Posted by

– Your face is like an open book

Abstract – The concept of ‘Cyberphrenology’ reflects a fundamental approach to datafication of human-computer interaction, which is harvested from CCTV surveillance, facial recognition software and online social configurations. The technology makes it possible to predict,  influence, and modify future behaviour of individuals.

In a benevolent perception facial recognition technology constitutes a practical tool for law enforcement priorities. effective policing and crime prevention efforts. However, in a belligerent application facial recognition becomes the basic premise for digitized mass surveillance that affect established methods of proof,  change the content of legal norms, and challenge civil liberties.

The chilling effect of the cyberphrenological paradigm increasingly penetrate both commercial and governmental data collection. Facial recognition exposed to machine learning is the wet dream of any Machiavellian prince. Facial recognition becomes the ultimate weapon of mass seduction.

KeywordsFacial recognition; phrenology; Lombroso; behavioural hygiene; platform economy; surveillance; regulatory gaps.


In 2016 two researchers at Shanghai Jiao Tong University, Xiaolin Wu and Xi Zhang, developed an AI based on photographic facial recognition.

Their claim is that with their algorithm they can teach a machine to read images of human faces and determine whether the depicted person has a criminal character. Their research demonstrates that their AI with 89.5% accuracy can determine whether a person is a criminal or has a criminal character.

Phrenological Chart of Faculties

The Shanghai study proposes that a person – with tangible mathematical certainty (89.5%) – can be identified as a criminal, even before he or she is convicted of a criminal offense.

This claim suggests that facial recognition technology can do much more than recognising faces. Facial recognition opens up the possibility of a worrying dimension in the implementation of ‘crime prevention’ and policing. It becomes possible to determine human character and predict future behaviour purely from analysis of digital facial recognition. Predictions make it possible to introduce the concept of preventative ‘behavioural hygiene’.


   The German-Austrian doctor, Franz Joseph Gall (1758-1828) is considered the founder of phrenology. Based on the surface of a person’s skull, Gall could make assumptions about that person’s fundamental faculties and therefore their character.

Cesare Lombroso (1835-1909)

Based on the principles of phrenology, the “Father of Criminology”, Cesare Lombroso (1835-1909), claimed that criminal behaviour was hereditary and innate.

‘The Born Criminal’ could be identified through studies and measurements of physical traits and Lombroso carried out a number of anthropometric studies focusing on measuring human skulls. Lombroso’s anthropometric profiling of criminals underpins the pseudoscientific claim that it is possible, based on human physics, to identify persons with criminal inclinations. Lombroso’s allegations inspired the Eugenics movement, theories of racial hygiene and behavioural genetic practices such as forced sterilization – a practice that was enforced in Denmark until 1967.

Despite the discriminatory and racist starting point, the study from Shanghai signals a resurrection of phrenology, criminological anthropometry and behavioural genetics in a digital disguise. Facial recognition technology revives cremated physiognomic and phrenological claims and restores the theoretical anthropometric genome in a new disguise: Cyberphrenology.


    Cyberphrenology digitizes the claim that it is possible to measure, record and explain a person’s character and personal characteristics based on an automated algorithmic computation of biometrics and person-related data that shape the individual’s digital fingerprint.

    The emergence of cyberphrenology can be attributed to the global spread of the internet and the birth of an era of Big Data, which considers data as objective, pure, innocent, and conceived in a divine intervention where data and facts are converted into ultimate universal truths.

    Cyberphrenology is the fundamental element of concealed datafication processes that create stereotypes, or cognitive schemes, where simplified and uncompromising digital representations are transferred to physical reality in the form of indisputable facts.

    The politicization, securitization, and commercialization of data is difficult to understand or theorize. We have not fully mapped the social consequences of digitization. Technological developments occur too fast. The technology is non-transparent. Practice is veiled in secrecy and anonymity. Data creates invasive profit accumulation through ‘surveillance capitalism’. Psychosocial and behavioural psychological influences are insidious and difficult to detect or deter over time. Digitization creates cognitive bias and shifts the individual’s judgement in making rational choices. Everybody become victim of cunning rhetorical misinformation and deliberate political, cultural, or commercial deception.

Cyberphrenology comes with a stern health and safety warning.

    From a criminological perspective, the concept of ‘cyberphrenology’ is product of a hermeneutic philosophical-archaeological method. The method starts in phrenology, but reveals a chronological conceptual debt that spans criminological anthropometry, racial hygiene, behavioural genetics, biometrics to socio-technology, where cyberphrenology is the latest off-spring.

