans purely benevolent actions, or prevents one from building one's own personal cosmos, and so thwarts practical reasoning. 21 But freedom from contradiction may play a decisive role in the construction and execution of intentions. To intend something is to be differentiated from merely wanting it by the normativity integrated into the concept of intending. In her "Consistency in Action", Onora O'Neill formulates requirements of rationality that intentions have to meet. For example, "it is a requirement of rationality not merely to intend all necessary and some sufficient means to what is fundamentally intended but also to seek to make such means available when they are not" (O'Neill 2001: 312). Already as children, we learn that intending something comes along with self-regulation and renunciation. To intend something and to take the necessary steps towards accomplishing it are closely connected to our ability to reason practically. We have to acknowledge this ability in order to deploy it.
Whether it is preparing a talk, producing an artwork, or fulfilling any long-term plan, so long as we intend something, we have to target a certain freedom from contradiction in our intentions, and to value our capacity to reason, in order to fulfill two necessary conditions of success. Thus, valuing one's own and others' capacity to form intentions that are sufficiently free from contradiction to be successfully pursued marks appropriate areas of application for deontological ethics.

Consequentialism
According to consequentialism, whether an act is morally right depends on its consequences: "an act is right if that act maximizes the good, that is, if and only if the total amount of good for all, minus the total amount of bad for all, is greater than this net amount for any incompatible act available to the agent on that occasion" (Sinnott-Armstrong 2019). Its paradigm, advocated by Bentham, Mill, and Sidgwick, is Utilitarianism, i.e. Hedonistic Consequentialism. Since then, many refinements and specifications of consequentialism have been undertaken; luckily, for present purposes it will suffice to focus on the thought that the rightness of an act depends contradictions and has to be valued if we want to speak about (moral) decision making in a meaningful way. How important freedom from contradiction is in philosophical thinking and in terms of the individual's inner constitution is controversial. 21 On the one hand, we have to admit that if we have too many conflicting personal principles, we can no longer be oriented by them; moreover, a principle is only a principle if it is not only applicable once but applies in several sufficiently similar contexts. Of course, Kant continues far beyond this point: noncontradiction is the touchstone for the moral approval of the use of a maxim. On the other hand, there are indications that contradictions are worth maintaining or at least not to be resolved by all means. Rousseau declares that he prefers to be a man of contradiction to a man of prejudices. Emerson insists that we should not even try to be consistent: "a foolish consistency is the hobgoblin of little minds, adored by little statesmen and philosophers and divines. With consistency a great soul has simply nothing to do" (Emerson 2015(Emerson [1841). It seems quite plausible that, in order for inner coordination to provide orientation, freedom from contradiction is needed. Nonetheless, to make every contradiction disappear would mean to diminish our inner versatility. on its consequences, in order to think about the accompanying two-phase nature of the concept.
What is it that makes consequences good, bad, or worth being realized? Clearly, we need an axiology. But how do we identify which criterion should be decisive for the evaluation of consequences? The answer to this question ultimately depends on anthropological assumptions. Some classic representatives, such as Mill and Sidgwick, clarify this issue by invoking a non-moral good, namely the greatest happiness of the greatest number. The decisive role of assumptions about the way we are and the consequences of what we do for our own lives and those of our fellows is connected to the impression that, if we want to live in a better world than we do, it doesn't suffice to concentrate on one's own virtues, or follow the rules of practical reasoning; rather, we have to take concrete decisions leading to actions that have a positive impact on the world we live in.
In order to make the world a better place for (as far as possible) all sentient beings, an agent has to have a view to the real-world possibilities and conditions. One important challenge is to identify good or even the best consequences; another challenge is to discern means to attain these objectives. 22 Our options and circumstances are in part determined by structural features of domains in which joint action is needed, but equal participation is impossible. The two-phase nature of consequentialism makes it an optimal theory inter alia for those spheres of life in which an antecedently determined definition of the target is available, and a person has to find out by instrumental reasoning what to do to reach that goal. That is why consequentialism is adequate for spheres in which someone has to live up to a prospective responsibility, for instance, if she or he holds political office. From this point of view, consequentialism captures -among other things not mentioned here -the deliberative structure of some role-based reasoning.
To be sure, many other aspects of the several moral theories could be emphasized, and they have been painted with a broad brush. Moreover, some theories may belong to more than one sphere. Nonetheless, I suggest thinking of moral theories as being more or less good if they supply moral guidance in a certain kind of situation, i.e. if they fit a question.
If we assume that every moral theory provides its own tools and concepts with which to describe and evaluate the morally relevant aspects of a decision situation, we have to recognize that, for one thing, each theory comes with a specific view of morality, and that, on the other hand, we cannot draw from it an independent and self-evident criterion to choose between different theories. In Schmidtz's words: "if opposing players are disputing a rule, we cannot settle the dispute by consulting a player. We need a referee" (Schmidtz 2007: 441). Simple functionality, as we have characterized it so far, cannot be the criterion we seek: what functionality means in this context is concretized in many different ways, and different moral theories will imply different renditions of it. Thus, one outstanding and crucial question is what mode of thinking is to handle overlaps.

