Examination of Witnesses (Questions 40-59)
Professor Clive Norris and Dr David Murakami Wood
28 NOVEMBER 2007
Q40 Chairman: Are you able to tell
us what information there is that CCTV has been as effective in
deterring and detecting as was originally envisaged?
Professor Norris: "Little" I think
is the short answer. When CCTV was first introduced in this country,
it was not subject to systematic evaluation. It was introduced
on the basis that practitioners thought that it was effective.
Over the last ten years, studies have been carried out by academics
and particularly the work by Jason Ditton in Glasgow and the work
of Professor Martin Gill at the University of Leicester, which
suggests overall that it has a very, very weak influence on reducing
crime. The Gill study was published in 2004; it was the first
major Home Office sponsored evaluation. Not only did it show that
CCTV had very limited impact in reducing crime but it had very
limited impact in reducing fear of crime. The evidence in terms
of general reductions in crime and general reductions in fear
of crime appears to be very, very weak. There are studies that
do show a reduction in specific places. For instance, the same
team that looked at Glasgow, the Ditton team, looked at Airdrie
and, in Airdrie, they did find a reduction. In Glasgow, they found
that crime increased when CCTV was introduced. One further study
that is worth mentioning is the Farrington and Walsh meta-evaluation
which also found very weak evidence for CCTV as a crime reduction
measure; this again was sponsored by the Home Office. If I remember
correctly, they suggested at best about a three per cent reduction
mainly in car parks and very little evidence that in town centre
space you would see a reduction, but that street lighting seemed
to be a rather more effective form of prevention.
Q41 Baroness O'Cathain: I have a
very simple question particularly relating to your point about
the fear of crime and showing that reductions in crime have not
been affected by CCTV. Do you have any statistics at all about
the reliability of these CCTV cameras? What proportion do you
actually think are working? How many of them break down? Where
do the manufacturers get a licence to produce them? Is there a
special code or a specification for putting these up in the first
place saying that they have to reach certain standards or can
anybody string together some sort of camera and pretend that it
is a CCTV camera?
Professor Norris: The answer to the first
part of the question as to how reliable they are and whether there
are statistics to tell us that, I do not know of any broad-range
statistics on how reliable they are. Clearly, if you look at the
Gill study of the implementation of range of systems, there were
problems with reliability of systems and they were part of it,
although I do not think they were necessarily wholly undermining
of the systems but there were technological problems. The second
part ... ?
Q42 Baroness O'Cathain: Is there
any organisation which actually looks at the manufacture of them,
the actual physical specification?
Professor Norris: Certainly the Home
Office has tried to issue guidelines.
Q43 Baroness O'Cathain: Tried to?
Professor Norris: Yes. I cannot answer
your question with any certainty other than to say that they do
issue guidelines as to what would be necessary in the technical
sense. I think that the problem is that the range of possibilities
is actually rather great, so specifying very exactly in any particular
case what you can put in place and where is not such an easy job.
One of the problems that has beset in a sense partly the expansion
of systems is the problem of inter-operationability: different
systems even in the same town and even run by the same council
have different technical requirements, they do not integrate properly,
and this still besets the industry.
Q44 Chairman: Do we understand from
your answer to the first point of Lady O'Cathain's question that
the reports we get in the newspapers of the number of cameras
and the number of times we are all photographed are guesses and
not based on any statistical evidence?
Professor Norris: I have to put my hands
up to this because I am the originator of both these numbers.
The number of 300 times a day that we are captured on film was
included in a book I wrote called The Maximum Surveillance
Society. Is it a guess, just a guess? I would say that it
is a guesstimate. How I came to that figure was that I took a
person in London moving around the City from early in the morning
until late in the evening and I constructed a journey that intersected
with known CCTV systems. So, this was not a fantasy in that sense,
this was a journey. I think that I wrote this in 1998, so nine
years ago, and I think that the estimate of 300 cameras was perfectly
justifiable on what I knew about each of the systems that they
intersected with.
Q45 Lord Rowlands: I am a little
surprised by your initial answer because, for example, I travelled
in on the number 24 bus this morning and inside the bus was a
noticing saying that there was a CCTV camera and that there had
been 60 prosecutions for vandalism. If you had polled that bus
this morning, I would have thought that the vast majority of us
would have said that it was an acceptable form of surveillance.
Professor Norris: I was not saying that
it was acceptable or unacceptable, it was a question of how many
there are.
Q46 Lord Rowlands: Yes, but you also
implied that it was of very little value.
Professor Norris: I am saying that the
best scientific evidence that we have does not suggest that CCTV
surveillance is very effective at reducing general levels of crime.
