Hostility and the Minimization of Suffering
by Socrethics
|
Table of Contents
1. Introduction 2. Risk Ethics 4. Buddhism 7. Metaphor 8. Conclusion
|
|
Abstract
Type of problem It seems that expansion at the cost of the quality of life is a characteristic of the utility function of biology and that the moral values of Maximin and negative utilitarianism (which do not favor this expansion) conflict with biological forces. Does this conflict necessarily lead to a hostile theory?
Hostility 1) Negative utilitarianism and Maximin devaluate (volatile) chances in order to avoid suffering and are therefore risk-averse ethics. Every form of invariably risk-averse ethics has the following characteristics: a) It is hostile in a pessimistic scenario (i.e. when preference-frustration increases with evolution) b) It is inferior in the competition with risk-tolerant ethics. 2) Monastic Buddhism is a form of invariably risk-averse ethics. It doesn’t believe in an optimistic scenario and is therefore hostile. 3) Prioritarianism mediates in the conflict between the biological goal to expand and the cultural goal to avoid suffering. It doesn’t decide on a specific scenario.
Rationality 1) A specific kind of population ethics is considered to be rational by the majority, if it corresponds to the risk-profile of the majority. 2) The risk-profile of the majority is risk-tolerant because (in the course of evolution) risk-tolerance proved to have a higher survival value.
|
1 Introduction
1.1 Type of Problem
Classical utilitarianism has the deficiency, that it favors expansion at the cost of the quality of life:
Example: A population Z, consisting of 500 billion individuals, each with a life that is barely worth living, is better than a population A consisting of 1 billion individuals, each having lives that are of extremely high quality – as long as the sum of happiness (welfare) is greater in Z than in A. See Derek Parfit's Repugnant Conclusion.
It seems that expansion at the cost of the quality of life is a characteristic of the utility function of biology (the maximal proliferation of genes) and that the moral values of Maximin or negative utilitarianism (which do not favor this expansion) conflict with biological forces. Does this conflict necessarily lead to a hostile theory?
1.2 Terminology
Axiology
Axiology is roughly synonymous with value theory. Axiology can be thought of as primarily concerned with classifying what things are good, and how good they are (Value Theory, Stanford Encyclopedia of Philosophy)
In population ethics a distinction is made between neutral and person-affecting axiologies. This paper defines the terms as follows.
1) Neutral axiologies:
a) values are assigned by a neutral observer
b) view from outside of the populations
c) static comparison of populations
2) Person-affecting axiologies:
a) values are assigned by persons who take decisions and thereby affect other persons
b) view from inside of a population
c) dynamic view, valuation of changes, exposure to risk
The term person-affecting axiology as used in this paper differs from the term neutral axiology in the same way as the concept of marginal utility differs from a utility function, or Pareto optimality differs from the valuation of an economical state.
Hostility
An ethics is called hostile
1. if it asks for a reduction of the population-size in order to reduce suffering
2. if it allows painless killing without the consent of the concerned individuals
Rationality
The following definitions are taken from Rational, Wikipedia:
1. Rationalism is a movement which asserts that reason and evidence should be held superior to religious teachings.
2. A rational argument is logically valid, i.e. does not lead to contradictions
3. Individuals are rational if they act optimally in pursuit of their goals, whatever those goals may be
The definition used in this paper is analogous to the one of artificial intelligence and corresponds to the third of above definitions:
The behavior of an individual is rational, if it maximizes his/her utility under the given level of cognition.
Reason
The term reason, as used in this paper is an enhancement of the term rationality. The parameters in the definition of rationality can be examined and changed by reason:
1. The preferences which determine utility can be examined
2. The risk-profile which guides the preferences can be questioned
3. The level of cognition can be raised
2 Risk Ethics
2.1 Terminology
Risk
Risk is a state of uncertainty where some of the possibilities involve a loss, catastrophe, or other undesirable outcome (Risk, chapter “Risk vs. uncertainty”, Wikipedia)
Obviously, in order to define risk, one has to define situations like loss, catastrophe or undesirable outcome. Risk can be expressed in terms of suffering, preference-frustration, negative utility, negative welfare etc. This definition makes clear, that risk (in contrast to uncertainty) can only be valuated relative to a goal. Risk can take two directions:
1) The risks of active behavior, e.g. the risks in the fight for dominance and expansion
2) The risks of passive behavior, e.g. the risks of stagnation in a changing environment (Stillstand = Rückschritt)
Risk-aversion
1) Risk-aversion is the reluctance of a person to accept a bargain with an uncertain payoff rather than another bargain with more certain, but possibly lower, expected payoff. The opposite of risk-aversion is risk-tolerance.
2) A person behaves risk-neutral if he/she doesn’t demand a premium for risk-taking. The person tolerates risk but doesn’t seek it.
