Ethics may be defined broadly as the study of values as they relate to human conduct. The subject matter is extremely broad, and the field has historically been divided into through main categories:

• Normative ethics is the study of what should be done, the study of what it is that makes an action `right’ or `wrong’. This study could be described as the search for a moral framework within the context of which moral judgments can be made.
• Descriptive ethics is the study of the moral choices that people actually make. It is a `value-free’ approach in that it makes no attempt to evaluate the actual `morality’ of any given action, merely to describe how people make moral decisions. We may include under this heading the study of how people come to make moral choices in the first place, including fields such as evolutionary ethics which attempt to understand to what extent moral behaviours may have evolved as a result of natural selection.
• Meta-ethics is the study of what ethical terms actually mean, and attempts to address the question `what does it mean to describe an action as “good” or “bad”?’

The question of values as they relate to human conduct is fundamental and exerts a significant influence on the human experience. From purely individual concerns such as questions of `what should I do in this situation?’ through to social policy matters including enshrining an ethical code of conduct in a legal justice system, the effect of ethical judgments, whatever their nature, is felt everywhere. It is therefore a subject with which the student of the human condition needs to be familiar.

Because we are interested in the ideas underlying the concept of ethics, we will focus in this essay mainly on normative ethics and meta-ethics, although we will make some references to descriptive ethics as we proceed. To begin with, we will look at some normative ethical theories to provide a groundwork of understanding and example which we can then use to proceed to an examination of meta-ethics.

The search for a framework

As we mentioned, normative ethics can be seen as the search for a framework under which moral choices can be made, and a fundamental assumption is this search does indeed exist and can be found. The soundness of this assumption will be examined later, but for now we will let it pass unchallenged for the sake of illustration. The nature of that framework, or of this principle or set of principles, is what lies at the root of the fundamental disagreements between various schools normative ethicists. We can begin by examining a number of these schools to illustrate the differences.

The school of thought which will perhaps be most familiar to modern secular readers is that of consequentialism. This school holds that the morality of a given action — the `rightness’ or `wrongness’ of that action — is determined by the consequences that such an action has. Under this idea, actions that lead to what we may describe as `desirable’ consequences are considered to be `good’, and actions that lead to what we may describe as `undesirable’ consequences are considered to be `bad’. The question of exactly what we mean by `desirable’ and `undesirable’ consequences is also one that the consequentialist school seeks to address.

As an illustration, utilitarianism is probably the best known type of consequentialism. Utilitarianism holds that human beings fundamentally seek to maximise pleasure or happiness, and seek to minimise pain or sadness, a position known as hedonism. Desirable consequences therefore are those which increase pleasure and decrease pain, and undesirable consequences do the opposite. `Good’ actions are those that therefore result in increased pleasure and reduced pain, and `bad’ actions are those that do the opposite, and the greater the extent to which the amounts of pleasure and pain are changed, the `better’ or `worse’ an action is said to be.

This is an intuitively satisfying approach, particularly as regards `wrong’ actions. Actions such as assault, murder, theft, kidnapping, false imprisonment, cruelty to animals, drug peddling, and the like, may be viewed as `bad’ because they inflict pain, suffering, unhappiness and loss on their victims. While most people will often accept that there are circumstances in which inflicting pain can be justifiable — such as in certain types of `just war’, or when protecting against personal assault, or when punishing a criminal, or during an emergency medical procedure — those same people will also generally accept that causing `unnecessary’ harm — without, at this point, specifying what we mean by `unnecessary’ — is bad, or wrong, and is it wrong precisely because of the fact that harm is caused without good reason. Such actions are therefore considered wrong because of their consequences.

As with most ethical theories, however, there are problems with pure utilitarianism. For instance, what are often termed `victimless crimes’ would, if they are indeed victimless, have to be labelled as positively `good’ rather than `bad’. For instance, if a corporate employee could steal a small amount of money from a large corporation, so small that it would fail to be noticed by anyone, then we could argue that nobody is harmed by that theft. Yet, if such a theft was sufficiently significant to the individual that they could stave off financial ruin or the repossession of their home, and the benefit they get from it is potentially great. We would therefore have an action which causes harm to nobody, but which results in a benefit to at least one person, and therefore overall has positive consequences.

