Thoughts on Ethics

• A dead person’s property is transferred to the heir as stated in the will.
• A dead person (or rather: his heirs) will be compensated for any offense against that person, if the offense was committed before the death and has not yet been resolved.
• A person owns his body, so a person may state what shall happen to his body after his death.
• The dead person’s intellectual property remains protected against plagiarism.
• A dead person cannot be insulted. However, if something insulting is said about the dead person, and if this statement is provably wrong, the perpetrator may be charged with lying.

• The probability of S2=“Alonso paints his car green on this very day” in the state S1=“Alonso’s car is red on this day” is 1.0
• The probability of S2=“Alonso paints his car green” in the state S1=“Alonso’s car is red” is 0, because this is the first time that Alonso changes the car’s color from red to green.
• The probability of S2=“Alonso paints his car” in the state S1=“Alonso’s car has some color” may be quite high if Alonso often changes the color of his car.
• The probability of S2=“Somebody paints his car” is even higher.

• in the German Civil Code, a person is the cause of a state if the action “leads to the state according to common life experience” (adequacy-theory). So grandma is safe — unless she lives in a criminal society in which the majority of people steals.
• in the German Criminal Law, one distinguishes between “causing” and “being responsible for” (equivalence-theory). Although someone may be the cause of a damage, he is only responsible for it if his action caused the danger of damage, as defined by the law. In the example, the grandma is the cause of her grandson stealing a car. However, she is not responsible for it, because having a grandson is not seen as a danger of damage in the law. The problem is that someone might find a way to cause a damage without producing a danger of damage as defined in the law.

The notion of double causality is necessary to avoid the problem of cumulative actions: Imagine that Alonso and Bertram each give a deadly portion of poison to Cedrick (2 actions). Cedrick dies. Each of the two actions can be left away without helping Cedrick. But if both actions are left away, Cedrick survives. Hence, both Alonso and Bertram are the agents of Cedrick’s death. Double causality includes the case in which it cannot be determined which action exactly caused the damage. The standard example is the car accident, in which a car hits a biker and causes deadly injuries. Some seconds later, a lorry passes by and also causes deadly injuries. The biker is brought to a hospital, but dies. It cannot be determined which injuries exactly caused the death. So both the lorry driver and the car driver count as agents.

Different cases of direct agents can be imagined:

• There is no direct agent at all (like for instance for natural disasters)
• There may be one agent (this is the easy case)
• There may be multiple agents if the agents acted simultaneously or if double causality applies. This is, e.g., the case when two cars bump into each other. In this case, the agent that could have avoided the damage most easily receives a bigger share of the responsibility. There are two special sub-cases of this scenario:
  • If one person wilfully causes harm to another person (e.g., Alonso wilfully bumps his car into Bertram’s), then obviously the harming person receives all resposibility, because he could have avoided the damage very easily by refraining from the action.
  • If someone creates a dangerous situation, where the other party is forced to react (e.g., if Alonso blindfolds himself and then rides his car), then the reacting party becomes the direct agent, but can still shift the burden by invoking bad surprise.
• The direct agent can also be an animal, child, or mentally ill person. In that case, the resolution may turn to their guarantor for compensation of damage.

The data can be shared in an altered form: the phrasing can be changed, the resolution can be different, or the data can be transmitted in an aggregated form. The different types of damages that rely on the notion of sharing will impose different constraints on such modifications: declaration damages protect the receiver of the data. They are concerned with statements. Personality rights protect the person that the data is about. They apply independently of any transformation. Finally, intellectual property protects the creator of the data. It extends to all transformations of the data, as long as these are sufficiently complex.

In times of large language models, declarations can be created also by automated means. In that case, the declaration has to make clear that this is the case. If no such clarification is given, the receiver may assume that the declaration was made by a human. If that turns out to be false, the person who sent the declaration can be charged with lying.

If the speaker says that his statements are hypothetical, or that he does not take responsibility for their truth, then he does not lie. This can happen through the form of presentation. A novel, e.g., is not supposed to represent the truth and is hence not a lie.