Q47 Lord Smith of Clifton: Was this
journey a journey which a number of people make or was it in search
of CCTV cameras? Was it a deep search as opposed to a journey
from Richmond to the City which a stockbroker might make?
Professor Norris: It was a busy day in
London and it was trying to make a point so, in that sense, it
was a piece of rhetoric. However, let us take my journey yesterday
from the University of Sheffield to my hotel in London. Every
stage of that journey was captured on a CCTV system. My university
system captured me; on the bus that I caught to go into Sheffield
to get to the station; as soon as I arrived at the station; I
was captured when I got off at St Pancras; I was captured when
I walked through to King's Cross, I was on their system; I walked
into Smith's and I was on their system; I got into a taxi in London
and that had a CCTV camera; I got dropped off at my hotel and,
as soon as I walked into the entrance of my hotel, I was captured
on a CCTV camera.
Q48 Lord Smith of Clifton: That was
roughly 20 times; it was not 300.
Professor Norris: How many cameras are
there? If we talking about the number of cameras that could have
seen me, in the Underground there are thousands of cameras; in
the stations there are thousands.
Lord Peston: I would like to ask a technical
question following on from Lord Rowlands. Surely on Lord Rowlands'
number 24 bus coming down from Hampstead no doubt.
Lord Rowlands: Coming from Pimlico.
Q49 Lord Peston: No one at the time
you are travelling is a vandal, so that really does not count
as evidence. The real point is that the vandals will be getting
on later and they are not affected by those cameras. They are
drunk, hooligans or what-have-you. Therefore, it is quite compatible
that none of you were misbehaving but that the cameras had no
impact on those who had a high propensity to misbehave. We do
not know that, but we have to do the research and what we are
being told is that the research shows that those who have a propensity
to vandalise buses are not affected by the cameras. That does
not surprise me at all.
Dr Murakami Wood: It is important to
stress that neither Professor Norris and myself would argue that
cameras are ineffective at adding to the weight of evidence or
being used in court. I do not think we are saying that. The question
was about prevention and the claims that were made for cameras
when they were first introduced to actually reduce or prevent
crime and I think it is quite clear from the evidence that we
have seen that there is not enough evidence to suggest that there
is any statistically significant effect on the rates of crime
or any kind of crime prevention and that is the important distinction.
It is up to you to make the judgment on whether that is important
or not.
Q50 Baroness Quin: What evidence
is there for displacement? In other words, if you have cameras
in one place, crime just moves elsewhere. I can certainly think
of an area that I knew quite well where crime was reduced by a
very effective if somewhat intrusive CCTV system but at the same
time crime rates just down the road rather increased.
Professor Norris: There is evidence for
both displacement and for the halo effect. Certainly from the
study in Doncaster conducted by David Skint city centre crime
did reduce but it spread to the outer lying townships and there
was statistically significant evidence to that effect. There have
also been arguments for which you will also find some statistical
evidence that, if you put a system in a particular geographical
area, it could have effects on the surrounding areas which do
not have cameras. Overall, the main level of effect is actually
not very much.
Q51 Lord Morris of Aberavon: Does
it not give a perception of safety to people when there are CCTV
cameras?
Professor Norris: If you look at the
evidence from the Gill study which is the largest study conducted,
a three/four-year study funded by the Home Office employing a
large number of researchers, their conclusions were that it did
not increase people's feelings of public safety.
Q52 Lord Morris of Aberavon: It is
the perception of people that I am asking about.
Professor Norris: Your feeling of safety
is a perception. It did not increase people's perception that
they were safer. In a way, we can see that that has been recognised.
The whole of the city centre warden movement to having in a sense
a visible authoritative presence on the street that is not necessarily
police is about responding to that public demand that what they
want is people not machines and technology and it is people who
make people feel secure rather than machines.
Dr Murakami Wood: I have carried out
a great deal of work in Japan and it is a useful comparison in
this case because Japan is a society traditionally regarded as
having a high level of social trust and very low crime rates in
comparison to western countries. What I found from talking to
people there, especially womenand CCTV is only now being
introduced in public spaces with government supportis that,
when they saw cameras, they felt less safe and not more safe because
that made them think that the area was dangerous, that there was
something they should watch out for. Whereas no cameras made them
feel their normal level of trust in other people. It really depends
on the level of social trust which you have in a society and I
think that cameras are probably a mark of the decline in social
trust and indeed may increase further that decline in social trust
as we rely more and more on technology to replace and compensate
for the decline in trust that exists overall in society.
Q53 Chairman: I have one final question
on this before I ask Lord Smith to come in. Are you able to say
what the evidence is of the effect of CCTV on detection?
