The Nature of Morality

Moral philosophy is an investigation of moral phenomena. However, morality is a very elusive notion and R. Perry had certainly a point when he complained that nothing is more familiar, and nothing is more obscure in its meaning than ‘morality’.1 It is thus generally accepted that it is one of the main aims of philosophical ethics to clarify this concept.2 In this paper I try to answer the question what morality is. Unfortunately, a discussion of the many other attempts to solve this problem would require too much space. I will therefore concentrate on a presentation and explanation of my view which, I think, avoids the mistakes of other accounts while preserving their insights.3 In order to explain what morality is, I will investigate what conditions a being must fulfil if it has a morality. In my view, without understanding what it means to have a morality we cannot adequately understand the nature of morality. Morality is a quality of humans and if we want to understand moral phenomena, we must above all understand what it means to have a morality. It is this latter notion I will try to explain in this paper. The definition I propose is that

S has a morality if and only if attitudes of others are to S relevant as such.

The Sections (I)—(VI) explain this definition and in (VII) I will give a concise argument for my account.


In this paper my main concern is with the nature of morality. We speak, however, in at least two different ways about morality. Usually we use the term evaluatively (’morality’ refers then to what is moral as opposed to what is immoral). But we also speak of morality, moral problems, moral judgements, moral virtues, moral emotions, and so on in a not evaluative way. We mean then that these things belong to morality and not, for example, to law or convention. The opposition of moral is then non-moral or amoral (but not immoral). It is this latter use of ‘moral’ and ‘morality’ that concerns us here. This means, I do not try to elucidate such normative questions as when a person or any other being is morally good or bad and when actions are morally right or wrong. My question is rather about the necessary and sufficient conditions for morality in the non-evaluative sense. My aim is therefore to demarcate morality from other domains which resemble it in some respects as, for example, legal systems, aesthetics, prudence or religion do. I try to demarcate morality from these areas by asking when a person or any other being has a morality. The definiendum of the given definition is therefore ‘S has a morality’.

If the nature of morality must be determined by investigating what it means to have a morality, the next question is, who can be said to have a morality. We can say that individuals have a morality but also that societies have a morality.4 In my view, the primary subjects of morality are individuals and we can only say that a society has a morality if a sufficiently high percentage (which I will not try to specify here) of its members have a morality.5 The ’S’ in the definition refers therefore to individuals by whom I mean primarily humans. But also other beings are not excluded. Even some animals can have a morality if they satisfy the definiens. Whether any animal can do this, cannot be determined by philosophical means, but the definition is open for it.6


According to the definition given, morality has to do with the attitudes of others. ‘Attitude’ is in ethics a technical term, meaning essentially that we are for or against something. In short, if we are for peace, we have a positive attitude towards it and if we are against wars, we have a negative attitude towards them.7 As I have argued elsewhere,8 having an attitude towards something is the same as valuing it. Because of this identity, we can also say that a person has a morality if valuations of others are to her relevant as such (or for its own sake).

It is, of course, true that a variety of other things have been proposed to be materially basic for morality (for instance, happiness, pleasure, well-being, or informed desires) and I should therefore make clear why I regard attitudes/valuations as basic. In other words, I should explain why I do not think that a person has a morality only if, for instance, the happiness of others is to her relevant as such. For lack of space, a few remark must suffice to make plausible why I regard the evaluative attitudes as basic for morality. First, all other candidates (such as happiness or pleasure) are valued. They are regarded as good, which means that people value them positively. Second, I think it is obvious that all other candidates are morally relevant only because they are valued. Some consider happiness as basic for morality because it is (in their view) intrinsically valued and hedonists regard pleasure as basic for morality because (as they see it) only pleasure is intrinsically valued. This means, in other words, that the valuation makes something a possible candidate for being materially basic for morality. From both considerations together follows that if something is valued, it is morally relevant, and if it is not, then it cannot be basic for morality. This, however, suggests that being valued is necessary and sufficient for anything that can be reasonably considered as morally basic. It is necessary because something cannot be basic for morality if it is not valued and it is sufficient because it is relevant for morality, if it is valued. It is only this apparently obvious proposition, which I need for a defence of my definition. These short remarks will become clearer when I explain now what it means that an attitude is relevant.

(a) Even if only the attitudes of others are basically relevant for morality, beings that have a morality are not affected by these attitudes themselves but by their beliefs that others have them. What causes a moral reaction of a person is her belief that someone else has an attitude and not the fact that this other being has in fact an attitude. This can easily be seen because the attitudes of others and our beliefs that they have such attitudes are independent of each other. One can exist even though the other does not. We can thus distinguish four cases. (1) A person A believes that B wants her to visit him and B does want it.9 If A has a morality, this belief should have an effect on her. (2) A believes that B wants her to visit him, but B does not want her to do so. In this case the moral reaction of A should be the same as in (1). Let us now (3) assume that A does not believe that B wants her to visit him, but B is longing for it. In this case it does not speak against A’s morality if she has no intention to visit him. (4) Finally, if A does not believe that B wants her to visit him and B really does not want this visit, the case is the same as in (3). This simple example shows that it is not crucial for morality that a person is influenced by the attitudes of others, what counts is rather that she is affected by her belief that others have an attitude. Saying that the attitudes/valuations of others are relevant to a person means nothing else than saying that she is affected by her belief that someone else has (had, will have, or could have) an attitude towards something. If a person has a morality, attitudes of others or things that are (putatively) related to them are relevant to her, but she is (at least sometimes) affected by her beliefs about them.