3) A person is risk-seeking if it is attracted to risk, i.e. he/she prefers an investment with a lower expected return but greater risk, to a no-risk investment with a higher expected return. Example: A bungee-jumper pays for risk.
In a rapidly changing environment stagnation becomes a major risk with regard to survival. In order to avoid this risk a dynamic, adaptive behavior is required; a behavior which is usually not associated with the term risk-aversion. If not otherwise mentioned in this paper the term risk-averse relates to
1. the avoidance of suffering (in hedonistic ethics)
2. the avoidance of preference-frustration (in preference-based ethics)
if necessary at the cost of Darwinian fitness.
The perception of risk
The perception of risk depends on the goal and may be distorted. There is no universal definition of a rational risk.
Examples:
1) The biological goal is to survive and procreate or, more precisely, to maximize the proliferation of genes (see Dawkins, God’s utility function). Nature designed the hedonistic mechanism in such a way that it serves survival. Consequently there are two kinds of risks:
a) Hedonistic risks involved in the struggle for survival
b) The risk to miss the biological goal
The second kind of risk doesn’t have to be conscious since the organism is guided by the hedonistic mechanism.
2) In many cultures the hedonistic mechanism serves survival as well as in biology. Asceticism originally was a strategy for survival (example: Spartans). But in contrast to animals, humans can reflect and deny goals. Asceticism developed a second meaning in the context of hedonism. In monastic Buddhism e.g. the goal to avoid suffering overrules the goal to survive. Intense (but volatile) happiness is devaluated in order to avoid suffering. This kind of ethics may be called hostile but not irrational. Cultural rationality is different from biological rationality.
3) In high-technology cultures we observe a similar phenomenon. Technology originally improved the Darwinian fitness, but then it developed an independent meaning in the context of hedonism. Examples: Transhumanism, Paradise-engineering, The Abolitionist Society. The goal to live eternally without pain overrules the goal to have a biological and human existence. This kind of ethics is neither hostile nor irrational, but certainly deviates from biological rationality.
Since the biological nature of humans is characterized by unrealistic optimism, the perception of technological as well as the natural risks might be distorted (see On the Perception of Risk and Benefit)
Risk-Profile
The risk-profile of a person is the entirety of the person’s attitudes towards risk (in different areas of decision-making).
2.2 Definition
Risk ethics
1) Risk ethics investigates the general question under which conditions a person is permitted to expose him/herself or others to a risk [Ericson].
2) Risk ethics is a form of consequentialism.
3) Risk ethics investigates the basis for rational decision-making in ethics.
4) The most important criteria to valuate risks/chances are Bayes and Maximin [Philosophisches Seminar, 18].
Risk-averse ethics
Risk-averse ethics devaluates (volatile)
1. happiness in order to avoid suffering
2. preference-satisfaction in order to avoid preference-frustration
Data base
1. Number-based risk ethics assumes that the risks are cardinally measurable and interpersonally comparable
2. Preference-based risk ethics doesn’t make this assumption.
2.3 Bayes
Definition
The Bayes criterion represents a utilitarian position. It maximizes the (subjective) expected value of a decision’s consequences. The expected value is calculated as follows:
1. in cases of chance: the product of probability times profit (utility)
2. in cases of risk: the product of probability times loss (negative utility)
Obviously the Bayes criterion corresponds to the definition of rationality used in this paper (chapter 1.2)
But what does it mean to maximize utility? In real life we are confronted with a mixture of risks and chances, i.e. we have to take chances and avoid risks at the same time.
For the following investigation we assume that
1. chances (utility) can be expressed in terms of preference-satisfaction
2. risks (negative utility) can be expressed in terms of preference-frustration
Under the assumptions given in Preference-frustration and the Hedonistic Scale these terms can be mapped to the hedonistic scale (i.e. to terms like happiness and suffering)
Individual level
What makes the difference between the maximization of preference-satisfaction and the minimization of preference-frustration?
1. If we consider a single preference, its satisfaction and the avoidance of its frustration is just a different view on the same phenomenon.
2. If we consider a hierarchy of preferences, then the frustration of a lower level preference may be desirable, as far as it serves the satisfaction of a higher order preference. But if the higher order preference is satisfied or if its frustration is avoided is the same thing.
The difference between the two strategies concerns the creation and elimination of preferences.
1) Expansive strategy
The abundant creation of new preferences is an adaptation to situations where preferences can easily be satisfied, i.e. where satisfaction is realizable with little energy. Each new preference increases the chances to increase satisfaction. The expansive strategy is limited by the environment. The lack of energy leads to the elimination of less important preferences.
2) Contractive strategy
The elimination of preferences is an adaptation to situations where preferences are difficult to satisfy. By this adaptation energy can be saved which in turn improves the chances to survive. Each eliminated preference diminishes the risk of frustration, i.e. there is a tendency for contraction. The contractive strategy is limited by the biological preference to survive and expand. Available energy leads to the reactivation of preferences.
