EVOPEDIA

← all systems · see it in The System

Surveillance

Surveillance is the systematic observation and monitoring of a person, population, or location, with the purpose of information-gathering, influencing, managing, or directing (src:wp:surveillance). Surveillance studies widen the dictionary sense of 'close observation, especially of a suspected person': much surveillance is applied categorically, to places, networks and categories of person, and the new surveillance is 'the use of technical means to extract or create personal data' (src:marx-2002-new-surveillance). Walked as the whole practice and its relation to the watched: the watchers (states, police and intelligence agencies, companies, health authorities, employers and peers), their channels (eyes and tails, informants, cameras, interception of calls, post and internet traffic, phone location, transaction records and data brokers, platforms' behavioural data, biometrics, disease reports), the files and databases they fill, the identification, analysis and sorting of people, the decisions taken under law and oversight, and what surveillance does: deters, intervenes, targets, notifies; together with the watched's side as the literature describes it: the felt watch, self-policing and chilling, and watching back (sousveillance, countersurveillance, leaks).

Properties published scienceour proposal

Sourced fields, and our own readings.

aliasesSurveillance and tracking (state and commercial)
born atcontested (C-1778): religious and wartime surveillance before electronics; UK state surveillance in public consciousness since the 19th century; CCTV 1942; the new surveillance in the late 20th century
born precisionera
boundaryInside: the watchers (governments, police, intelligence agencies, corporations, data brokers, employers, health authorities, peers); their collection channels (human eyes, tails and stakeouts, informants and infiltrators, cameras, wiretaps and interception of post and internet traffic, phone location, transaction and public records, platforms' behavioural data, biometric capture, disease reports); the recordings, files and databases; identification and linking (facial recognition, record linkage, profiles); analysis and sorting (data mining, social network analysis, keyword triggers, risk sorting); decisions under law and oversight (warrants, courts, committees); the outputs (intervention, deterrence by visible watching, targeting, health notifications); and the watched's side the literature treats as part of surveillance: the feeling of being watched, self-policing, chilling of speech and reading, and watching back (sousveillance, countersurveillance, leaks). Outside, as neighbours: privacy (the withholding side, walked), signals intelligence (sibling: foreign intercept), espionage (covert, illegal to the observed party), digital identity (the credential surveillance reads), law and courts (which restrain it), the internet (carrier), cryptography (the tools of countersurveillance).
nested levelA practice of states, firms and people (rung 11), carried by institutions (police, intelligence agencies, corporations, health services) and by networks and devices (rung 12); restrained by law; running on the traces people give off in daily life.
scalesocial

Measures (26) published science

characteristic clock: lifetime of the Stasi1950-1990 (founded 8 February 1950; files opened after reunification 1989-1991) years
characteristic clock: data retention (GCHQ)up to 2 years
share of cameras in Asia (2016)about 65 %
surveillance cameras worldwide (2016)about 350 million cameras
CCTV cameras in China (2018)over 170 million (400 million more expected in three years) cameras
cameras in the United Kingdom (2011, reliable estimate)around 1.85 million (a 2002 figure of 4.2 million overstated) cameras
Stasi full-time staff and informants (1989)91,015 staff; 173,081 informants in the GDR; 1,553 in West Germany people
Stasi informants identified by 1995about 174,000, almost 2.5% of those aged 18-60 people
Stasi spy density by county (1980s)0.12 to 1.03, mean 0.38 % of population
points needed to single out a person in phone mobility data4 points identify 95% (hourly, antenna resolution; 1.5 million people, 15 months) spatio-temporal points
trust lost per one standard deviation of spy densityabout 4 percentage points, from a 14% average
gender-classifier error rate, darker-skinned women vs lighter-skinned menup to 34.7 vs at most 0.8 % error
uniqueness of mobility traces against resolutionabout the 1/10 power of resolution exponent
hand hygiene compliance, overt vs covert observation78 vs 55 (31,522 opportunities) % compliance
NHS COVID-19 app use and notifications16.5 million users (28%); 1.7 million notifications; 4.2 per index case users / notifications
cases reduced per percentage point of app uptake0.8 (modelling) or 2.3 (statistics) % per percentage point
National Security Letters (2003-2005)more than 140,000 letters
share of honeybee colony offspring from the queen100 (females), 99.9 (males), of up to 50,000 females that can lay eggs %
location data requests to one carrier8 million requests per year
contributions with an image of eyes vs controlnearly three times as much ratio
crime reduction with CCTV (2017 review)24-28 (streets and subways); none shown in parking facilities in that review % less crime
share of time a city CCTV system worked (Philadelphia, 2013)32 % of the time operational
NSA documents copied (estimate)50,000 to 200,000 documents
US fusion centres43 planned, $370 million centres
face scans deleted when Facebook shut face recognition (2021)more than one billion users' face scans
mail opened in one CIA/FBI programmemore than 215,000 communications