Entering a train

The damage is trespassing the service provider’s territory. The declaration accepted by the customer are the terms of service.

Entering an auction hall

Again, the damage is trespassing. By trespassing, the customer accepts that by raising his hand he buys the item that is being auctioned.

Taking a newspaper out of a sell box

The damage is theft. The customer agrees to pay the price noted on the sell box.

Entering private property

If a sign at the entrance of a museum prohibits photography, then the customer makes a promise not to take pictures by entering the museum.

A contract can also be established without writing. Since declarations can be quite informal, a simple verbal consent already constitutes a contract. Since a declaration can also be made implicitly (e.g. by entering a restaurant), contracts may even spring to life without a single spoken word (see the definition of declaration).

The technically cleanest form of a contract is the following:

The above definition of a contract differs from the definition in the German Civil Code, where a contract consists of “two coinciding declarations of will”. With this definition, though, lying, breaking a promise and breaking a contract have to be treated separately.

The concept of heritage poses the question whether a death duty is reasonable. Without a death duty, the heir of a millionaire does not have to work, because he could simply have his father’s money work for him. Although this does not cause damage to anybody (on the contrary), it might be a fair assumption that everybody should work to earn his livings. Under this assumption, we may give part of the heritage to a charity. This would greatly benefit the community, but it would not cause damage, because, under the assumption, earning his livings is not a damage to the heir. Since people can circumvent death duties by transferring property before their death, one could envisage that people can never own land, but only lease it from the state for a fixed period of time (say, 50 years). After that period has elapsed, the lease has to be prolonged. This avoids the accumulation of land in one family’s hands. This principle is implemented, e.g., in Singapore.

Data that can be memorized by a human is not protected. This allows citations and summaries of intellectual property. It also disallows copyrighting words or patenting trivial mechanisms. Human nature as a yardstick may seem highly arbitrary as a choice. However, the argument goes as follows: In early times, there was no notion of copyright. People would just copy other people’s ideas, stories, or musical tunes. The problem appeared when recording became available. Then, creative works could become so complex that people expected to be paid for it. We would like people to pay for books, CDs, or software. So, in order to draw the line where complexity begins, this essay says that anything that requires recording is worthy of protection.

Intellectual property encompasses all transformations of such data (change of resolution, modification, combination, the creation of machine learning models, etc), as long as the result is also sufficiently complex. In this aspect, this essay differs from the copyright law of the United States: The US copyright sees a sufficiently different transformation of some material as a work on its own. This essay, in contrast, argues that the transformation could not exist without the original. Hence, the transformation is the intellectual property of both its creator and the creator of the original material.

This applies in particular to the training of large language models: both the model and its output are transformations of the input. If the output generated from one piece of intellectual property is sufficiently complex, it is intellectual property by the creator of that intellectual property. It is difficult to establish if some output of the model was generated by some specific input. However, this can be assumed if the output is nearly identical to the input.

This essay protects intellectual property data against unlawful sharing. Different from classical copyright, this sharing can also be transient: When a bystander listens to the music played by a street band, there is no act of “copying” — but the event still counts as sharing. Since this essay does not distinguish between “observing something” and “making a private copy of something”, it will allow people to keep a copy of whatever they see or hear (“You own what you see”). Thus, if someone plays his (legally obtained) music to his friends, his friends have the right to keep a copy of this music. However, as soon as the friends then let other people listen to the music (or give away a copy of it), the copyright is infringed.

There is some debate as to whether “fair use” of a copyrighted work should permit posting excerpts of it — such as a 10 second excerpt of a song on Wikipedia, a 2 page excerpt of a book in a magazine, or a 30 second teaser of a movie on Youtube. This essay does not permit such a possibility. The reason is that if the creator envisages such a possibility, he can always publish the excerpt on his own without any strings attached. Then Wikipedia, the magazine, or Youtube can pick it up. If the creator makes no such choice, then others should not do it for him. The reason is that such excerpts can be put together to form the whole. They can also misrepresent the work.