Note that the question of social policy is entirely separate to this conclusion, and it does not necessarily follow that victimless crimes should remain unpunished, even under strict utilitarianism. It may be argued, for instance, that if such victimless crimes were not pursued they would become sufficiently commonplace that, in the aggregate, sufficient harm would be caused as to render it worthwhile to society as a whole to prevent such acts. Yet, any individual act of this nature would still be considered `good’ under a utilitarian position. If the preceding argument was considered sound, the implication would be that even though individual acts of this nature are `good’, a policy of preventing them would be `better’, since the overall benefit of preventing them exceeds the sum of the personal benefits of the individual acts.

And, indeed, the question of the `overall’ good presents more problems for the utilitarian position. At first there is the practical problem of determining and then ranking the amount of `pleasure’ or `pain’ that a given act would cause to different individuals. But even with that problem solved, there can be conclusions that most would find unpalatable. Most notably, if it could be determined that publicly executing innocent people would lead to a significant decrease in violent crime, for instance, and the overall benefit from that reduction exceeds the pain caused to the executee, then strict utilitarianism would be forced to deem that action good. Similarly, if forcibly removing the assets of the wealthy and redistributing them to the poor could be shown to have an overall positive effect, then that too would be a `good’ act according to utilitarianism.

It seems, then, that strict utilitarianism, although intuitively pleasing in concept, is not sufficient to explain the moral choices that people actually make, and is unsuitable as the kind of comprehensive moral framework sought by the normative ethicist. This conclusion also highlights a linkage between normative ethics and descriptive ethics. We have stated that some of the conclusions of utilitarianism appear to contradict with what most people would consider to be moral. When developing a theory of normative ethics, then, it appears that we already have some idea of what constitutes `right’ and `wrong’, and that any satisfactory normative theory should produce conclusions which broadly align with that idea. The philosopher Peter Singer is famous for drawing some controversial conclusions from his theory of `preference utilitarianism’ — which holds that the morally right thing to do is that which satisfies the preferences of the greatest number of people — including the idea that if it is ethical to perform medical experiments on live animals it must be equally acceptable to perform medical experiments on young or brain-damaged human beings which equivalent mental capabilities, and the idea that `abortion’ may be permissible several weeks after birth in the case of severely disabled children. Singer is unapologetic about these views as he demands consistency, and agrees with Socrates that `we must follow the argument where it leads.’ Yet most people would intuitively think that if a moral theory leads to conclusions which we morally disagree with, then there must be something wrong with the theory. In other words, it is all well and good constructing an elaborate theory based on the principle of, say, maximising the amount of overall pleasure in the world, but if that theory does not align well with the ethical judgments that people actually make then most would reject it. Thus, we could say that while descriptive ethics attempts to describe the moral choices that people make, normative ethics should seek to explain them, to abstract those choices into a set of principles which could be said to adequately describe the `moral sense’ that people actually do have. Under this idea, although it may seem intuitively pleasing conceptually to suggest that `good’ can be equating with `increasing pleasure’, if the theory arising from such a concept does not adequately explain the moral choices that people actually make then we would be forced to reject — or at least to modify — that concept.

An alternative framework to consequentialism in general is deontologism. This school argues that the morality of an action is wholly independent of its consequences, and that morality really amounts to `doing our duty’. Thus, if part of our duty was `not punishing the innocent’ then executing innocent people would never be a `good’ or `moral’ action, regardless of how much overall benefit it might bring to society.

This general approach is also intuitively satisfying in many cases. We can imagine a typical `good Samaritan’ case, where a member of the public stops to help someone in need. Most people will agree that offering such aid cannot really be described as `moral’ is it is done in the expectation of some kind of reward, for instance. Others may go further and say that it also fails to qualify as `moral’ even if it is done out of a sense of compassion, since easing our own discomfort at the sight of the suffering of others really amounts to a form of self-interest. Only if the act is done out of a sense of duty, the deontologists argue, only if it is done from a feeling that people should help out their fellow men, does it qualify as `moral’. Thus, from a deontological perspective it is the motive for an act which determines its morality, and only duty-based motives count.

