Copyright Sociological Research Online, 2001


David Lyon (2001) 'Surveillance after September 11'
Sociological Research Online, vol. 6, no. 3, <>

To cite articles published in Sociological Research Online, please reference the above information and include paragraph numbers if necessary

Received: 13/11/2001      Accepted: 30/11/2001      Published: 30/11/2001


The aftermath of terrorist attacks on September 11 2001 includes widespread tightening of surveillance. The responses are a prism that puts several things in perspective. One, it is premature to see decentralised and commercial surveillance simply supplanting nation-state power. Rather, the nation-state now draws upon an augmented surveillant assemblage for its own purposes. Two, reliance on high tech surveillance methods is undaunted by the low-tech attacks or the failure of high tech security systems already in place. While they may not work to curb terrorism they are likely to impede civil rights for citizens who will be even more profiled and screened. Three, the struggle to make mushrooming surveillance systems more democratically accountable and amenable to ethical scrutiny is being set back by panic regimes following September 11.

Biometric; CCTV; Civil Rights; Political Control; Security; Surveillance

The September 11 2001 'terrorist'[1] attacks on New York and Washington prompted a series of responses from military retaliation on Afghanistan, the country said to be harbouring Osama Bin Laden, to extensive anti-terrorist legislation aimed at domestic protection. Among the latter, one of the most prominent ongoing reactions is to enhance surveillance operations on a number of fronts and there has been no lack of proposals concerning the best way to achieve this. Public money is being poured into policing and security services, and high- tech companies are falling over themselves to offer not just 'heartfelt condolences' for the attack victims but technical fixes to prevent such attacks from happening again.[2]

Sociologically, this raises many important and urgent questions. With surveillance, as in many other areas, it is frequently suggested that 'everything has changed', an idea that will stir the hairs on the back of any sociologist's neck. This sometimes reduces to a list of new gizmos on the everyday landscape, like iris scanners at airports, closed circuit television (CCTV) cameras on downtown streets and squares, and so on, or it can refer to a 'new era' of political control that overrides previous legal restrictions on monitoring citizens. (Curiously, in this context, commentators in Britain and the USA have each warned against the 'police state' tactics of the other! <> and <http://www.nytimes .com/2001/10/07/magazine/07SURVEILLANCE.html> So, has everything changed, or not? I shall argue that the answer is yes and no. The underlying continuities in surveillance are at least as significant as the altered circumstances following September 11.

Focussing on the aftermath of September 11 is a worthwhile reminder that big events do make a difference in the social world. As Philip Abrams wisely said, an event ' is a portentous outcome; it is a transformation device between past and future; it has eventuated from the past and signifies for the future' (Abrams 1982: 191). To see events -- and what I examine here, their aftermath -- as sociologically important rescues our experiences in time from being merely moments in a meaningless flux. But the event is also, says Abrams, an 'indispensable prism through which social structure and process may be seen' (Abrams 1982: 192).

To take a notorious example, figures such as Hannah Arendt and, perhaps more sociologically, Zygmunt Bauman (1987), have helpfully viewed the Holocaust as revealing not merely the human capacity for evil but also some of the key traits of modernity itself. The triumph of meticulous rational organization is poignantly and perversely seen in the death camp, making this not just an inexplicable aberration from 'modern civilization' but one of its products. The reason that this example springs to mind in the present context is that today's forms and practices of surveillance, too, are products of modernity, and thus carry a similar ambivalence.

Surveillance is not inherently sinister or malign. But the focused attention to persons and populations with a view to influencing, managing or controlling them -- that we call 'surveillance' -- is never innocent either. At times, the need for greater vigilance becomes apparent and in that sense the events of September 11 acted as a 'wake-up' call. Heightened surveillance is not in itself questionable in terms of justice or freedom. Think, for example, of the use of closed circuit television cameras in police interrogation rooms which may foster, not impede, fair treatment of suspects (Newman and Hayman 2001). But under some circumstances, intensified surveillance may have socially negative effects which mean that proscription takes precedence over protection, control over care. The current anxious and tense situation following September 11, I shall argue, is helping to create a potentially dangerous augmentation of surveillance in several countries.

