[MeMe]

@ Potemkin, Truth To Power, Sceptic

it violates the sense we have of the incommensurability of individual human life.

Different people's utilities are not commensurable. ... It is impossible to sum quantities that are not commensurable. ... Utilitarianism is incoherent because utility can't be measured except for an individual.

Who the hell divides happiness into units and then makes out like it's a measurable phenomenon

These remarks have not been debated in the thread. Nevertheless, they merit some reflections. The economists Edgeworth and Jevons do believe that utility can simply be measured. However, it is clear that this is a gargantuan task, because utility (or happiness, or satisfaction, or well-being) is a subjective variable. Moreover, the neoclassical economists were desparately trying to develop an exact science, and therefore they wanted to keep out the psychology. Thus they managed to develop a formalism, which does not depend on utility. For convenience, many introductory textbooks state that utility can not be quantified (called cardinal utility, as opposed to ordinal utility, which is qualitative). But this statement is controversial, and imhb it is false.

In fact utility (or satisfaction) has always been used, even within the economic science. Immediately after WWII Von Neumann and Morgenstern developed a popular theory about expected utility, which supposes that utility can be quantified. Measurements of satisfaction have been done since the sixties of the last century. This domain of economy uses techniques from sociology, in particular opinion polls. The interrogated person values his satisfaction by means of a number. In this way the satisfactions of two individuals can be compared. Evidently, the answers are not necessarily reasonable or rational.

Moreover, it turns out that this approach yields meaningful results. For instance, it can be shown that on average the satisfaction depends on the personal income. And since that connection can be quantified, it is even possible to attach a money value to satisfaction. Of course the relations hold for the representative (average) person, and not for particular persons. But still this makes them a good measure for the development of policies. The representative person does not have peculiarities, and therefore it behaves in a rational manner, at least as far as the human nature and urges allow for rationality.

In principle the measurement of satisfaction can also be employed in the valuation of human life under certain circumstances. For instance, the happiness as a result of life-saving operations can be calculated. And this can be compared with happiness as a result of other investments. In case that someone objects to this approach: in reality politicians do it all the time.

Evidently it can be argued that morals yield happiness of a higher order, which can not be compared with the happiness due to money. Then one has a pluralistic utility. There is some merit in this standpoint, since it can be used to defend for instance the absolute validity of the natural human rights. However, morals are cumbersome in cases where a practical policy must be developed. For the material resources are scarce, and therefore costs do matter and policies must be efficient. I like to believe that efficiency herself is a virtue. So politicians must always weigh the various policy options, and in general all options are negotiable, even the moral choices, even efficiency. I think that is is more fruitful to suppose that the utility is monistic, and to value certain morals with an extremely high price.

Finally, it must be admitted that the measurement of satisfaction (or utility) is still in its infancy. A huge problem is the dynamics in the human preferences. Individual preferences depend on the personal situation, and may change when the situation alters. And it is a difficult task to identify all relevant variables, which are dominant in the influence on the satisfaction. If this is done in a flawed manner, then hidden variables may disturb the correlations of the included variables. Anecdote: the short-term level of satisfaction is influenced by the results of the national soccer team. But these are practical problems, which can in principle be overcome.

[MeMe]

@ emmit, Eran

Quantitative research methods are used to find out about people's subjective well-being. If you only used qualitative methods, you'd be doing interviews which would be analyzed afterwards. But that's not the way it's done (most of the time).

Not on a single scale that allows comparisons and trade-offs.

I do not understand these remarks. Often interviews are structured in such a way, that the analysis does yield quantitative results, for instance with regard to satisfaction. Perhaps emmit refers to research, that employs imaging of neural processes?

[Truth To Power]

In fact utility (or satisfaction) has always been used, even within the economic science. Immediately after WWII Von Neumann and Morgenstern developed a popular theory about expected utility, which supposes that utility can be quantified. Measurements of satisfaction have been done since the sixties of the last century. This domain of economy uses techniques from sociology, in particular opinion polls. The interrogated person values his satisfaction by means of a number. In this way the satisfactions of two individuals can be compared. Evidently, the answers are not necessarily reasonable or rational.