The maximization of utility is a complex process consisting of
1) Maximizing satisfaction:
a) The direct satisfaction of preferences
b) The satisfaction produced by the elimination of preferences (liberation from responsibilities, painful attachments etc.)
c) The satisfaction produced by the creation of preferences (spontaneous live out of biological needs etc.)
2) Minimizing frustration
a) The direct frustration of preferences
b) The frustration produced by the elimination of preferences (suppression of biological needs etc.)
c) The frustration produced by the creation of preferences (responsibilities, painful attachments etc.)
Since the development of the environment is unknown, an individual is always exposed to the risk of preference-frustration. Some risks are perceived in the same way by a vast majority and are therefore called objective. But basically the perception of risk is an individual matter. The adaptation is rational if it serves the goal of the individual:
1) If the goal is to survive and procreate then a growth-oriented concept like Nicomachean ethics (with all its implied risks) is rational.
2) If the goal is to avoid frustrations then a retreat-oriented concept like the ethics of Epicurus (with all its missed chances) is rational.
Risk-averse ethics is as rational as risk-tolerant ethics as long as it serves the goal of the individual. For a specific individual perception of risk (i.e. individual definition of utility) risk-averse ethics maximizes utility as well as risk-tolerant ethics.
Society level
Depending on the definition of utility, the application of Bayes on the society-level results in the following theories:
1) Hedonistic ethics:
a) In classical utilitarianism utility is defined as happiness
b) In negative utilitarianism utility is defined as the avoidance of suffering
2) Preference-based ethics:
a) In preference utilitarianism utility is defined as preference-satisfaction
b) In negative preference utilitarianism utility is defined as the avoidance of preference-frustration
Utilitarianism considers society is considered as an entity (complex organism) which is exposed to risk. For such an entity the same reflections apply as for the individual above. Again we can say that the maximization of utility is a complex mixture of expansion and contraction. It is plausible to assume that this mixture is influenced by the biological nature of humans. Since the biological goal is characterized by a maximization function (see God’s utility function) it is also plausible to assume that it favors the expansive strategy. Classical utilitarianism has a tendency to support the biological mechanism; it favors the expansive strategy even in a scarce environment and assigns a moral value to the creation of lives barely worth living (see Parfit’s Repugnant Conclusion). By doing so the biological preference to expand life at any price (which has been questioned by cognition) is reinstalled at the cultural level.
1) On the biological level the goal to propagate genes is encoded in relatively rigid mechanisms which do not consider long-term risk or radical changes in the environment. This lack of flexibility can lead to the destruction of basic life resources and to the downfall of the species. But in slowly changing environments the risk-tolerance of biological mechanisms proved to be superior to risk-aversion.
2) From an evolutionary point of view technological progress represents a quickly changing environment. The biological nature of humans, which is adapted to slowly changing environments, has a tendency to underestimate technological risks (see on the Perception of Risk and Benefit). In a technological environment the risk-tolerant strategy doesn’t necessarily have the best survival value. Overpopulation and over-consumption may lead to the loss of sustainability and to the collapse of the environment which is necessary for survival. We call this a destructive (but not a hostile) potential.
2.4 Maximin
Definition
The Maximin criterion maximizes the minimal utility (instead of accumulating utilities) and is therefore non-utilitarian. On the other hand it can be considered as an extreme case of Bayes: The more weight is given to the worst case, the more the maximization of utility converges to Maximin (A Drink from the Group, Livia Levine).
Individual level
Harsanyi questions the Maximin criterion. He postulates that it is irrational to make your behavior dependent on some highly unlikely unfavorable contingencies, regardless of how little probability you are willing to assign them [Philosophisches Seminar, 20]
There are two objections to this claim:
1) Maximin may be plausible in cases where the probabilities are unknown. Harsanyi recommends to assume equiprobability in those cases [Philosophisches Seminar, 22]
2) Maximin may be plausible in cases where the loss is immense [Philosophisches Seminar, 21]
It appears plausible to argue that many rational people do not wish to gamble, especially if their lives are at stake. A small probability is not a sufficient criterion to take a deadly risk [Shrader-Frechette, 107].
This justification of Maximin however is misleading. We can only decide about rationality, if we know the goal of the concerned persons. The decision to take a deadly risk doesn’t identify the decision-maker as an irrational. Deadly risks are only irrational if combined with the goal to survive at any price.
Society level
1) Rawls promotes the Maximin criterion on the society level. He objects that we cannot carry the (utilitarian) Bayes criterion (on the individual level) to decisions on the society level e.g. to decisions about risky technologies.