How to Choose the Suitable Theory
If you do not take as moral only those problems that are addressed by a given moral theory, you may allow that different contexts and respective questions are best dealt with by different theories -for "[n]o one moral theory adequately addresses the full range of moral problems" (Brennan 2007: 61). In other words, and falling back on the analogy: there is no single language that enables us to express everything equally well, and that should be henceforth spoken by everyone. 23 It is good to master one language -to speak two or more languages is even better, not least because it extends one's world view, one's ability to act and one's scope of activity.
As long as we find contexts in which a moral theory provides helpful strategies, we should retain it in our toolkit. But what does it mean for a moral theory to give "helpful" recommendations? This is again a normative question. Not everything that helps or works is necessarily good. And as long as we have no criterion for determining what kind of "good" we are looking for, a great deal of work is still to be done. One question is how different moral theories can be combined; another is criteria for appropriate application.

Possibilities of Combination
If every single moral theory has its own strengths and domains of competence, there are at least two different ways to combine them: either we combine the different aspects of each into a single moral theory, thereby endorsing pluralism; or we retain them as distinct and advance an overarching theory assigning the subsidiary moral theories their respective spheres. I recommend an approach to be distinguished both from pluralism, at the level of normative theories, and from attempts at the level of metatheory to combine different theoretical approaches into one unified blend. In contrast to pluralist theories, rather than a normative theory involving several nonreducible values, the recommendation is to deploy several moral theories, without attempting to unify them. Thus, I am rejecting the approach in which these theories would converge on one principle, as in Parfit's Triple Theory. 24 This rejection is due to something that becomes apparent in a limitation of the analogy between language and moral theories: there is no common universal structure shared by all 23 With Esperanto, there was a famous attempt to create a common world language; it is by now clear that the initiative was unsuccessful. Remarkably, Wittgenstein notes: "Esperanto. The feeling of disgust we get if we utter an invented word with invented derivative syllables. The word is cold, lacking in associations, and yet it plays at being 'language'" (Wittgenstein 1980: 52e). 24 Cf. Parfit 2011. Another example is the attempt of von der Pfordten: "An adequate normative ethics contains central elements of each of the four families of ethical theory" (von der Pfordten 2012). For an approach of this kind in applied ethics, see Beauchamp & Childress (2009). In contrast to the thoughts presented here, these are attempts to pick out specific parts of the different theories to create a further theory. Here, the aim is to preserve the different theories as such. moral theories, that could serve as the analog of a common, underlying "universal grammar", and would allow us to combine their strengths in one joint approach.
In Chomsky's view, we have to assume that there is an innate universal grammar, which all languages embody and from which particular grammars are derived. If there were something like a "universal grammar of morality", we could try to transfer Chomsky's treatment of grammars to morality. But, as we saw, the questions to which different moral theories propose their conflicting answers are not the same. They are sufficiently different to suggest that there is no common basic point of origin or objective, and that we should not hope for a common underlying structure. Even though there are some commonalities (in almost every moral theory we find an expression of the principle of no harm, as we find in every grammar a differentiation