So what aspects of social structure and process may be seen through the prism of surveillance responses to September 11? I suggest that the prism helps to sharpen our focus on two matters in particular: One, the expanding range of already existing range of surveillance processes and practices that circumscribe and help to shape our social existence. Two, the tendency to rely on technological enhancements to surveillance systems (even when it is unclear that they work or that they address the problem they are established to answer). However, concentrating on these two items is intended only to mitigate claims that 'everything has changed' in the surveillance realm, not to suggest that nothing has changed. Indeed, I think it safe to suggest that the intensity and the centralization of surveillance in Western countries is increasing dramatically as a result of September 11. Such systems, once in place, are harder to dismantle than to instal.

The visible signs of putative changes in surveillance have both legal and technical aspects. The USA and several other countries have passed legislation intended to tighten security, to give police and intelligence services greater powers, and to permit faster political responses to 'terrorist' attacks.[3] In order to make it easier find (and to arrest) people suspected of 'terrorism', typically, some limitations on wiretaps have not only been lifted but also extended to the interception of e-mail and to Internet clickstream monitoring. In Canada (where I write) the Communications Security Establishment may now gather intelligence on terrorist groups, probably using 'profiling' methods to track racial and national origins as well as travel movements and financial transactions. Several countries have proposed new national identification card systems, some involving biometric devices or programmable chips; others have brought forward more limited ID card systems, such as the new Canadian Immigration Card ('New ID cards for landed immigrants, Toronto Star, 11 October 2001) or the 'smart ID' for asylum seekers in the UK (The Guardian, 30 October 2001, < hitehall/story/0,,583304,00.html>).

Some have questioned how new, while others have questioned how necessary, are the measures that have been fast-tracked through the legislative process. Sceptics point to the well-established UKUSA intelligence gathering agreement, for example, and to the massive message interception system once known as CARNIVORE, that already filtered millions of ordinary international e-mail, fax, and phone messages long before September 11. Debates have occurred over how long the legal measures will be in force - the USA has a 'sunset clause' that phases out the anti-terrorist law after a period of five years - but few have denied the perceived need for at least some strengthened legal framework to deal with 'terrorist' threats.

In some respects bound up with legal issues, and in others, independently, 'technical' responses to September 11 have also proliferated. High-tech companies, waiting in the wings for the opportunity to launch their products, saw September 11 providing just the platform they needed. Not surprisingly, almost all the 'experts' on whom the media call for comment are representatives of companies. Thus, for instance, Michael G. Cherkasky, president of a security firm, Kroll, suggested that 'every American could be given a "smart card" so, as they go into an airport or anywhere, we know exactly who they are' < /18/national/18RULE.html>. Or in a celebrated case, Larry Ellison, president of the Silicon Valley company Oracle, offered the US government free smart card software for a national ID system <>. What a commercial coup that would be! He failed to explain, of course, what price would be charged for each access to the Oracle database, or the roll-out price-tag on a national smart card identifier.

Other technical surveillance-related responses to September 11 include iris-scans at airports -- now installed at Schipol, Amsterdam, and being implemented elsewhere in Europe and North America as well; CCTV cameras in public places, enhanced if possible with facial recognition capacities such as the Mandrake system in Newham, south London; and DNA databanks to store genetic information capable of identifying known terrorists. Although given their potential for negative social consequences <> <><>) there is a lamentable lack of informed sociological comment on these far-reaching developments, where such analyses are available they suggest several things. One, these technologies may be tried but not tested. That is, it is not clear that they work with the kind of precision that is required and thus they may not achieve the ends intended. Two, they are likely to have unintended consequences that include reinforcing forms of social division and exclusion within the countries where they are established.