These beliefs need not be rational. If I believe that my house-plants literally suffer from the bad air in my room and I open the window to give them some relief, this behaviour may speak against my rationality but not against my morality. However, for a moral reaction some cognitive processes must be involved. If we imagine a being that is affected by attitudes of others through a purely physical causal chain (without any involvement of cognitions), it would have a morality only in a metaphorical sense.

(b) Valuations or attitudes can be dispositional or manifest. Relevant are in my view basically only manifest or actual attitudes. To my mind, dispositional attitudes count only insofar as they can become attitudes we actually have. Some examples may illustrate this. A father has an aversion towards pop music which his children love. Out of consideration for their father, they do not listen to their favourite music if he is at home. One day, when their father is out, his son is about to play a CD of his favourite pop group, but is urged by his sister not to do so because their father cannot stand this kind of music. The son, however, need not have any moral scruples to listen to his preferred music, even though his father has also a dispositional aversion towards it while he is absent. On the other hand, the children’s decision to buy their father a new record of Bach’s St. John Passion is a moral decision, if it is based on their belief that he likes this piece of music. But it is not so because of his dispositional attitude but because of the expected actual enjoyment (which is an attitude) when he listens to this music. This can be seen if we assume that the children believe that their father is deaf. In this case, the children’s decision does not any longer show that they have a morality, even though their father has still the same dispositional attitude.

(c) The attitudes can be present, past or future. Florence Nightingale is said to have regarded it a sin to whisper in a ward in which patients were dying. In my opinion, she did so because she believed that the whispering could worry the dying since they may see it as a sign that things were not looking good for them. On this interpretation, Nightingale was motivated by an anticipated future attitude of the patients and if her motivation was based on these attitudes as such, it was clearly a moral reaction. But also past attitudes of others are morally relevant. If I learn today that I offended yesterday someone needlessly and I feel therefore regret today, then this reaction shows that I have a morality. My regret is caused by my belief of my yesterday’s action, but only because this past offence (an evaluated state) is relevant to me.

(d) Some authors seem to think that only rational attitudes of others are morally relevant.10 This view may be justified for normative ethical problems, but it is incorrect regarding the question whether or not someone has a morality. If one of my friends wants to climb a dangerous mountain in winter and I am convinced that this plan is unreasonable (for example, because of the minimal chance of success), then I should nevertheless not be indifferent to his plan. This does not mean that I should advise him to take the risk or that I should try to persuade him not to risk it (both belong to normative ethics), but it means that I am not indifferent even towards irrational attitudes of others if I have a morality. Even impossible attitudes (meaning that they cannot be satisfied) are morally relevant. For example, if an old man wishes to be young again and a person therefore feels pity for him, this reaction shows that she has a morality.

(e) The belief that someone has an attitude can prove the morality of a being even if this attitude does not exist at all. A person, who refuses to kill an insect because she believes this would inflict pain on it, has a morality even if this insect is incapable of feeling pain. Even if a person were the only living being in the entire universe, she could nevertheless have a morality, if she believed that others exist and if their attitudes were relevant to her. Also in these cases are the attitudes of others relevant, but the persons are affected by their beliefs that they exist. Since these beliefs can be wrong, there can be moral reactions without attitudes of others. For this reason it is also a sign of a person’s morality if she is affected by the mere possibility of an attitude. Let us assume that A believes his friend, a certified bachelor, could enjoy a married life, if only he would decide to marry. If this belief motivates him to advise his friend to marry, then he shows herewith that he has a morality even if his friend never marries and therefore never enjoys a married life. It is sufficient that A believes his friend could enjoy it if he married.


According to the definition, relevant attitudes must be valuations of others. This point is important as it makes my concept of morality social. We have only a morality if we also take others into consideration. Morality is the negation of pure selfishness. The ‘others’ can be any beings about whom a person believes that they can have attitudes and who are not identical with her; for instance, God or gods, humans, animals, but also plants and even inanimate things. As said, these entities need neither exist nor need they be capable of attitudes if they exist. It is sufficient that they are believed to exist and being sentient. This means my concept of morality is not social in the sense that it requires the factual existence of others, but only insofar as it presupposes their putative existence.

Others have their attitudes towards different objects. It is not important for morality which things they value. These objects can be events, motives, feelings, they are often consequences of actions but they can also be any other things. What counts is that others have putative attitudes towards something, not what this something is.

Even if a person has only a morality if attitudes of others are relevant to her, this does not mean that she must care about the valuations of each other being. It would, for example, be arbitrary to deny persons a morality if only members of their own tribe count for them as such (for their own sake). Morality need not be universalistic; it can be particularistic.11 For example, the morality of the Israelites (at least in their earlier periods to which the older parts of the Old Testament refer) was particularistic. Seen from a normative ethical standpoint, their morality may have been defective, but it is clearly unjustified to claim that they did not have a morality at all (what, however, would follow, if one proposed a universalistic conception of morality). Also a particularistic morality is a morality. Similarly, it is not necessary that each attitude of others is relevant. It is sufficient that there are attitudes which are relevant.12 If someone is indifferent to a person’s desire for wealth but not to her wish not to starve, then he has a morality. It is also not a necessary condition for morality that the attitudes of others are equally relevant. Impartiality is no requirement for having a morality. It is, of course, true that many authors defend the ‘requirement of impartiality’.13 To my mind, they can only do so because they have not sufficiently distinguished between the evaluative and the non-evaluative concept of morality. As has been said, we are here only concerned with the latter notion. If members of our families are more important to us than strangers, this does not mean that we have no morality. Thus, my conception of morality is neither universalistic nor egalitarian. This point of the definition also does not say that only attitudes of others must be important for us, it claims rather that attitudes of others must also count for us as such. Our attitudes are always relevant to us and this fact does not speak against our morality. But if a person cares as such exclusively for her own attitudes, we must say that she has no morality, that is, she is an amoral being.


