ArticlePDF Available

The Nature of Collective Resilience: Survivor Reactions to the 2005 London Bombings


Abstract and Figures

Accounts from over 90 survivors and 56 witnesses of the 2005 London bombings were analysed to determine the relative prevalence of mass behaviors associated with either psychosocial vulnerability (e.g. 'selfishness', mass panic) or collective resilience (e.g. help, unity). 'Selfish' behaviors were found to be rare; mutual helping was more common. There is evidence for (a) a perceived continued danger of death after the explosions; (b) a sense of unity amongst at least some survivors, arising from this perceived danger; (c) a link between this sense of unity and helping; and (d) risk-taking to help strangers. We suggest a novel explanation for this evidence of 'collective resilience', based on self-categorization theory, according to which common fate entails a redefinition of self (from 'me' to 'us') and hence enhanced concern for others in the crowd.
Content may be subject to copyright.
Drury: Reactions to London bombings
International Journal of Mass Emergencies and Disasters
March 2009, Vol. 27, No. 1, pp. 66–95.
The Nature of Collective Resilience:
Survivor Reactions to the 2005 London Bombings
John Drury
University of Sussex, UK
Chris Cocking
London Metropolitan University, UK
Steve Reicher
University of St Andrews, UK
Accounts from over 90 survivors and 56 witnesses of the 2005 London bombings were
analysed to determine the relative prevalence of mass behaviors associated with either
psychosocial vulnerability (e.g. ‘selfishness’, mass panic) or collective resilience (e.g.
help, unity). ‘Selfish’ behaviors were found to be rare; mutual helping was more
common. There is evidence for (a) a perceived continued danger of death after the
explosions; (b) a sense of unity amongst at least some survivors, arising from this
perceived danger; (c) a link between this sense of unity and helping; and (d) risk-taking
to help strangers. We suggest a novel explanation for this evidence of ‘collective
resilience’, based on self-categorization theory, according to which common fate entails
a redefinition of self (from ‘me’ to ‘us’) and hence enhanced concern for others in the
Keywords: London bombings, resilience, panic, self-categorization
On July 7th 2005, a series of coordinated terrorist bomb blasts hit London’s public
transport system during the morning rush hour. The London bombings saw the largest
mass casualty count in the UK since World War II (Aylwin et al. 2006). The present
paper presents an analysis of survivors’ behaviors and experiences as collected in
contemporaneous newspaper data, personal accounts, and interviews. Patterns in each of
these datasets are analysed quantitatively and qualitatively in order to evaluate two
Drury: Reactions to London bombings
influential explanatory frameworks that are characteristically applied to the mass
psychology of emergencies and disasters.
On the one hand, there is the vulnerability framework, which emphasises the
psychosocial frailties of, and risks to, the public (Durodié and Wessely 2002; Furedi
2007)1. This framework suggests that, within an emergency, people are collectively prone
to pathological, irrational and maladaptive responses, particularly ‘mass panic’ (Dynes
2003). The notion of ‘mass panic’ shares with classical ‘crowd science’ the assumption
that the crowd is less intelligent and more emotional than the lone individual (Le Bon
1895) and hence reactions to an emergency will be disproportionate to the actual danger
(Smelser 1962). Simple ideas and sentiments are said to spread quickly through the
crowd in a process of ‘contagion’ (McDougall 1920). However, rather than the
emergence of ‘mental unity’(Le Bon 1895) as in the ‘rioting’ crowd, ‘mass panic’ is
understood as the dissolution of unity and social bonds in the crowd (Freud, 1921). In an
emergency, it is suggested, ‘instincts’ for personal survival override socialized responses
(Strauss 1944). The result is said to be uncoordinated and competitive behavior, such as
individuals pushing and trampling each other to reach personal safety (Schultz 1964).
In the field of mass emergency and disaster research, the notion of mass panic has
been largely discredited by the finding of orderly, meaningful mass behavior in disasters
(e.g., Sime 1990). However, some influential practitioners, including crowd modellers in
the fields of engineering and design, still draw upon the notion (e.g., Helbing, Farkas and
Vicsek 2000; see Sime 1995). Its assumptions also still influence social policy (Dynes
2003) and persist in the form of off-the-shelf clichés and popular representations of
disasters (Tierney, Bevc and Kuligowski 2006).
On the other hand, however, there is the resilience framework. This emphasizes
collective self-help, community resources, and survivors’ ability to recover and function
in the face of adversity (e.g., Dynes 2003; Furedi 2007; Wessely 2005b). For example, in
analyses of the 2001 World Trade Center disaster, resilience was used to conceptualize
the endurance of citywide social organization through the planned provision of resources
(Kendra and Wachtendorf 2001), and the improvisation of effective cooperation amongst
emergency teams through spontaneous use of informal social networks (Tierney 2002). In
some form, the concept is embodied in a number of government agencies and
organizations in the UK and USA which were set up to deal with the threat of terrorism
(e.g. the London Resilience Team, Birmingham Resilience).
Each framework clearly has a number of policy implications. If the public are
inherently ‘vulnerable’, one corollary is that they will need to be actively protected from
‘risk’ through the withholding of information which might lead to mass panic (Furedi
2007). The threat of mass panic and social disintegration also serves to justify
paternalistic social control policies and the implementation of a mistrustful ‘homeland
security’ approach, rather than social policies which might encourage public initiative
and independence (Dynes 2003).
Drury: Reactions to London bombings
