Which ethical theory seeks the greatest good for the greatest number of people?

Pragmatic objections against the greatest happiness principle are many. The most basic objection is that happiness cannot be defined and that all talk about happiness is therefore mere rhetoric. The second objection is that happiness cannot be measured, so that we can never establish an absolute degree and number for happiness. A third objection holds that lasting happiness of a great number is not possible, that we can at best find some relief in fleeting moments of delusion. The last claim is that we cannot bring about happiness. These objections will strike many readers of this journal as outdated. Still it is useful to consider the evidence for and against for each of them.

Can Happiness be Defined?

The word happiness has different meanings and these meanings are often mixed up, which gives the concept a reputation for being elusive. Yet a ‘confusion of tongues’ about a word does not mean that no substantive meaning can be defined. Let us consider what meanings are involved and which of these is most appropriate as a final policy aim.

Four Qualities of Life

When used in a broad sense, the word happiness is synonymous with ‘quality of life’ or ‘well-being’. In this meaning it denotes that life is good, but does not specify what is good about life. The word is also used in more specific ways, and these can be clarified with the help of the classification of qualities of life presented in Table 1.

Table 1 Four qualities of life

Full size table

Vertically there is a difference between chances for a good life and actual outcomes of life. This distinction is quite common in the field of public-health research. Pre-conditions for good health, such as adequate nutrition and professional care are seldom confused up with health itself. Yet means and ends are less well distinguished in the discussion on happiness.

Horizontally there is a distinction between ‘external’ and ‘internal’ qualities. In the first case the quality is in the environment, in the latter it is in the individual. This distinction is also commonly made in public health. External pathogens are distinguished from inner afflictions. Yet again this basic insight is lacking in many discussions about happiness.

Together, these two dichotomies mark four qualities of life, all of which have been denoted by the word ‘happiness’.

Livability of the Environment

The left top quadrant denotes the meaning of good living conditions. Often the terms ‘quality-of-life’ and ‘wellbeing’ are used in this particular meaning, especially in the writings of ecologists and sociologists. Economists sometimes use the term ‘welfare’ for this meaning. ‘Livability’ is a better word, because it refers explicitly to a characteristic of the environment and does not carry the connotation of Paradise. Politicians and social reformers typically stress this quality of life.

Life-Ability of the Person

The right top quadrant denotes inner life-chances. That is: how well we are equipped to cope with the problems of life. This aspect of the good life is also known by different names. In biology the phenomenon is referred to as ‘adaptive potential’. On other occasions it is denoted by the medical term ‘health’, in the medium variant of the word.Footnote 2 Sen (1992) calls this quality of life variant ‘capability’. I prefer the simple term ‘life-ability’, which contrasts elegantly with ‘livability’. This quality of life is central in the thinking of therapists and educators.

Utility of Life

The left bottom quadrant represents the notion that a good life must be good for something more than itself. This presumes some higher value, such as ecological preservation or cultural development. In fact, there is a myriad of values on which the utility of life can be judged. There is no current generic for these external turnouts of life. Gerson (1976: 795) referred to these kinds as ‘transcendental’ conceptions of quality of life. Another appellation is ‘meaning of life’, which then denotes ‘true’ significance instead of mere subjective sense of meaning. I prefer the more simple ‘utility of life’, admitting that this label may also give rise to misunderstanding.Footnote 3 Moral advisors, such as your pastor, emphasize this quality of life.

Satisfaction with Life

Finally, the bottom right quadrant represents the inner outcomes of life. That is the quality in the eye of the beholder. As we deal with conscious humans this quality boils down to subjective appreciation of life. This is commonly referred to by terms such as ‘subjective wellbeing’, ‘life-satisfaction’ and ‘happiness’ in a limited sense of the word.Footnote 4 There is no professional interest group that stresses this meaning, and this seems to be one of the reasons for the reservations surrounding the greatest happiness principle.