A third and larger dimension of the technological aspect of surveillance practices is that seeking superior technologies appears as a primary goal. No matter that the original terrorism involved reliance on relatively aged technologies - jet aircraft of a type that have been around for 30 years, sharp knives, and so on - it is assumed that high- tech solutions are called for. Moreover, the kinds of technologies sought - iris scans, face-recognition, smart cards, biometrics, DNA -- rely heavily on the use of searchable databases, with the aim of anticipating, pre-empting, preventing acts of terrorism by isolating in advance potential perpetrators. I shall return to this in a moment, but here it is merely worth noting that Jacques Ellul's concept of la technique, a relentless cultural commitment to technological progress via ever-augmented means seems (despite his detractors) to be at least relevant.[4]

So, what do these post-September 11 surveillance developments mean, sociologically? Before that date, surveillance studies seemed to be moving away from more conventional concerns with a bureaucratic understanding of power relations (Dandeker 1990) that in fact owes as much to George Orwell as to Max Weber. This puts a fairly high premium on seeing surveillance as a means to centralised power as exemplified in the fictional figure of Big Brother - the trope that still dominates many scholarly as well as popular treatments of the theme. Although some significant studies, especially those located in labour process arguments about workplace monitoring and supervision, see surveillance as a class weapon (Braverman 1980), this view is often supplemented with a more Foucaldian one in which the Panopticon plays a part.

Within the latter there is a variety of views, giving rise to a lively but sporadic debate (Boyne 2000). One fault-line lies between those who focus on the 'unseen observer' in the Panopticon as an antetype of 'invisible' electronic forms of surveillance, but also of relatively unobtrusive CCTV systems, and those that focus more on the classificatory powers of the Panopticon (an idea that is worked out more fully in relation to Foucault's 'biopower'). The latter perspective has been explored empirically in several areas, including high-tech policing and commercial database marketing (Gandy 1993, Ericson and Haggerty 1997). Either way, data-subjects are seen in this perspective to be 'normalized' by surveillance, which is still thought of by many as an identifiable system of power.

While both aspects of the Panopticon offer some illuminating insights into contemporary surveillance, the latter has particular resonance in the present circumstances. In this view, persons and groups are constantly risk-profiled which in the commercial sphere rates their social contributions and sorts them into consumer categories, and in policing and intelligence systems rates their relative social dangerousness. Responses to September 11 have increased possibilities for 'racial' profiling along 'Arab' lines in particular, the consequences of which are already seen in the American detention of several thousand 'suspects' and a FBI trawl of more than 200 campuses to collect information about 'Middle Eastern' students < /12/national/12STUD.html>.

Both the Weberian-Orwellian and the Foucaldian perspectives depend on a fairly centralised understanding of surveillance. However, given the technological capacities for dispersal and decentralization, not to mention globalization, some more recent studies have suggested that a different model is called for. The work of Gilles Deleuze and Felix Guattari (1987) offers some novel directions, suggesting through some striking metaphors that the growth of surveillance systems is rhizomic; more like a creeping plant than a central tree trunk with spreading branches. This has persuaded some to see surveillance as a looser, more malleable and flowing set of processes - a 'surveillant assemblage' - rather than as a centrally controlled and coordinated system (Haggerty and Ericson 2000). This also relates to our daily experiences of surveillance, which occur in mundane moments rather than in special searches. As Nikolas Rose puts it, 'surveillance is "designed in" to the flows of everyday existence' (Rose 1999: 234).

In the assemblage, surveillance works by abstracting bodies from places, splitting them into flows to be reassembled as virtual data-doubles, calling in question once again hierarchies and centralized power. One important aspect of this is that the flows of personal and group data percolate through systems that once were much less porous; much more discrete and watertight. Thus, following September 11, surveillance data from a myriad of sources - supermarkets, motels, traffic control points, credit card transaction records and so on - were used to trace the activities of the 'terrorists' in the days and hours before their attacks. The use of searchable databases makes it possible to use commercial records previously unavailable to police and intelligence services and thus draws on all manner of apparently 'innocent' traces.

