Surveillance & Society
CCTV Special (eds. Norris, McCahill and Wood)
2(2/3): 216-229
http://www.surveillance-and-society.org/cctv.htm
© 2004
Surveillance & Society and the author(s). All rights reserved. ISSN: 1477-7487
The Legal Regulation of CCTV in Europe
Marianne L. Gras
1
Abstract
This paper explores the recent history of CCTV system regulation in England and Wales questioning
whether recent additions to the law can be regarded as providing for effective regulation, in particular, of
camera numbers. It goes on to explore the legal landscape relating to public and private use of CCTV to
subject publicly accessible space to surveillance in Germany as well as giving an overview of the regulatory
systems in France, the Netherlands and Sweden. Drawing from this analysis, minimum standards for effective
regulation are explored in terms of fulfilling both the letter and the spirit of laws across Europe
.
CCTV and Legal Regulation in Britain
What is the legal regulation of CCTV? Simply, a body of legal norms to regulate surveillance
cameras and their use to observe individuals in a public space. A few years ago, in the mid
1990s when CCTV systems were starting to be installed in town centres across England and
Wales any English (or Welsh) lawyer could more or less happily have said regulation isn’t
necessary according to English law. He or she would have pointed out that there is no right to
privacy (at least not independent of property rights) in this country, so there is little a person
unhappy about being subjected to surveillance can do. The lawyer might have pointed out that
where local authorities spend money on CCTV systems, they have a legislative basis for doing
so (para. 163 of the Criminal Justice and Public Order Act 1994, London Local Authorities
(No. 2) Act 1990) and furthermore, that a possibility for real control, namely the need for
planning permission, (Town and Country Planning (General Permitted Development) Order
1995 Part 33, SI 1995 No. 418) had been deliberately abolished in relation to camera
installation. Those more concerned with privacy, may have pointed to article 8 of the European
Convention on Human Rights which gives even British subjects a right to privacy. A certain
amount of jurisprudence from the Human Rights Court could also have been cited to support this
line of argument. One could, however, also point to other judgements of the same court
underlining the difficulty of satisfactorily proving a measure (already installed) as unnecessary in a
democratic society. The more cynically inclined could point to reactive British legislation born of
1
Abteilung Kriminologie der Juristischen Fakultät der Georg-August-Universität Göttingen, Germany.
mailto:mgras@uni-goettingen.de
Gras: The Legal Regulation of CCTV in Europe
Surveillance & Society
2(2/3)
217
(at least expected) condemnation from the Strasbourg court, making the actual situation worse in
the name of fulfilling the letter of the European Convention on Human Rights. One could,
however, be certain of one matter, no British court could ever demand legal regulation of CCTV
based on the legal situation at the time.
That the tenor of debate has changed, since the 1998 Human Rights Act, there is no doubt. Even
British subjects (or are we now citizens?) have a right to privacy, also in public places. With
reports of over 4 million cameras in use in Britain (McCahill and Norris, 2004), European
lawyers seem uneasy. How can so many cameras be regarded as satisfying a test of
proportionality. The Home Office, however, pronounces itself ”confident” that British CCTV
practice is in line with the Human Rights Act (Henderson 2001). The recent hearing of Peck v.
United Kingdom in Strasbourg, in spite of the complaint for treatment which can only be
described as an ultimate breach of privacy (at least in continental European terms) being upheld,
seems to confirm this view. The Court did not appear overly keen to criticise the prolific use of
CCTV in Britain. Indeed there is reason to believe that the mere use of CCTV surveillance will
not suffice to invoke article 8 (see e.g.: Taylor, 2002: 76)
We have also seen the 1995 European Data Protection Directive put into practice by the 1998
Data Protection Act. By 2003 all CCTV systems controllers were required to register with the
Information Commissioner and to ensure they are operating in line with data protection
principles. These were made explicitly applicable to all CCTV systems and specifically defined in
a Code of Practice for CCTV. Although the legal status of these rules isn’t entirely clear, anyone
operating a system would be unwise not to adhere to them. Breach of the law, as defined by
these guidelines, is a crime and the principles laid down sound rather good: open, fair,
proportional. Surely that’s regulation of CCTV and quite good too?
Well it is. And yes, one could describe British systems which work by the rules as well
regulated. The fact that the regulation came after a great deal of the systems had been installed is,
to European minds, a bit peculiar, but then history works in strange ways and it’s the end result
that counts. To be fair, the end result cannot yet be seen and there are signs that the changed
legal situation is making a difference. In June 2003 a High Court ruled that ”name and shame”
campaigns using CCTV recordings could be illegal. The law contains potential to alter the (mal-)
practices which appeared to have been frequent in the past.
