Ir para o conteúdo
Life Security

Pesquisar no site

Encontre páginas e seções.

Why privacy matters

Última atualização:

Privacy matters because information about people gives others power over them, and the real question is who holds that power. Law professor Neil Richards makes this case in Why Privacy Matters (2021). When someone can learn and use facts about you, they can influence what you’re offered, how you’re judged and what you feel free to do.

People often say privacy is already gone. Mark Zuckerberg once declared that the age of privacy was over. Richards disagrees and says its future is up for grabs, which means it depends on choices still being made.

This page defines privacy, security and anonymity the way researchers use the terms, shows what goes wrong when privacy is missing, and answers two common objections: that privacy is just secrecy, and that it’s just a matter of personal control. It doesn’t tell you what to do about any of it. The rest of the digital section covers that.

What privacy means

People use privacy, security and anonymity as if they were one thing. Researchers keep them apart, and the differences explain why a tool can be excellent at one and useless at another.

Privacy

No single definition has won. The most influential comes from Alan Westin’s 1967 book Privacy and Freedom. He described privacy as a claim by individuals, groups or institutions to decide for themselves when, how and to what extent information about them reaches others. The International Association of Privacy Professionals (IAPP) notes that nearly every data protection law written since then reflects this idea, including the OECD guidelines of 1980, the EU’s 1995 directive and the APEC framework of 2004.

Richards offers a plainer version: privacy is the degree to which information about people is neither known nor used. Daniel Solove, a law professor at George Washington University, calls privacy “a concept in disarray” and, instead of forcing one definition, catalogues the distinct harms that privacy problems cause. We return to his list below.

These definitions differ, but all three are about how information moves between people and institutions. None is about hiding things.

Security

Security has a more formal standard. US federal law, repeated in the glossary of the National Institute of Standards and Technology (NIST), defines information security as protecting information and information systems from unauthorized access, use, disclosure, disruption, modification or destruction. The aim is to keep them confidential, intact and available when needed. Those three goals are often called the confidentiality, integrity and availability triad.

NIST’s wording for confidentiality includes protecting personal privacy, and that’s where the two ideas touch. Security is a means to privacy, not the same thing. A website using HTTPS (the encrypted version of the web protocol) stops someone on your network from reading or altering what you send. That is security. It says nothing about what the site’s owner later does with the data.

Anonymity and pseudonymity

The terminology most often cited in research comes from Andreas Pfitzmann and Marit Hansen. They define anonymity as the state of not being identifiable within a set of possible subjects, called the anonymity set. All else being equal, the larger that set, the stronger the anonymity. Being anonymous therefore doesn’t mean having no name. It means being one of many people who could have performed an action.

Pseudonymity is different: you use a persistent identifier, such as a username, that isn’t tied to your legal identity. People can recognise your handle across posts without knowing who you are.

How the three overlap

An encrypted messenger gives you security, because outsiders can’t read or alter messages, and privacy for the content. It doesn’t give you anonymity if the service knows who talks to whom. A post under a pseudonym on a public forum is pseudonymous but not private, since anyone can read it. A data leak is a security failure that causes a privacy loss.

Most real situations need some mix of the three, and you rarely get all of them. Deciding which matter for you is the job of threat modeling.

What goes wrong when privacy is missing

Solove’s 2005 A Taxonomy of Privacy sorts privacy harms by the activity that causes them. Three of them explain most of what happens to personal data online.

Aggregation is combining information about a person. In September 2026, Dismislab reported buying three kinds of data from sellers in Bangladesh: a copy of a national ID for Tk 500, delivered in 17 minutes, three months of call records for Tk 1,050, and a phone’s recent location. Each item has limited value alone. Together, they show who someone talks to and where they spend their time. Dismislab verified the data with the people concerned.

Secondary use is reusing information for a purpose different from the one it was collected for, without the person’s consent. In 2024 the US Federal Trade Commission took action against three data brokers, alleging they sold location data gathered through phone apps that revealed visits to places such as health clinics and places of worship. The agency said the practice exposed people to stigma, discrimination and violence.

Exclusion is not being able to know or influence how your information is used. In its case against another broker, Kochava, the FTC said people are often unaware their location data is being bought and sold, and have no control over it.

Richards sums up what is at stake as four values: identity, power, freedom and trust. Human rights law reached a similar conclusion much earlier. Article 12 of the Universal Declaration of Human Rights (1948) says nobody should face arbitrary interference with their privacy, family, home or correspondence, and that the law should protect people against it.

Privacy is not secrecy

A common reply to privacy concerns is that people with nothing to hide have nothing to fear. Solove wrote a 2007 essay on this argument. He says it only works if privacy means concealing wrongdoing, a narrow view that has nothing to say about the other problems caused by collecting and using data. In his 2005 taxonomy he calls this the secrecy paradigm: the idea that information which isn’t hidden carries no privacy interest.

Everyday life shows why that fails. You tell a doctor about a symptom, a bank about your income and a friend where you live. None of it is secret from the person you told. You’d still object if all three facts were put in one file and sold to a stranger. That’s aggregation and secondary use, and it doesn’t depend on you having done anything wrong. The test for privacy isn’t whether something is hidden. It’s whether information flows only to the people and purposes you expected.

Is privacy about control?

Westin’s definition is a control model, and it shaped data protection laws around the world. It’s appealing because it treats you as the decision-maker about your own information.

Its weakness is scale. Real control would mean making informed choices with every app, website, mobile operator and clinic that holds data about you. Privacy Guides, a non-profit privacy project, says it once defined privacy as control and dropped that view. It argues that cookie banners and settings dashboards give an illusion of control while pushing the burden onto individuals.

We think that is partly right. Control is a good right and a poor strategy on its own. US law gives a hint of the same problem. The FTC treats a practice as unfair when it is likely to cause substantial injury that consumers can’t reasonably avoid themselves. Harms that you can’t avoid by choosing differently need rules, not only choices. Richards makes the same argument at book length: protecting identity, freedom and trust takes privacy rules that apply whatever any one person clicks.

Sources

Perguntas frequentes