Intellectual theft differs from declaration damages such as lying in several ways: First, intellectual theft is not concerned with simple statements. The data has to be more complex to be protected. Second, the victim is the source of the data, not its recipient. Third, the perpetrator is the person who shares the data, not the one who created it. Finally, the damage happens when the data is shared, not when it is perceived by someone.

We could argue that true artists (and real programmers) produce their works not for profit, but for the enjoyment of it. There are many free software products, e.g., which are developed by the community, and there are many artists who just give away their work for free. However, at the same time, many artists and software companies do not give their work away for free. And even if there are free alternatives, people still use the non-free works. For example, there are free operating systems. Still, 90% of consumers use Microsoft Windows or Mac OS. There is also free music by amateur artists. Still, people buy CDs. This indicates that the non-free works are worth having. Furthermore, if we protect intellectual property, we do not prohibit that people make their works available for free. Creators can still do that.

One way to protect intellectual property is to count on the honesty and generosity of the users. For example, many artists make their music available for free, and then invite the user to make voluntary donations. Likewise, some software is available as “donation-ware”. This means that it can be used for free and that the client pays a voluntary amount of money to the programmer, if he decides so. This concept, however, works only because of a general awareness that the creator deserves something for his work. If we remove the concept of intellectual property from the law book, then this awareness will most likely go away. Already today, very few people are actually aware that they are using other people’s work, and that they should reward these people. For example, if we find a song for $1 on Amazon, but for free on some other Web site, then few people hesitate to download the free version. Few people have ever thought that they could pay for their virus scanner. This seems to indicate that the awareness for intellectual work is weak unless it is enforced.

The solution is to limit the protection of intellectual property in some way. Several methods can be envisaged: We could protect the property only against commercial use. The creator of the artwork would be entitled to benefits if his work is used in cinemas, concerts, books, CDs, and so on. Any other, private, use would be free of charge. This would mean, however, that the work itself has no financial value. The value would only be induced by an additional commercial service on top of it. That would mean that a work that requires no such mediation by an additional commercial service would have no financial value. E-books, e.g., can be copied and read without any commercial mediation. Music by smaller artists can be copied and played on any computer without the need for buying a CD. Software, likewise, can be copied without any commercial mediation in between. Furthermore, we will still have to come up with terms for the intellectual property, because we do want free commercial exploitation at some point of time.

Another solution is to allow the creator of the intellectual property to earn, say, 100 times the amount of money that he has spent for the creation of the work. After that, the work would be in the public domain. This would encourage people to create intellectual works, and at the same time limit the price that humanity has to pay. This solution, though, has 3 problems. First, it is difficult to estimate how much money the creator invested in the work. Does a trip to China count as an investment, if the artist needed it for inspiration? Second, the approach will allow people to sit on their intellectual property forever, if they decide to not make money from it (or do so in an inefficient way). That would be against the goal of giving the work quickly into the public domain. Third, and maybe most importantly, it would reward creators of great work with the same amount as creators of bad work. If a good author spends one year and writes a fantastic book, he should earn more money than a mediocre writer who spends one year and writes a bad book.

Therefore, a common solution is to limit the term in which intellectual property is protected. Current copyright law protects artwork for 70 years after the death of the creator. This is against the harm principle of this essay: The creator cannot be hurt by events that happen after his death. Hence, his intellectual property is not worthy of protection after his death. This essay proposes shorter terms of protection, in particular in the light of the large amount of money that media companies make. The essay limits the term of protection explicitly for some types of intellectual property. It leaves the terms for other, potentially future, types of works undefined. Such a term restriction allows copyright holders to make money from their work. At the same time, it transfers the work into the public domain quickly, even if the the creator does not make commercial use of it.

This essay protects artwork for a fixed period of time, no matter whether the creator is alive or not. This allows creators to sell the rights in the work to a company. The company can then count on income from the intellectual property no matter whether the creator dies or not.