The law also contains requirements such as proportionality which, as we will see, could be a key
to setting quite different standards and ultimately requiring that systems currently in place in
Britain might have to be dismantled. The law has this potential, if it is interpreted and enforced in
this way. And here lies the crux. The essence of the threat many regard as emanating from
CCTV is not captured well by legislation concerning data protection. A growing net of cameras
with the potential to surveil an individual day in, day out may in fact do no more than give that
individual the perception of being under surveillance. That is what CCTV systems and even
dummy cameras intend to do in order to deter that individual from committing a crime.
Regulation dealing with factual gathering of personal data will not be invoked in the vast majority
of cases. This is particularly true in light of the recent Court of Appeal decision in Durant v. FSA
([2003] EWCA Civ 1746) limiting the definition of ‘personal data’.
Gras: The Legal Regulation of CCTV in Europe
Surveillance & Society
2(2/3)
218
Even beyond legal argument and assuming CCTV systems will remain fully subject to data
protection legislation, for very practical reasons, like the actual resources available to enforce
these rules, serious doubts remain that Britain, has
effective regulation. Regulatory bodies in the
UK notoriously lack resources and are reliant upon the good-will and co-operation of those they
are regulating and their activities are frequently triggered by a specific complaint (see e.g.:
Sanders and Young, 2000: 365). These issues become all the more relevant in the data
protection field where breaches of regulation will not leave any physical trace.
Let us however assume that CCTV systems in the UK work according to data protection
regulation and are within the ambit of what is regarded as necessary in a democratic society as
defined by the European Convention. In that case, the legal situation in England and Wales could
be said to regulate CCTV. It is, however, no more than regulation of work with CCTV. It is by
no means regulation of the main issue surrounding CCTV. Namely legal control of whether a
camera may be installed or not. How could it be? The cameras were there first. We cannot
conclusively say that it never will effectively regulate CCTV, although the idea of installed
systems being dismantled due to these regulations doesn’t fit easily with the development in
England and Wales so far. Any ruling coming after the fact – unless it is effectively forbidding it –
must surely lead to far-reaching changes before it can be claimed to be doing anything more than
just clearing up the mess and damming in the damage.
The regulation of CCTV in Britain can at most be regarded as potentially interesting for others in
terms of good practice. But as far as the most basic issue of regulating CCTV is concerned,
namely regulating the number of cameras installed, English law cannot be said to be of any
interest to another country.
As the above analysis has demonstrated, in Britain we have not been able to rely on European
privacy law to regulate CCTV use. Although the issue is controversial, it is difficult to
conclusively declare the British situation as compatible with or in breach of European legislation.
Thus the search for European regulation becomes a question of whether other European
countries have regulated CCTV at a national level. As Britain is still regarded as world-wide
leader as far as camera surveillance is concerned, it would appear that Britain is either quicker
than the others or that they have more effective regulation.
In terms of avoiding high camera numbers, Denmark certainly has stronger regulation than
Britain. CCTV surveillance is generally forbidden there. There are exceptions for owners of
certain kinds of property, such as petrol stations. They are subject to a strong requirement to
inform those subjected to surveillance about it. Public authorities and the police are permitted to
use CCTV. The police can also do so covertly. These exceptions may appear far reaching, but if
one considers the proportion of British cameras which were installed and are owned by nonpublic
bodies, but which make up such a significant part of the surveillance net, it is not difficult
to imagine how different the situation might be, if a general ban on such systems had been in
place during the 1990s. Denmark’s position is however unique.
Gras: The Legal Regulation of CCTV in Europe
Surveillance & Society
2(2/3)
219
Before I can begin an more in-depth analysis of other European legal systems, some closer
definition of the subject matter would appear necessary. As has frequently been pointed out,
CCTV surveillance is a diverse phenomenon (see e.g.: Norris and Armstrong, 1999: 55). For
this paper’s purpose, it is sufficient to define it as camera surveillance of publicly accessible
space. Publicly accessible space can be a street or square, but also includes what is factually
private property, open to the public, such as banks, petrol stations, etc. A CCTV system can be
run by a public authority, like the police, or by a private owner or body. As we will see, the
regulation of CCTV within Europe usually differentiates between private and public, in particular
police, CCTV.
Germany
Germany is a case in point. The situation there is complicated by the country’s federal structure.
As far as video surveillance is concerned, it is sufficient to say that the installation of cameras to
surveil an area permanently, in so far as this is done by a public authority, is governed by state or
‘Länder’ police law. This means that each of the 16 German Länder can have differing situations.
In fact, there are not 16 different solutions; about 3 can be easily identified: a) Länder with no
legal basis for CCTV surveillance as British people know it, b) states allowing video surveillance
by the police in certain, relatively well-defined situations and c) states with a wider ranging legal
permit for police and occasionally other public authorities, to install CCTV systems, to make
recordings and to store these for a relatively long period of time (Weinbrenner, 2001).
