Freedom of speech is one of the core values of TIHOMA. In any democratic public space, digital or otherwise, people should be free to express themselves, which also means that they may offend others and may be offended by others. That being said there are also some limits to freedom of speech which ought to be as minimum as possible. In TIHOMA these limits are set on the condition that some act causes unjustified significant harm to a person, group of people or non-human life-forms. Any act that obstructs a being from living and expressing itself freely or causing non-consensual physical, mental or emotional pain to it is considered as harmful. Harm is considered unjustified when it is caused for any other reason than minimizing harm and protecting the rights of humanity and the rest of the life-forms when there is no better alternative.
Criteria of harm
As the notion of harm is vague and can be very subjective, some as objective as possible qualities of harmful acts are provided which on the other hand also take into account the current social context. They are designed to both filter harmful content but also to protect content that should not be censored, such as satire, provocative art and journalism from being labeled as harmful. They are all prerequisites so that some content can be deemed inappropriate in the platform. The harm caused by the act has to be:
- Clear: The harm caused has to be specific. For example, a vague threat of someone watching the acts of another with some implications is not clear. A threat, that clearly states or implies that someone will attack someone else in their house is clear.
- Direct in causality (even if not apparent): It cannot be something that just has some indeterminate contribution, it has to play central role. For example expressing a negative opinion about a minority, right or wrong, does no direct harm. Calling the public to acts of violence against them does.
- Significant: The harm caused has to be impactful. For example, expressing a general mistrust against vaccines without evidence is less impactful. Spreading misinformation about a vaccine during a lethal pandemic can cost the lives of many people. Threatening someone to throw a water-balloon at their house is no big deal. Threatening to burn it is.
- Unjustified: There rarely actions that cause no harm if any. Based on a chosen ethical system and a certain historical context, some types of harm are acceptable. For example, some harm, like environmental damage from building an airport, might be justified if it prevents greater harm to society. Conversely, trophy hunting wild animals is entirely unjustified.
Cases of harmful content
Below are the main cases where content or content creators can be banned from the platform or have some kind of penalty if the harm falls within the aforementioned qualitative criteria:
- Slander.
- Threat.
- Blackmail.
- Harassment.
- Repeated harmful behavior, even if individual actions are minor.
- Making public, private or secret information, unless it is for justified public interest as in investigative journalism or whistle-blowing.
- Calls to crimes against people and groups or non-human life-forms, such as inciting violence, hate speech or provoking acts of animal cruelty.
- Spread of misinformation-disinformation that is harmful to society.
- Child pornography or depictions of torture, unless intended for journalistic or educational purposes to expose injustice or inform the public.
Overall, there is some relativity in what constitutes harm according to different historical, cultural and personal contexts. Some cases, such as child pornography, are easy to identify while others, such as harassment, are more difficult. For example, in some societies which are organized more communally, a house visit without notification could be normal, while in others could be viewed as intrusive. In cases where objective judgment is difficult, the community and moderation team should review the situation individually whenever possible or apply context specific protocols.