If a term ends, the intellectual property can be used freely. This does not mean, however, that the creator of the work would have to provide the work to the public. For example, if a blog writer replaces his blog article by a slightly modified version of the blog article right before the end of the term, then a new term starts. The old article becomes free for use, but if no one has downloaded it, then no one can use it.

This offense concerns the person who receives the work, and who is misled by the sender. In addition, and in a separate case, the creator of the embedded work can demand credit for his work.

It is worth considering whether data on the Internet should in general be available for arbitrary use. Yet, the Internet will most likely be the mainstream medium in the future. It will not be an exceptional space of interaction, with different laws, but the normal space of interaction. Therefore, all arguments that we bring forward to the protection of data in the real world should also apply on the Internet.

This is sad, because the pressure comes solely from the stigma that our society attaches to nudity. In the ideal world, we would re-educate our society so as to de-stigmatize nudity. However, this is unlikely to happen in the near future. Therefore, this essay declares sharing such pictures an offense. This also means that it is an offense to use them for blackmailing.

Normally, one would expect people to be in private when they are nude. However, even pictures taken legitimately in private have to be protected against further sharing. Moreover, moments of suffering, injury, and even nudity can appear also in public. The offense of exposure provides a comprehensive protection against such embarrassments.

We could also completely prohibit the creation or possession of embarrassing material. However, if someones observes another person in an embarrassing situation, then this event becomes part of the observer’s life as well. Thus, the observer shall have the right to record this event (think of bodycams). However, he may not share this information. This constitutes a compromise between the right of privacy for the depicted person, and the right to document one’s own life for the observer.

It is highly controversial whether we may take pictures of people and use them. DougNZ provides a survey of regulations in many countries. These vary from very restrictive (virtually no use of pictures, as in Germany) to very liberal (everything is allowed, as in the US). We could say that everything that happens in a public space can also be distributed in a public space. Following this argument, it is no different whether I see a person in public, or whether someone else sees that person in public and sends me a picture of them. Furthermore, future on-body devices will most likely monitor anyway everything we see and hear. It would be implausible to require that pictures of other people be removed from these data streams. An extrapolation of this setting is discussed by Antoine Amarilli in his blog “Privacy in public space”. On the other hand, people do not like it if their picture is distributed in public. This holds even more so if the person is known in public, and if the picture shows the person in an unpleasant situation. Therefore, this essay prohibits the public sharing of portraits.

Independently of the rights of the portrayed person, the creator of the portrait has intellectual property rights in the image — as he does with any other image he creates.

This essay does not distinguish between a photo and a recording of someone’s voice. Thus, this essay permits recording someone’s voice as long as this is not shared publicly. This is a very controversial stance. In Germany, for example, it is prohibited to record someone’s voice. This essay acknowledges the controversy, but follows a liberal interpretation of justice: By talking, we influence someone else’s life. Therefore, we become part of that person’s life, and that person has a right to the content of the conversation. In the same spirit, the receiver of a letter, a mail, or a picture may share that piece with others. If we want to make sure that the receiver treats the letter as a secret, we need a prior consent. For example, we can make the receiver promise to keep the letter secret, and then send it.

Personally identifiable information also covers brand names, logos, and trademarks. In this way, it ensures that such items are not misused. However, the items can be used to facts, fiction, or an opinion about a corporate body.

The notion of knowledge is central to define denial of assistance: A person can only help somebody in an emergency if he knows about the emergency. The notion is also used as one way to make a declaration: A person can let another person know that something happened (or will happen) by showing physical evidence for it, without actually uttering a sentence. This act still counts as a declaration.

As the notion of knowledge, the notion of desire is a mental notion. It is hard to define it convincingly. Therefore, this essay simply assumes that a person desires a state if certain conditions are given. Naturally, a person may also desire a state if these conditions are not given. For this essay, though, it is only important to define when we can be sure that a person desires something. It does not matter if the conditions do not catch all cases where people desire something.