CCTV can of course not only be installed by public bodies (and for the most part in Germany,
discussion about allowing public instalment of CCTV focuses almost exclusively on the police)
but also by private persons or entities. Where private property is open to the public, this too falls
under the law. As in other continental legal systems, however, a private person wishing to use
cameras to watch over his or her own property is regarded as entirely different to the police
wishing to observe a public road or square.
Surveillance by Private Persons or Entities
A person may install and use CCTV on the basis of his or her rights as the owner of property
(Hausrecht). This right is, however, not unlimited, as his or her actions touch upon the human
rights of those affected, namely their right to develop their personality freely (Recht auf freie
Entfaltung der Persönlichkeit). This right is protected by article 2 I of the German constitution in
connection with article 1 I. This general right to one’s own personality includes the right to one’s
own picture (BVerfGE 34, 238, 246, 248). The legality of an interference with this right, is
judged according to the ‘Sphären’ or ‘Drei-Stufen-Theorie’ (sphere or three steps theory) of the
Federal Constitutional Court. According to this theory, every person has three levels of privacy.
She or he has a right to privacy, but as a social being is also required to accept a certain
limitation of this right. Centrally, there is the intimate sphere (Intimsphäre) or the core area
(Kernbereich) of his or her personality, which receives absolute protection. Beyond that, there is
the basic private sphere (die schlichte Privatsphere) and the individual sphere (Individualsphäre).
These can be interfered with on the condition that Verhältnismaßigkeit - proportionality - is
preserved in doing so. This has particular relevance in relation to CCTV, because any recording
Gras: The Legal Regulation of CCTV in Europe
Surveillance & Society
2(2/3)
220
made can be used before a court, which is regarded as state (i.e. a higher level of) interference.
The property owner’s or (where recordings are to be used before court) general public’s interest
is weighed against the individual’s right to privacy and the former must outweigh the individual’s
in order for surveillance to be permitted, i.e. the general public’s interest in prosecuting crime
must be weightier in order for recordings to be admissible as evidence (BVerfGE 34, 238, 247).
Interference by CCTV will often be considered as comparatively light as it only takes place for
as long as one is within a certain space. Surveillance is even more likely to be regarded as
proportionate, if one gives persons who become the objects of surveillance a choice whether or
not to subject themselves to it by informing them of this surveillance. In this case, the arguments
against CCTV are regarded as fairly weak because there is implicit consent to CCTV
surveillance. The fact that a person who, for example, needs to travel may have to enter a train
station with CCTV to do so and, therefore effectively has no choice, has yet to be addressed. It
is likely to become more controversial as camera numbers increase. This situation cannot really
said to be one of satisfactory regulation and is the subject of frequent criticism in particular by the
Data Protection Commissioners of the federal states and at national level. The new Federal Data
Protection Act does nothing to improve the situation. As the number of cameras grows,
however, the pressure for more regulation is growing too.
On a positive note, one should mention that the courts do perform a real balancing evaluation.
Where people are being filmed regularly near their homes or where the court has deemed milder
measures more suitable, an order will be made to dismantle a camera. In the most famous case
the Federal Court ordered one party in a conflict between neighbours to remove the camera he
had installed to document high noise levels being caused by his neighbours. The court surmised
that those affected were subjected to a feeling of being watched permanently and that the
interests of the other party could not justify this, particularly as other less intrusive methods,
could achieve any documentation deemed necessary (see BGH-Urteil NJW 1995, 1955).
Furthermore, although there is a central weakness: that illegally obtained pictures are also
weighed against the individual’s interest, one recent court decision, shows that this type of
regulation does have some effect. In this case, the court refused to admit recordings of a
customer apparently swapping price labels in order to purchase a purse at a significantly lower
price, because the department store detective had used the CCTV (which was considered
secret because no signs drew attention to it) to secure evidence of a crime and not to prevent
damage being caused to the owner by it. In other words, secret CCTV would be permissible if
used to take immediate steps and to prevent crime. Where this purpose is not served, individual
privacy is more important. Where CCTV is used more as a form of entrapment to secure
evidence, the courts will declare this use unfair and illegal, if no warning of potential surveillance
is given.
Nevertheless, as far as privately owned property is concerned, the ability to film and surveil using
CCTV is far-reaching and recordings made will often be admissible before a court. Even illegally
made recordings may well be considered to be admissible evidence of a crime (Beulke, 2002:
474).
Gras: The Legal Regulation of CCTV in Europe
Surveillance & Society
2(2/3)
221
The use of CCTV by private persons and entities accounts for the vast majority of installations.
This is true both in Germany and in the UK. A recent study illustrates very clearly how these
installations contribute to the surveillance net and it should, of course, not be forgotten, that these
cameras also present a resource to the police (For clear illustration of the role and extent of such
surveillance to protect buildings and traffic networks in London and Berlin see Töpfer
et al.,
2003: 33-40; and McCahill and Norris, 2002: 6-10, in particular). Where the police work
actively to intigrate such cameras into their strategy (as e.g.: in Münich, see: Simon
, 2000) this is
all the more true.