2) According to Harsanyi the Maximin strategy is irrational because it is based on some highly unlikely unfavorable contingencies. He maintains that the distinction between small scale and large scale is irrelevant in this context.
3) Shrader-Frechette votes in favor of Rawls:
„Situations of individual risk are voluntarily chosen, whereas situations of societal risk typically are involuntarily imposed“. There is no analogy between the two situations [Shrader-Frechette, 105]:
a) Fairness: With the (utilitarian) Bayes criterion each person gets equal weight in decisions on the society level. If some persons profit from such a decision whereas others are harmed, then it may be unfair to equally weigh them.
b) Human rights: With the Bayes criterion even human rights could be violated in order to maximize utility on the society level.
[Philosophisches Seminar, 21].
Let’s consider a concrete example like a predicted asteroid impact with a small probability. Let’s further assume that the predicted impact destroys humanity if it occurs:
1) If the goal is survival at any price then, according to the (risk-averse) Maximin strategy, all resources must be concentrated on the deflection of the asteroid. The strategy is rational because it makes every effort to save humanity, but it is at the same time Bayes-irrational because it consumes all available resources for an unlikely event.
2) If the goal is hedonistic, then, according to the (risk-tolerant) Bayes strategy, we can take the small probability of the impact into account and allocate much fewer resources to the deflection of the asteroid. The strategy is rational because resources can be allocated to promising projects with a high probability, but it is at the same time Maximin-irrational because it takes the (small) risk to be annihilated.
If two nations are in competition and each of them is threatened with a different asteroid, then the one with the risk-tolerant strategy will succeed with a high probability. The risk-averse strategy of the other nation proves to be inferior in the context of competition. In most practical examples the Maximin strategy cannot be uncoupled from the context of competition.
Maximin cannot be the basis for a neutral axiology for the following reason:
Assume that a population A consists of a very large number of people with blissful lives and one person suffering terrible pain. In another population B, everybody suffers terrible pain but slightly less than the poor person in A. According to Maximin , B is better than A. One could say that Maximin imposes a dictatorship of the worst off [Arrhenius, 101].
Conclusion
In many practical examples, there is not just one worst case but several cases with serious loss, each one with a specific probability. These examples can be investigated with Bayes. Maximin is as an extreme case within Bayes, but not the basis for a general ethical theory.
2.5 Prioritarianism
Definition
1. Prioritarianism is a compromise between Bayes and Maximin.
2. Prioritarianism is related to NU insofar as the ethical priority increases with the degree of suffering.
Example: A prioritarian welfare function is a possible basis for progressive tax rates.
3. The convergence from Bayes to Maximin can be implemented by Prioritarian Welfare Functions.
As we have defined the Priority View, it implies the Repugnant Conclusion since it ranks populations according to the total sum of people’s transformed welfare. But the core idea of the Priority View – that gains in welfare matter more, the worse off people are, and losses in welfare matter less, the better off people are – can be combined with other aggregation methods. No specific method for aggregating the (transformed) welfare of different lives seems to follow from the core idea of the Priority View and, hence, it is hard to see how this idea could affect our evaluation of different-number cases. It seems that the Priority View is an idea mainly about how to distribute welfare among a fixed number of people. [Arrhenius, 110].
The term Prioritarianism, as used in this paper combines the core idea of the Priority View with the utilitarian aggregation method and therefore implies the Repugnant Conclusion. Prioritarianism is also a possible interpretation of the NU slogan “The least suffering for the highest number”. This slogan is the converse of Bentham’s “Greatest happiness for the greatest number”. To search for happiness doesn’t mean to forget about suffering and vice-versa. The original NU which doesn’t give any weight to happiness is not the only possible interpretation of the slogan.
Rationality
Prioritarianism preserves the efficiency of utilitarianism and a particular concern for those badly off. Prioritarianism, with all its mathematical subtleties is based on the intuition of sympathy [Lumer, 2]. But the degree of sympathy which is encoded in the formulas can be challenged by more or less deviating concepts. Since there is no universal consensus on the desirable degree of sympathy, the normative force of such an approach is limited. Prioritarianism is rational for those who agree with the encoded degree of sympathy.
Hostile potential
A society is threatened
1) by the limitation of natural resources (which makes the risk-tolerant utilitarianism destructive)
2) by the competition with other societies (which makes the risk-averse Maximin strategy disadvantageous)
Prioritarianism defines a strategy between the mentioned risks, i.e. the utilitiarian expansionism is partially devaluated in order to avoid the worst case.
Example: If a further expansion of the population-size increases the risk of global warming, then
1) the risk-tolerant utilitarianism continues the expansion until lives are barely worth living (see Repugnant Conclusion)
2) the adherents of the Maximin strategy reduce the number of their children (which puts them in disadvantage relative to classical utilitarians)
Prioritanians attempt to negotiate between the two groups.