Which of these four meanings of the word happiness is most appropriate as an end-goal for social policy? I think the last one. Commonly policy aims at improving life-chances by, for example, providing better housing or education, that is, in the upper half of Table 1. Yet more is not always better and some opportunities may be more critical than others. The very problem is that we need a criterion to assign priorities among the many life-chances policymakers want to improve. That criterion should be found in the outcomes of life, so in the lower half of Table 1. There, ‘utility’ provides no workable criterion, since external effects are many and can be valued differently. ‘Satisfaction with life’ is a better criterion, since it reflects the degree to which external living-conditions ‘fit’ with inner life-abilities. Satisfaction is also the subjective experience Jeremy Bentham had in mind.Footnote 5

Four Kinds of Satisfaction

This brings us to the question of what ‘satisfaction’ is precisely. This is also a word with multiple meanings and again we can elucidate these meaning using a simple scheme. Table 2 is based on two distinctions; vertically between satisfaction with ‘parts’ of life versus satisfaction with life ‘as-a-whole’, and horizontally between ‘passing’ satisfaction and ‘enduring’ satisfaction. These two bi-partitions yield again a four-fold taxonomy.

Table 2 Four kinds of satisfaction

Full size table

Pleasures

Passing satisfaction with a part of life is called ‘pleasure’. Pleasures can be sensoric, such as a glass of good wine, or mental, such as the reading of this text. The idea that we should maximize such satisfactions is called ‘hedonism’.

Part-Satisfactions

Enduring satisfaction with a part of life is referred to as ‘part-satisfaction’. Such satisfactions can concern a domain of life, such as working-life, and an aspect of life, such as its variety. Sometimes the word happiness is used for such part-satisfactions, in particular for satisfaction with one’s career.

Peak-Experience

Passing satisfaction can be about life-as-a-whole, in particular when the experience is intense and ‘oceanic’. This kind of satisfaction is usually referred to as ‘peak-experience’. When poets write about happiness they usually describe an experience of this kind. Likewise religious writings use the word happiness often in the sense of a mystical ecstasis. Another word for this type of satisfaction is ‘Enlightenment’.Footnote 6

Life-Satisfaction

Enduring satisfaction with one’s life-as-a-whole is called ‘life-satisfaction’ and also commonly referred to as ‘happiness’. Elsewhere I have delineated this concept in more detail and defined happiness as ‘the overall appreciation of one’s life-as-a-whole’ (Veenhoven 1984, 2000a).

In my view, life-satisfaction is most appropriate as a policy goal. Enduring satisfaction is clearly more valuable than passing satisfactions and satisfaction with life-as-a-whole is also of more worth than mere part-satisfaction. Moreover, life-satisfaction is probably of greater significance, since it signals the degree to which human needs are being met. I will come back to this point in Sect. 3.2.1.

Components of Happiness

There are different views on what ‘satisfaction’ with life is precisely. Some authors see that as an affective state and equate happiness with good feelings (e.g. Wessman and Ricks (1966: 240/1). Others see satisfaction rather as a cognitive judgment, e.g. McDowel and Newell (1987: 204) describe life-satisfaction as a “Personal assessment of one’s condition compared to an external reference standard or to one’s aspirations”. Several authors assume that both affect and cognition are involved. Diener defines ‘Subjective Well-Being’ (SWB) as being satisfied with life (cognition), while feeling good (affect) (Diener et al. 1997: 25). Likewise Sumner (1996: 145/6) describes ‘being happy’ as: “having a certain kind of positive attitude toward your life, which in the fullest form has both a cognitive and an affective component”.

I follow that latter view, assuming that humans are capable of evaluating their life in two ways. We have in common with all higher animals that we can appraise our situation affectively. We feel good or bad about particular things and our mood level signals overall adaptation. As in animals these affective appraisals are automatic, but unlike other animals, humans can reflect on this experience. We have an idea of how we have felt over the last year, while a cat does not. Humans can also judge life cognitively by comparing life-as-it-is with notions of how-it-should-be.

Most human evaluations are based on both sources of information, that is: intuitive affective appraisal and cognitively guided evaluation. The mix depends mainly on the object. Tangible things such as our income are typically evaluated by comparison; intangible matters such as sexual attractiveness are evaluated by how it feels. In the case of our life-as-a-whole, these appraisals do not necessarily coincide. We may feel fine generally, but nevertheless be aware that we failed to realize our aspirations. Or we may have surpassed our aspirations, but nevertheless feel miserable.

