Memorandum submitted by Professor Marian
Fizgerald
KEY POINTS
Prevalence
— Knife crime is probably three to
four times more prevalent than gun crime but it is much less well
evidenced athough there may be a link between the two.
— The homicide figures do not point
to a rise in knife crime but other sources suggest knife carrying
and knife crime have been increasing over several years.
Patterns
— Knife crime is not a crime in its
own right but a dimension of other offences—in particular
robbery and the full range of other offences of violence.
— The profile of knife crime will vary
by area and by time of day in terms of the types of offence involved
and the characteristics of victims and suspects.
— There is a danger of focussing too
narrowly on the problem of knife crime among school age children
and particular ethnic groups. The majority of suspects are likely
to be young adults and older people (depending in part on the
type of offence involved) and, though the problem may disproportionately
affect particular communities, the majority of victims are white.
Solutions
— Tackling the problem requires short-,
medium- and long-term solutions; but a sine qua non is reliable
statistics. So the recent announcement by the Home Secretary that
knife crime figures will now be kept on the same basis as gun
crime is particularly welcome.
— Police searches are indispensable
for discovering knives; but a balance has to be struck between
their use and the danger of alienating groups who may themselves
be at greatest risk of victimisation, especially when these could
also be valuable sources of intelligence.
— These tensions may be offset by better
use of technology (and in particular search wands) by the police
and by others, including teachers; but a systematic evaluation
of their effectiveness, as well as their acceptability is needed.
— Taking young people's victimisation
more seriously and providing them with reassurance (including
by more effective targeting of those they most fear) may be more
effective in persuading them to stop carrying knives than relying
mainly on educating them about the criminal sanctions if they
are caught.
BACKGROUND
The following observations are based on my own
work over the last eight years, including:
— focus groups with secondary school
pupils in different areas of London in relation to a range of
research projects including a study of young people's involvement
in street crime (FitzGerald, Stockdale and Hale 2003);
— analyses of relevant published sources
of data for a study which is soon to be published by the Youth
Justice Board about young people, `gangs' and weapons; and
— an evaluation in 2006 of initiatives
to tack knife crime which was funded by Government Office London.
This involved analyses of knife crime figures collected by the
Metropolitan Police Service (MPS) with a particular focus on three
boroughs, two of which had undertaken detailed analyses of the
profile of knife crime locally.
PREVALENCE OF
KNIFE CRIME
The only reliable trend data in recorded knife
crime are details of method of killing in the homicide statistics,
albeit these subsume knife crime within a wider category of "sharp
instruments". The number of such killings varies from year
to year but there has been no discernible upward trend over the
last 10 years.
However, homicides represent only a minute proportion
of all known acts of violence; so it may not be safe to draw inferences
from these figures about overall trends in knife crime. During
my field work for the street crime study in particular (ie as
far back as 2001-02) both young people and professionals working
with them, including staff in Youth Offending Teams spontaneously
raised concerns about the increase in knife carrying among young
people. Recently talking to a group of sentencers they also believed
they over time they had observed an increase in cases of violence
involving more serious injury because weapons (and, in particular,
knives) were now being used in altercations which in the past
would have been settled using fists.
Corroboration for perceptions that knife carrying
has increased may be inferred from at least two statistical sources.
One is a Home Office study published in 2004.
It reported the findings of interviews with adult arrestees, including
self-reported gang membership; and some of the questions related
to their use of weapons. As Table 1 shows, the proportion of past
and present gang members who reported carrying guns was similar.
However, present gang members were much more likely to report
carrying "weapons" more generally and it is safe to
assume that part (and possibly a large part) of this increase
will have been accounted for by knives.
Table 1
WEAPON CARRYING BY GANG MEMBERS, PAST AND
PRESENT
| % of current
gang members
| % of past gang
members | % of all gang
members
|
| Ever possessed weapon during an offence |
63 | 44 | 49 |
| Ever possessed a gun during an offence |
33 | 31 | 32 |
(Source: Bennett and Holloway, 2004)
The second is trends shown in the published figures for s1
searches by the police (Home Office 2006). Between 2000-01 and
2004-05 the total number of arrests from s1 searches fell by just
under 1%; but arrests for offensive weapons from searches rose
by 16%[1].