Conclusion:
1. Prioritarianism is risk-averse, but less so than NU and Maximin.
2. Since Prioritarianism doesn’t ask for an unconditional reduction of the population it cannot be called hostile. It is, however, inferior to classical utilitarianism in a world of competition.
3 Negative Utilitarianism
3.1 Terminology
Suffering
1) Popper uses the terms pain and suffering as synonyms, as well as pleasure and happiness. For a definition of these terms see Preference-frustration and the Hedonistic Scale.
2) In order to aggregate suffering across a population (including animals), it has to be measured. The assessment of suffering is a complex and unsolved problem Nevertheless it should be possible to arrive at a consensus concerning the worst cases of suffering, so that the most important ethical decisions aren’t troubled by epistemic questions.
3) Suffering is called extreme in this paper
a) if the concerned individual wishes not having been born.
b) if it causes irreversible damage to the psyche and destroys the capability to compensate.
Preference
1) Preference (as used in this paper) is a synonym for desire, need, interest, attachment or a goal to be reached. Preferences are tied to emotions and control behavior. For a metaethical investigation of the term see Preferences by Christoph Fehige and Ulla Wessels.
2) In contrast to hedonistic theories preference-based ethics doesn’t imply that utilities are cardinally measurable and interpersonally comparable. But under certain assumptions the two kinds of ethics can be linked as defined in Preference-frustration and the Hedonistic Scale.
3) Preference-based ethics rejects the Experience Requirement.
4) The goal of preference utilitarianism is to maximize welfare. For a detailed definition see Consequentialism.
Compensation
Concerning the terms compensation in the wider and narrower sense; see Preference-frustration and the Hedonistic Scale.
Welfare and utility
The term quality of life, well-being or welfare is the net result of all preference-frustrations and -satisfactions. Satisfied preferences are counted positively and frustrated preferences negatively.
1) It corresponds to positive welfare, if the sum of all preference-satisfactions exceeds the one of all frustrations.
2) It corresponds to negative welfare, if the individual has lost his/her preference to exist. In this case we speak of a life not worth living.
3) In utilitarianism the term welfare corresponds to utility.
3.2 Background
History
The idea to formulate an ethical goal negatively is more than 2000 years old (it originates in Buddhism), but in the 19th century it is attributed to Karl Popper.
Popper's idea could have had the following sources:
1. the asymmetry of pain and pleasure (see below)
2. his epistemological work
3. the failure of happiness-promoting philosophies
4. his personal experiences: sixteen of his closest relatives became victims of Hitler, partially in Auschwitz, some committed suicide (from Die Erkenntnistheorie und das Problem des Friedens)
The negative goal in the words of Popper:
1. It adds to clarity in the fields of ethics, if we formulate our demands negatively, i.e. if we demand the elimination of suffering rather than the promotion of happiness. (Karl R.Popper, The Open Society and its Enemies, London 1945, I 9 n.2).
2. Similarly, it is helpful to formulate the task of scientific method as the elimination of false theories (from the various theories tentatively preferred) rather than the attainment of established truths, see The Political Thought of Karl Popper, London 1996
The asymmetry of pain and pleasure
The asymmetry between suffering and happiness seems to be an individual perception at first sight. It could however have its roots in physics. Life is subordinated to the law of thermodynamics and destined to decay.
1. Suffering is unavoidable because of accidents, defeats, illnesses and aging. Happiness is avoidable; it can be terminated at any point in time.
2. There is a kind of suffering (in this paper called extreme) which causes irreversible damage to the psyche and destroys the capability to compensate.
3. The goal to maximize happiness has no limits, in can never be satisfied. The goal to minimize suffering can be satisfied.
4. It is easy to make someone unhappy but much less easy to make that person happy again, see Bad Is Stronger Than Good. It is easier to produce suffering than to produce happiness.
5. Procreation, the biological climax of happiness, is a short event. Also amorousness is usually a short period within lifetime. The goal of nature is not to make people happy.
6. There are genetic defects which cause immense suffering. No corresponding phenomenon is known which causes immense happiness.
A different source of asymmetry is the following intuition:
1. There is no moral duty to be happy.
2. There is a moral duty to help a suffering individual, but there is no moral duty to make an individual happy.
3. We are responsible for the probable misery of future people, but we have no moral duty to procreate because unborn people do not suffer from missed chances.
This usage of the term asymmetry will be investigated in chapter 5.
Compensation
From the moral point of view, pain cannot be outweighed by pleasure and especially not one man’s pain by another man’s pleasure (Karl R. Popper, The Open Society and Its Enemies, London 1945, I 9 n.2).
1. NU primarily denies that the suffering of a person can be compensated by the happiness of another person. Similar thoughts can be found in Parfit’s claim about compensation [Parfit 1984, 337] and in Wolf’s misery principle [Wolf, 63].