Open contentions (7) open contention

Where sources or readings disagree. Both stay on the map.

Cross-scale patterns seen here (22) our proposal

Shapes that recur when the same kind of problem is solved at different scales, and how each shows in Surveillance. Each links to every other system that shows it.

Valence: what it trades, and with whom (16) published science

Each exchange with a neighbour, read from both sides: who gains, who pays, and what carries it.

Partners and parts published science

Built from / builds (10)

Sits inside / holds (3)

expressed by (20)

floor filled by (18)

Kinds (14)

Notes published science

Layers

The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.

surveillance as surveillance studies describe it: people's lives and the traces they give off, up to the managing, influencing and directing of conduct (reader's view)

  1. L1: People, populations and places, and the traces they give off: movements, speech, transactions, bodies
  2. L2: Collection: human eyes and informants, cameras, taps, phone location, records, biometric capture; the new surveillance extends the senses
  3. L3: Recording, transmission and storage: footage, transcripts, location records, files and databases
  4. L4: Identification and linking: biometric matches, mobility traces and records tied to one person; profiles
  5. L5: Analysis and sorting: filtering, data mining, social network analysis, categories of interest; the watched's felt watch
  6. L6: The institution's decision under law and oversight
  7. L7: Influencing, managing and directing conduct; self-policing

the boardroom (shared floors)

  1. L5: Analysis and sorting: automated filtering hands the 'interesting or suspicious' to human analysts; data mining and profiling; social network analysis; sorting people into categories of interest and risk (social sorting)
  2. L5: The watched person's regulation: the feeling of always being watched and self-policing; chilling of reading and speech; behaviour changed under overt observation (the Hawthorne effect)
  3. L6: The watching institution's decision under law and oversight: agencies decide to act; laws, warrants, courts and committees restrain them; leaks and public debate feed back
  4. L7: What surveillance is for: influencing, managing and directing conduct (security, order, profit, health), and, on the watched side, conduct that polices itself

way in: Being watched (the watched person's channel)

  1. L1: A visible camera, a guard, an image of eyes
  2. L2: Eyes and gaze picked up; even an image of eyes counts
  3. L3: Skin conductance up and heart slowed to direct gaze, only when believed watched
  4. L4: The experience of being watched (port)

way in: Bodies as identifiers (face, fingerprint, DNA, gait, voice)

  1. L1: Fingerprints, DNA, faces, gait and voice
  2. L2: Scanners and cameras; face recognition needs the least cooperation
  3. L3: Biometric databases (an FBI database of $1 billion)
  4. L4: A match, with errors up to 34.7% for darker-skinned women in gender classifiers (port: the match)

way in: Cameras (CCTV)