Police Surveillance
As far as surveillance by public authorities (and in Germany this usually means the police) is
concerned, we have an entirely different situation. There is a very complex constitutional
argument about whether mere observation of a scene in which no individual person is identifiable
via a camera, is an interference with a German citizens’ basic rights and therefore requires
specific legal regulation. This debate has, however, become unimportant in practice because: a) it
is now almost superfluous, as the majority of Länder and the Federal government have
introduced permissive legislation for CCTV surveillance and b) because CCTV as it is used in
the UK, and as the police in Germany are finding they wish to use it, is not about observing
unidentifiable groups of people, a central aim is precisely the ability to observe an individual and
to enable his or her identification. The latter is unquestionably an interference with a person’s
basic rights according to German constitutional law.
We can, therefore, assume that using CCTV without a permissive legal basis is illegal. For this
reason most Länder have introduced legislation more or less like that of Saxony, for example,
where § 38 2 of the Saxon Police Act allows police to surveil and record in various (broadly
defined) geographic areas where actual indications justify the assumption that, in this type of
place or at objects of this sort, offences will be committed which endanger persons, objects or
assets (so-called Vermögenswerte).
These laws are not quite as broad as they sound at first as they are subject to the constitutional
principles of Verhältnismäßigkeit and Erforderlichkeit - as are all state institutions in the exercise
of their power. Verhältnismäßigkeit or proportionality means that the method chosen may not be
disproportional to the problem to be solved. Accordingly, some experts regard the interference
with constitutional rights as illegal in order to combat crimes of a trivial nature. Indeed some
states expressly forbid this.
Erforderlichkeit means that the method chosen must be appropriate and necessary, i.e. it must be
one which promises to solve the problem and must be the mildest possible method in the sense
that it interferes as little as possible with the rights of those affected.
The Impact of Legislation
The practical effects of this legislation, and in particular of the debates which surrounded the
creation of permissive regulation, have been that the police and those legislating for police forces,
appear to feel a need to justify the use of CCTV. Thus an installation will often be preceded by a
crime rate analysis to prove the need for CCTV and this analysis will continue whilst the cameras
Gras: The Legal Regulation of CCTV in Europe
Surveillance & Society
2(2/3)
222
are in action. Persons running CCTV systems might well argue that no legal regulation is
necessary to achieve this and they may well be right. Well run systems should be doing this
anyway. A legal requirement to do so is still a different matter. Furthermore, it should be stressed
that German law requires a justification and legislative basis for interference with an individual’s
rights. This may not seem like much in the face of the potential threat to these rights some of us
regard CCTV as being, particularly since the police can and do still install. Nevertheless, I think
it is safe to say that so far the legal requirements have lead to police and decision makers being
cautious about the installation of CCTV. This is reflected in the relative limited number and size
of CCTV systems in Germany (which usually consist of less than 10 cameras) and the fact that in
Leipzig one camera was dismantled after it was deemed no longer to be necessary. A system, in
Bielefeld, also stopped operating for some time for the same reason. Comments can also be
found in the press which illustrate a greater awareness of potential controversy, thus one can
read for example, that a chief superintendent (in Leipzig) has no interest in linking police to
private cameras, not because of technical difficulties but because “we do not want an English
situation” (Holzer and Krischer, 2000). There is evidence of general awareness that there is a
conflict of interests.
The legal regulation in Germany is one matter and its power should not be over-estimated. In
spite of all I have said, the use of CCTV is increasing and it is not always being used in ways one
would regard as compatible with the stringent enforcement of the principles mentioned above.
Nevertheless, in comparison to the British situation it must be regarded as advantageous. The
need for permissive legislation meant that parliamentary debate had to ensue, relatively strong
data protection institutions were able to give these debates a serious critical flavour, which even
hard-line CCTV supporters were forced to acknowledge. In some cases, they led to legal
restrictions in the police use of CCTV (Northrhein-Westphalia), in others they did not. But at
least an atmosphere was created from the very beginning in which police were aware that some
citizens would question their use of CCTV. Whilst the use of CCTV will certainly continue to
grow, potentially excessively, especially if surveillance by private owners is not regulated in some
way other than a post-hoc appeal possibility before courts, one can say that legal regulation of
CCTV exists in Germany. As far as use by public owners in concerned, this regulation has been
relatively successful, so far, in ensuring that CCTV is used - I would like to say only in situations
in which it is truly necessary, that would however be overly optimistic - only in fairly limited
circumstances. This may change as post 9.11 legislation comes into use. So far, however, the
spread has not been extensive. Public systems are small and still relatively low in number.
As far as the regulation of privately owned systems are concerned, there are other examples of
regulation in Europe which require an installation to be approved. This would appear to be a
potentially effective means of regulation.