2. The suffering of a person in different stages of his/her life is often perceived like the suffering of different persons. It is sometimes impossible to compensate actual suffering with the happiness of previous or later periods. This may have been Poppers’ intuition. Furthermore he used the terms pain and suffering, not the terms inconvenience and disturbance when he spoke about the moral dubiety of compensation. He never claimed that absolutely no kind of inconvenience can be compensated. The radical interpretation of his statement above doesn’t accord with his intentions. Unfortunately it became the official interpretation of NU.
3.3 Original Version (NU)
3.3.1 Definition
Individual level
1. The only thing that has moral value is the avoidance of suffering.
2. Happiness has no intrinsic value; the non-existence of happiness is morally neutral.
3. Happiness gets an indirect moral value as far as it avoids suffering
The term suffering applies to all sentient beings, not only humans
Society level
The best state of affairs is the one with the smallest sum of suffering.
The community is considered as an entity. The individuals with the highest degree of suffering have the highest marginal utility and get the most attention. The degree of suffering is defined by a certain combination of intensity and duration. See Negative Utilitarianism, Dan Geinster.
Principle of consequentialism
An agent should perform the act, which leads to the best consequences or state of affairs.
Since sentient animals are included in the best state of affairs, their interests have to be represented by human agents.
1. From the point of view of the suffering individual the result of an ethical decision is more important than the attitude, which is at the bottom of the decision. One can achieve a bad result with a good attitude if one isn’t aware of counterproductive mechanisms.
2. Normally the attitude is important because it aims at the desired result and tries to correct errors. Nevertheless the attitude is valuated according to the result and not vice-versa. Example: A passive strategy (like Theravada Buddhism) is correct according to NU, if and only if it can be shown, that all violent actions (including the monopoly of power of the state) are counterproductive.
3.3.2 Comparison with Classical Utilitarianism
Individual level
1. In contrast to classical utilitarianism happiness does not have an intrinsic moral value. Happiness gets an indirect moral value when it helps to avoid suffering. But the creation of new sources of happiness is morally neutral, if there is no suffering from the absence of these sources.
2. In classical utilitarianism, the happiness of every period in one’s life is equally weighed to calculate the value of a life. Negative periods are compensated by positive ones [Broome]. The NU denial of compensation is based on the idea that the same person in different stages of his/her life may (in many instances) have to be regarded like a different person. The happiness of adolescence often cannot relieve (compensate) the suffering of old age. A single event may be so horrible, that all previous joy is completely devaluated. The equal weighing of all periods of a life is a descriptive point of view. Popper said that from a moral point of view, this valuation is questionable. Previous happiness is often no help for the person in distress. Certain kinds of suffering are intolerable, independent of previous happiness. But the claim that absolutely no kind of suffering can be compensated is indefensible. For more information on compensation see Preference-frustration and the Hedonistic Scale.
Society level
Any amount of suffering can be outweighed by a sufficient amount of happiness. There may be no imaginable amount of happiness that can outweigh the horrors of Auschwitz. But there might be an unimaginable amount that can do it (see Negativity, utilitarian.org).
Compensation on the society level is the subject of a theory of justice and NU is first of all a concept of justice. It says that an increase in happiness including all modern comforts has no moral value, as long as certain kinds of suffering exist.
In classical utilitarianism it is theoretically possible that the suffering of the Auschwitz victims could be outweighed by the happiness of the post war generations. In NU such compensation is unacceptable. More on this issue in Negative Utilitarianism and Justice.
Functional equivalence
The man who would give up all his life's joy, rather than have it contain also the slightest degree of suffering, can do so from a classical utilitarian position. The negative utilitarian position is functionally equivalent to a certain type of classical utilitarian position - a position where pleasure holds very little value (Negativity, utilitarian.org).
There is no classical utilitarian position where pleasure holds very little value. In classical utilitarianism the weight of pleasure (and pain) is defined by the concept of marginal utility. The devaluation of pleasure is a characteristic of NU.
3.3.3 Rationality
NU is applied when priorities have to be calculated for several cases with serious loss. The chances are devaluated as well as in Maximin but probabilities are taken into account.
Example: Resources have to be allocated in order to avoid the following disasters:
1. Asteroid impact
2. Climatic catastrophe
3. Nuclear terror
Each of the three events has a devastating potential but the probabilities are different. There is no single clearly defined worst case.
The reflections concerning rationality are similar to the ones mentioned under the Maximin principle.
3.3.4 Hostile Potential
The preference for the empty world
The doctrine of negative utilitarianism is counter-intuitive, not least insofar as it entails that from a purely ethical perspective it wouldn't matter if nothing at all had existed, or if everything ceased to exist (David Pearce, Negative utilitarianism).