Fundamental to this idea of morality is the argument that we can only be held morally responsible for those things we can control. Consequentialism, it is argued, can never provide a sound basis for moral judgment, because consequences are not always within our control; if we try to save a dying person, for instance, but inadvertently accelerate that person’s death through our actions, the deontologist would argue that our action was still a moral one, even though it ultimately had negative consequences. Similarly, acting out of a sense of compassion cannot be considered moral because we do not always have control over our emotions; if we are emotionally `coerced’ into action because our emotions prompt us, then that action would not be a result of our free choice, and is therefore something which we should not take `moral credit’ for. If we act out of compassion then, the argument goes, if for some reason we did not feel that compassion we would not act, and in the absence of a duty-based moral framework we would be morally blameless for failing to act.

There certainly appear to be elements of this idea in the moral judgments that people actually make. In the case of the victimless crime, most people would be of the opinion that action is still morally bad, because theft and dishonesty are just fundamentally wrong, regardless of whether anybody is directly harmed by those acts. Similarly, many would hold that punishing the innocent is always morally wrong and is not up for debate, regardless of what practical benefits such a policy may bring. Thus there are many examples in everyday life where people appear to agree that one is duty-bound to act in a particular way, or to refrain from acting in a particular way, regardless of what the consequences might be.

Naturally, there are as many problems with the purely deontological perspective as there are with the purely consequentialist perspective. Most notably, the very fact that morality becomes divorced from consequences is often unappealing, and the idea that people should be held morally responsible for their actions — as opposed to their intentions — is usually deeply ingrained; a woman who inadvertently drives her car into a tree while fixing her make-up in the rear-view mirror may be viewed as foolish, and even a source of comedy, but if she drives through a group of infant schoolchildren on her way to that tree she will be viewed in a very different light. To many people, the consequences of actions do have in influence on their morality, regardless of any philosophical arguments about moral culpability. In addition, different duties may conflict, and the consequences may be the deciding factor between them; for instance, it may be considered to justified to lie to save someone’s life, or to steal to feed one’s family if no other option is available.

Further, the idea that compassion, guilt, remorse, sympathy and other emotions are completely divorced from the idea of morality makes little sense to some. Many people would indeed view compassion, for instance, as a very moral emotion, and would praise the morality of actions motivated by it.

It appears, then, as if both consequentialism and deontologism have their place in framing the moral judgments that people make, but that neither of them are able to tell the full story by themselves. It may be that by combining the two somehow — `rule utilitarianism’ is one such approach, which combines the general principle of maximising happiness, or minimising harm, with duty-based rules such as `never punish the innocent’ — could provide a satisfactory framework. But, we are also forced to consider that the very foundation upon which normative ethics is founded may be flawed, and that there may be no coherent and consistent framework upon which moral choices actually are made. It is possible — arguably likely — that actual moral judgments arise on a case-by-case basis as a haphazard combination of evolution and conditioning.

The source of the framework

Another way of thinking about the study of normative ethics is to consider the question of where this `framework’ would actually arise from. We have stated that people tend to have a `moral sense’ which they can use to assess the conclusions of any given normative ethical theory, but where does this `moral sense’ originate, and can we consider it to be reliable? This is an important question. Evolutionary ethics, for instance, would suggest that our `moral sense’ arises from natural selection in the same way that our other senses have, and embodies a set of principles which are most consistent with the survival and reproduction of our genes. Imagine two distinct groups of primitive humans. Suppose that through random genetic mutation one of those societies has a tendency to both trust his fellows and to refrain from cheating them, while the other group is purely selfish and will take advantage of his fellows whenever he can. (In reality, complex behaviour patterns such as this could not spontaneously arise from genetic mutation without the accompanying pressure of natural selection, and certainly could not propagate to an entire community without this pressure.) We may assume that the first group will be more successful in dealing with their environment, since cooperation makes possible a degree of success that pure individualism cannot. If the second group found that whenever they tried to cooperate they were taken advantage of, they would learn not to do it, and the community as a whole would be less successful as a result. In this way, we can see how natural selection could result in the propagation of tendencies towards cooperation which could be the basis of a moral sense. This is intuitively pleasing, because it explains how tendencies towards altruistic behaviour could arise even though such behaviour might be detrimental to any given individual. Since the propagation of genes depends not upon how happy or satisfied individuals are, but on how successful a community or gene-pool is able to survive and reproduce, the well-being of specific individuals is not a barrier to the survival and reproduction of genes.