It is not sufficient that a person believes that others have attitudes. They must also be ‘relevant’ to her. I have already used this term several times, it is therefore time to explain it. Attitudes of others (in the sense explained) are for a person relevant, if she is capable of being affected by her belief that they have these attitudes. As has already been said, what really affects us is our belief that others have attitudes. (For brevity’s sake we can sometimes say that we are influenced by attitudes of others, but strictly speaking we should say by their putative attitudes.) Philosophers have regarded this capability of being affected by the putative attitudes of others as a mysterious phenomenon. Schopenhauer, for instance, calls it the mystery of ethics.14 Indeed, that valuations of others are to us relevant as such (not only as means for our purposes) is a very peculiar phenomenon, which is in need of an explanation. There are attempts to account for it on different levels. Psychologists have investigated ontogenetic factors, biologists have given genetic explanations and philosophers such as Schopenhauer have been convinced that there can only be a metaphysical explanation for morality.15 In this paper I am concerned with the conditions that must be satisfied for having a morality, but not with the problem of its explanation. Therefore, we can ignore these questions here.

The effect of our belief that another being has an attitude can be of various kinds. Other authors have usually concentrated only on one kind. According to Schopenhauer (1977), the distress of others should trigger our pity. The British empiricists Hutcheson, Hume and Smith refer to similar emotional reactions.16 Albert Schweitzer, on the other hand, has concentrated on the concept of reverence and has thus emphasised an attitude.17 In my view, these conceptions are too restricted. Our beliefs that others have attitudes can influence all spheres of our personality. If I see a blind beggar on the roadside and her putative misery causes me to feel pity, then this reaction shows my morality. But I need not react emotionally. If my experience causes an intention to give the beggar some money, then this attitude shows also my morality. But also considerations what I could do to help her would prove my morality, which shows that also cognitive reactions can indicate the morality of a being. This means that I defend a holistic concept of morality, meaning that any reaction can show the morality of a being, if it has the right genesis.

By ‘right genesis’ I mean that the belief that someone else has an attitude is as such (for its own sake) an explaining reason for a reaction (e.g. an emotion or motivation). What makes a reaction a moral reaction is its explaining reason, not its kind and not its content. It is therefore not sufficient if, for instance, my intention is in accordance with the will of someone else. If my neighbour wants me to help him and I have the intention to do so, then this intention proves that I have a morality only if I have it because of my neighbour’s want. Cooperation between people is not necessarily a sign of their morality because the conformity of their wills can be only accidental.


If a person has a morality, attitudes of others are to her relevant ‘as such’ (or for their own sake). This as-such clause is an essential element of morality. Everything I have said so far is compatible with egoism because the attitudes of others are also relevant to enlightened egoists, but they are not relevant as such.18 If a person has a morality, there must be attitudes (valuations) of others which are to her relevant as such, that is, relevant independently of all conditions of these attitudes and all effects which they (that is, their satisfaction or frustration) could have. It must be only the belief that another being has an attitude, which has an effect on her. Be it noted, however, this as-such condition does not demand that valuations of others must be considered only as such. Therefore, it is sometimes better to explain this point negatively by saying that an attitude is to S relevant as such if it is not only externally (because of its causes or effects) relevant. If a mountaineer asks me for the best route to the peak (which I know) and I give him the right answer because he wants this information, then this behaviour proves that I have a morality even if my desire to show my competence was a further motive.19

Lest this point be misinterpreted, I will now explain it further by distinguishing it from similar cases. (a) Suppose a person A is on a mountain hike with some others. At a difficult spot, she would like to go back but does not do so because she thinks her companions would despise her for it. A is here influenced by putative attitudes of others, but they do not count as such. What A really affects, is her own wish not to be despised. (b) Many people (e.g., parents, teachers, and politicians) claim that their strategies and policies are determined by the well-being of others. Often, however, they are determined by their own ideas of what these others should be like. In these cases their own ideals regarding others count as such but not the attitudes of the others. Examples are those ambitious parents who see their children as famous future artists, sportsmen or scientists and do everything to get their children to meet these ideals. (c) Let us assume that A is interested in the music of Franz Schubert and I like her for this reason. It is not a sign that I have a morality if I like A only because my interests correspond with her interests.20 I have not been affected by A’s interests as such, but by the congruity of my interests with hers. (d) Cases of attitude incongruity are analogous. If A calls her fellow worker B a bore because he is more interested in books than in sports, then A’s attitude towards B is influenced by one of B’s interests. It is, however, not this interest as such which causes A’s attitude towards B but rather the incongruity with her interests. (e) We are often influenced by the person who has an attitude and not by this attitude itself. If, for example, a movie star promotes AIDS research, many of his fans will also be in favour of it. Often, however, they are neither affected by the attitude of this actor nor by the plight of the AIDS victims but only by the fact that their hero supports this research. They are influenced by a person, not by her attitudes. (f) Sometimes it is the reason for an attitude and not the attitude itself that is relevant to a person. Suppose A learns that B appreciates the works of Anton Chekhov because he describes the life of the ordinary people in nineteenth century’s Russia and that A likes B for this reason. It is then not B’s appreciation of Chekhov which affects A, but the reason for it. If A has a positive attitude towards B only because she is also interested in Russian society of the nineteenth century, her attitude does not show that she has a morality. These remarks show that in spite of the simplicity and inconspicuousness of the as-such clause, some care is required to interpret it correctly.