France
In France for example a 1995 law and a 1996 decree require that the Prefect in each French
Département
(administrative region) be informed of any plans to install a CCTV system and that
he or she decide whether the installation is to be approved. In making this decision, the Prefect
Gras: The Legal Regulation of CCTV in Europe
Surveillance & Society
2(2/3)
223
must consult a local committee, the CDV, which is presided over by a judge. Alongside the
judge, the CDV is manned by another magistrate, a representative elected by the local trade
chamber, an elected politician such as the mayor and a person with technical knowledge (who
may, however, not be a serving police officer). Their decision is made by majority vote and is a
recommendation to the Prefect, which he or she will, however, almost always go by, because the
CDV is more competent to make the decision than he or she is. The Prefect’s decision is final
but can be contested before an administrative court (i.e. is subject to judicial review). An
applicant must prove that the area one wishes to protect by CCTV is an object particularly liable
to theft or attack.
By the end of 1999, 38.520 CCTV installations had been approved. In 1999, 4500 applications
to install CCTV were filed, 4200 were, approved, 300 were refused, because they violated the
law. In other words a considerable number of CCTV systems are applied for and only a small
minority do not gain approval (most of whom are told what needs to be changed in order to gain
approval). This has been the subject of particular criticism. It should, however, be noted that of
these figures, a large number of applications stemmed from banks, petrol stations and shops.
Areas in which, in Britain but also in Germany, CCTV installations have long taken place and are
regarded as relatively uncontroversial. In how far this will prove to be an effective control of
CCTV remains to be seen. Clearly the legal and institutional potential to effectively regulate the
use of CCTV systems is there, in how far practice will do this, is another matter. The failure to
include police cameras in this legal framework is a subject of considerable controversy in France
(Oqueteau, 2001).
Sweden
Another, at least apparently, well-regulated system is the Swedish one. Here, again, there is a
kind of licensing system. All potential CCTV users (except for those who wish to use cameras to
observe a ‘protected object’, to improve a drivers sight when installed onto a vehicle, if the
national road authorities wish to observe traffic flows, weather conditions etc., or where the
police use CCTV against speeding offences or to observe a particular location for up to one
month with grounded suspicion that a particular serious crime or accident will take place there)
must make an application to the County Administrative Board. This application must contain
detailed information about the planned system and surveillance area, as well as include
agreement to surveillance by any employees affected. There is a general requirement to inform of
the surveillance by signposting outside of the surveillance area. Installation may only be for crime
prevention and detection reasons and the cameras must be fixed and may not have a zoom
function. An application will be approved if the interest in the surveillance in question outweigh
the individual’s interest in not being surveilled in those particular circumstances. There are some
circumstances, e.g. banks, were owners only have to inform of surveillance, i.e. where it is
always considered acceptable. A representative of the area to be surveilled must be heard. An
application can be granted in part or as a whole or only for a limited period of time. An appeal
can be made against the Board’s decision before an administrative court.
Gras: The Legal Regulation of CCTV in Europe
Surveillance & Society
2(2/3)
224
This sounds like a very privacy protecting system and comparatively, it certainly is. Actual
practice has, however, been criticised because the County Administrative Boards are also
required to supervise surveillance systems and have been found not to be doing so. For example,
of an estimated 11 500 cameras, the Stockholm Board had only inspected 400 between 1998
and 2000. When the Swedish Helsinki Committee did a random survey of surveillance sights,
they found many CCTV systems to be in breach of the law, in particular due to a lack of
signposting, but also due to filming a larger or different area than was allowed. A complaint
about this state of affairs to the Parliamentary Ombudsman resulted in a statement that one could
not oblige the County Administrative Boards to perform their legal duties as long as they are
financially incapable of doing so, which they claimed to be. A potentially attractive regulatory
solution must of course be adequately financed (Hårdh, 2001).
It should be noted, however, that the requirement even for the police to have to apply for CCTV
installations, may still be considered an effective control and is likely to enhance some of the
beneficial effects found in Germany further. Also one should remember that in 2000 Sweden was
estimated as having 30 000 CCTV cameras installed, with systems consisting of 3,2, cameras on
average. A very different situation than in Britain. To enable a comparison; this would mean
approximately one camera per 300 Swedish citizens, as opposed to roughly 1 per 14 British
citizens.
The Netherlands
Another interesting case is that of the Netherlands where CCTV was first installed in the late
1990s. In the meantime, 80 of 550 municipalities have at least one installation. Public authorities
must make an application to install to the
gemeenteraad (municipal council) which is considered
according to local needs. The view in the Netherlands was that international treaties, the Dutch
constitution and the European Data Protection Directive provided a basis for a comprehensive
regulatory scheme. This was regarded as requiring that surveillance must not be secret (unless
required to be so for detection of a specific crime), be for a closely defined purpose to detect or
prosecute defined crime or behaviour and it must be necessary for the owner to perform his or
her duties. The duties of a private person are regarded as limited to his or her property. Public
places are the responsibility of the mayor assisted by the police. As in Germany, less intrusive
measures must be considered not only before installation but also periodically in reviewing
CCTV surveillance. A complaint about unsuitable surveillance can be made to the local council
and by civil writ to a court. In addition the Data Protection Board has a duty to supervise CCTV
surveillance and has inspection powers. In order to consolidate these regulations, the Dutch
government drew up laws specific to CCTV surveillance and has expressly forbidden the secret
use of CCTV surveillance in public places (Offens, 2001).