I refer to these components as hedonic level of affect and contentment and see them as sub-totals in the inclusive evaluation of life, which I call overall happiness. This distinction is discussed in more detail in Veenhoven (1984 ch2). It fits a common distinction in philosophy between three views on happiness: the ‘hedonistic view’, the ‘desire fulfillment’ view and the ‘hybrid’ view (Parfit 1984). This view is further developed in my paper ‘How do we assess how happy we are?’(Veenhoven 2009).

Hedonic Level of Affect

Hedonic level of affect is the degree to which various affects that someone experiences are pleasant in character and this reflects typically in ‘mood’. A person’s average hedonic level of affect can be assessed over different periods of time: a week, a year, as well as over a lifetime. The focus here is on ‘characteristic’ hedonic level. The concept does not presume subjective awareness of that average level.

Contentment

Contentment is the degree to which an individual perceives his/her aspirations are met. The concept presupposes that the individual has developed some conscious wants and has formed an idea about their realization. The factual correctness of this idea is not at stake. The concept concerns the individual’s subjective perception.

Preponderance of Affect

There is mounting evidence that affective experience dominates the overall evaluation of life. This fits the theory that affects are the basic orientation system in mammals and that cognition evolved later in evolution and functions as an addition rather than as a substitute. This view and the evidence for it is discussed in more detail in Veenhoven (2009). The related theory that hedonic level reflects gratification of universal human ‘needs’ is developed in Veenhoven (1991, 2005, 2009, 2010b).

In sum: Happiness can be defined as the ‘overall enjoyment of one’s life as-a-whole’. This encompassing judgment is based on both affective and cognitive appraisals of life.

Can Happiness be Measured?

A common objection against the ‘greatest happiness principle’ is that happiness cannot be measured. This objection applies to most of the above-discussed meanings of the word, but does it apply to happiness in the sense of life-satisfaction?

By definition, this kind of happiness is something we have on our mind and consequently we can measure it using questions. That is, simply asking people how much they enjoy their life-as-a-whole. Questions on happiness can be posed in various contexts; clinical interviews, life-review questionnaires and survey interviews. The questions can also be posed in different ways; directly or indirectly, and by means of single or multiple questions.

Self-Reports

There are many reservations about self-report measures of happiness: people might not be able to oversee their life, ego-defense might distort the judgment and social desirability could give rise to rosy answers. Because of this, early investigators experimented with indirect questioning. Happiness was measured by a clinical interview, by content analysis of diaries and using projective methods such as the Thematic Apperception Test. These methods are quite laborious and their validity is not beyond doubt. Hence, direct questions have also been used from the beginning. A careful comparison of these methods showed that direct questioning yields the same information at a lower cost (Wessman and Ricks 1966).

Common Survey Questions

Because happiness can be measured with single direct questions, it has become a common item in large-scale surveys among the general population in many countries. A common question reads:

Taking all together, how satisfied or dissatisfied are you currently with your life as a whole?

Many more questions and answer formats have been used. All acceptable questions are documented in full detail in the collection ‘Measures of Happiness’,Footnote 7 which is part of the ‘World Database of Happiness’ (Veenhoven 2010a).

Validity

Though these questions are fairly clear, responses can be flawed in several ways. Responses may reflect how happy people think they should be, rather than how happy they actually feel and it is also possible that people present themselves happier as than they actually are. These suspicions have given rise to numerous validation studies. Elsewhere I have reviewed this research and concluded that there is no evidence that responses to these questions measure something other than what they are meant to measure (Veenhoven 1984: chapter 3, Veenhoven 1998). Though this is no guarantee that research will never reveal a deficiency, we can trust these measures of happiness for the time being.

Reliability

Research has also shown that responses are affected by minor variations in wording and ordering of questions and by situational factors, such as the race of the interviewer or the weather. As a result the same person may score six in one investigation and seven in another. This lack of precision hampers analyses at the individual level. It is less of a problem when average happiness in groups is compared, since random fluctuations tend to balance. This is typically the case when happiness is used in policy evaluation.