Professor Norris: The simple answer to
that is, "No, I cannot, not with any certainty".
Q54 Lord Smith of Clifton: What do
you see as the key adverse effects of state surveillance? Do they
go beyond the deleterious effects on individual privacy?
Professor Norris: Yes. One needs to think
about this mainly in terms of mass surveillance rather than individualised
and targeted surveillance. There are four broad issues here. Firstly,
there is the issue that mass surveillance promotes the view in
a sense that everybody is untrustworthy. If we are gathering data
on people all the time on the basis that they may do something
wrong, this is promoting a view that as citizens we cannot be
trusted, and I think that that is a general issue. A second problem
is that once you are into a surveillance solution, it becomes
in a sense expansionary to a huge degree. If you see that information
is what you need to solve a problem but you do not quite know
what that problem is and you do not know what future events you
are going to be responding to, the temptation is to collect all
information about all people, and that is in a sense partly the
way that things have gone. If one thinks about the new criminal
records system which will integrate the databases of all police
forces including all the intelligence files on 11 million people,
I think it is 65 million records that will be integrated, all
information comes to bear and then there is the idea that we have
to join this all up. So, the information held in health fields,
education fields and welfare fields all becomes part of the resource
to solving a particular problem. The expansionary nature of a
surveillance system is a problem if it does not have checks and
brakes to it. The next point relates to what we said earlier.
I think that there is an undue faith in technological solutions
to the problem of crime, security and order. The best evidence
is that order, crime and security are best promoted at a local
face-to-face negotiated level. The best way for police to solve
crime is if the public give them information freely. It is if
the public trust the police that there is that flow. That is about
a reciprocal relationship. One of the problems one has with reliance
on technological solutions is that we can create a distance between
police and public. We can see a police that actually see themselves
as standing outside the community and coming down in a sense from
the mountain to impose order rather than a police that are an
integral part of that community who have to negotiate, sometimes
with discretion and toleration, with various communities and individuals
but, in that process of trade-off, what one does is build up trust
and consent and consent is at the heart. I feel that the faith
in technological solutions may actually lead to, in a sense, a
shift from one of the fundamental principles of British policing.
That would be my third point. My fourth and perhaps actually in
a way the most serious issue is that, as one creates a mass surveillance
system, as this personal information becomes more and more available,
what we are seeing is the idea of risk assessments becoming more
and more prevalent in various aspects of certainly criminal justice
management but also within education and so forth, and the risk
assessment provides the basis for pre-emptive intervention. I
think that this is a really serious issue. The issue of course
is that we normally talk about intervening with people in the
criminal justice sense on the basis of individualised reasonable
suspicion. Indeed, the PACE Codes of Conduct actually say that
you cannot stop and search someone merely on the basis of a category
such as their race. You have to have a better reason than that.
Where you collect information and you say that if an individual
who in a sense shares the characteristics of other individuals
who deviate in some way and therefore are seen as criminal or
whatever, that gives us the right to intervene with them and their
families with various social programmes, some of which may have
punitive elements to them. This seems to change and challenge
in some ways and I am not sure that I understand all the ways
it does, but ideas of reasonable suspicion and the presumption
of innocence, for instance. Something is going on here that I
think represents a fundamental shift which our concepts have not
quite caught up with. I think that is how I would see the main
adverse effects, but again I am particularly thinking about the
mass targeted.
Q55 Lord Smith of Clifton: I would
like to press you on this and it seems to me that you began to
allude to this. Are there some categories of individuals or social
groups who are adversely affected more than others?
Dr Murakami Wood: First of all, the thing
to say is that both ends of the social spectrum, the most wealthy
and the worst off, are both subject to high levels of surveillance.
There is a big difference. Those at the top end of society tend
to get the protective and inclusive benefits of this. This is
surveillance that is voluntarily entered into for protection and
for social inclusion in volunteering for systems like the iris
scanning at Amsterdam Airport to speed you through immigration.
You get better security, gated communities and things like that.
At the bottom end however, there is significant deleterious effects
on people's lives and Clive will detail some of these.
Professor Norris: If you look at the
studies done on the operation of CCTVand I think this raises
one question about all these systemsgenerally these systems
have elements of discretion built into them. They are not just
automatic systems following automated routines; they involve people
making choices. CCTV operatives have to make choices about who
to target and the evidence is that they are most likely to target
young males particularly if they are from ethnic minority communities.
In terms of the way that that has an impact, actually it is not
so much in the public sphere of the town centre, it is more in
evidence in the private sphere of the shopping mall with what
sorts of people tend to get excluded from those areas. It is not
just that they get excluded for criminal infraction, they are
getting excluded because youths in a shopping mall are hanging
about and they are not shopping and they are asked to move on.