However, this suggests that `morality’ is merely a result of evolution, of the behavioural urges which have arisen as a result of natural selection as those which are most conducive to the surival of genes. This description of morality is unsatisfactory to some. For instance, it may be that these behavioural urges result in a tendency to cooperate with one’s own family or tribal group, but to be positively aggressive or even murderous towards other competing groups. If this is all there were to morality, then genocide could be an arguably `moral’ action, since it is consistent with and arises out of those evolved behavioural urges. Yet this would contradict a moral sense that racism is wrong, and a sense that murder is always wrong even if it involves out-groups. As a theory of ethics, it may also have difficulty explaining where this moral sense that racism is wrong arises from, if there appears to be no evolutionary pressure to generate it. It may be the case that the original tendency to care for one’s close fellows may have somehow become extended to humanity as a whole, so the evolutionary theory may still have validity, but it is unlikely that an evolutionary theory of morality could in itself explain how or why this expansion occurred.

To others, the whole idea of morality being a result of biological urges is unsatisfactory. To them, morality is almost by definition something that must sometimes contradict biological urges, a set of principles to which one’s biological urges, self-interest, and interest of one’s own close group must sometimes be subordinated. Accordingly, our own `moral sense’ may therefore sometimes just be wrong, and it may be misguided to judge an ethical theory by how closely it aligns with our own moral opinions. It may be that morality is something `out there’ to be discovered, rather than a sense we already have which merely needs to be abstracted and explained. We would not, it can be argued, refuse to believe the scientific theory that the earth orbits the sun merely because we already have an inbuilt `sense’ which tells us the sun orbits the earth, so why, it may be asked, would we reject an ethical conclusion because it contradicts our inbuilt moral `sense’? This accords with Singer’s idea, that if we can accept a set of principles as a sound basis for a moral theory, then we should accept the conclusions of that theory, and if our inbuilt moral sense balks at the conclusions then it is our moral sense which is faulty, not the theory.

History can be used to support such an argument. We can look into the past and see times where slavery, genocide, and racism were common and accepted practices, yet they are now commonly rejected in the west. It can be argued that this represents some kind of `maturing’ of our moral sense, that in the past we `did not know that slavery was wrong,’ but that now we do. The idea that `people didn’t know any better in the old days’ does indeed suggest a set of objective moral principles which can be discovered, and does indeed suggest that our inbuilt moral sense can be incorrect in the moral judgments that it makes, that morality is something we have to learn and discover over time, rather than something that we already intuitively `know’ inside ourselves.

Of course, the variability of moral sense over time — and geographically — can suggest something quite different; it could suggest that morality is not something `out there’, and that our moral sense simply changes according to time and convention, and that there is nothing inherently `more moral’ about believing that slavery is unacceptable than believing it is, merely that more people happen to believe that it is unacceptable today than in the past, for whatever reason. This would both support the idea that there is no comprehensive and coherent moral framework which can explain human morality, and support the idea that even if evolution can explain the basis of our moral sense, it cannot account for all of even most of it.

The question of the source of such a framework is, therefore, of crucial importance to the study of normative ethics. If morality is something objective to discover, then assessing a normative theory based on how intuitive we find its conclusions is a fundamentally misguided approach. It would raise a further problem in that if our moral sense cannot be trusted to lead us to moral conclusions, then we would not even be able to assess the principles upon which a normative theory would be constructed. Thus, a philosopher like Singer might argue that we should follow the conclusions of `preference utilitarianism’ even if our moral sense balks, because that moral sense is not a good guide to evaluating those conclusions. Yet, if it is not a good guide to evaluating those conclusions, then why should it be any better a guide to evaluating the idea that `satisfying the preferences of the greatest number of people’ is an appropriate moral principle in the first place?