Comparability

Still, the objection is made that responses on such questions are not comparable, because a score of six does not mean the same for everybody. A common philosophical argument for this position is that happiness depends on the realization of wants and that these wants differ across persons and cultures (Smart and Williams 1973). Yet it is not at all sure that happiness depends on the realization of idiosyncratic wants. The available data are more in line with the theory that it depends on the gratification of universal needs (Veenhoven 1991, 2009, 2010b). I will come back on this point in the later discussion on the ‘signal function’ of happiness in Sect 3.2.1.

A related qualm holds that happiness is a typical western concept that is not recognized in other cultures. Yet happiness is recognized in facial expression all over the world (Ekman and Friesen 1975) and words for it exist in all languages. Likewise, there is hardly any difference in ‘don’t know’ responses to survey questions about happiness, the percentage being below 2% in almost all countries of the world (Veenhoven 2010b).

Another objection is that happiness is a unique experience that cannot be communicated on an equivalent scale.Footnote 8 This qualm roots also in a constructivist view of man. Yet from an evolutionary point of view it is unlikely that we differ very much. As in the case of pain, there will be a common human spectrum of experience. In the related functional ‘signal’ view of affect it is not plausible either that happiness is something idiosyncratic.

The data also tell a different story. If happiness cannot be communicated on an equivalent scale, there will be little correlation between subjective happiness and objective living conditions. Yet research shows several sizable correlations, some of which are presented in the Tables 4, 5 and 6 of this article. Particularly revealing is the 75% explained variance at the bottom of Table 4, which denotes a close to perfect correlation between objective quality of society and average life satisfaction of its members.Footnote 9 Likewise, incomparability of self reported happiness should manifest in a low correlation with objective indicators of how well we thrive. In that context, the relation with longevity is illustrative: follow-up of individuals shows that happiness is a strong predictor of how long we live (Veenhoven 2008b) and comparison across nations shows a +.70 correlation between average happiness and statistical life expectancy.Footnote 10

Lastly there is methodological reservation about possible cultural bias in the measurement of happiness, due to problems with translation of keywords and cultural variation in response tendencies. Elsewhere I have looked for empirical evidence for these distortions, but did not find any (Veenhoven 1993: chapter 5). In this context the 75% explained variance in Table 4 is also telling. Measurement error can at worst be responsible for 25% of the observed difference in average happiness across nations The error margin is probably much lower, since we cannot quantify all societal qualities exhaustively and because there is also inevitable measurement error in the measurement of objective characteristics of nations. Probably the error in the measurement of happiness in nations is no more than 5–10%.

The comparability of happiness is discussed in more detail in: Diener and Oishi (2004) VanPraag and Ferrer-i-Carbonell (2004) and Veenhoven (2001, 2008a).

In sum: Happiness as life-satisfaction is measurable with direct questioning and is well comparable across persons and nations. Hence happiness of a great number can be assessed using surveys.

Is Happiness Possible?

Aiming at happiness for a great number has often been denounced as ‘illusionary’, because long-term term happiness is a fantasy and certainly not happiness for a great number.

This criticism has many fathers. In some religions the belief is that man has been expelled from Paradise: earthly existence is not to be enjoyed, we are here to chasten our souls. Classic psychologists have advanced more profane reasons. Freud (1929) saw happiness as a short-lived orgasmic experience that comes forth from the release of primitive urges. Hence he believed that happiness is not compatible with the demands of civilized society and that modern man is therefore doomed to chronic unhappiness. In the same vein, Adorno believed that happiness is a mere temporary mental escape from misery, mostly at the cost of reality control (Rath 2002).

The psychological literature on ‘adaptation’ is less pessimistic, but it too denies the possibility of enduring happiness for a great number. It assumes that aspirations follow achievements, and hence concludes that happiness does not last. It is also inferred that periods of happiness and unhappiness oscillate over a lifetime and that the average level is therefore typically neutral. Likewise, social comparison is also seen to result in a neutral average and enduring happiness is only possible for a ‘happy few’ (Brickman and Campbell 1971).

If all this is true, utilitarians can still go for the least possible unhappiness, but must miss out on the progress optimism from which the idea emerged.

Enduring Happiness

Figure 1 presents the distribution of responses to the 10-step question on life-satisfaction in the UK. The most frequent responses are 7, 8 and 9 and less than 10% scores below neutral. The average is 7,2. This result implies that most inhabitants of the UK feel happy most of the time. That view has been corroborated by yearly follow-up studies over many years (e.g. Ehrhardt et al. 2000) and by studies that use the technique of experience sampling (e.g. Schimmack and Diener 2003).