• desired, as defined in “desire”
• detested, i.e. the person desires the contrary of the state
• neutral or unknown, i.e. we do not know whether the person desires or detests the state

We need the notion of expected damage to quantify how dangerous a behavior is.

Any reduction of the expected damage is valuable, no matter whether the damage is actually prevented in the real course of time.

The compensation for refundable assistance is the damage caused to the helper (see provocation of assistance). This includes any loss of property as well as the effort itself (which counts as obstruction) and any offenses committed by the helper.

The assistance should not cause more damage than the expected damage, i.e. the effort of the helper is bound from above by the reduction of the expected damage — unless the damage affects the perpetrator of the expected damage directly. This direct effect means that the perpetrator must be the immediate victim of the damage, if the damage caused by the assistance exceeds the expected damage. This definition is necessary because any damage caused by the assistance affects the perpetrator on the long run, since he has to refund the assistance. Severe damages count more than other types of damage. Severe damages among themselves are compared as usual. As a result, any assistance that prevents severe damage is adequate, as long as it does not cause more severe damage. If severe damage is used, its benefit must be immediate. It is not allowed to use severe damage to coerce somebody into doing something (torture is prohibited). The reason is that it may be impossible to determine whether the person can fulfill the demand or not. Torturing thus always bears the danger of unjustly causing severe harm in vain.

• You do not have to accept severe damage to help somebody.
• You have to help only if the victim can’t do it for himself.
• You have to help only if you can expect your efforts to be refunded.

The obligatory assistance is an “obligatory purchase”: The helping person is obliged to help and the victim is obliged to “buy” this assistance, i.e. to compensate the effort. If an offense is prevented, the wannabe-perpetrator will have to pay for the prevention.

The compensation for a property violation is, as the default definition suggests, giving back the thing and making up for the period of loss.

Littering is an offense against Mother Nature, i.e., the ensemble of plants and animals in that area. Because mother Nature is non-human, it will be represented by a lawyer and the penalty will go to a charity, see resolutions.

Obstruction raises the question of whether livestock may be kept, because keeping the animals means locking them up. Admittedly, it is disputable whether mankind shall keep livestock at all, given that one could in principle live without animal products. However, at this point of time, it is unrealistic to require mankind to give up animal products. Therefore, this essay allows keeping and eating livestock (see Food Killing). We ask only that farmers keep their livestock in a reasonable environment, and that customers buy products that have been produced without harming animals.

The German Criminal Law does not punish killings of animals unless it provokes public anger. Apart from the fact that it is a pity that killing animals usually does not provoke public anger, the killing itself should be punished regardless of the public reaction. Due to the distinction of humans and animals, special laws in the criminal law are needed to protect endangered species.

Note that killing oneself is a desired damage and hence not unjust. People are allowed to commit suicide.

See the next definitions, Food Killing, Abortion, and Mosquito Killing, for exceptions to the general prohibition of killing.

The main reason against abortion is the protection of unborn life. The definition of “life”, however, is disputable: An egg is not a chicken. This holds even if the egg is fertilized — as are, indeed, a large number of the eggs that we eat. Yet, we do not consider these eggs chicken. Similarly, a human ovum, even if fertilized, is not a human.

Other arguments against abortion are of religious nature. A common Christian argument goes that abortion is a destruction of life against the will of God. However, Christianity had different stances towards abortion. The Bible, e.g., states that abortion is not murder (Bible / Exodus 21:22). Saint Augustin, an important Christian scholar, held that an early abortion is not murder, because the soul had not yet entered the fetus. This belief got passed into canon law, and was upheld by Saint Thomas Aquinas, Pope Innocent III, Pope Gregory XIV, and the Council of Vienne [Wikipedia / Christianity and abortion]. Muslims, likewise, mostly condemn abortion today. However, during the Golden Age of Islam, abortion was not prohibited. It was thought that the soul had not yet entered the fetus, and that hence the killing of an unborn was not murder. This interpretation was based on the quranic descriptions of the development of the fetus (Quran 23:12-14 and 22:5), and on a Hadith [Wikipedia / Ensoulment]. In any case, around 20% of pregnancies end with a natural premature death of the fetus, and we know no way to avoid that. If God is the cause of everything, then God is also the cause of these deaths. This makes God the most active abortionist.