More generally, the homicide statistics indicate that killings
with sharp instruments are three to four times as prevalent as
killings using firearms. In addition, the figures collected by
the MPS not only confirm the greater prevalence of knife crime
compared to gun crime, two of the boroughs covered in my recent
study were boroughs where gun crime is higher than average. In
both, the trends for knife crime seemed to run in parallel with
those for gun crime, suggesting that they may in some way be linked
(see Figure 1). A common sense understanding of this would suggest
that while many more individuals are involved in knife crime than
in gun crime, they are not necessarily different individuals;
and those who are involved in gun crime do not start out carrying
guns but other, more readily available weapons including knives.
So an increase in one will often be reflected in an increase in
the other, especially in areas like these where higher than average
levels of violent crime are reflected in a higher than average
use of weapons in general (ie not only knives and guns but probably
other types of weapons also).
Figure 1a
Borough 2
Knife crime vs gun crime (rolling averages)
Figure 1b
Borough 3
Knife crime vs gun crime (rolling averages)
PATTERNS
AND UNDERLYING
FACTORS
Like gun crime, knife crime is not a discrete offence. Knives
may come into play in a range of different offence contexts—most
obviously
— general offences of grievous and actual bodily
harm;
— robbery (where the threatened use of a knife may
be instrumental even where no knife actually exists); and
They may also feature in cases of property crime (including
burglaries) where violence or the threat of violence is an aggravating
factor.
Detailed analyses undertaken by two London boroughs suggest
that the type of offence which most commonly involves knives may
vary from one part of the borough to another, as well as by time
of day. Violence related to the night time economy often features
significantly and produces a peak in figures in the late night/early
morning; but this tends to be highly concentrated in a limited
number of hotspots which attract the largest numbers of people
to pubs and clubs. Most of this late night violence is likely
to be affray and inter-personal violence but may also include
some robberies. However, in the two boroughs which had plotted
knife crime by time of day, this also tended to show a much smaller
peak in the mid- to late-afternoon (ie around the end of the school
day) which seems almost certainly to reflect robberies and other
altercations between school pupils.
However, in addition to the very locally-specific dimension
of knife crime, it can also be very mobile. Many borough hotspots
are around transport hubs; and British Transport Police (BTP),
as well as Transport for London will hold information on which
particular transport routes might equally well be designated as
hotspots. These, of course, will cross borough boundaries; and
BTP claim[2] to have found
instances of a large number of incidents in different parts of
London which were carried out by juveniles from one particular
area, facilitated by the entitlement of young people to Oyster
cards which now allow them to travel widely throughout the capital
free before they have legal access to their own transport.
The profile of victims in the two boroughs which had analysed
their own figures varied in ways which reflected differences in
the ethnic make up of the two boroughs (Figure 2a). However, Borough
1 had also broken down their analyses by the type of offence involved.
This showed some difference in the ethnic make-up of victims depending
on whether the offence was robbery or assault, with black people
(IC3s) more likely to be victims of knife-related assaults than
of robberies. Still more significant were variations in the age
group of the victims according to the type of offence. The overall
profile of victims in Borough 2 is somewhat older than in Borough
1; but Borough 1 also suggests that the average age of victims
of knife-related robberies is much younger than that for victims
of assaults, which cover a wider age span (Figure 2b).
Figure 2a
Victims of knife-related crime in two boroughs
Figure 2b
Age of victims of knife crime in two boroughs
Both boroughs suggest a high degree of black involvement
in knife crime in their areas; but again, the figures in Borough
1 suggest that this is more pronounced in the case of robbery
(Figure 3a) where the age profile is also very much lower (Figure
3b).
Figure 3a
Suspects' ethnicity in two boroughs
Figure 3b
Suspects' age in two boroughs
A further important source of information on this point is
the surveys of young people undertaken by MORI for the Youth Justice
Board, the latest (and last) of which was in 2004. This casts
light for young people aged 10-16 on:
— the characteristics of those most likely to carry
knives;
— their reasons for doing so; and
— the relationship of knife carrying to knife crime.
The survey included a question on whether respondents had
carried a weapon of any sort in the last year. A majority of the
interviewees who had been excluded from school had done so; and
this was true of a sizeable minority of those in mainstream education.