In this Section, I make some additional remarks which will further clarify my conception of morality.

  1. In contemporary philosophy it is usual to distinguish between a broad and a narrow concept of morality.21 Morality in the broad sense comprises everything that is necessary for a good and a rational life. It includes also the domains of prudence and egoism. This concept of morality is nowadays often defended,22 especially by philosophers who have been influenced by ancient Greek ethics.23 The narrower concept of morality, on the other hand, includes necessarily other beings and their well-being.24 For this reason, some call this narrower concept ‘altruistic morality’,25 which is somewhat misleading because it suggests incorrectly that according to this concept of morality only the others count as such or others count more than the acting person. Of course, there are many variants of the broad and the narrow conception. Since my concept of morality belongs to the narrower group, I will now briefly explain why I think that a narrower conception of morality is more adequate.

Some think it is pointless to ask whether the broader or the narrower concept is more adequate because both are used in every day discourse and it is only important to distinguish between them to avoid one of the many sources of ethical confusion. Nevertheless, I hold that the narrower concept is more adequate and I will give here two arguments for my view. (a) If we take morality in its broad sense, the question ‘Why be moral? ‘ becomes meaningless because it asks (at least in its usual interpretation) whether or not a person should do what morality demands, if she considers all things. It is, however, exactly this question what we should do if we consider all things, which is the basic moral question according to the broad concept of morality. For this reason, ‘Why be moral? ‘ turns out to ask the meaningless question why we should do, all things considered, what we should do, if we consider all things. But it is widely accepted that the above question is not meaningless in this way. (b) Apart from that, some reflection about morality reveals that problems of our own well-being (which belong to the broad concept of morality) must be distinguished from moral problems. Having a good life is different from leading a good life. If someone is unable to promote his own welfare, we may question his prudence, but it is no reason to question his morality. Similarly, people who have been seen as moral ideals acquired this reputation because of their commitment to others, but never because of their skill to promote their own happiness. It is obvious that we are dealing here with different concepts and it seems to me that we should refer to the narrower concept when we use the word ‘morality’.

  1. For a further clarification of my concept of morality, it seems to be useful to determine its relation to egoism and altruism. Many philosophers have held the view that morality and egoism exclude each other.26 This has been claimed for the normative theory of ethical egoism, which is then not seen as a moral theory, but for the motivational theory of psychological egoism as well (and it is only this latter one that concerns us here).

Since many different conceptions of psychological egoism can be found in philosophical and psychological literature, I will explain here only the one form of egoism with which my concept of morality is indeed incompatible. If a person has a morality, it is impossible that no attitude of others is to her relevant as such (for its own sake). This, however, excludes only a very extreme form of egoism, which in my opinion can hardly ever be found. Because for such an ‘amoral egoist’ no being and no wishes, wills, or intentions would exist that are to him relevant as such to any degree. For this extreme egoist all attitudes of all other beings would be only instrumentally relevant. This means, however, that having a morality is compatible with most concepts of psychological egoism. For example, egoists, who always favour themselves, can nevertheless have a morality, and the same goes for those who take attitudes of others only sometimes into consideration or count only some as such. Thus, having a morality and being an egoist exclude each other only if one is an extreme egoist as characterized above.

Also altruism can be seen as a normative ethical theory and a psychological theory.27 We are here only interested in psychological altruism; but since this doctrine is defined in different ways, I will here only show that most concepts of altruism (if it is not only defined through the effects of actions) are narrower than my concept of morality. There are at least three reasons for this. (a) Concepts of altruism often imply a partiality for others,28 which is not excluded by my concept of morality but also not implied. (b) Altruism is usually defined motivationally.29 If such a motivation is intrinsic, a person has a morality, but morality does not necessarily require an altruistic motivation. As shown in Section IV, also other propositional attitudes can prove the morality of a person. (c) Altruism implies often a benevolent attitude towards others or affection for them. My concept of morality does not rule out affection, but it is not required by it. Being influenced by the attitudes of others is quite different from feeling affection for them. We can see this difference when we are affected by the misfortune of persons who we do not know or who we even dislike. In summary, taking into consideration the many concepts of altruism it can be said that an altruistic person has a morality but a person who has a morality need not be altruistic.

