When the State Is Liable Even Though It Did Not Act: Government Omission
This page is a translation of the original Portuguese text, provided for convenience. It deals with Brazilian law. If there is any discrepancy, the Portuguese version prevails. Read the original in Portuguese
The State is liable both for what it does and for what it fails to do, but the way liability works changes depending on whether the harm comes from an affirmative act or from the absence of a service the government was obligated to provide. Understanding this difference determines who bears the burden of proving fault and which defenses the Administration can raise to escape liability.
The constitutional basis of State liability
The civil liability of the State is expressly established in Article 37, paragraph 6, of the Federal Constitution. The provision states that public-law entities, as well as private-law entities that provide public services, are liable for harm that their agents, acting in that capacity, cause to third parties. They retain the right of recourse against the responsible agent in cases of intent or fault.
Legal scholars draw from this text what is known as the administrative risk theory. Under it, the State’s liability is strict: the victim does not need to show that the public agent was at fault. It is enough to prove the State’s conduct, the harm suffered, and the causal link between the two. The idea is that those who benefit from administrative activity must bear the risks it creates for private individuals.
Strict, however, does not mean unlimited. The administrative risk theory recognizes defenses that break the causal link, such as force majeure, an external fortuitous event, and the victim’s sole fault. In those cases, the State’s liability is eliminated, in whole or in part, even though the harm occurred in the context of a public activity.
Harm caused by action versus harm caused by omission: the difference that changes everything
The sensitive question arises when we ask whether this strict liability applies equally to harm caused by action and harm caused by omission. This is not an academic detail: the answer changes what the victim will have to prove in court and, often, decides whether the lawsuit succeeds.
In harm caused by action, also called commissive conduct, a public agent performs a positive act that injures a private individual. Examples include an official vehicle that collides with a car, an improper shot fired by an armed public servant, or the mistaken demolition of a lawfully built property. Here there is consensus: the strict liability of administrative risk applies, with no inquiry into fault.
In harm caused by omission, the situation is different. The State does not directly cause the injury. It results from a service that should have been provided and was not, or that worked poorly or came too late. Examples include a flood made worse by uncleaned storm drains, an inmate killed by another inmate in an unguarded prison, or an accident caused by an unmarked hole in a public road.
For much of the classic scholarship, led by Celso Antônio Bandeira de Mello, omission triggers fault-based liability. The reasoning is logical: if the State did not act, the harm was produced by a third party or by nature, and there is reason to hold the government liable only when it had a legal duty to act and failed to meet that duty through negligence, recklessness, or lack of skill.
This approach ties omission to the theory of administrative fault, also known by the French expression faute du service, or “failure of the service.” The idea is that liability arises when the service does not work, works poorly, or works late, creating a duty to pay compensation even without identifying which specific agent failed. This is known as anonymous fault, attributed to the Administration as a whole.
In action, the State answers for the risk it created; in omission, it answers for the failure it should have prevented.
In practice, though, this division is not rigid in the courts. Case law has refined the issue and now distinguishes situations of omission that deserve different treatment. This requires a careful look at the nature of the duty that was breached before deciding which rule applies to the case at hand.
The idea is that those who benefit from administrative activity must bear the risks it creates for private individuals.
Specific omission and generic omission
The distinction most often used to resolve this impasse separates specific omission from generic omission. It was systematized in the scholarship of Sérgio Cavalieri Filho and is widely adopted by the higher courts. It works as a filter for deciding whether liability will be strict or fault-based.
Specific omission exists when the State, through an earlier act or inaction, creates the conditions for the harm and had a concrete, individualized duty to prevent it. A prisoner in custody is the clearest example: the government assumes the duty to guard and protect the physical integrity of a person who is detained, so harm suffered by that individual gives rise to strict liability, because the State occupied the position of guarantor.
Generic omission exists when the duty to act is diffuse, owed to the community as a whole, and the State has not taken on any special position toward the victim. A lack of police patrols that failed to prevent a street robbery, or the absence of broad inspection, illustrates the point. In these cases, fault-based liability prevails, and the victim must show that the service failed through fault and that this failure could reasonably have been expected to be avoided in that context.
The procedural consequence is significant. Faced with a specific omission, the victim only needs to prove the harm and its link to the breached duty of protection. Faced with a generic omission, the burden grows: the victim must show that the Administration could and should have acted, that it had the means to do so, and that its inaction was decisive in producing the harm.
What the victim must show and which defenses are available
Whether liability is strict or fault-based, the causal link is the central element of any claim against the State. Without a connection between the State’s conduct or inaction and the alleged loss, there is no duty to pay compensation. The causal link is precisely where the government’s defense tends to concentrate its efforts.
Liability defenses work by breaking that chain of causation. Force majeure and external fortuitous events, which are unforeseeable and unrelated to administrative activity, eliminate the duty to compensate. The victim’s sole fault also eliminates liability, and concurrent fault proportionally reduces the amount of compensation, dividing the burden between the parties.
Another decisive point is the time limit. Compensation claims against the Treasury (Fazenda Pública, the government as a litigant) are generally subject to a five-year statute of limitations under Decree 20.910/32. The five-year period usually starts on the date of the harmful event or when the victim becomes unmistakably aware of the injury and its extent, which underscores the importance of seeking legal advice without delay.
In practice, the lawyer should, from the very first contact, classify the case: determine whether the harm resulted from action or omission and, if it was an omission, whether it was specific or generic. This classification guides the evidence strategy, determines which documents and witnesses are essential, and anticipates the defenses the public entity will certainly raise in its answer to the complaint.
Frequently Asked Questions
Is the State’s liability for omission always fault-based?
No. The classic scholarly rule associates omission with fault-based liability, requiring proof of fault in the service. However, case law distinguishes specific omission, in which the State occupies the position of guarantor and is strictly liable, from generic omission, tied to diffuse duties, in which the requirement to show the Administration’s culpable failure remains.
What is the difference between the administrative risk theory and failure of the service?
The administrative risk theory is the basis for strict liability for harm caused by action. It does not require proof of fault and allows defenses that break the causal link. Failure of the service, an expression of administrative fault, applies mainly to omissions and requires showing that the service did not work, worked poorly, or worked late, which is what is called the government’s anonymous fault.
What is the deadline to seek compensation from the government?
A compensation claim against the Treasury (Fazenda Pública) generally expires after five years, under Decree 20.910/32. The period usually runs from the harmful event or from unmistakable awareness of the harm and its extent. Because disputes can arise over the starting date and possible suspensions of the period, it is advisable to assess your specific situation as soon as possible so you do not lose your right to sue.
Receba novidades por e-mail
Cadastre-se gratuitamente para receber nossos novos artigos.
Seus dados estão protegidos conforme a LGPD.
Ficou com dúvidas? Fale com um advogado especialista.
📱 Falar pelo WhatsAppAs informações deste artigo são de caráter informativo e não substituem consulta jurídica individualizada. Cada caso possui particularidades que exigem análise específica.