  1. L1: An area, its people and vehicles, in view
  2. L2: Surveillance cameras: about 350 million worldwide in 2016
  3. L3: Recording device or IP network, watched by a guard or officer
  4. L4: Automated software that makes footage a searchable database; facial recognition (port: the video analytics)

way in: Disease reports and contact records

  1. L1: Disease and conditions in a territory
  2. L2: Facilities' reports; app proximity records between users
  3. L3: Continuous, systematic collection of health data
  4. L4: A positive case linked to its recent contacts (port: the index case)

way in: Informants and infiltrators

  1. L1: What members of a group say and do among themselves
  2. L2: Infiltrators and members pressed to inform; in the GDR 173,081 unofficial informants in 1989
  3. L3: Reports to officers; tens of thousands of unofficial collaborators
  4. L4: Personal files on millions of citizens (port: the file)

way in: Interception of calls, post and internet traffic

  1. L1: Phone calls, VoIP, email, web traffic and post
  2. L2: Taps built into carriers by law; mail opened (215,000 communications in one programme)
  3. L3: Speech turned to machine text and stored for automated analysis
  4. L4: Traffic flagged by trigger words, sites and contacts (port: the selector)

way in: Law, courts and oversight bodies

  1. L1: Constitutional limits on search
  2. L2: Courts hear challenges (Riley 2014, Carpenter 2018)
  3. L3: Rulings: historical cell-site records need a warrant
  4. L4: Agencies' procedure changed (port: the warrant requirement)
  5. L5: Legislative inquiry weighs the programmes (Church Committee)

way in: Phone location

  1. L1: A phone's signals to nearby towers, even when not in use
  2. L2: Towers, or an IMSI catcher posing as one
  3. L3: Carriers' location records, requested 8 million times in a year from one carrier
  4. L4: A mobility trace tied to one person: four points single out 95% (port: the trace)

way in: Platforms' behavioural data and public posts

  1. L1: What people post, like, search and whom they link to
  2. L2: Platforms log behaviour; personal data used to target consumers
  3. L3: Massive data sets analysed to reduce uncertainty
  4. L4: Maps of social networks (port: the network map)

way in: The watched watching back (sousveillance, leaks, journalism)

  1. L1: Police and surveillance workers at work
  2. L2: Citizens' cameras and wearable devices; insiders
  3. L3: Recordings and documents (50,000 to 200,000 NSA documents copied)
  4. L4: Journalists receive them (port: the press)
  5. L5: Public debate about civil liberties

way in: Transaction and public records, bought or compelled

  1. L1: Bank machine, card, phone, library and rental transactions
  2. L2: Electronic records, cheap to collect and store
  3. L3: Gathered from third parties by request, subpoena or purchase; pooled in fusion centres
  4. L4: A detailed profile of one person (port: the profile)

way in: Watchers' eyes (tails, stakeouts, guards)

  1. L1: The watched person, their movements and contacts, in view
  2. L2: A tail or a stakeout team watching covertly
  3. L3: Report on the movements and contacts
  4. L4: Evidence on a person of interest (port: the case file)

way out: Countersurveillance and watching back (the watched's outlet)

  1. L1: Intercepted content arrives as 'gibberish'; police recorded from below
  2. L3: Encrypted messengers, privacy coins; phones built to resist taps
  3. L5: Decide to avoid or hinder the watch

way out: Covert disruption (Zersetzung) and removing network nodes

  1. L1: The person withdraws from activity
  2. L3: Psychological harassment to 'switch off' perceived enemies
  3. L6: Choose targets: 'hostile-negative' persons or important nodes

way out: Health notifications and isolation

  1. L1: Contacts isolate; about one case averted per consenting case
  2. L4: Exposure notifications sent (about 1.7 million)
  3. L7: Plan and evaluate public health action

way out: Intervening on people (arrest, stop, remove)

  1. L1: The person stopped, arrested or removed
  2. L3: Police directed to the person; arrests (19 with warrants found at Super Bowl XXXV; about 250,000 arrests with informants' help in the GDR)
  3. L6: Decide to act on a person flagged

way out: Self-censorship (the watched's outlet)

  1. L1: Fewer views of sensitive articles; dissent unspoken
  2. L3: Reading and speech withheld
  3. L5: Self-policing under the felt watch

way out: Targeting (ads, prediction products, content)