Fig. 1

Life-satisfaction in the UK 2006. Source: European Social Survey 2006

Full size image

Happiness for a Great Number

The high level of happiness is not unique to the UK. Table 3 shows similar averages in other western nations. In fact, average happiness tends to be above neutral in most countries of the world. So happiness for a great number is apparently possible. All this is in flat contradiction to Freudian theory, which predicts averages below four everywhere and also defies adaptation theory that predicts universal averages around five.

Table 3 Life-satisfaction in nations around 2006 Average scores on scale 0–10

Full size table

In sum: Enduring happiness for a great number of people is possible.

Can Happiness be Manufactured?

The observation that most people can be happy does not mean that they can be made happier by public policy. Like the wind, happiness could be a natural phenomenon beyond our control. Several arguments have been raised in support of this view.

A common reasoning holds that happiness is too complex a thing to be controlled. In this line it is argued that conditions for happiness differ across cultures and the dynamics of happiness are of a chaotic nature, and one that will probably never be sufficiently understood.

The claim that happiness is not makable is also argumented with a reversed reasoning. The reasoning is then that we understand happiness sufficiently well to realize that it cannot be raised. One argument is that happiness depends on comparison and that any improvement is therefore nullified by ‘reference drift’ (VanPraag 1993). Another claim in this context is that happiness is a trait-like matter and hence not sensitive to any improvement in living conditions at all.

All this boils down to the conclusion that planned promotion of happiness is an illusion.

Can We Know Conditions for Happiness?

As in the case of ‘health’, conditions for happiness can be charted inductively using epidemiological research. Many such studies have been performed over the last decade. The results are documented in the earlier mentioned World Database of Happiness (Veenhoven 2010a) and summarized in reviews by Argyle (2002), Diener (1999), Layard (2005) and Veenhoven (1984, 1997). What does this research teach us about conditions for happiness?

External Conditions (Top Left Quadrant in Table  1)

Happiness research has focused very much on social conditions for happiness. These conditions are studied at two levels, at the macro level there are studies about the kind of society where people have the happiest lives and at the micro level there is a lot of research about differences in happiness across social positions in society. As yet there is not much research at the meso level. Little is known about the relation between happiness and labor organization, for example.

Livability of Society In Table 3 we have seen that average happiness differs greatly across nations. Table 4 shows that there is system in these differences. People live happier in rich nations than in poor ones and happiness is also higher in nations characterized by rule of law, freedom, good citizenship, cultural pluriformity and modernity. These data suggest ways for advancing happiness that fit well with current policies.

Table 4 Happiness and society in 146 nations around 2006

Full size table

Figure 2 depicts the relationship between economic development and average happiness in more detail. The scattergram reveals not only a positive correlation, but also a pattern of diminishing utility. This means that investment in economic development is not the best way for promoting happiness in rich nations. Figure 3 depicts the relation with government effectiveness, which appears to be linear and this means that more happiness can be gained in this way.

Fig. 2

Average happiness and economic development in nations around 2006

Full size image

Fig. 3

Average happiness and government effectiveness in nations around 2006

Full size image

Not everything deemed desirable is related to happiness however. Income equality in nations appears to be unrelated to average happiness (Berg and Veenhoven 2010) and there is no relationship either between expenditures for social security and average happiness (Veenhoven 2000b). Likewise, democracy does not seem to add to happiness in all conditions, in particular not in conflict ridden developing nations.

There is much interrelation between the societal characteristics in Table 4; the most affluent nations are also the most free and modern ones. It is therefore difficult to estimate the effect of each of these variables separately. Still it is evident that these variables together explain almost all the differences in happiness across nations; R 2 is .75!

These findings fit the theory that happiness depends very much on the degree to which living conditions fit universal human needs (livability theoryFootnote 11). They do not fit the theory that happiness depends on culturally variable wants (comparison theory) or that happiness is geared by cultural specific ideas about life (folklore theory). I have discussed these theoretical implications in more detail elsewhere (Veenhoven and Ehrhardt 1995).