The German Criminal Law states that making penetrant noise is a crime — although, sadly, this does not apply to vehicles, which contribute most of the noise in our environment.

The compensation for disturbance means ending the disturbance.

Breaking a contract and breaking a promise are special cases of lying.

Often, threats are bound to certain requests, like for instance in extortion. In this case, the threat itself counts as damage, independent of whether the victim obeys the perpetrator’s conditions or not. Furthermore, the threat counts as a declaration, i.e. the victim may assume that the perpetrator fulfills the threat (by the Dixit-Principle, see lying). Hence, if the victim obeys the perpetrator and does as he requests, the victim performs refundable assistance to rescue himself. Thus, the victim will get a compensation for all damage he suffered while following the perpetrator’s request. This may be direct damage (like giving money to the blackmailer) as well as indirect damage (like committing an offense and having to pay the compensation). However, while following the perpetrator’s request, the victim shall not cause damage that is greater than the damage he is threatened with. In this case, the victim is expected to disobey and to suffer the damage he is threatened with. The definition of refundable assistance makes sure that this does not apply for severe damage.

If the victim does not obey the perpetrator and the perpetrator makes his threat come true, the perpetrator will be held responsible for both the threat itself and the actual damage caused to the victim (as covered by the other definitions).

The compensation for a threat means revoking the threat and making up for the distress.

The offense differs from the other offenses in that it refers to an external law enforcement system. If that system is not in place, dangerous items are permitted. If it is in place, dangerous items may be used only by the respective professions.

If the sender wants to criticise the suffering (or the distribution of such graphical violence), then he may share such descriptions. However, only moral subjects can be the target then, not children.

Sometimes people tell their secrets to other persons. These secrets are not protected by this law on privacy violation. If the person tells the listener afterwards not to disseminate the secret, this request has no legal value. Instead, the person should ask the listener to promise to keep silent (and it is even better to do so before telling the secret).

This definition also protects people from having their data or behavior analysed by automated means in order to construct psychological profiles (or any other type of summary). Any data gained from such an analysis is private data.

If a privacy violation is necessary to unveil an offense, the privacy violation may be committed. It will be compensated by the perpetrator of the offense as refundable assistance. This entails that perpetrators of offenses are not protected against privacy violations concerning their own offenses.

Anybody who receives the data can share the data in their personal circle. For example, you can show an email that you received to your friends. However, your friends may not share the email any further. The principle that “You own what you see” means that we cannot prevent them from recording or storing what they hear or see for themselves (in their email account, or through a body cam), but they cannot share the data with others then.

If you gain hold of private information through a privacy violation, then the case will be treated as a privacy violation, and not as intellectual theft. After an intellectual theft, the perpetrator also loses his right to the data. This step is necessary because the “destruction of the data from the domain of the receiver” is meaningless if the receiver did not store the data. The perpetrator will usually be the owner of the source domain, but it could also be a third person.

The perpetrator can get away with such statements if he can prove that the statement is true. Different from declaration damages, the victim of the insult is not the receiver of the message, but the person who is being talked about. The perpetrator is the sender of the statement.

If you witness an embarrassing situation, you may record it, because in this case the victim himself shares the information. However, you may not share it with anyone else. If you do, you lose the right to keep the data in your domain. If someone else shares the information from your domain, then you can keep the data.

It is not necessary that the embarrassing information contains personally identifiable information. This is because even if the victim cannot be identified by permanent identifiers, he could still be identified by the circumstances in which the embarrassing information was recorded.

It is not sufficient for the offense of exposure that the material is funny or awkward. It has to be so embarrassing that a reasonable person would be willing to pay money to prevent the sharing.