  1. That persons who have a morality need not to be altruistic follows also from my view that even evil persons can have a morality. Philosophers have been divided on the question whether morality must have a direction in the sense that it is necessarily something good (for an individual or a society). The question in here not whether, for instance, also the members of a street gang can have a morality (of course they can), but whether we can say that a person has a morality only if her moral reactions have a positive value; that is, if, for instance, the plight of others motivates her to help and not to add more suffering or evokes her pity and not gloating. Many philosophers hold that morality must have such a direction.30 I think, however, they a have been mislead by the evaluative concept of morality. When we speak about a moral person, a moral action, a moral motive, etc., we take ‘moral’ usually in its evaluative sense, meaning morally good. But when we ask whether or not someone has a morality, we are not concerned with this evaluative concept. Therefore, I side with those philosophers who deny that it is necessarily good to have a morality.31 To have a morality means to be capable of being affected by the putative attitudes (valuations) of others as such. However, this requirement is also satisfied by malicious beings. If we conceptualise the Devil as a being which desires the evil of humans as an end in itself, then also the Devil has a morality. The Devil is an immoral being but not an amoral one. The morally good and the morally bad are contraries and as such opposite poles of morality. Since I have defined this latter concept, also the morally bad is included in the non-evaluative concept of morality.32

The following argument can summarize my view on this point. (1) A being whose only other-regarding motives are to harm them, has morally bad motives and is an evil being. (2) If someone has morally bad motives, then he or she has moral motives (in the consequent of this conditional ‘moral’ is taken in the non-evaluative sense). (3) If someone has moral motives, he or she has a morality. It follows therefore (4) that also evil beings have a morality.

Morality is a complex trait that includes the good but the bad as well. That we have a morality is a cause of happiness, of mutual help and comfort, in fact it is indispensable for entire cultures and civilizations. But it is also the reason for suffering, misery and many forms of crime. We can call this the fundamental ambivalence of morality.

  1. Morality is a many-dimensional trait because the moralities of people differ in a variety of aspects. I can sketch here only some of them. (a) There is a difference between persons regarding whose attitudes are (primarily) relevant to them. Some care only for beings close to them or for those they feel emotional bonds; to others also strangers are relevant. For most people the beings that are especially important to them are humans, some prefer animals or even other entities. (b) There is also a difference between persons in how many are relevant to them. The morality of some is very particularistic, some are universalistic and the morality of most is somewhere in between. (c) Another question is how much others count. I think it is fair to say that for most people their own attitudes are more important than those of others. But there may be people who regard the attitudes of others as equally important or count them even more. In addition, we tend to give more weight to the attitudes of some at the cost of others. Also on this partiality dimension we will hardly find two persons whose morality is the same. (d) There is also a difference in the kind of attitudes we are receptive to. I think most people are more affected by the immediate and observable misery of others than by the more abstract plight of people far away and even if I cannot agree with Schopenhauer that we are only affected by the suffering of others,33 it seems to me true that we are more affected by it than by their happiness. (e) There is also a difference in the morality of people regarding the circumstances under which they are receptive to attitudes of others. Some are almost always prepared to listen to the problems of their fellows, others harden their heart against the needs of the disadvantaged if they are in trouble themselves. Psychological investigations show how much our openness to others is modified by personal and situational factors. (f) There is also a difference in which way we primarily react. Many react primarily emotionally (for example, by feeling pity), others more conatively (e.g. by being prompted to take measures against the plight of others) and some react primarily cognitively. It is true that almost all react holistically but the priorities are interpersonally different.

The intention of this incomplete outline of some aspects of our moralities has been to strengthen the plausibility of my view that our moralities are different in many respects and that the morality of a person is probably as unique to her as her fingerprints.


According to my definition, a being has a morality if and only if attitudes of others are relevant to her as such. This is logically equivalent to the claim that this relevance is necessary and sufficient for morality. Reasons for this latter claim are thus also an argument for the adequacy of my definition. In what follows, I will thus concisely argue that the proposed definiens is indeed necessary and sufficient for morality.

(a) I think nobody can seriously deny that a person has a morality if attitudes of others are to her relevant in the explained way. This has even been conceded by proponents of the broad concept of morality. They would say that this is a very narrow morality, but also a narrow morality is a morality. There may be reservations if we consider borderline cases (for instance, if only one other being counts for a person as such) but in such cases a precising definition need not be in accordance with our conceptual intuitions because they are not reliable for extraordinary cases. The view that the definiens is sufficient for morality seems therefore unproblematic.

(b) But is it also necessary for morality that attitudes of others are relevant as such? Imagine a person who lacks this capability, for example, an extreme egoist who is quite indifferent to all other beings or has only instrumental relationships to them. It seems to me quite wrong to say that this person has a morality, which suggests that the definiens is indeed necessary. Morality has not only to do with relationships to others but with relations that are not purely instrumental. This view is held by many philosophers even if their terminology may be different.34 If we accept that the explained relevance of attitudes is sufficient and necessary, we must also accept the proposed definition of morality because it follows logically from it.

In conclusion, some of the characteristic features of my concept of morality may be summarized. According to my definition, morality is (a) materially defined (by attitudes of others) but is (b) nevertheless normatively neutral (since it does not imply any normative ethical theory) and has (c) no direction (it is not necessarily good). It is (d) a social phenomenon (because related to attitudes of others), but (e) neither necessarily impartial (it allows a preferential treatment of some others) nor (f) necessarily universalistic (because it can include only a fairly limited number). Morality is (g) a personality trait (not e.g. a system of rules), which is (h) not ‘pathological’ in Kant’s sense (because it is not based on accidental affections). It is (i) something holistic (i.e. including cognitive, affective and conative aspects) and is therefore (j) an action-guide (because it is also motivational). In short, morality is a unique quality of humans (and possibly also of other beings).