    Thus, cyberphrenology becomes the common conceptual denominator for phrenology and criminological anthropometry in a digital disguise. Cyberphrenology recognizes the fundamental anthropometric quantitative premise of phrenology.

    Cyberphrenology accepts this historical conceptual heritage and argues that it is possible to determine a person’s mental disposition, temperament, social status, economy, sexuality, political or religious standpoint from an automated digital calculation of Big Data, biometric data and individual digital behavioural patterns on social media.

    All these concepts include serious latent warnings, if they are put into practice. Cyberphrenology comes with a stern health and safety warning. The concept reflects a predatory approach to datafication of human-computer interaction: Numbers and data harvested from facial recognition are processed to predict, influence, and modify future behaviour.

Photo by Thomas Windisch on


Any digital human-computer interaction produces data. Over time, more and more data is generated about the individual, but personal data is controlled by others.

    In the Wild West Web, individuals are not in control of their own digital data. In totalitarian regimes, the state basically takes centralized ownership of all data. In liberal democracies, actual ownership of data is less obvious. While public entities gather and process data is a commodity that is owned by public-private platforms. Data is harvested, processed and traded, mostly to 3rd parties,

    Data analysis can identify effective psychosocial or behavioural psychological stimuli that lead to predictions, which can be translated into political influences or profits. However, it is not possible to strictly separate ‘state surveillance’ – undertaken for the purpose of law enforcement or delivery of security –  and ‘commercial surveillance’(undertaken for profit reasons), as both types of monitoring tend to merge and collaborate,

Digital behaviour of individuals is an inexhaustible source of free raw material that provides a new type of industrial manufacturing process, where success is defined by industry’s ability to predict human behaviour in the future.

Shosana Zuboff, 2016

    The cyberphrenological paradigm of data analysis is basically a formidable tool for modifying human behaviour. It is used in both totalitarian regimes and in liberal democracies.


    Profiling individuals on the basis of their digital fingerprint is the fundamental element of the business model in platform economy. Profiling involves collection and processing of data on behaviour harvested from digital platforms. Profiling generates a number of cognitive bias, allowing a population to be divided into different categories used to create various ranking systems. Processing data on behaviour can be judged with an aim to reward the good or punish the bad. However, the social repercussions of profiling, data analysis, and exponential digital progress are subjected to both belligerent and benevolent interpretations.

Chinese police with the new advanced sunglasses that have built-in facial recognition.
Foto: – / Ritzau Scanpix

    In a benevolent perception, profiling makes our life easier. Profiling is a way to save costs, increase efficiency and maintain competitive advantage. The use of data as a resource is interpreted as a common good. We readily and willingly let profiling platforms govern our lives.

    In a belligerent perspective, profiling could involve pragmatic manifestations of prejudice or racist arguments to initiate systematic behavioural modification, or suppression of religious or political opinions, abnormal behaviours, and segregation of unwanted individuals – or entire communities. In this interpretation, profiling forms the basis for social exclusion of individuals, groups or specific behavioural patterns.

    From a cyberphrenological perspective profiling can be used to introduce digital behavioural hygiene through modification of behaviour using tools ranging from behavioural design, manipulation, isolation over exclusion to ‘elimination’.

In totalitarian regimes, the classical tools for modifying behaviour are often very direct and hard-handed. In liberal democracies, however, the tools are often much more difficult to distinguish.

    Behavioural modification can be difficult to detect and even harder to deter. The paradox is that both agents of the state, corporate actors and criminals adopt similar methods and instrumentation for the processing, exploitation and trade of data.

    Profiling is used in various forms of behavioural design. The result is a form of ‘social engineering’, ‘tactical behavioural hygiene’ or ‘Weapons of Mass Persuasion’ that can affect many, but also has a great potential for targeted precision towards specific groups or individuals.

    Profiling can improve delivery of personalized content – or exclude you from access to content.

Numbers are the basic building blocks of data and digital technology. Numbers don’t lie. They can’t. Lying is a human behaviour. On the other hand, numbers do not always tell the whole truth. Yet many decisions are based on blind faith in data and digital technology.

    Applied in a setting of criminological profiling, the cyberphrenological paradigm opens up opportunities for identifying individuals who, on the basis of an analysis of data on  social background, family relationships, geographical upbringing, interests, habits and values, could be expected to perform crime in the future.

    The predictive character of profiling represents a serious concern, if employed unregulated. We have come full circle: Cyberphrenology revives the idea of ‘The born criminal’ in a digital disguise.