between subject and object, for example), the point remains that they serve to highlight different aspects of a situation as morally relevant. Jason Brennan introduced the concept of "Theory Multiplism", arguing that we should take the extant moral theories as conceptions that supplement each other. In his view, "some contexts call for simplicity, some for impartiality, and some for rigor" (Brennan 2007: 88), and "the multiplist might employ each [moral theory] for different purposes" (Brennan 2007: 89). We are now in a position to tighten up his concept: if we need orientation for decision-making in groups, we should play the contractualist card; if we need orientation in the intimate sphere, we should play the virtue ethics card, and so forth. 25 There are situations that one theory can help to resolve and another not. 26 So, to answer the question of which moral theory is the right one in a given context, the first step to take would be to follow the guidelines concerning different strengths for different contexts. We can extract from the roots of virtue ethics a fruitful conception of virtue: virtues help us to arrange a more or less well-ordered personal cosmos. We can extract from a version of contractualism the idea that equality combined with a joint project presupposes common decisionmaking that integrates valuing the community and therefore paying attention to the common will. We can learn from Kant to always value the ability to reason practically and to act out of our rational capacities. And we can extract from consequentialism the idea that instrumental reasoning is the kind of practical reasoning we are committed to in cases of fulfilling an antecedently determined goal.
So far, we could be tempted to say: A moral theory is good (or, better than another one) if it ensures orientation in decision-making when needed. But here the analogy to language runs out; it is questionable whether the justification of a moral theory can and should ultimately be purely pragmatic. Perhaps what enables us to get along best in practice is (morally) mistaken? 25 The proposed model could be named a "puzzle model", to differentiate it from a structurally different way of combining different moral theories, such as a "multilevel model". Dagmar Borchers is developing a multilevel model that sees, for example, the strength of consequentialism in the weighing in case of conflicts between duties, and thereby proposes another way to think about different theories jointly. 26 A further point which we won't discuss here: it might be argued that we need the different, opposing views to keep thought and discussion alive and critical. See Mill's picture of the danger of uniformity in On Liberty (Mill 1859).

How to Choose the Suitable Theory
What we have so far is not enough: we need a metatheory, to tell us when to release the clutch, work the gear shift, and bring a different moral theory to bear. We need a criterion that lets us decide when one or another theory should go into effect. How do we decide which one to choose?
To ask for the specific strengths and functions of a moral theory is to ask what it is good for. But that again is a normative question. Exclusion is sometimes called for on the basis of consistency or elegance, sometimes of the scope of application, but often it will actually be due to normative considerations. Thus it may seem we need to think about normative criteria for counting a theory as moral. 27 But that would only move the initial question to another level and -again -ask what morality is in order to find a systematization of it. Given that we do not have a precise and perfected understanding of morality to start with, I propose to go at it from a different angle: What can we say about criteria for excluding or choosing a theory? The remaining pages will serve to determine three minimal aspects the criteria we are after have to address: It must help us to set respective scopes of application to the moral theories; it must allow for further and new moral theories; and at best it would be self-limiting. 28