If the reliability of the moral sense is to be called into question, then the normative ethicist is forced to propose an alternative source of moral judgment. One obvious alternative source which has been commonly proposed is a `god’. Christian ethicists, for instance, argue that the Christian Bible is the source of morality, and that the morality of actions should be assessed on how well they conform to the ethical standards described in the Bible. There are serious problems with this approach beyond the obvious preponderance of evidence against the existence of the supernatural in general and the Christian God in particular. Euthyphro’s dilemma, translated into modern language, asks `does God command something because it is right, or is it right because God commands it?’ If the latter, then morality is completely arbitrary, and if God wished it he could command that the rape and torture of babies is `good’, and he would be correct in saying that. If the former, then morality does not come from God at all, but somehow independently exists outside of him. Neither of these alternatives is acceptable to the Christian ethicist, rendering the idea of morality arising from the Bible almost impossible to accept even if we do accept that the Christian god exists and that the Bible is an accurate account of his word. Of course, none of these objections matter to the believer since his acceptance is a matter of `faith’ which cannot be shaken by mere logical analysis or rational criticism, but this does mean that the idea is unable to take up a place in any sensible examination of morality.

A second alternative source is the reason itself. Immanuel Kant in particular argued that it was possible to develop a moral theory as a result of pure reason, without requiring recourse to a moral sense. His `categorical imperative’ argued that moral laws were maxims `whose universality as a law you can at the same time will,’ and which are the only maxim `under which a will can never come into conflict with itself.’ Kant thought, therefore, that one could deduce rationally what moral actions are by distinguishing those actions which are in accordance with rules that one could rationally desire to apply to everybody. In other words, Kant defending his conception of morality by saying that it simply does not make sense to describe an action as `moral’ if it is not one which you would wish everybody to do, and if it is one which you would wish everybody to do, then it’s a moral action. Thus, he believed he could develop a framework for moral action simply by thinking about what qualities actions must have to qualify as `moral’, without needing to appeal to a separate `moral sense’ to judge them. Kant expanded this formulation of his idea by essentially requiring that rational beings, within the framework of universalisability, treat both themselves and other rational beings as `ends’ in themselves rather than `means’, to pre-empt anybody who wanted to claim that they were perfectly happy for `kill anybody in the street you happen to come across’ to be a universalisable principle, but the most significant problem with his idea that he was never able to actually develop a sufficiently large set of principles that would actually be of any use in solving moral questions. In other words, he was able to develop a coherent framework, but was not able to fill it with actual maxims, such that it remains today a mere theoretical rather than practical scheme. It also has the potential to suffer from some serious conflicts; if `never tell lies’ and `never let people come to harm if you can prevent it’ were both universalisable principles, for instance, then the individual would be put into a bit of a pickle if he was ever in a situation where he could only save the life of another by telling a lie.

The actual sources of moral judgments that people make are likely to be varied and complex. It is very likely that there is an evolutionary tendency towards some types of moral action, but the greatly varying moral opinions that people hold over time and space demonstrates that evolutionary ethics cannot come even close to being a complete description of our moral sense. It is further likely that reason plays a significant role. Over the last few hundred years in the west, the idea of the rightness of individual liberty and equality — consistent, at least, with other varying principles such as causing no harm and `social justice’ — has been growing, and some moral development may have been the result of reasoning off the back of principles like this. Certainly we have seen an expansion, for instance in who is allowed to vote. We have further seen a reasoning that if everybody really is to be equal, then they must be given equal opportunity to education, government services, and protection from discrimination. A general principle may therefore give rise to new moral opinions as a result of people thinking about what actions must be performed if that principle is to be actualised. Many moral opinions may also result purely from convention, or influence; different people hold widely differing views on the morality of subjects such as capital punishment, abortion, homosexuality, and others. It seems unreasonable to suppose that such disparity in opinion could result either from an `innate’ sense of morality or from rational argument, but simply from æsthetic preference, personal values, or indoctrination from others.