  • Baier, K. (1958). The moral point of view: A rational basis of ethics. New York: Cornell
  • Baier, K. (1981). “Defining morality without prejudice”. Monist, 64(3), 325-341.
  • Baier, K. (1995). The rational and the moral order: The social roots of reason and morality. Chicago and La Salle/Ill.: Open Court.
  • Batson, C. D. (1998). “Altruism and prosocial behavior”. In D. T. Gilbert, S. T. Fiske & G. Lindzey (Eds.), The handbook of social psychology (4th ed., Vol. 2, pp. 282-316). Boston: McGraw-Hill.
  • Becker, L. C. (1973). “The finality of moral judgments: A reply to Ms Foot”. Philosophical Review, 82, 364-370.
  • Bennet, J. (1995). The Act Itself. Oxford: Clarendon Press.
  • Brandt, R. B. (1979). A theory of the good and the right. Oxford: Clarendon Press.
  • Broad, C. D. (1971). “Emotion and sentiment”. In D. R. Cheney (Ed.), Broad’s critical essays in moral philosophy (pp. 283-301). London: Allen & Unwin.
  • Chang, K. C. (1996). Moral agoraphobia: The challenge of egoism. New York: Peter Lang.
  • Clarke, D. S. (1985). Practical Inferences. London: Routledge & Kegan Paul.
  • Cooper, N. (1981). The diversity of moral thinking. Oxford: Clarendon.
  • Donagan, A. (1977). The theory of morality. Chicago: University of Chicago Press.
  • Ewing, A. C. (1953). Ethics. New York: The Free Press.
  • Falk, W. D. (1960). “Morality and convention”. The Journal of Philosophy, 57, 675-685.
  • Flanagan, O. J. (1991). Varieties of moral personality: Ethics and psychological realism. Cambridge/Mass.: Harvard University Press.
  • Frankena, W. K. (1980). Thinking about morality. Ann Arbor: University of Michigan
  • Frankena, W. K. (1983). “Moral-point-of-view-theories”. In N. Bowie (Ed.), Ethical theory in the last quarter of the twentieth century (pp. 39-78). Indianapolis: Hackett Publishing Company.
  • Friedman, R. Z. (1988). “Morality and the morally informed life”. Midwest Studies in Philosophy, 13, 149-160.
  • Gauthier, D. (1963). Practical reasoning: The structure and foundation of prudential and moral arguments and their exemplification in discourse. Oxford: Clarendon Press.
  • Gewirth, A. (1978). Reason and morality. Chicago: University of Chicago Press.
  • Griffin, J. (1986). Well-being: Its meaning, measurement, and moral importance. Oxford: Clarendon.
  • Hare, R. M. (1963). Freedom and reason. Oxford: Oxford University Press.
  • Hare, R. M. (1981). Moral thinking: Its levels, method and point. Oxford: Oxford University Press.
  • Harris, G. H. (1981). “Frankena and the unity of practical reason”. Monist, 64, 406-417.
  • Harsanyi, J. C. (1982). “Morality and the theory of rational behavior”. In A. Sen & B. Williams (Eds.), Utilitarianism and beyond. Cambridge.
  • Hoffman, M. L. (1981). “Is altruism part of human nature?” Journal of Personality and Social Psychology, 40(1), 121-137.
  • Hoffman, M. L. (2000). Empathy and moral development: Implications for caring and justice. Cambridge: Cambridge University Press.
  • Hume, D. (1964). A treatise of human nature (2. vols.). London: Everyman’s Library.
  • Hume, D. (1984). Eine Untersuchung ueber die Prinzipien der Moral. Stuttgart: Reclam. (Originally published 1751: An enquiry concerning the principles of morals)
  • Hutcheson, F. (1986). Ueber den Ursprung unserer Ideen von Schoenheit und Tugend. Hamburg: Felix Meiner. (Originally published 17262: An inquiry into the origins of our ideas of beauty and virtue, in two treatises)
  • Kant, I. (1964). Groundwork of the metaphysics of morals. (trans. H. J. Paton). New York: Harper & Row. (Originally published 1785: Grundlegung zur Metaphysik der Sitten)
  • Louden, R. B. (1992). Morality and moral theory: A reappraisal and reaffirmation. New York-Oxford: Oxford University Press.
  • MacIntyre, A. (1987). Der Verlust der Tugend: Zur moralischen Krise der Gegenwart. Frankfurt-New York: Campus. (Originally published 1981: After virtue: A study in moral theory)
  • Milo, R. D. (1984). Immorality. Princeton: Princeton University Press.
  • Oakley, J. (1992). Morality and the emotions. London-New York: Routledge.
  • Perry, R. B. (1954). Realms of value. A critique of human civilization. Cambridge/Mass.: Harvard University Press.
  • Rachels, J. (1986). The elements of moral philosophy. New York: Random House.
  • Schopenhauer, A. (1977). Die beiden Grundprobleme der Ethik. (Zuericher Ausgabe. Werke in zehn Baenden, Bd. 6) Diogenes: Zuerich. (Originally published 1839/40)
  • Schweitzer, A. (1982). Die Ehrfurcht vor dem Leben: Grundtexte aus fuenf Jahrzehnten (3. durchges. Aufl.). Muenchen: Beck.
  • Schweitzer, A. (1990). Kultur und Ethik. Muenchen: Beck. (Originally published 1923)
  • Smith, A. (1994). Theorie der ethischen Gefuehle. Hamburg: Meiner. (Originally published 17906: The theory of moral sentiments)
  • Spielthenner, G. (2003). Grundlagen der Ethik. Frankfurt/Main: Peter Lang.
  • Sprigge, T. L. S. (1970). “Definition of a moral judgment”. In G. Wallace & A. D. M. Walker (Eds.), The definition of morality (pp. 119-145). London: Methuen. (Original erschienen 1964)
  • Staub, E. (1978). Positive social behavior and morality (Vol. 1): New York: Academic Press.
  • Staub, E. (1986). “A conception of the determinants and development of altruism and aggression: Motives, the self, and the environment”. In C. Zahn-Waxler, E. M. Cummings und R. Iannotti (Eds.), Altruism and aggression: Biological and sociological origins (pp. 135-164). Cambridge: Cambridge University Press.
  • Stevenson, C. L. (1944). Ethics and language. New Haven and London: Yale University Press.
  • Thomas, L. (1988). “Moral motivation. Kantians versus Humeans (and evolution)”. Midwest Studies in Philosophy, 13, 367-383.
  • van Ingen, J. (1994). Why be moral? The egoistic challenge. New York: Peter Lang.
  • Warnock, G. J. (1967). Contemporary moral philosophy. London: Macmillan.
  • Weizsaecker, C. F. v. (1980). Der Garten des Menschlichen: Beitraege zur geschichtlichen Anthropologie. Frankfurt am Main: Fischer.
  • Williams, B. (1978). Probleme des Selbst: Philosophische Aufsaetze 1956-1972. Stuttgart: Reclam. (Originally published 1973: Problems of the self: Philosophical papers 1956-1972)
  • Wilson, E. O. (1975). Sociobiology. The new synthesis. Cambridge/Mass.: Harvard University Press.
  • Wispé, L. (Ed.). (1978). “Altruism, sympathy, and helping: Psychological and sociological principles”. New York: Academic Press.