  1. L1: The targeted ad or offer on a person's screen
  2. L4: Personal data used to target consumers more precisely
  3. L7: Sell predictions of behaviour

way out: Watching made visible (cameras, signs, eyes)

  1. L1: The camera or eyes seen by the watched
  2. L2: Visible cameras and resources
  3. L7: Show the watch to deter and reassure

Sources (25) published science

Privacy (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01)
Surveillance (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01)
Panopticon (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Daymude J.J., Axelrod R., Forrest S. (2025) Strategic analysis of dissent and self-censorship. PNAS 122(45). doi:10.1073/pnas.2508028122 (behind its links)
Surveillance capitalism (Wikipedia, via Wikimedia Enterprise) (behind its links)
Wymant C. et al. (2021) The epidemiological impact of the NHS COVID-19 app. Nature 594:408-412. doi:10.1038/s41586-021-03606-z (behind its links)
Carpenter v. United States (Wikipedia, via Wikimedia Enterprise) (behind its links)
Mass surveillance (Wikipedia, via Wikimedia Enterprise) (behind its links)
Stasi (Wikipedia, via Wikimedia Enterprise) (behind its links)
Lichter A., Löffler M., Siegloch S. (2015) The Economic Costs of Mass Surveillance: Insights from Stasi Spying in East Germany. IZA Discussion Paper 9245 (behind its links)
de Montjoye, Hidalgo, Verleysen and Blondel 2013, Unique in the Crowd: the privacy bounds of human mobility, Sci Rep 3:1376, doi:10.1038/srep01376 (PMC3607247) (behind its links)
Buolamwini J., Gebru T. (2018) Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification. PMLR 81:77-91 (behind its links)
Wu K.-S. et al. (2018) Identifying heterogeneity in the Hawthorne effect on hand hygiene observation: a cohort study of overtly and covertly observed results. BMC Infect Dis 18:369. doi:10.1186/s12879-018-3292-5 (behind its links)
Ratnieks F.L.W., Helanterä H. (2009) The evolution of extreme altruism and inequality in insect societies. Phil Trans R Soc B 364:3169-3179. doi:10.1098/rstb.2009.0129 (behind its links)
Bateson M., Nettle D., Roberts G. (2006) Cues of being watched enhance cooperation in a real-world setting. Biol Lett 2:412-414. doi:10.1098/rsbl.2006.0509 (behind its links)
Closed-circuit television (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Edward Snowden (Wikipedia, via Wikimedia Enterprise) (behind its links)
Facial recognition system (Wikipedia, via Wikimedia Enterprise) (behind its links)
Marx G.T. (2002) What's New About the 'New Surveillance'? Classifying for Change and Continuity. Surveillance & Society 1(1). doi:10.24908/ss.v1i1.3391 (behind its links)
Chen T., Helminen T.M., Linnunsalo S., Hietanen J.K. (2024) Autonomic and facial electromyographic responses to watching eyes. i-Perception 15. doi:10.1177/20416695231226059 (behind its links)
Verhelst HM, Stannat AW, Mecacci G (2020) Machine Learning Against Terrorism: How Big Data Collection and Analysis Influences the Privacy-Security Dilemma. Sci Eng Ethics, doi:10.1007/s11948-020-00254-w (behind its links)
Public health surveillance (Wikipedia, via Wikimedia Enterprise) (behind its links)
Mann S., Nolan J., Wellman B. (2003) Sousveillance: Inventing and Using Wearable Computing Devices for Data Collection in Surveillance Environments. Surveillance & Society 1(3). doi:10.24908/ss.v1i3.3344 (behind its links)
Chilling effect (Wikipedia, via Wikimedia Enterprise) (behind its links)
Andrejevic M. (2004) The Work of Watching One Another: Lateral Surveillance, Risk, and Governance. Surveillance & Society 2(4). doi:10.24908/ss.v2i4.3359 (behind its links)