The Spirit of the Laws in Europe
The law in France, Sweden and the Netherlands requires that attention be drawn to CCTV
surveillance by signs placed outside of the surveillance area so that a person be made aware of
Gras: The Legal Regulation of CCTV in Europe
Surveillance & Society
2(2/3)
225
entering it. In Germany signposting significantly lowers the level of interference with constitutional
rights CCTV surveillance is seen as causing. In the UK the advent of data protection legislation
applicable to CCTV systems led to a proliferation of signage. Across Europe it would seem,
there is agreement that those subjected to surveillance should be made aware of it. There can
only really be one reason for this; where an area is generally under surveillance any person
entering it having been informed of the surveillance implicitly agrees to it. The general stance
would appear to be, if privacy rights are affected by surveillance at a general level (and it would
appear the jury is still out on this one), implicit consent by the individual under surveillance
negates any breach.
The legal concept of consent is, however, not a simple matter. There is agreement that to be
valid, consent must be full and free (Alexander, 1996: 165). This indicates that the person giving
consent must have the capacity to do so based upon sufficient information and must have a
choice in doing so. There is plenty of controversy here. Given the doubts expressed in relation to
CCTV’s publicly propagated image as an effective crime prevention tool (see e.g.: Coleman and
Norris 2000: 166) or that the general public know the workings of CCTV systems (see e.g.:
Squires and Measor, 1996: 8), it is difficult to argue that consent is full. As the numbers of
camera installations grow and especially when a surveillance density such as in the UK is
reached, the issue of freedom rears its head. If I cannot purchase food without coming under
surveillance, how can my consent to that surveillance be said to be free? If I can only avoid
surveillance by leading an self-sufficient life on my own private property, surely my consent to
surveillance cannot be considered to be free? Anything which forces me to make such a choice
must surely be interfering with my human rights? The obvious response is of course that no
ordinary reasonable man would respond in this way because it is disproportionate. One might
even try to argue that there is no breach of any rights or that interference is only slight. In which
case we return to the original question; why the requirement to signpost? Why is consent to the
surveillance by those subjected to it regarded as at least desirable?
Furthermore given that the European Human Rights Court have refused to exclude very public
spaces and actions from the realms of privacy law (Niemetz v Germany (1992) 16 EHRR 97),
surely such a level of surveillance must affect my right to privacy if this is to afford any protection
to social interaction even in public space?
It has been pointed out that the right to privacy as defined over time is of little help when it
comes to regulating low level, general surveillance (see: Stalder, 2000) and indeed CCTV
surveillance is a case in point. The weakness of focusing on privacy in a data protection sense
becomes obvious. The German notions of spheres of privacy, which can also be seen in the
European Court’s doctrine (see: Taylor, 2002: 73-) are more helpful. And indeed the Federal
Constitutional Courts judgement highlighting the right to informational self-determination, the socalled
Volkszählungsurteil, focused not only on a right to determine what happens to one’s data
but also on the data subject’s behaviour in response to the potential collection of data (BVerfGE
34, 238; see also: Bäumler, 2000: 3). Following this line of legal argument, an interference with
the right to privacy can not only be caused by gathering data but by causing a person to adapt
his or her behaviour. A focus on the reaction to surveillance means that privacy issues arise not
Gras: The Legal Regulation of CCTV in Europe
Surveillance & Society
2(2/3)
226
only where the person is in fact under surveillance but also where she or he believes s/he might
be. Thus a dummy camera or even a sign indicating surveillance becomes privacy relevant.
There can be little doubt, that surveillance will be allowed in many cases. Deterring crime or
using cameras to assist police work are doubtless legitimate aims necessary in a democratic
society. However, whether this will remain the case once surveillance density passes a certain
point, can be seen as a different matter. Where public space is increasingly subjected to
surveillance, the nature and weight of the privacy rights of those within it must change if the
notion of some kind of privacy protection in it is not to become a farce. This argument only
becomes more weighty in interaction with issues of consent.
Issues of this kind are often not apparent when reading the letter of the law but the spirit of
European and national law and jurisprudence provides some foundation for such argument. The
increase of CCTV use in countries with a long-term constitutional commitment to privacy is likely
to alter the terms of debate surrounding CCTV.