Problem is, people's expressed opinions about satisfaction of their desires do not match their actual experience of it. The classic example is lottery winners, who wanted the money enough to pay (often a great deal of money) for tickets, but when they won, found that their personal level of satisfaction and happiness declined.

Moreover, it turns out that this approach yields meaningful results. For instance, it can be shown that on average the satisfaction depends on the personal income. And since that connection can be quantified, it is even possible to attach a money value to satisfaction. Of course the relations hold for the representative (average) person, and not for particular persons. But still this makes them a good measure for the development of policies.

Refuted above.

The representative person does not have peculiarities, and therefore it behaves in a rational manner, at least as far as the human nature and urges allow for rationality.

The representative person is wholly fictional.

In principle the measurement of satisfaction can also be employed in the valuation of human life under certain circumstances. For instance, the happiness as a result of life-saving operations can be calculated. And this can be compared with happiness as a result of other investments. In case that someone objects to this approach: in reality politicians do it all the time.

The fact that politicians do something shows it is rational??!? Don't THINK so....

However, morals are cumbersome in cases where a practical policy must be developed. For the material resources are scarce, and therefore costs do matter and policies must be efficient. I like to believe that efficiency herself is a virtue.

Efficiency implies a measurable standard, and utility is almost useless (;^) for that purpose.

So politicians must always weigh the various policy options, and in general all options are negotiable, even the moral choices, even efficiency. I think that is is more fruitful to suppose that the utility is monistic, and to value certain morals with an extremely high price.

But now you are second-guessing actual utility.

Finally, it must be admitted that the measurement of satisfaction (or utility) is still in its infancy. A huge problem is the dynamics in the human preferences. Individual preferences depend on the personal situation, and may change when the situation alters. And it is a difficult task to identify all relevant variables, which are dominant in the influence on the satisfaction. If this is done in a flawed manner, then hidden variables may disturb the correlations of the included variables. Anecdote: the short-term level of satisfaction is influenced by the results of the national soccer team. But these are practical problems, which can in principle be overcome.

IMO there are intractable theoretical as well as practical problems with utility.

[Hong Wu]

The other day I gave a homeless Korean guy in Seoul some money. He was selling some shit or something and I went through his shit to see if I wanted to buy any of it (I didn't) before I gave him the money. He was more happy to be taken seriously and treated with respect than he was to receive the money. Utility is limited as a vector for agreement because people can't agree on what is utility or on what matters in the first place.

[MeMe]

@ Truth To Power

Problem is, people's expressed opinions about satisfaction of their desires do not match their actual experience of it. The classic example is lottery winners, who wanted the money enough to pay (often a great deal of money) for tickets, but when they won, found that their personal level of satisfaction and happiness declined.

It is possible to measure the adaptation of happiness for the representative (average) person. So the adaptation process can be taken into account for policy decisions.

The representative person is wholly fictional.

All theories and models are fictional. Nevertheless, it is noteworthy that the utilities of the representative person in various situations can be compared in an objective manner, by means of opinion polls and the subsequent statistical analysis of their results. This is useful for policy decisions. I agree with you that interpersonal comparisons of utility between two real individuals are purely subjective. They are just personal opinions.

The fact that politicians do something shows it is rational??!? Don't THINK so....

It requires talent and an enormous effort to become a successful politician. So this objection is not very convincing, and is better left out. My point here is that political decisions require the weighing of various moral policy options. Life-saving operations must be compared with fair trials; etcetera. In the weighing process the policians focus on the average voter.

IMO there are intractable theoretical as well as practical problems with utility.

Again we agree. The theoretical and practical problems are gargantuan. However, I still entertain great hopes of the monistic utility approach, notably since it is a fairly exact science. The results of utility measurements can be tested and verified. The approach has already yielded valuable new insights, such as the adaptation phenomenon. This does not require the total rejection of models with pluralistic utilities, which assume incommensurability, for instance between consumer gratification and morals. You may call this rule utilitarism. Ethical debates about the death penalty, abortion, euthanasia etcetera remain necessary. But I do think that models of pluralistic utilities (such as the I & We model of Etzioni) are not very fruitful for decisions in daily life. They are too complicated.

[MeMe]

@ Hong Wu

Utility is limited as a vector for agreement because people can't agree on what is utility or on what matters in the first place.

