Skip to content

Alegre 2021 – Protecting Freedom of Thought in the Digital Age

Year

The citekey says 2022, but the document is CIGI Policy Brief No. 165 from May 2021 (PDF p. 2). Rebekka should check which version she cites.

In brief

A policy brief by a human rights lawyer, arguing that digital technology's impact on how we think, feel and behave calls for a new regulatory perspective: protecting freedom of thought and opinion, not only privacy and data protection (p. 2) [U]. Freedom of thought has three elements: the right to keep thoughts private, the right not to have them manipulated, and the right not to be penalised for them (pp. 2, 4) [U]. These are absolute rights, which may require banning technology that interferes unlawfully with our minds (p. 3) [U]. States have a positive obligation to protect the right against businesses through legal and regulatory frameworks (pp. 2, 4) [U].

Research question and main claim

  • The brief asks what freedom of thought means in the twenty-first-century data economy, and how innovation can support our inner freedoms (p. 2) [U].
  • Main claim: regulation has so far focused on privacy, data protection and freedom of expression, but a new perspective is needed to protect the ‘forum internum’ – the inner space of the mind (p. 2) [U].
  • Regulators need new frameworks that respond to the wider societal impacts of data use ‘rather than focus on control of personal data’ (p. 6) [U].

Theoretical perspective

  • International human rights law: ICCPR articles 18 and 19, the UN Human Rights Committee, the European Court of Human Rights, UN Guiding Principles on Business and Human Rights (pp. 3–4) [U].
  • Zuboff's surveillance capitalism and the ‘human futures’ market (p. 4) [U].
  • The concept of the ‘attention economy’ (Herbert Simon) (p. 5) [U].
  • Earlier work by the author and others on freedom of thought (Alegre 2017; Aswad 2020; Vermeulen 2006) (p. 4) [U].

Empirical material and context

  • No systematic empirical material. Examples: Facebook ‘likes’ predicting personality, targeted advertising to vulnerable young people, mental health websites sharing data, Facebook's emotional contagion experiment, Cambridge Analytica, predictive policing, an insurance company's personality-based pricing app (pp. 4–5) [U].
  • Regulatory examples: the Spanish Constitutional Court on political parties' profiling, UN General Comment 25 on children's rights, the US FTC's orders to social media companies (2020), the European Commission's AI proposal (2021) (pp. 6–7) [U].
  • Context: published by the Centre for International Governance Innovation (CIGI), a Canadian think tank (pp. 2, 11) [U].

Method

  • Normative legal argument / policy brief with recommendations [T – characterisation].

Key concepts

Concept How the source defines or uses it Page U/T
freedom of thought Legal right in international human rights law; absolute in the forum internum 2–4 U
forum internum The inner space of the mind 2, 4 U
three elements Keeping thoughts private; not having them manipulated; not being penalised for them 2, 4 U
absolute right Unlike privacy or freedom of expression, cannot be restricted 3–4 U
inferences Insights about what we think, inferred from metadata, accurate or not 4–5 U
privacy as gateway right Privacy is a gateway, but freedom of thought is at the heart of being human 8 U

How privacy and private data are understood

  • Privacy is a gateway right, but not enough on its own (p. 8) [U]. Regulation focused on privacy and data protection does not capture the impact on freedom of thought (pp. 2, 7) [U].
  • Privacy, unlike freedom of thought, can be restricted in certain circumstances (p. 4) [U].
  • The central concern in the data economy is not names or phone numbers, but the sharing and sale of inferences about what we think, based on metadata (p. 4) [U].
  • Personal data are used to predict and curate behaviour in the ‘human futures’ market (p. 4) [U].
  • Inferences are used in both the public and private sectors to make automated decisions about people, for example predictive policing (p. 5) [U].
  • The author calls for regulation that looks at wider societal impacts rather than control of personal data (p. 6) [U].

Arguments and findings

  • Companies may have many data points about us even if we have never used their platforms (p. 5) [U].
  • Data analysis of how we feel in real time is valuable because it makes it possible to exploit weak moments for commercial or political gain (p. 5) [U].
  • AI that claims to analyse mood or predisposition to crime is a tool for punishing people based on assumptions about what goes on in their heads (p. 5) [U].
  • States have so far failed to take serious action (p. 6) [U].
  • Recommendations: ban micro-targeted political ads and technology that infers political views from biometric data; outlaw emotion-analysis facial recognition, lie detectors and black-box risk algorithms in criminal justice; protect children's freedom of thought (pp. 6–7) [U].
  • The EU's AI proposal is welcome, but its effectiveness depends on the details (p. 7) [U].

Limitations the source itself mentions

  • Regulating from the perspective of freedom of thought is new and complex (p. 6) [U].
  • What the FTC will uncover remains to be seen (p. 7) [U].
  • Otherwise: Not found in the text.

What the source does not cover

  • Norway [T].
  • Empirical documentation beyond examples and news reports [T].
  • Detailed analysis of how existing data protection law (GDPR) would or would not cover the concerns [T].

Possible relevance for the thesis

  • Offers a rights-based understanding where privacy is only a ‘gateway’ to something more fundamental; this can be compared with how Norwegian documents justify privacy [T].
  • The focus on inferences rather than personal details (p. 4) may help ask which data the documents see as sensitive [T].
  • The brief explicitly says the public sector uses inferences in automated decisions (p. 5), which connects to the state's use of data [T].

Connections to other sources

  • Fourcade 2017 – thematic: data economy and behavioural prediction [T].
  • Richardson et al. 2019 – thematic: micro-targeting in political campaigns (p. 6) [T].
  • Hankey et al. 2018 – thematic: data and democracy [T].
  • Solove 2011 – thematic contrast: Solove argues privacy is a plurality of problems, Alegre that privacy is a gateway to freedom of thought [T].

Key quotes

“There are three key elements to the right to freedom of thought and all are affected by technology” (p. 2)

“But a more fundamental concern in the data economy is the sharing and sale of inferences and insights about what we are thinking, whether they are accurate or not.” (p. 4)

“Doing so will require new regulatory frameworks that recognize and respond to the wider societal impacts of the use of data rather than focus on control of personal data.” (p. 6)

“Privacy is a gateway right, but the right to think for ourselves is at the heart of what it means to be human.” (p. 8)

Open questions for Rebekka

  • Which version do you cite – the 2021 policy brief or a 2022 publication?
  • Do Norwegian documents discuss inferences and profiling, or mainly collection of personal data?
  • Is freedom of thought a concept that appears in your primary documents at all?
  • Is this a source for theory, for context, or as an example of a rights-based position?