If they suggest that they have a right to be there, they are told
that they do not. It is private space, so maybe they do not have
a right to be there. Then they are excluded. If they argue too
much, they will be banned from the shopping centre and the cameras
and the security officers will enforce that ban. So, there is
a form of exclusion that can go on which tends to target particular
social groups and not generally us, as it were. If I may take
another example, we have introduced mandatory drug testing in
prisons. One of the features of such systems which seem to me
to make them at least have elements of fairness in them is that
they are random and, when they are random, everyone has an equal
chance of being subjected to them. Unfortunately, there is also
a little bit that says if a prison officer thinks that you warrant
drug testing, then you will get it. This introduces again a human
discretionary element to it. What I do not know is the extent
to which that may be based on discriminatory bases. We do not
know the answer to that question. As soon as you do that, you
have that potential. I think that the DNA Register is one where
this is really very serious. The over-representation of black
men in the DNA Register is a serious issue and cause for concern
and part of that over-representation is because they are more
likely to be arrested by the police and in some ways that over-representation
in arrest statistics may represent an over-representation in certain
forms of crime but, in other ways, what it represents, as we know
that those people are more likely to be arrested without charge,
more likely to be acquitted and so forth, is that there is evidence
that this is not just on the basis of good evidence. So, we have
a system that is disproportionately including someone on a register
which will affect their life chances in ways in the future which
is based on forms of differentiation and I have suggested perhaps
at times forms of discrimination.
Q56 Lord Morris of Aberavon: Am I
getting the wrong impression? Is it that neither of you are keen
on any form of surveillance or is that wrong? In your written
evidence, you refer to "the emergence of a `safety state'
obsessed with security and stability, and increasingly favouring
the precautionary surveillance of groups, categories and individuals
... " What are the main dangers of this kind of approach?
Dr Murakami Wood: First of all, I think
that it is very important to stress that we would never say that
surveillance itself is a bad thing. If you read our report which
we wrote on the surveillance society for the Information Commissioner,
we are quite clear that surveillance is often about the best intentions
regarding care and indeed many of our functions in a welfare society
would not be able to work without surveillance. Indeed, safety
and security of the Realm are also assured by surveillance in
many cases. We would like to make it quite clear in the record
that we are not suggesting that all surveillance is wrong or that
surveillance necessarily has negative effects and Clive will talk
about what we mean when we talk about precautionary surveillance.
Professor Norris: Again, it seems to
me that there is this problem of if one is gathering information
pre-emptively on a citizenry on the basis that they might commit
future crimes, one is widening and changing the nature of the
contract. If you look at the document on transformational government,
it is clear that what is envisaged is basically a merging of all
the data held by government in various forms. Information sharing
and taking down the silos are key elements of that report. One
of the questions for me here is that we have a regulatory system
that has been built up on the principle that you give information
for a particular purpose, but you give information in a context
and it is to be used in that context. We now have a situation
where it appears that what is emergingand it is emergingis
that the context is merely governance, that you give information
at one point of the system. So, as a child you have information
recorded about youI am not sure that you freely give it
but it is certainly recorded of you and from youand that
can then become available at another point in the system, a criminal
justice context for instance. This seems to be a change in the
nature of how we have traditionally thought about information
and about the extent to which people have the right to control
information about themselves and how it is used. I think that
that represents a significant shift. Does that answer your question?
Q57 Lord Morris of Aberavon: Up to
a point only. You mentioned DNA testing. Presumably you would
take an adverse view of the collating of information. You mentioned
classes of people who will get on that register. What about the
balance of advantage which might occur when people who have committed
an offence 20 years ago are apprehended on the basis of information
that happened to be stored? Would you put the ID card in the same
category? Would you put the collating of health information, a
study about which a large number of doctors are refusing to take
part in, in the same category? Are they all in the same bag, as
it were?
Professor Norris: No. In a sense, I think
that the issue of the DNA register raises some very interesting
questions. If we are as a society prepared to acceptand
we seem to have beenthat the police may arrest somebody,
not charge them, take them DNA and store it on a register, then
I am slightly concerned because actually I think that the issue
becomes, if merely arrest is the criteria for being on the register,
(1) it gives the police a perverse incentive to arrest people
because I think there is advantage to the police for having the
register, it has definitely to be shown to be
Q58 Lord Morris of Aberavon: An advantage
to all of us maybe.
Professor Norris: But then I think the
question becomes, if it is so advantageous, we should all be on
the register. That is something that one might have to consider.
Q59 Lord Morris of Aberavon: Why
not?
Professor Norris: I am not saying "Why
not?" I think that is the debate to be had.
|