However, most of those who admitted carrying a weapon said they
had never used it (see Table 2). [3]
Table 2
PROPORTION OF YOUNG PEOPLE CLAIMING TO HAVE CARRIED WEAPONS
IN THE LAST YEAR
| Young people in school
| | Excluded young people
|
Base: all young people | (4,715)
| (2,460) | (2,225) | (687)
| (502) | (174) |
| % | %
| % | % | % |
% |
| I have carried a weapon but never used it |
21 | 31 | 11 | 47
| 52 | 31 |
| I have used a weapon against another person |
3 | 5 | 1 | 14
| 17 | 9 |
| I have threatened another person with a weapon
| 3 | 5 | 2 |
21 | 24 | 12 |
| I have taken a weapon to school to defend myself
| 2 | 3 | 1 |
10 | 12 | 7 |
| I have taken a weapon to school to use against another pupil
| 1 | 1 | - |
4 | 6 | 1 |
| I have never used a weapon | 62
| 47 | 78 | 24 |
17 | 46 |
Source: YJB 2004
In both cases, the weapon most commonly carried was a pen
knife, though this was followed closely by a ball bearing gun.
The survey also confirmed the gendered nature of weapon use more
generally. Whereas one in three boys in mainstream education claimed
to have carried a weapon at least once in the last year, only
one in 10 girls did so. Also, while white young people were more
likely to have carried penknives than black or minority ethnic
groups (26% against a figure of 20%), a higher proportion of black
young people admitted to having carried a flick knife than white
or Asian young people (15% against 9% and 8% respectively).
Figure 4
Potential weapons carried by young people
(% 'ever' carrying in the previous year)
YJB 2004
Importantly, the survey not only confirms more general findings
(including the study by Lemos and Crane) that young people carry
weapons for `protection'. It specifically suggests that there
is a link between knife carrying and young people's prior experience
of victimisation:
Over a third (36%) of young people in mainstream education
who have been a victim carry a knife, compared with 18% of those
who have not been a victim of crime. Similarly, 62% of young people
in excluded projects who have been a victim of crime carry a knife,
compared with 51% who have not been a victim.
YJB 2004
WAYS FORWARD
Especially in the wake of the most recent high profile knife-related
murders, there is a danger that the problem of knife crime may
be cast too narrowly as an inner-city problem which predominantly
affects young people and which is primarily of concern to black
communities. The borough-level analyses at Figure 3b suggest that
young people aged 15 and under account for a very small proportion
of all knife crime. The largest age group involved is 16-20 year
olds but this may be especially true in the case of robbery, whereas
most knife-related assaults involve suspects between the age of
16-35. It is also important to recall that public awareness of
the problem of knife crime among young people was first ignited
in 2003 with the killing of Luke Walmsley, a white teenager stabbed
to death by a fellow pupil in a school in Lincolnshire. More recently,
a 22 year old white man from Lancashire was found guilty of the
stranger murder of a young white student returning from university
in Scotland on a train in May 2006. According to the most recent
s95 figures, white victims accounted for 73 per cent of all homicides
involving `sharp instruments' in the years 2001-02 to 2003-04
inclusive.
Improved statistics and better statistical analyses should
provide a better basis for interventions to tackle knife crime;
but it is also important to recognise that—even though the
two are linked—there is a distinction to be made between
knife carrying and knife crime. This distinction has implications
for the types of intervention which are needed and they also require
immediate responses to the problem of knife crime to complement
medium- to long-term approaches to tackling some of the underlying
causes.
Statistics
Survey data, at best, provide a useful snapshot of knife
crime and knife carrying and they can probe some of the reasons
for the picture they show. However, surveys can produce contradictory
results and may be unreliable as indicators of trends. The YJB
surveys, for example, have now been superseded by the Home Office's
Offending, Crime and Justice Survey, which, in the same year (2004)
found that only 4% of 10-17 year olds had carried a knife. Sample
sizes also mean that surveys are unable to provide the fine-grained
detail which is essential to inform local strategies or to monitor
their impact.
The Home Secretary's recent announcement that knife crime
is to be recorded on the same basis as gun crime is therefore
particularly welcome. Nonetheless, it is important to recognise
not only that it will take some time before the system beds down
and begins to show trends but that some level of uncertainty will
always hang over the figures for a number of reasons. One is that
they depend on reporting to the police; but the other is that
recording depends on officers remembering to "flag"
the use of a knife when they record the substantive offence in
which a knife has been:
— threatened (but neither used nor seen).