    There’s a great chance that the concept of behavioural hygiene and modification will shape your life in the future. Completely without you noticing it. Whatever you say or do can make your life easier – or be used against you years from now. The question is: What will you do about it?

    Here’s one suggestion: We need to create new forms of governance that focus on governing people through platforms.


    Platforms are sets of rules that enable instantaneous social interaction. The business model is pivotal in a new digital platform economy that aims towards monopolistic positions.  We must learn to govern an economy that is made up of platforms. 

    Platforms compute user behaviour, collect and process data, capitalize on datafication and distribute feedback loops.

Photo by Pixabay on

    From one perspective, platforms are sets of rules and practices to collect data, which in turn helps to create even better rules and practices for data collection. Rules help in the production of physical goods, services, or value in the form of hyper connected social feedback – loops.

    Any platform is a collection of deals where users barter ownership of data, how it is processed, and whether data is accessible to a broader audience. All deals can be either fair or unfair. It is of crucial importance that future regulatory architecture govern exponential digital progress and ensures platforms provide everyone with a fair deal.

    Platforms accumulate not only data, assets and capital, but also rights. This unilateral redistribution of rights sustains a privately administered compliance regime of rewards and penalties. The current complex architecture of self-regulation in platforms provides a level of murky non-transparency that leaves regulatory regimes beyond the comprehension of the average user, remaining largely free from detection or sanction.

    In this interpretation platforms operate without meaningful mechanisms of traditional form of consent – often without  adherence to conventional perceptions of democratic oversight as expressed in law and regulation.

Photo by PhotoMIX Ltd. on

    Users happily sign-off proprietary rights to their own data to obtain an instant reward in the present that could shape their lives in the future. The redistribution of ownership to data, copyrights to content, privacy and freedom of speech become pivotal for discussions on governance of ethics and regulation of rights in platform economies. However, deregulation has become a permanent fixture on government agendas in recent years. This development is opening a regulatory gap.


The common  argument underpinning governmental deregulation supports the idea that increased freedom from governmental influence gives companies and individuals more opportunities for self-actualization, which is believed to lead towards a more prosperous society. This forms a paradoxical situation, where governments are turning away from regulation, at the same time as inherently monopolistic platform corporations focus on creating their own regulatory systems, which users and non-users agree to adopt. This creates a gap in governance: people are willingly accepting to be governed to gain access, but platforms set their regulatory standards with their own business in mind.

    Platforms have become key governance structures in their own right. Business models and praxis in the current platform economy appear to ignore social needs or evade social responsibilities.

    Influential platforms, such as Google, Twitter. and Facebook, have realised that they can’t behave as they like detached from the context in which they thrive. Platforms police themselves and sanction unacceptable behaviour in their own realm. However, self-regulation is still clouded in secrecy.


    In sum, the concept of cyberphrenology serves as a stern warning.  The scourge of predictions on human behaviour through facial recognition exemplifies belligerent latent attributes in digital technology.

    To bridge the regulatory gap, public-private collaboration must create regulatory frameworks that are designed to work through the self-regulatory mechanisms used by platforms.

The regulatory architecture of tomorrow must address redistribution of ownership of data to ensure everybody get a fair deal.

    Future regulatory architectures must determine the way public and private actors capitalize data as a resource – both in terms of influence and rights to profit.

    Tomorrow, platforms could pay basic income to its users, who actually upload content or agree to transfer ownership of personal data to the platform.

    Platform economy needs to strike a balance between regulation of datafication and  promoting innovation or disruption in exponential digital progress.

    To avoid belligerent social repercussions,  future governance is required to regulate algorithmic analysis in a way that determine accepted boundaries in the use of autonomous automation for profiling and predictions on human behaviour.


Cybercriminology – Introduction



Deindividuation in Cybergaps


Wu, X. and Zhang, X. (2016) Automated Inference on Criminality using Face Images. Shanghai: Shanghai Jiao Tong University

Danish Research Leads Facial Analysis Research, Technical University of Denmark

Olivia Solon (2019) Facial recognition’s ‘dirty little secret’: Millions of online photos scraped without consent, NBC News, 12 March 2019

Zuboff, S. (2016) The Secrets of Surveillance Capitalism, Frankfurter Algemeine,

Skriv et svar

Please log in using one of these methods to post your comment: Logo

Du kommenterer med din konto. Log Out /  Skift )

Google photo

Du kommenterer med din Google konto. Log Out /  Skift )

Twitter picture

Du kommenterer med din Twitter konto. Log Out /  Skift )

Facebook photo

Du kommenterer med din Facebook konto. Log Out /  Skift )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.