Exceptions are granted for anything that related to a public speech. Such speeches can be freely related and criticized. Exceptions are also granted for talking about corporate bodies. Basically everything can be said about a corporate body — as long as facts, fictions, and opinions are distinguished. It remains prohibited to use the name of a company for any other purpose — such as to advertise, to suggest approval, or to appropriate.

Independently of the offense of disclosure, wrong or insulting information may not be shared.

Giving false credit is not just a lie, it is also an offense of missing credit. The victim of the lie is the recipient, and the victim of the missing credit is the creator.

This law puts the question of whether homosexuals may adopt children into the hands of science. Should science find that an upbringing with two fathers or two mothers is harmful, then it becomes illegal. Otherwise, it is not. A divorce, likewise, is considered harmful. It cannot always be avoided, but it is considered unjust damage to the child. Incest, likewise, is punished under this law, if it leads to predictable and known congenital defects.

The victim is the child. This means that the child is entitled to damages if the parents mess up his youth. The law also implies that we may have to take away the child from his parents, if these are unwilling or incapable of executing their duties. This is to be done (1) in the best interest of the child and (2) seeking consent of the old parents, the new parents, and the child (as far as the persons are capable of deciding).

Denial of assistance is a strange offense, because it does not create damage by itself (the damage happened anyway). Therefore, the compensation goes to a charity. The victim still receives the apology, and the perpetrator is still punished.

The compensation increases with the damage. It decreases with the effort that is required for the assistance. It also decreases if the helper did provide some kind of assistance, but not the full required assistance. By default, this essay suggests a compensation that is equal to 1/3 of the compensation of the damage.

In general, a victim can never be obliged to provide obligatory assistance to himself, because he cannot commit an offense against himself. However, if the victim knows that the perpetrator is unaware of the damage he causes, he has to notify him (see silent suffering).

Note that, by the above definition, denial of assistance includes cognisance: If someone helps planning a crime, but does not participate in it, then he can still be accused of denial of assistance.

Provocation of assistance is not modular in the sense of the Principle of Modularity. I.e. if victim and perpetrator differ, the helper can demand compensation not from the victim but from the perpetrator. This is because the victim may be unknown, see the remarks at “expected damage”.

This rule also proves quite useful for minor cases. Suppose, e.g., that Alonso eats some jam and does not close the top of the jar properly. Bertram comes and takes the jar by the top, the jar falls down and crashes. Here, the state that a jam-jar is not closed properly is less common than somebody taking it. Hence it was Alonso’s fault.

Conveniently, conventions may be used to establish a high probability. For example, it is a convention in Germany to drive on the right-hand side of the road. Thus, anybody who uses the left hand side performs an uncommon action and can be held responsible for eventual damage.

The compensation of complicity is computed as follows: Be IA the indirect agent. Be d(IA) the compensation for the hypothetical damage caused by IA alone. Be DA the direct agent. Be d(DA) the compensation for the hypothetical damage caused by DA alone. Be D the compensation and apology for the actual damage. Then the compensation of complicity is D*d(IA)/(d(IA)+d(DA)). This is the compensation that the direct agent DA may expect from the indirect agent IA.

Note that the joint damage D may well exceed the sum of the respective hypothetical damages! It is necessary to treat complicity apart from bad surprise, because bad surprise relies on the probabilities of the actions, which would not make sense for complicity. Furthermore, bad surprise cannot allow proportional sharing of the damage (potential division by zero), whereas complicity relies on proportional sharing on principle.

In this essay, corruption is a special case of breaking a contract, which is by itself a special case of lying. The government official has an employment contract with the government. If he gives an illegal advantage to someone, he breaks his employment contract. This makes him liable to prosecution under the laws of this essay. This holds regardless of whether the official obtains money for his actions or not.

The party who bribes the official is conscient of the misbehavior of the official, but does not prevent it. Therefore, this party commits a denial of assistance to the employer of the offical (which is the government). This way, both parties have to be punished.

By default, the victim is the perpetrator. However, someone else may be the guarantor for the damage. Assume for example that a young child falls down the stairs. There is no direct agent for this damage. Hence the child himself is the perpetrator. The resolution will then turn to the parents of the child for compensation.