The interviews include the question: "All in all, how satisfied are you with your present life?" The respondent can choose an integer value between 0 and 10. It turns out that everybody, all over the world, is able to answer this question. Furthermore it is found that on average the satisfaction depends on (correlates with) specific variables, such as the personal or family income, health, age (yes, the midlife crisis is again confirmed!), etcetera. In general researchers do not distinguish between utility, satisfaction, happiness, or well-being. The results appear to be reproducible. Note though that some researchers try to stretch the method to its limits, for instance by studying the effect of the political system or of aircraft noise on satisfaction. So some suspicion with regard to some studies is justified.

[Saeko]

@ Hong Wu

The interviews include the question: "All in all, how satisfied are you with your present life?" The respondent can choose an integer value between 0 and 10. It turns out that everybody, all over the world, is able to answer this question. Furthermore it is found that on average the satisfaction depends on (correlates with) specific variables, such as the personal or family income, health, age (yes, the midlife crisis is again confirmed!), etcetera. In general researchers do not distinguish between utility, satisfaction, happiness, or well-being. The results appear to be reproducible. Note though that some researchers try to stretch the method to its limits, for instance by studying the effect of the political system or of aircraft noise on satisfaction. So some suspicion with regard to some studies is justified.

The interviews include the question: "All in all, how many unicorns have you seen during your lifetime?" The respondent can choose an integer value between 0 and 10. It turns out that everybody, all over the world, is able to answer this question. Furthermore it is found that on average the number of unicorns depends on (correlates with) specific variables, such as the personal or family income, health, age (yes, the midlife crisis is again confirmed!), etcetera. In general researchers do not distinguish between unicorns, UFOs, sasquatches, or loch-ness monsters. The results appear to be reproducible. Note though that some researchers try to stretch the method to its limits, for instance by studying the effect of the political system or of aircraft noise on unicorn sightings. So some suspicion with regard to some studies is justified.

[Stegerwald]

@ MeMe
For a practical union man, a personality of action, like I am, your arguments make a rather abstract impression. It is a mixture of sociology and psychology. Apparently in the nineteenth century the moral statistics has been introduced (others called it social physics). A scientist like Quetelet supposed that personal qualities (courage, etcetera) can simply be measured. In this way the character of the average man (homme moyen) can be determined. Unfortunately, the number of personal qualities is endless. I remember that the famous German social-democrat ideologist Kautsky made a distinction between instinctive (selfpreservation and propagation) and social urges. He even wrote a book about it, with the title Ethik und materialistische Geschichtsauffassung. This was an attempt to apply historic materialism to the moral statistics. Each society has her own morals. He also summed up the leading social urges, namely altruism, courage, loyalty, disciplin, authenticity, and ambition. But this is just a start, and it will be cumbersome to measure all of them. Moreover, other respected sociologists such as Comte and Durkheim were critical about the moral statistics. They had a preference for stories, that is to say, qualitative models about man and his society. Stories are convincing, and you can truly believe in them. What is your standpoint?

[MeMe]

@ Stegerwald

other famous sociologists such as Comte and Durkheim were critical about the moral statistics. They had a preference for stories, that is to say, qualitative models about man and his society. What is your standpoint?

I understand your doubts. Nowadays, the sociologist Etzioni thinks along the lines of Kautsky, even though he is neither an adherent of historical materialism nor a socialist. Imho selfpreservation stands out above the other urges, although there may be natural exceptions. Certainly in economics selfpreservation or egoism plays a vital role in market transactions. Therefore I defend the monistic utility in this thread, albeit within an institutional framework (such as laws). Now you may argue that utility or satisfaction is an abstraction, which has no counterpart in real life. Only real phenomena can be measured. However, then you may also question the existence of sound, light or heat. For satisfaction is a cognitive state, which can be felt and quantified in numbers by humans. Perhaps in the far future neuro science will also be able to perform satisfaction measurements, independent of opinion polls. Moreover, happiness research has already discovered various important human regularities, for instance the adaptation process. Thanks to this find people can be warned, that their expected satisfaction (or utility) is usually larger than their experienced satisfaction at a later time. This knowledge may induce them to make different decisions. Satisfaction has practical significance, also for you. So even if it is hard to define or interpret satisfaction as a cognitive and neural state, it is still a relevant concept.