In addition, the figures will only really come into their
own where the police are prepared to analyse them to produce their
own particular local profile of knife crime (ie by type of offence,
location, time and suspect) and to use this to inform policy and
monitor its impact. Even in the MPS which has been collecting
figures at borough level for several years, this type of approach
has been rare and, even where the will has existed, the commitment
of resources has rarely been justified because—as many respondents
told me—knife crime has not been a priority and analysts
are often over-stretched to provide management information on
issues to which targets are attached.
Tackling knife crime and knife carrying
In the short term, reducing knife crime will depend mainly
on the success of tactics for reducing the offences in which knives
most commonly come into play. That is, effective measures for
reducing robberies and violence associated with the night time
economy will pay off faster than measures to reduce knife carrying,
especially in view of the likelihood that many of the knives which
are discovered would never have been used.
Reducing knife carrying, in turn, will depend on striking
the right balance between enforcement and other measures, especially
in view of the importance in the medium to long term of breaking
the cycle whereby young people in particular feel the need to
protect themselves by carrying knives.
Enforcement in this context depends to an unusual degree
of stop and search. My own 1999 study of s1 searches in London
discovered that the majority of arrests for offensive weapons
in the capital came from searches. This poses particular dilemmas
since searches have long been a significant source of tension
between the police and sections of the public, in particular young
people and members of minority ethnic groups. Searches are always
potentially adversarial encounters which may often be inflamed
by the attitudes of either or both of the parties but only a minority
of which result in an arrest. Their disproportionate impact on
particular groups may harden attitudes against the police in such
a way as to make it less rather than more likely that these groups
will report crimes to the police or trust the police to protect
them. That is, inasmuch as these same groups may also disproportionately
carry knives, an escalation in police searches may prove counter-productive
in terms of breaking the cycle even though searches are essential
to discovering knives.
Improved technology may go some way to squaring this circle.
The use of search arches has received a lot of attention in this
context; and these have the obvious advantage of being generally
acceptable. Members of the public have become used to them in
many other situations and they are seen as non-discriminatory
since (in principle) everyone in a given location is equally subject
to being scanned. Indeed, seeing them in use may even be reassuring.
Arches, however, are cumbersome and resource intensive and figures
from BTP suggest that the yield from search arch operations at
transport hubs in terms of the number of knives detected is relatively
low.
Search wands, by contrast, are very much more flexible and
they have the advantage in traditional search contexts of obviating
the need for police officers physically to touch the subject of
the search (which, of itself, can increase the inherent tension
in these encounters). Nor do they have to be used exclusively
by the police. Some areas report their successful routine use
by pubs and clubs as a condition of entry; and some schools have
also been using them. However, the wands can produce both false
positives (metal objects other than knives) and false negatives.
For a wider issue which needs to be considered is that sharp instruments
will include many non-metal, everyday objects which may be as
capable of causing serious injury as knives, including sharpened
pencils and combs. Nonetheless, it would be worth systematically
evaluating the use of search wands for discovering knives in terms
of both their acceptability and their effectiveness.
In the medium to long term, breaking the cycle will depend
on more than education about the risk of detection and the likely
sanctions for being caught in possession of a knife. As long at
young people in particular feel unprotected and know that the
people they fear are still carrying their knives, it may be difficult
to persuade them unilaterally to disarm. Building trust in the
police is essential—not least inasmuch as young people may,
in principle, be the best source of intelligence to ensure police
interventions are effectively targeted. Hence the need referred
to above in relation to police searches to avoid scatter-gun approaches
which make all young people feel de facto suspect; but with this
also goes a need to be seen to take the victimisation of young
people seriously and to recognise young people's very justifiable
but largely neglected "fear of crime". One of the most
effective ways of providing them with the necessary reassurance
may be not only to catch those individuals they most fear and
to ensure they receive effective disposals as a result of intelligence-led
policing but also to ensure that other young people get to know
that this has happened.
26 March 2007
1
Over this period, there had been a 19% increase in the total number
of s1 searches; and the proportion of these where the grounds
were recorded as "offensive weapons" rose from 7%-9%.
Arrests for offensive weapons from searches over the same period
rose from 9%-10%. Back
2
Personal communication. Back
3
Only a minority of mainstream pupils (26%) admitted to having
ever committed an offence of any sort, though this was true of
60% of excluded pupils. Out of this subsample of offenders, the
proportion who said that one of the offences they had committed
was carrying a knife was 30% and 51% respectively. Back
| 