It is unfortunate that if the perpetrator and the victim agree on a compensation, the perpetrator evades the penalty. The penalty prevents the perpetrator from repeating the deed. However, the liberal principles of this essay prohibit a resolution for a case that is already solved.

Compensation does not cover other unjust damages.

Example: Alonso’s car is stolen. Alonso decides to walk instead of driving, falls into a drain and breaks his leg. This damage is not covered by the compensation for the car theft, because in reasonable courses of time, people do not suffer accidents (which fall under the definition of unjust damages).

Compensation covers consequential losses.

Example: A machine in Alonso’s factory is destroyed. Hence Alonso cannot use the machine for production. However, he is assumed to use the machine in the closest reasonable course of time without the destruction. Hence the loss in productivity is covered by the compensation.

Compensation does not cover unreasonable behavior.

Example: A machine in Alonso’s factory is destroyed. Alonso decides to cease production completely, although the remaining machines could continue production. However, he is assumed to continue production with the remaining machines in the closest reasonable course of time despite the destruction. Hence this additional loss is not covered by compensation.

Compensation covers contingent events.

Example: A fire-extinguisher in Alonso’s house is stolen. A fire breaks out and the house burns down. In the closest reasonable course of time without the theft, the fire would have broken out as well. However, in this course of time, Alonso would have extinguished the fire with the fire-extinguisher. Hence the compensation for the theft covers the house. That means: Whoever stole the fire extinguisher is liable for the destruction of the house.

There is no apology for provocation of assistance, because helping people in need is assumed to be a civic duty by this essay. There is no apology for complicity because the indirect agent is not morally worse than the direct agent. There is no apology for killing because the victim cannot benefit from it.

The penalty shall include all fruits that the perpetrator harvested from the offense. These fruits do not go to the victim (because the victim is compensated by the compensation and shall not benefit from the perpetrator’s diligence) and they cannot remain with the perpetrator (because this would encourage him to commit the offense again). Hence they go, with the remainder of the penalty, to a charity.

In case of an offense by a corporate body, a penalty shall be applied to both the corporate body and to the individual person who caused the damage. This individual responsibility avoids that the boss of a company engages in illegal behavior and then makes the company pay for it.

The penalties that this essay establishes do not include the death penalty (or in fact any other corporal punishment). There are several reasons against the death penalty (do a Google search). The main argument is that juries may err in their decision. In particular, laws can change. In this case, the perpetrator has to be able to challenge his conviction. If he is killed, this is no longer possible. Therefore, this essay resorts to a wide range of measures with the aim of putting the perpetrator back on track.

In general, all necessary efforts by investigators (including offenses) and all necessary efforts by the lawyer and the victim (including prior, failed searches for the guarantor) are to be refunded by the guarantor. In case of doubt, another resolution is necessary on this issue. Note that the jury does not get paid.

Since this essay relies on the Principle of Modularity, indirect damages entail two resolutions instead of a single one (e.g. bad surprise). This modularity ensures that the victim can turn to a single guarantor for his claims (instead of having to resolve the problem of shifted responsibility). Furthermore, the principle ensures that the direct agent of the damage (e.g. the victim of a bad surprise) works together with the jury, even if he will not be held responsible in the end. Nevertheless, it may be reasonable to join two resolutions, if the offenses are related.

The lawyer will get the apology upon conviction, any person involved in the search for the guarantor will be refunded, the jury promised to act to the best of their capabilities and the suspect will want to proof his innocence. Hence, all people involved should have an interest in a speady resolution. If the victim proves uncooperative, the case is dismissed (see No sulking). If a suspected guarantor proves consistently uncooperative, it may be assumed that he is indeed responsible for the offense.

This essay distinguishes between the perpetrator of an offense, and the guarantor of the perpetrator. The jury may split the compensation and apology between both. The jury may also impose separate penalties on both, always with the goal of avoiding repetitions of the offense. These considerations apply only to human perpetrators.