Determining Scopes of Applications
Discussions of classical moral theories reveal not only strengths, but also weaknesses. Time and again the thought reappears that Kantian deontological ethics, with its reputed rigidity and inelasticity, cannot do justice to an environment that demands versatility, and to the moral questions and dilemmas with which we find ourselves confronted; to apply deontological ethics in some contexts thwarts moral action because the procedure is too rigorous -it is important to leave room for exceptions of benevolence, for example. The answers found by virtue ethics can blind us to those rules that should be followed by everyone -sometimes generalizability matters. Consequentialism leads to coarseness when we fail to value autonomy and personhood in contexts in which they are at stake. Just as there "is no difficulty in proving any ethical standard whatever to work ill, if we suppose universal idiocy to be conjoined with it", as Mill famously wrote in his Utilitarianism, there is no ethical standard that cannot prove unhelpful and indefensible. But finding the limits of a theory doesn't mean rejecting it. On the contrary, to find their limitations is to ensure their applicability. To see the scope of its application and to find its defeaters helps us use and understand a moral theory appropriately. 27 It is encouraging that we often seem to agree about the weak point of a moral theory. But the phenomenon still needs to be investigated. We do not have a metatheoretical theory of the acceptability of moral theories, but we do nonetheless often agree on propositions like "such and such is an important weakness of utilitarianism / deontology / ..." and "That cannot count as a moral theory". 28 Without doubt we find important discussions of further criteria a moral theory has to satisfy in order to be a convincing account, as for example "explanatory" and "normative adequacy" (cf. Korsgaard 1996). It is not the place to discuss them here: I want to sketch out only those criteria closely connected to the argumentation developed here; the listed criteria certainly can and have to be complemented.

Allowing for New Moral Theories
If I'm right, and moral theories answer different questions, the moral theories we have are not exhaustive. New developments will raise new questions, and so press us to find new guidelines for acting -and future moral theories will reconceive morality in fundamentally different ways. It is to be expected not only that new variations on the classical moral theories will arise, but also that fundamentally novel concepts will emerge. Humanity's upcoming challenges will be different from anything we can picture, and so other kinds of moral theories will be needed. We must make room for further moral theories and for practical reasoning that enables us to find and invent new values. Who knows which moral questions and (approaches to) answers may prove beneficial, in the situations in which humans will find one day themselves, situations we cannot even imagine today? And that again means that we need not only to assess the existing different theories with their strengths, weaknesses and appropriate field(s) of application, but to make room for new moral theories. Openness is mandatory.

Being Self-Limiting
These two complications suggest a first approach which leads us to a necessary condition on choice between moral theories. My tentative final proposal is: selflimitation matters; and freedom is a good example for a relevant and promising candidate.
It has been proposed to assess moral theories on the basis of fit for particular problem-solving domains. If we turn to the question of metatheory, that is to the question of which criteria to invoke to choose between moral theories, the corresponding thought would be to choose the one best suited to that domain, that is, to pick out one rather than another moral theory for a particular application. At this point we want to avoid at least one thing: an infinite regress. We do not want to find various criteria that are, firstly, finite, concrete, and exclusive, and that, secondly, come with a claim to be all-embracing. The challenge of choice on the first level arose because moral theories are generally supposed to be mutually exclusive and all-embracing; if the selection criteria were self-limiting, we would at least prevent this regress. A self-limiting criterion is one that entails constitutive limits, i.e. selfregarding reasons to restraint. Freedom is self-limiting in the sense that, even if there were no value worth restricting our freedom for, for freedom to prevail it has to limit itself: The best reason to limit freedom is the enabling of freedom.

A Preliminary Conclusion
When we have to choose between different moral theories, we should give priority to the theory that preserves openness and freedom. As long as humans are so multifaceted, under development, and undefined, a moral theory must guarantee that we remain free enough to step back and engage in moral reasoning -whatever "morality" may mean.
There are various contexts and situations in which more than one moral theory is equally helpful. If there is a drowning child to be saved easily, you will come to the conclusion that you should rescue it, whichever moral theory you consult. But given the different areas in which we are engaged in our everyday and moral life, and which ask for different foci, we should embrace polyphony and be thankful that we have several moral theories at our disposal.
When we face different demands, depending on which theory we apply, we should look to their strengths and weaknesses, and turn to the option with the most adequate moral assumptions for the specific context, while making sure that we promote and respect the possibility of inventing new ones. It seems advisable to be oriented towards selection criteria that comply with maintaining openness and meet selflimitation -as, for example, freedom does.