The doctrine is counter-intuitive for a happy person, but not for a suffering person, at least not if suffering exceeds a certain degree. In is also plausible to replace “person” by “population” in this sentence.
Painless killing
The most controversial implementation strategy of NU is a violent reduction of the population-size in order to improve the state of affairs. A radical pursuit of this thought leads to the extermination of humanity or life as a whole and therewith to a major disagreement between negative and classical utilitarianism (see Introduction to utilitarianism, utilitarian.org):
1) A negative utilitarian believes that, if it was possible to exterminate all life in the universe instantly and painlessly and permanently, it would be correct and ethically required that we do so in order to prevent any future cases of suffering
2) A classical utilitarian might decide either way, depending on his estimation of the relative amounts of future suffering and happiness.
There is an empirical argument against violence: Even contemplating and planning a project for the violent reduction of the population-size would provoke distress. In addition, it would classify the supporters of the project among the worst kind of terrorists. A violent extermination of mankind is not feasible in practice. The result would be an increase in suffering and therefore contradict the utility function.
The pinprick argument
1) The Pinprick Argument puts the idea of a painless extermination to an extreme and makes the empirical argument against violence even stronger. In order to defend the pinprick argument one has to assume, that the term painlessly not only refer to the victims of destruction but also to the agents - a scenario which is far from reality.
2) The pinprick argument is also applied to the creation of the world:
Would it really be better that life had never arisen if the only unpleasant experience that would otherwise occur would be a pinprick? (Pinprick Argument by David Pearce)
a) Here the argument is based on the false assumption, that a human being can evaluate the world without being part of the world. In reality, in an empty world there is no regret about the loss of happiness.
b) Under the assumption that there is a human observer, the world is not empty and the emotions of the observer have to be accounted for. Weighing the suffering produced by annihilation against the pain of a pinprick should result in a clear wish to survive. Only the assumption that the observer can painlessly annihilate himself allows defending the pinprick argument.
c) But what about major suffering and the tolerable limit? There is no general answer to this question, because it is an individual matter. Some individuals like to take risks, others not. It can only be said that a negative utilitarian (out of his commitment to rationality) would refuse to create a life of extreme suffering, a life where he wishes not having been born. In a negative utilitarian world suffering would be accordingly limited.
Childlessness
There is an empirical argument against childlessness: it produces suffering. If consequentialism is not restricted, then there are two scenarios to be valuated
1) if suffering cannot be besieged, then the accumulated suffering of all descendants will surpass the one of the childless would-be-parents
2) if suffering can be besieged one day (see transhumanism below) then the accumulated suffering up to this point may be smaller than the one of a world without this transformation.
The transhumanist belief in progress
Indeed, if the option were humanly available, the logic of negative utilitarianism morally obligates bringing the world to an end were this the only way to banish the suffering endemic to it. Happily, there is a much better way to rid the natural world of its endemic nastiness. This is to use biotechnology to eradicate aversive experience in all sentient life, David Pearce, Negative utilitarianism
1. If above vision becomes true, then suffering could be seen as a limited, intermediary state associated with the birth of an ecstatic or spiritual world. By this interpretation suffering receives a sense and a purpose. Evolution could be regarded as a project which reduces or eliminates extreme suffering, whereas otherwise it would persist or increase. The goal of NU is to minimize suffering across all generations, not only for the actual one. The actual generation is morally obliged to sacrifice itself, if it serves the long-term purpose to minimize suffering.
2. The transhumanist belief in progress stands in the tradition of enlightenment and shares its optimism. A corresponding NU constitution should prevent expansion at the cost of the quality of life (see introduction). It should avoid suffering rather than create new desires. In the words of Karl Popper: The role of the state is not to make people happy but to relieve avoidable suffering.
Conclusion
1) Hostility is not irrational in NU. In a theory which considers the empty world as the best state of affairs, survival doesn’t have the highest ethical priority.
2) Although the pinprick argument is not convincing, the claim that absolutely no kind of suffering can be compensated is untenable. It is more plausible to allow compensation and only disapprove those lives where the result is negative, i.e. lives which are not worth living.
3) NU population ethics is hostile in a pessimistic scenario and life-friendly in a transhumanist scenario.
4) A violent reduction of the population-size remains a theoretical option, as long as the suffering produced by the reduction is smaller than the suffering eliminated. Even if the optimistic transhumanist implementation of NU is supported by the majority, mankind could be exterminated by a pessimistic and militant minority. In addition the theory allows to kill a single suffering person without his/her consent, if the suffering produced by the killing is smaller than the suffering eliminated.
For these reasons NU never gained practical importance and is considered to be refuted. Incidentally Karl Popper was neither pessimist nor militant, as can easily be derived from his public statements. Example: Die Erkenntnistheorie und das Problem des Friedens.