This brief survey[5] of surveillance studies shows how the once-dominant model of centralised state informational power has been challenged by socio-technical developments. The result is newer models that incorporate the growth of information and communication technologies in personal and population data processing, and more networked modes of social organization with their concomitant flexibility and departmental openness. But it is a mistake to simply leave the other kinds of explanation behind, as we move up (to the next plateau) using something like Wittgenstein's ladder. To illustrate this, I shall simply offer a series of questions that once again allow the prism of September 11 aftermath to point up aspects of structure and process that relate in particular to surveillance.

Is surveillance best thought of as centralized power or dispersed assemblage? The responses to September 11 are a stark reminder that for all its changing shape since World War Two the nation-state is still a formidable force, especially when the apparently rhizomic shoots can still be exploited for very specific purposes to tap into the data they carry. Though the Big Brother trope did not in its original incarnation refer to anything outside the nation-state (such as commercial or Internet surveillance that is prevalent today) or guess at the extent to which the 'telescreen' would be massively enhanced by developments first in microelectronics and then in communications and searchable databases, it would be naive to imagine that Big Brother type threats are somehow a thing of the past.

Draconian measures are appearing worldwide as country after country instates laws and practices purportedly to counter 'terrorism'. Attorney General Ashcroft warned Patrick Leahy, one of the only American senators seriously to raise his voice against the new police powers, that 'talk won't prevent terrorism' (quoted in < >. Panic responses such as these, that both silence critical discussion and impose restrictions on civil liberties through policing and security crackdowns, are likely to have long-term and possibly irreversible consequences. They permit extraordinary measures, which include appropriating data on everyday communications and transactions - phone calls, e-mail, the Internet - while implicitly discouraging the use of these media for democratic debate. The surveillant assemblage can be coopted for conventional 'strong state' purposes.

With regard to the experience of surveillance it is worth asking, is intrusion or exclusion is the key motif? In societies that have undergone processes of steady privatization it is not surprising that surveillance is often viewed in individualistic terms as a potential threat to privacy, an intrusion on an intimate life, an invasion of the sacrosanct home, or as jeopardising anonymity. While all these are understandable responses (and ones that invite their own theoretical and practical responses), none really touches one of the key aspects of contemporary surveillance; 'social sorting' (Lyon, 2001). It is hard to get an adequate theoretical handle on this, and it does not help that no compelling metaphor - such as 'Big Brother' - has yet been proposed to give it popular cachet.

Yet the increasingly automated discriminatory mechanisms for risk profiling and social categorizing represent a key means of reproducing and reinforcing social, economic, and cultural divisions in informational societies. They tend to be highly unaccountable - especially in contexts such as CCTV surveillance (Norris and Armstrong 1999) - which is why the common promotional refrain, 'if you have nothing to hide, you have nothing to fear' is vacuous. Categorical suspicion[6] has consequences for anyone, 'innocent' or 'guilty', caught in its gaze, a fact that has poignant implications for the new anti-terror measures enacted after September 11. It is already clear in several countries that 'Arab' and 'Muslim' minorities are disproportionately and unfairly targeted by these measures.

The experience of surveillance also raises the question of how far subjects collude with, negotiate, or resist practices that capture and process their personal data? Surveillance is not merely a matter of the gaze of the powerful, any more than it is technologically determined. Data-subjects interact with surveillance systems. As Foucault says, we are 'bearers of our own surveillance' but it must be stressed that this is not merely an unconscious process in which we are dupes. Because surveillance is always ambiguous - there are genuine benefits and plausible rationales as well as palpable disadvantages - the degree of collaboration with surveillance depends on a range of circumstances and attitudes. Under the present panic regime (towards the close of 2001) it appears that anxious publics are willing to put up with many more intrusions, interceptions, delays, and questions than was the case before September 11, and this process is amplified by media polarizations of the 'choice' between 'liberty' and 'security'.[7] The consequences of this complacency could be far-reaching.