  1. See Perry, R. B. (1954). Realms of value. A critique of human civilization. Cambridge/Mass.: Harvard University Press. ↩︎

  2. Compare, for instance, Frankena (1980) who claims that the question about the nature of morality is often taken to be the main topic of moral philosophy … (p. 16). ↩︎

  3. As I see it, there are two major problems with these other accounts. (1) Some are too broad because they cannot demarcate morality from other realms that are obviously different from morality (for example, aesthetics). This can, for instance, easily be shown for the formal accounts of Brandt (1979) and Hare (1963; 1981). (2) So-called material definitions, on the other hand, tend to be too narrow, often because they include or imply normative ethical concepts. This can be seen in the account given by Rachel (1986). His definition requires impartiality between the interests of each individual who will be affected by one’s conduct: The definitions of other authors include the concept of universality (see, e.g. Harsanyi (1982), Schweitzer (1982; 1990). Impartiality and universality may be conditions for adequate normative ethical theories but it is not appropriate to include them in an account of the nature of morality. The morality of the Israelites (at least in their early period) lacked both qualities; it was particularistic and partial. It may well be that their morality was (normatively seen) faulty, but it would be quite arbitrary to deny them a morality at all. An adequate account of morality must thus avoid the mistake of defining morality only formally (which makes the account too broad), but it must also avoid normative ethical criteria (which make it too narrow and often ethnocentric). ↩︎

  4. It is therefore usual to distinguish between a ‘personal morality’ and a ‘social morality’, see e.g. Brandt (1979) or Frankena (1980). ↩︎

  5. I agree here with Frankena (1980) who says that a society has a morality ‘when some kind of majority of its individual members, including some but not necessarily all of its leaders have a morality…’ (p. 29, see also p. 46). ↩︎

  6. Since there are different types of definitions, I would like to stress here that the proposed definition is a precising definition. This means that it makes a vague and ambiguous term, which is taken from ordinary usage but also employed in science, more precise. Thus, the extension of my concept of morality need not fully agree with the vague ordinary concept (and as a more precise concept it cannot do so) because by making it more precise, not only the intension of this term changes but also its extension cannot be exactly the same. The given definition is therefore neither only a description of the meaning of our ordinary term ‘morality’ nor a pure stipulation of a new meaning for this word, it rather starts from the existing unclear meaning of this term and makes it more precise and herewith more useful. The form of the given definition is an equivalence. It is easy to state only necessary or sufficient conditions for morality. The proposed definition gives both, necessary and sufficient conditions for having a morality, which makes it more fruitful but also more vulnerable to criticism. ↩︎

  7. ‘Positive attitude’ and ‘negative attitude’ are very broad terms. There are, of course, many different kinds of such pro- or con-attitudes. Stevenson (1944) mentions e.g. ‘purposes’, ‘aspirations’, ‘wants’, ‘preferences’, ‘desires’, ‘interests’, ‘approving’, ‘favour’, ‘ideals’ and ‘aims’ — together with their contraries. Clarke (1985, 43) lists ‘desire’, ‘wish’, ‘hope’ and ‘be interested in’. One could also add e.g. ‘recommendation’, ‘attraction’, ‘admiration’, ‘esteem’ and ‘indignation’, together with their respective contrary, but every list is necessarily incomplete because in every ordinary language new terms emerge which can express attitudes. ↩︎