Providing Effective Legal Regulation
From the examples of legal regulation above, one can gain a good impression of what good
regulation should provide for. The most important element would appear to be to provide an
opportunity for serious debate. This can only be achieved where there is some legal requirement
preventing those who wish to install CCTV from being able to do so without consultation. This is
best served either by a need for a permit or a constitutionally anchored requirement for
permissive legislation.
Whether it would be sensible and practicable to require police to apply for each individual
system is bound to be controversial. The police have the legal duty and right to interfere even
with the most important of human rights under certain circumstances so it may well be regarded
as sufficient if, in relation to CCTV, these circumstances were defined by a democratically
legitimated process. A definition of the parameters of CCTV and the acceptability of its use is,
however, a if not the central pre-condition. This requires a debate of some sort as to what level
of surveillance is acceptable altogether, what crimes CCTV can legitimately be used to fight,
how etc. These then need to be moulded into clear laws, where necessary augmented by clear
guidelines at a local level, including provisions (in so far as the law does not provide) for a review
of each installation before it takes place, but also with mechanisms to review the need for CCTV
surveillance on a regular basis. Given that the police are regarded as serving not their own but
general public interest, this must be defined. Since the police are concerned with fighting crime
and public order issues, it seems inappropriate to require or expect them to consider privacy and
human rights issues and to get the balance right. This is a job for the legislature.
As far as private systems are concerned, the issues are quite different: the conflict is between
many individuals privacy and usually the owner’s interest in protecting his or her property. Whilst
the legitimacy of protecting property needs to be considered, it would seem wrong to neglect the
public interest element entirely, e.g. by only recognising a conflict where one individual complains
Gras: The Legal Regulation of CCTV in Europe
Surveillance & Society
2(2/3)
227
that his rights have been interfered with in a specific case. Simply balancing one individual’s
property rights with one other’s privacy rights is only part of the issue. The property rights must
be balanced against a larger number of individuals’ privacy rights. This and effective legal
regulation can only realistically be achieved by a permit system. Where, however, such a large
number of applications to install cameras are made that one would consider an area as a whole
to have too much CCTV, there must additionally be a possibility of rejection on these grounds,
i.e. the ability to review each application to install a camera in a greater context of how widespread
surveillance already is. A first come first serve basis for deciding who may install would,
however, in itself be constitutionally problematic and so we return to issues of planning.
Ultimately legal regulation in this field also requires the pre-condition that decisions have been
made as to how far CCTV surveillance is acceptable.
Again installations should be subject to continual reviews as to their necessity and it must be
possible to order a system to be dismantled.
Legal regulation on a micro-level is relatively straight forward and has been provided for in many
European countries. However it is more difficult to see how the spirit of constitutional or human
rights doctrine can be worked into this scheme, particularly as data protection law may prove to
be inappropriate. There is nothing new in technology challenging the law. Nor in the
modernisation of legal concepts to fit the times. If we are not only to witness an ever increasing
proliferation of CCTV, the authorities permitting installation must not only have the right to
dismantle on the grounds that the cameras are no longer necessary to achieve the aim for which
they were installed but also because the area in which they are installed now has so many
cameras that issues of consent and possibly the nature of the privacy rights of those under
surveillance has changed. This of course requires the political will to define and confer such
powers.
Concluding Comments
Many European states explicitly acknowledge that CCTV surveillance in public space creates a
conflict with the right to privacy – and thus regulation is both necessary and desirable. A permit
system appears to be fairly common for private users wishing to surveil a public space. The
police are frequently granted a wider ambit in which to use CCTV surveillance but this does not
necessarily mean that they feel free to install a large number of cameras. The law can effectively
limit CCTV use. This can of course only be true where it is effectively enforced in word and
spirit.
There can be no doubt that law will reflect public opinion and the pressure and desire politicians
feel to ‘do’ something about crime. The wave of panic unleashed by the 11
th of September 2001
saw several states in Germany broaden the permissive nature of their video surveillance police
laws.
Whether regulation in the European sense, would have made any difference to the British
development of CCTV use is a question which will always remain unanswered. Would any legal
Gras: The Legal Regulation of CCTV in Europe
Surveillance & Society
2(2/3)
228
argument have stood stead to the pressure perceived and generated after the Bulger case? It
might have done. It could have given critics a better forum in which to argue their case. The fact
is that legal regulation provides an opportunity. It provides a forum in which legitimate criticism
and concern about CCTV in general, but also about use in a particular case, can be stated.
Simply having to seriously address privacy issues when applying for permission to install CCTV
may influence the way operators use the technology later. A legislative and, particularly, a
licensing procedure (with the potential to reject an installation after taking a wide range of factors
into consideration) can provide a bulwark against panic reactions. This of course all depends on
there being persons willing to and in a position to present the arguments against CCTV.
Therefore, a well established data protection network (such as that of state and Federal Data
Protection Commissioners in Germany) will, for example, be necessary to provide the substance
within a legal framework. But argument is also necessary beyond the issues of data protection.