This diversity of sources compels us to be even more pessimistic of the possibility of developing a simple and coherent framework of morality which would be sufficiently satisfactory as to obtain widespread approval from the population, since the varying moral opinions of actual people do not seem to derive from such a simple and coherent set of principles. This may be an unsettling thought for some; in the absence of such a framework, we would really have no justification for labelling any given action `right’ or `wrong’. Others may consider themselves to have all the justification they need, and will regard all this as meaningless philosophical wrangling; `I know perfectly well that murder is wrong,’ they may say, `and no about of philosophising is going to change that. I don’t need your “framework”; I know the difference between “right” and “wrong” already. If someone disagrees with me on topics such as abortion and capital punishment, then I’ll either just keep on trying to convince them and trying to change the law, or I’ll just graciously accept their right to differ. This philosophy of yours has no relevance to me whatsoever.’ There is some validity in this latter approach. However much philosophical difficulty the subject of ethics may present, we’ve all been living with an ethical system of some form or other for thousands of years, and there is widespread agreement on the morality of actions such as murder, rape, theft, and assault on the one hand, and charity, compassion and kindness on the other. It may be argued that `the system works,’ and that philosophising about it, while interesting, may ultimately be of little practical benefit. People seem to have strong moral opinions on many topics which are unlikely to be affected by philosophy; indeed, some moral philosophy may be regarded as little more than an attempt to `prove’ one’s existing moral opinions. It is even possible to argue that moral philosophy may be positively dangerous; it may be better, it can be argued, to trust our existing, albeit imprecise, moral judgment than to rationally deduce the most `moral’ course of action and risk committing some enormous atrocities as a result of making some small logical error or basing one’s analysis on inaccurate or incomplete information. Some may point to the prevelance of historical atrocities committed by religious or other idealistic groups — such as the Nazis, or the communists — as evidence of how a fixed idea of morality can result in some truly horrendous occurences. Others may point to what they regard as an inappropriate extension of `scientific’ principles to the realm of morality, such as popularity of `social Darwinism’ in the late 19th and early 20th centuries.

The rejoinder, of course, is that there is no reason to suspect `moral ignorance’ will lead to any more acceptable results, and it is equally arguable — somewhat paradoxically, admittedly — that we have a positive moral obligation to investigate what we actually do mean by `morality’. Many would argue that the best way to combat undesirable moral idealism is through rational moral criticism, rather than hiding one’s head under the carpet and hoping it never arises. Further, it is perfectly possible to investigate morality without necessarily having to implement any of the moral conclusions we might arrive at along the way. If it is to be argued that our currently system of morality has not itself arisen through a process of philosophical investigation, then it seems inconsistent to argue that further philosophical investigation is likely to overturn it. The strength of one’s moral convictions will not necessarily be lessened as a result of examining them. Moreover, for those interested in looking deeper as an exercise in itself, the ostrich approach is unlikely to prove satisfactory.

Thus, although we may concede that there does appear to be widespread agreement on many moral issues, we can also observe that there is widespread and often fundamental disagreement on many others. There is a very practical aspect to moral philosophy, therefore. Who, it may reasonably asked, is right? Is there an objective and absolute moral standard against which all actions may be judged? Does morality actually come down to just a question of what is sanctioned by a particular majority? The idea of teaching children in particular `the difference between right and wrong’ is common, but what, precisely, is that difference, beyond two simple lists of acts which are believed to be `right’ and acts which are believed to be `wrong’? We have already seen that attempts to derive both a framework which can explain our moral judgments, and attempts to determine precisely what the source of moral judgment is, have been largely unsuccessful. It may be valuable to delve a little deeper to try to understand these difficulties, and to examine on a more fundamental level exactly what we mean by `right’ and `wrong’ in the first place. To do this, we must turn to meta-ethics.