  8. See Spielthenner (2003, 9-19). ↩︎

  9. Note that wanting is one of the many different kinds of attitudes. ↩︎

  10. Compare e.g. Griffin (1986, 40-55) or Hare (1981). ↩︎

  11. Many authors defend a universalistic conception of morality; e.g. Harsanyi (1982), Albert Schweitzer (1982; 1990), Warnock (1967) or Weizsaecker (1980). But for a definition of the non-evaluative concept of morality there is no justification to do so. ↩︎

  12. The definiens is therefore not ‘… the attitudes of others…’ but ‘… attitudes of others…’ Psychological investigations show that many factors have an influence on our receptivity to the attitudes of others (see e.g. Staub, 1986; compare to this also Hutcheson, 1986, 216-18). ↩︎

  13. For example Rachels, who claims that the basic idea of morality is ‘that each individual’s interests are equally important: from within the moral point of view, there are no ‘privileged’ persons; everyone’s life has the same value (1986, p. 9). Many others agree with this view, e.g. Baier (1958; 1981), Bennet (1995), Clarke (1985), Gewirth (1978), Griffin (1986) or Harsanyi (1982), to mention only a few. ↩︎

  14. See Schopenhauer (1977, 248, 268-9), compare also Hutcheson (1986, 272). ↩︎

  15. Compare e.g. Hoffman (2000), Wilson (1975) and Schopenhauer (1977). ↩︎

  16. See e.g. Hutcheson (1986, 159), Hume (1964, 287) and Smith (1994, LXI). ↩︎

  17. See, for example, Schweitzer (1990, pp. 80-90). ↩︎

  18. Other phrases instead of ‘as such’ are that attitudes of others must count ‘in themselves’, must be ‘ultimate’ or ‘final ends’ or the Kantian phrase that others must be ‘ends in themselves’. ↩︎

  19. That morality requires us to consider others as such is a view which has also been taken by other philosophers. See, for instance, Schopenhauer (1977, 247) who claims that others must be the ultimate end of my will (see also p. 268) and Kant’s end-in-itself formula of the categorical imperative which says, ‘act in such a way that you always treat humanity, whether in your own person or in the person of any other, never simply as a means, but always at the same time as an end’ (cf. Kant, 1964). Similarly Hume (1984) when he discusses the question whether or not morality is finally based on egoism, which he denies (see e.g. p. 141) and where he agrees with Hutcheson (1986); and Adam Smith for whom it is essential for morality to sympathize with others (see 1994, 1). But also contemporary philosophers defend this view; e.g. Baier (1958), Gauthier (1963) or Frankena (1980), to mention only a few. ↩︎

  20. Many psychological studies show these effects of attitude congruity and attitude incongruity. ↩︎

  21. See e.g. Frankena (1980, 83), Milo (1984, 200-2) or MacIntyre (1987, 60). ↩︎

  22. For example by Flanagan (1991) who says that he advocates ‘[a] very broad conception of morality, ‘ which includes not only ‘social relations’ (p. 17). Similar conceptions are held by Chang (1996), Cooper (1981), Louden (1992) and Oakley (1992). ↩︎

  23. Ancient Greek philosophers held almost unanimously this broad concept. Therefore one has to agree with Frankena (1980) when he writes, ‘[T]he Greeks were seeking the rational way to live, without making special mention of the moral way to live; their solutions do not center on our relations to other persons’ (p. 11). ↩︎

  24. Compare e.g. Frankena (1980) and Friedman (1988). ↩︎

  25. See Cooper (1981) and Thomas (1988). ↩︎

  26. See, for example, Hutcheson (1986), van Ingen (1994), Schopenhauer (1977), Williams (1978) and Frankena (1980) who summarizes this view by stressing that ‘playing by the rules of egoism is playing a very different kind of a ballgame from morality as morality is usually understood’ (p. 67). ↩︎

  27. Broad (1971) defines ethical altruism as ‘the doctrine that each of us has a special obligation to benefit others, as such…’ (p. 265). ↩︎

  28. See e.g. Wispé (1978) or Hoffman (1981). ↩︎

  29. For example by Batson (1998, 282) who defines: ‘Altruism is motivation to increase another person’s welfare.’ ↩︎

  30. See e.g. Baier (1981) who claims that ‘it is an essential characteristic of morality that it is a good thing for societies and individuals to have it … (p. 335), Milo (1984, 198), Schopenhauer (1977) or Warnock (1967), according to whom morality must be seen as something that promotes (among other things) the well-being of people. ↩︎

  31. Compare, among others, Hare (1963) or Frankena (1983) who says that morality ‘might take the form of sheer hate or malevolence … (p. 71). ↩︎

  32. An analogy may make this clearer. Hot and cold are opposite poles of temperature and are therefore included in the concept of temperature. Similarly, also the antonyms ‘morally good’ and ‘morally bad’ are included in the concept of morality. ↩︎

  33. See Schopenhauer (1977, 249-51). ↩︎

  34. See, for example, Williams (1978) and Ewing (1953) who hold that morality implies altruism. A similar view is defended by Cooper (1981). ↩︎