Effective regulation can only achieved if it is in place before CCTV has spread widely. In other
words, it is more likely to be achieved in countries where there is fundamental recognition of
privacy issues such as systems constitutionally requiring legislative authority before police can use
them.
The efficacy of regulation will depend not only on its quality, but far more on a country’s long
term commitment to privacy and the institutional safeguards it has already installed to protect it.
CCTV surveillance also raises problems as to the definition of privacy and will continue to be a
legal challenge long after the keenest issues of regulation have been solved.
Additionally, real, perceived or generated pressures on crime policy, and how far they are
allowed to erode legal safeguards, will be elemental. Ultimately, the question remains in how far
a population will be prepared to chose freedom from surveillance over the security it believes
CCTV can give it. Given the tendency in Europe to accept a great deal in order to combat
crime, the apparent pressure for CCTV will have to be combated by reasoned argument. Above
all this needs time and a forum which, as this paper argues, legal regulation can help to provide.
Legal regulation is only a first step. But it may be a key one.
References
Alexander, L. (1996) The Moral Magic of Consent (II),
Legal Theory, 2: 165-174.
Bäumler, H. (2000) Datenschutzrechtliche Grenzen der Videoüberwachung. In
Der Landesbeauftragte für den
Datenschutz Mecklenburg-Vorpommern, Grenzen und Risiken der Videoüberwachung
, Schwerin
Beulke, W. (2002)
Strafprozeßrecht. 6th ed. Heidelberg : Müller
Coleman, C. and C. Norris (2000)
Introducing Criminology. Cullompton & Portland, Willan Publishing
Gras, M. (2003)
Kriminalprävention durch Videoüberwachung - Gegenwart in Großbritannien, Zukunft in
Deutschland?
Baden-Baden, Nomos.
Hårdh, R. (2001) The Use of Public Video Surveillance in Sweden. Paper given at: Public Video Surveillance
as a crime prevention instrument - a European Comparison. Göttingen, Germany 22-24 February.
Gras: The Legal Regulation of CCTV in Europe
Surveillance & Society
2(2/3)
229
Henderson, R. (2001) Public Video Surveillance in the United Kingdom. Paper given at: Public Video
Surveillance as a crime prevention instrument - a European Comparison. Göttingen, Germany 22-24
February.
Holzer, K. and M. Krischer, (2000) Im Land der 1000 Augen,
Focus, 33/2000: 51-.
McCahill, M. and C. Norris, (2002) Urban Eye, Working Paper No. 6, CCTV in London,
http://www.urbaneye.net/results/results.htm
McCahill, M. and C. Norris (2004) From cameras to control rooms: The mediation of the image by CCTV
operatives. Paper given at: CCTV and Social Control: The Politics and Practice of Video-surveillance
-European and Global Perspectives, University of Sheffield, 8 -9 January.
Norris, C. and G. Armstrong (1999)
The Maximum Surveillance Society. The Rise of CCTV. Oxford, Berg.
Ocqueteau, F. (2001) Video Surveillance in France – Regulation and Impact on Crime. Paper given at: Public
Video Surveillance as a crime prevention instrument - a European Comparison. Göttingen, Germany
22-24 February.
Offens, H. (2001) Public Video Surveillance in the Netherlands. Paper given at: Public Video Surveillance as a
crime prevention instrument - a European Comparison. Göttingen, Germany 22-24 February.
Sanders, A. and R. Young (2000) Criminal Justice. 2
nd ed. London, Butterworths
Simon, S
. (200) Polizei will Videokameras mit Augenmaß einsetzen, Süddeutsche Zeitung, 9 February.
Squires, P. and L. Measor (1996) Closed Circuit TV Surveillance and Crime Prevention in Brighton: half yearly
report, Health and Social Policy Research Centre, Brighton
Stalder, F. (2002) Opinion. Privacy is not the antidote to surveillance.
Surveillance & Society 1(1): 120-
http://www.surveillance-and-society.org/articles1/opinion.pdf
Taylor, N. (2002) State Surveillance and the Right to Privacy.
Surveillance & Society 1(1): 66-
http://www.surveillance-and-society.org/articles1/statesurv.pdf
Töpfer, E., Hempel, L. & Cameron, H. (2003) Urban Eye, Working Paper No. 8, Watching the Bear,
http://www.urbaneye.net/results/results.htm
Weichert, T. (2001) Öffentliche Videoüberwachung aus der Sicht der Europäischen Datenschutzrichtlinie und
des deutschen Datenschutzrechts. Paper given at: Public Video Surveillance as a crime prevention
instrument - a European Comparison. Göttingen, Germany 22-24 February.
Weinbrenner, U. (2001) Videoüberwachung in Deutschland. Paper given at: Public Video Surveillance as a
crime prevention instrument - a European Comparison. Göttingen, Germany 22-24 February
No comments:
Post a Comment