I have mentioned technological aspects of surveillance several times, which points up the question, are these developments technologically or socially driven? To read some accounts - both positive and negative -- one would imagine that 'technology' really has the last word in determining surveillance capacities. But this is fact a fine site in which to observe the co-construction of the technical and the social (Lyon, forthcoming). For example, though very powerful searchable databases are in use, and those in intelligence and policing services are being upgraded after September 11, the all-important categories with which they are coded (Lessig 1999) are produced by much more mundane processes. Database marketers in the USA use crude behavioural categories to describe neighbourhoods, such as 'pools and patios' or 'bohemian mix', and CCTV operators in the UK target disproportionately the 'young, black, male' group. The high-tech glitz seems to eclipse by its dazzle those social factors that are constitutionally imbricated with the technical. At times, it also eclipses the fact that surveillance may occur with no added 'technology' at all.

Still on the technical, however, a final question would be, are the proposed new anti-terrorist measures pre-emptive or investigatory? Over the past few years an important debate has centred on the apparent switch in time from past-oriented to future- oriented surveillance. Gary T. Marx predicted in the late 1980s that surveillance would become more pre-emptive and in many respects he has been vindicated. This idea has been picked up in a more Baudrillardian vein by William Bogard who argues that surveillance is increasingly simulated, such that seeing-in-advance is its goal (Bogard 1996). A glance at any promotional platform of high tech surveillance devices confirms that prevention of future occurrences is the supposedly clinching claim. However, this kind of argument easily loses sight of actual data- subjects - persons - whose daily life chances and choices are affected -- often negatively -- in reality by surveillance (Graham 1998) as well as resting on dubious and seldom independent empirical tests.

Unfortunately, the attraction of new technologies that will be able to prevent future 'terrorist' acts is strong in policy circles. It would be nice to believe in this anticipatory and preventive capacity - and as one who was in mid-flight over North America at the time of the attacks I would love to think it true! - but the overwhelming evidence points in the other direction. Surveillance can only anticipate up to a point, and in some very limited circumstances. Searchable databases and international communications interception were fully operational on September 10 to no avail. The likely result will be that internal surveillance of citizens by the state will increase. And if 'terrorists' are apprehended it will be by other means.

Surveillance responses to September 11 are indeed a prism through which aspects of social structure and process may be observed. The prism helps to make visible the already existing vast range of surveillance practices and processes that touch everyday life in so-called informational societies. And it helps to check various easily made assumptions about surveillance - that it is more dispersed than centralised, that it is more intrusive than exclusionary, that data-subjects are dupes of the system, that it is technically-driven, that it contributes more to prevention than to investigation after the fact.

Sociologically, caution seems to be called for in seeing older, modernist models simply as superseded by newer, postmodern ones. For all its apparent weaknesses in a globalizing world, the nation-state is capable of quickly tightening its grip on internal control, using means that include the very items of commercial surveillance -- phone calls, supermarket visits, and Internet surfing -- that appear 'soft' and scarcely worthy of inclusion as 'surveillance'. And for all the doubts cast on the risk-prone informational, communications, and transport environment, naive faith in the promise of technology seems undented by the 'failures' of September 11. Lastly, in the current climate it is hard to see how calls for democratic accountability and ethical scrutiny of surveillance systems will be heard as anything but liberal whining. Yet democratic accountability starts with a willingness to listen to the voice of the other. And ethical scrutiny begins with care for the other, to relieve and to prevent suffering. The sociology of surveillance discussed above sees neglect of these as a serious mistake, with ramifications we may all live to regret.