Position in Society Next to the above discussed studies on average happiness across nations, there is much research on differences in individual happiness within nations. A common theme in this kind of studies is the relationship between individual happiness and social position. The main results are summarized in Table 5. Happiness is moderately related to social rank in western nations, while in non-western nations the correlations tend to be stronger.Footnote 12 Happiness is also related to social participation and this relationship seems to be universal.Footnote 13 Being embedded in primary networks appears to be crucial to happiness, in particular being married. This relationship is also universal.Footnote 14 Surprisingly, the presence of offspring is unrelated to happiness, at least in present day western nations.Footnote 15

Table 5 Individual happiness and position in society

Full size table

These illustrative findings suggest that happiness can be improved by facilitating social participation and primary networks, in other words, by creating ‘social capital’.

Internal Conditions (Top Right Quadrant in Table  1)

Happiness depends not only on the livability of the environment, but also on the individual’s ability to deal with that environment (see Table 6). What abilities are most crucial?

Table 6 Individual happiness and personal characteristics

Full size table

Research findings show that good health is an important requirementFootnote 16 and that mental health is more critical to happiness than physical health.Footnote 17 This pattern of correlations is universal. Intelligence appears to be unrelated to happiness, at least ‘school-intelligence’ as measured by common IQ tests.Footnote 18

Happiness is strongly linked to psychological autonomy in Western nations. This appears in correlations with inner-control, independence and assertiveness.Footnote 19 As yet we lack data on this matter from Non-Western nations.

Happiness has also been found to be related to moral conviction. The happy are more acceptant of pleasure than the unhappy, and they are more likely to endorse social values such as solidarity, tolerance and love. Conversely, the happy tend to be less materialistic than the unhappy.Footnote 20 It is as yet unclear whether this pattern is universal.

In sum: Conditions for happiness can be charted empirically; the available data is already very informative.

Can Happiness be Raised?

These findings suggest that happiness can be advanced systematically. Public policy can create conditions that appear conductive to happiness, such as freedom, while therapy and education can foster personal characteristics such as independence. Yet these empirical data will not convince the critics who believe in a theory that holds happiness as immutable.

One such theory is that happiness depends on comparison and that standards of comparison adjust to success and failure, although with some delay. In this view, happiness can at best be raised temporarily. Though this theory applies for some kinds of satisfaction, it does not apply for satisfaction with life-as-a-whole. It appears that life-satisfaction is not ‘calculated’ cognitively, but rather ‘inferred’ from unreasoned affective experience, which in its turn is related to the gratification of basic needs. The theory that happiness is relative is simply wrong (Veenhoven 1991, 2009).

Another theory holds that we are born either happy or unhappy and that policy interventions can change little as far as this is concerned. A collective variant of this theory is that happiness is a national character trait, for instance that Russian are chronically unhappy because of a cultural tradition of melancholy. This theory is also wrong, follow-up of individuals show marked changes over the long-term and trend studies of nations show also profound changes, such as in Russia, where happiness dropped dramatically in the late 1990s after the Rubel crisis (Veenhoven 1994, 2001).

Figure 4 presents the trend of happiness over 35 years in Denmark and Italy. The gain is considerable, the Danes gained 0,5 point on the 0 to 10 scale and the Italians 0,7. The gain in Denmark demonstrates that greater happiness is even possible in the happiest countries of the world. These countries are no exceptions; during this era average happiness has also risen in many other countries (Veenhoven and Hagerty 2006).

Fig. 4

Trends in average happiness in Denmark and Italy 1973–2008. Source: World Database of Happiness (Veenhoven 2010a), collection of happiness in nations, trend report 2005-1

Full size image

From a functional view it is also unlikely that happiness is immutable. As we will see in Sect. 3.2.1, there is good evidence for the theory that happiness serves as a meta-signal in the adaptation process and unresponsive signal systems tend to disappear in the course of evolution.

In sum: Happiness of the great number can be raised, just like public health can be promoted. At best there is an upper limit to happiness, analogous to the ceiling of longevity.

Which theory of ethics seeks the greatest good for the greatest number of people?

What is the strongest ethical theory?

What is the principle of the greatest good for the greatest number?

What is ethics for the greater good?