The logical attempts to rescue NU as a general ethical theory are the following:
1) Remove the pinprick-argument (i.e. replace the avoidance of suffering by the avoidance of preference-frustration) and remove the painless-killing argument (by refusing the experience requirement).
This solution is proposed by Antifrustrationism and Buddhism.
2) Allow compensation within the individual so that lives with positive welfare are possible, but disallow compensation across individuals:
This solution is proposed by Negative total utilitarianism (NTU).
3) Allow compensation within and across existing persons, but not within future persons:
This solution is proposed by the Prior existence view of preference utilitarianism
3.4 Negative Preference Utilitarianism (Antifrustrationism)
3.4.1 Definition
1. In antifrustrationism utility is defined as the avoidance of preference frustration. Satisfied preferences have no intrinsic moral value and frustrated preferences count negatively, a concept which avoids the Repugnant Conclusion.
2. Satisfied preferences can have an indirect moral value. There is a (high order) biological preference to unfold and expand; a biological program to maximize the proliferation of genes (God’s utility function). Countless satisfied (low order) preferences help to avoid the frustration of this high order preference, whereas frustrated (low order) preferences worsen the state of affairs. This process can be described as compensation, as well as in preference utilitarianism.
3. A life starts deep in the minus zone and the satisfaction of all secondary preferences only serves to diminish this initial negative welfare. Since a maximization function can never be satisfied, the axiology stays in the negative territory.
4. The satisfied preference to exist only has moral value, when it concerns an existing life, but not when it concerns a possible life.
5. The method of aggregating utilities is undefined. For the purpose of this paper we assume that it is utilitarian.
6. Antifrustrationism denies the Experience Requirement
The term antifrustrationism was introduced by [Fehige]
3.4.2 Rationality
1. The applications are the same as for NU. Only the definition of utility differs.
2. The reflections concerning rationality are similar to the ones mentioned under the Maximin principle.
3.4.3 Hostile Potential
The preference for the empty world
Antifrustrationism corresponds to a “Schopenhauerian” theory of well-being: preference-frustration cannot be compensated by preference-satisfaction. There are no lives with positive welfare. Practical lives have negative welfare and the theoretically best life has neutral welfare. This assumption leads to the Reverse Repugnant Conclusion:
According to antifrustrationism a universe with no individuals is better than a universe in which there are 10 billion individuals with extremely happy lives, but where one of these individuals experiences a minor preference-frustration. This is an absurd implication.
For an answer to the Reverse Repugnant Conclusion see On the Buddhist Truths and the Paradoxes in Population Ethics.
Painless killing
As compared to NU the term suffering is replaced by the term preference-frustration. The goal of antifrustrationism is not to avoid every kind of suffering, but only those kinds, which are caused by preference-frustrations. This has a deep implication concerning the hostility of the theory. In NU there are only empirical arguments against a violent reduction of the population-size, in antifrustrationism the arguments are of a theoretical nature: A reduction without the consent of the affected individuals frustrates their preferences and is therefore immoral (even if the suffering in the world could be eliminated).
Example: In NU a surgeon could let a patient (painlessly) die in order to stop somebody else’s suffering by organ transplantation. In antifrustrationism this requires the consent of all persons involved.
Childlessness
The Reverse Repugnant Conclusion says that the best antifrustrationist world is the empty world. This suggests that the theory promotes childlessness. But Fehige’s view is different:
The claim, that our world is worse than an empty world does not imply, that we should empty the world we live in; we can’t for there have been preferences already [Fehige, 522].
Antifrustrationism doesn’t say that mankind should stop procreating because future people will have frustrated preferences [Fehige, 523].
The fact, that Fehige doesn’t propagate childlessness means that he gives a chance to transhumanism. In a pessimistic scenario, the frustration caused by the childlessness of a single generation is certainly smaller than the accumulated frustrations of all future generations. Fehige abstains from predicting the future and only looks at a single step in the development. He investigates if additional people make a population better or worse, but there is no attempt to valuate the long-term consequences of the action [Fehige 527-9]. This restriction of consequentialism allows the actual generation to improve their situation at the cost of future generations
Conclusion
1. In analogy to NU hostility is not irrational in antifrustrationism. In a theory which considers the empty world as the best state of affairs, survival doesn’t have the highest ethical priority.
2. In analogy to NU antifrustrationist population ethics is hostile in a pessimistic scenario and life-friendly in a transhumanist scenario.
3. Antifrustrationism denies the experience requirement and thereby removes the hostile potential of NU with regard to painless killing. But it uses the same unfamiliar axiology and claims that there are no lives with positive welfare [Ryberg, 140-1].
For a profound criticism of Fehige’s work see Future Generations, chapter 5.2, by Gustav Arrhenius
3.5 Negative Total Utilitarianism (NTU)
3.5.1 Definition
1.