1Rather than attempt a definition of 'terrorism', I simply leave the word in quotation marks. It is a deeply ambiguous and, because of its tendency to demonize the other, dangerous. A helpful discussion of some aspects of recent 'terrorism' is: < >

2This may be seen on many web sites, e.g. <>

3The USA's PATRIOT Act was first, in October 2001, followed quickly by similar legislation in the UK and Canada (the Anti-terrorism Bill C-36; not yet law at the time of writing). Other countries had second thoughts on legislation as a result of September 11. In Germany, the draft of a new, more liberal immigration law was scrapped at the same time as laws regulating freedom of movement and requiring fingerprints in identity cards were tightened. See <http://www.nytimes. com/2001/10/01/international/europe/01GERM.html>

4Knowledge of Ellul's work is often limited only to the allegedly deterministic The Technological Society (New York: Vintage, 1964). But he saw his sociological work as integrated with his more theological writings that are anything but deterministic. It is misleading to see his most famous work out of the context of the whole corpus.

5A longer survey appears in David Lyon (2001).

6This elegant concept was first used by Gary T. Marx (1988). I discuss its commercial equivalent, 'categorical seduction' in (Lyon 1994).

7I experienced this, anecdotally, when an op-ed piece I wrote under the title 'Whither surveillance after bloody Tuesday?' was published in the newspaper as 'What price in liberty will we pay for security?' The Kingston Whig-Standard, September 28, 2001.


Thanks to those who gave helpful comment on the first draft of this article: Yolande Chan, Roberta Hamilton, Laureen Snider, Elia Zureik. It also benefited from discussion after a version was given as a talk at the University of Toronto, 12 November 2001.


ABRAMS, P. (1982) Historical Sociology, Shepton Mallet UK: Open Books.

BAUMAN, Z. (1987) Modernity and the Holocaust, Oxford and Malden MA: Blackwell.

BOGARD, W. (1996) The Simulation of Surveillance, Cambridge and New York: Cambridge University Press.

BOYNE, R. (2000) 'Post-Panopticism', Economy and Society, Vol. 29, No. 2, pp. 285-307.

BRAVERMAN, H. (1980) Labour and Monopoly Capital, New York: Monthly Review Press.

DANDEKER, C. (1990) Surveillance Power and Modernity, Cambridge: Polity Press.

DELEUZE, G. and GUATARRI, F. (1987) A Thousand Plateaus, Minneapolis: University of Minnesota Press.

ERICSON, R.V. and HAGGERTY, K. (1997) Policing the Risk Society, Toronto: University of Toronto Press.

GANDY, O.H. (1993) The Panoptic Sort: A Political Economy of Personal Information, Boulder CO: Westview.

GRAHAM, S. (1998) 'Spaces of surveillant simulation' Environment and Planning: Society and Space, No. 16.

HAGGERTY, K. and ERICSON, R.V. (2000) 'The surveillant assemblage' British Journal of Sociology, Vol. 51, No. 4, pp. 506-522.

LESSIG, L. (1999) Code, and Other Laws of Cyberspace, New York: Basic Books.

LYON, D. (ed., forthcoming) Surveillance as Social Sorting, London and New York: Routledge.

LYON, D. (forthcoming a) 'Surveillance technology and surveillance society' in Tom Misa, Philip Brey, and Andrews Feenberg (eds.) Modernity and Technology, Cambridge MA: MIT Press.

LYON, D. (2001) Surveillance Society: Monitoring Everyday Life, Buckingham: Open University Press.

LYON, D. (1994) The Electronic Eye: The Rise of Surveillance Society, Cambridge: Polity Press and Malden MA: Blackwell.

MARX, G. (1988) Undercover: Police Surveillance in America, NEWMAN, T. and HAYMAN, S (2001) Policing, Surveillance, and Social Control, Collumpton UK: Willan Publishing.

Berkeley: University of California Press.

NORRIS, C.A. and ARMSTRONG, G. (1999) The Maximum Surveillance Society, London: Berg.

ROSE, N. (1999) Powers of Freedom, Cambridge: Cambridge University Press.

Copyright Sociological Research Online, 2001