If people are naturally collectively resilient, however, rather than being treated as part
of ‘the problem’, crowds can be trusted with information during emergencies (Proulx and
Sime 1991) and communities should have greater involvement in their own defence and
psychological recovery (Jones, Woolven, Durodié and Wessely 2006; Wessely 2005a).
For example, some argue that being open over the nature of terrorist threats is the best
way to ‘vaccinate’ the public against panic (Glass and Schoch-Spana 2002).
In this paper, we ask which of the two explanatory frameworks—vulnerability or
resilience—has more resonance in the reported experiences and behaviors of those caught
up in the emergency events of July 7th 2005. Thus, based on the previous literature, a first
question concerns the extent to which survivors and witnesses describe the events as
‘panic’ rather than referring to calm or orderly evacuation behavior amongst survivors
and witnesses.
On the surface, greater reference to ‘panic’ than to ‘calm’ or ‘order’ would suggest
support for the vulnerability framework. Such references might therefore lead us to
expect extensive evidence of personally selfish behaviors of the type that characterizes
mass panic: uncoordinated, competitive acts, such as individuals carelessly neglecting
others in need, or pushing and trampling others to reach personal safety (Shultz 1964).
Yet, as mentioned, the term ‘panic’ is a commonsense cliché. The term is often used
when what is fact is being described is simply flight from the source of danger
(Quarantelli 1960). Its use by survivors and witnesses therefore may be a gloss rather
than a description of what people actually did. Thus the second question is whether any
such observations of personal selfishness were more or less common than observations of
mutual helping. If perceived helping outweighed personally selfish acts this would be in
line with the resilience framework.
While a finding that resilience prevailed during the London bombings might not
appear particularly novel, the present analysis seeks to go further than previous research
by exploring its social-psychological basis in crowds of survivors. Understanding the
psychological nature of resilience and the conditions that facilitate it will have profound
consequences for the planning of emergency evacuation procedures, the response of the
emergency services, and aftercare practices.
In the field of disaster research, resilience has a number of features including
improvised use of informal relationships to achieve goals and the ability of organizations
to provide backup and coordination (Aguirre 2006; Dynes 2003; Kendra and
Wachtendorf 2001; Tierney 2002). The question motivating the present research is
whether such improvised, adaptive sociality can occur not just in structured institutions
and organizations, but also in unstructured ad hoc crowds of survivors in the very midst
of an emergency.
There are a number of models of mass emergency behavior that suggest a social-
psychological basis for resilience: in particular normative approaches and affiliation. We
briefly outline these before describing our own approach, which is based on the principles
Drury: Reactions to London bombings
of self-categorization theory (Turner 1982; Turner, Hogg, Oakes, Reicher and Wetherell
Approaches to resilience in emergency crowds
Emergent norm theory (ENT; Turner and Killian 1957, 1972, 1987; Turner 1964)
represents one of the earliest attempts to transcend the longstanding and pernicious
irrationalist tradition in the general field of crowd psychology (e.g. Le Bon 1895). For
ENT, all social and collective behavior is a function of norms. But since ‘riots’ or
disasters are ‘extraordinary’ events, where everyday rules of conduct do not necessarily
apply, new norms have to be developed. These are said to emerge from a process of
interpersonal interaction (‘milling’, rumour and ‘keynoting’ by influential individuals),
until a shared definition of the situation is agreed (Turner and Killian 1972).
Early versions of ENT suggested that crowd unanimity was an ‘illusion’ and stressed
(‘personality’/motivational) variety within the crowd (Turner and Killian 1972). By
contrast, R.H. Turner’s most recent formulations (1994a, b, 1996) take into account new
developments in sociology (such as framing) to place a greater emphasis on unity in the
crowd and the role of shared ‘worldviews’, frames of reference, shared perceptions of
risk, and a shared ‘moral sense’, in explaining this unity. However, we agree with
McPhail (1991) when he argues that the essence of ENT has remained unchanged. As he
says, for ENT it is still interpersonal interaction that makes collective behavior possible.
Turner and Killian’s model of shared sociality is grounded in that of Sherif (1936), for
whom face-to-face discussion between individuals is the source of ‘groupness’.
Empirically, ENT suggests that the emergence of collective behavior will be a
relatively drawn out process: the larger the group, the longer the discussion to reach
agreement, and hence the greater the delay in acting during an emergency (Aguirre,
Wenger and Vigo 1998). Yet even some supporters of ENT admit that crowds can
sometimes remain united but shift norms very quickly in relation to changing contexts
(Wright 1978). As Reicher (1984) argues, extended milling and discussion is not always
necessary for the acceptance of new norms even in a novel situation.
From the 1970s onwards, ENT came to place more stress on the (constraining,
enabling) role of the (existing) normative order and social structure in shaping collective
behavior (though, as Weller and Quarantelli 1973 argue, Turner and Killian still saw
emergent norms as more important). This was still not enough for Johnson (1988),
however, who criticized their account in which ‘panics’ result from an emergent
definition of the situation in which norms of cooperation no longer apply and in which
selfish pursuit of individual ends are viewed as legitimate (Turner and Killian 1987).
Johnson’s is perhaps the dominant normative model today (Aguirre 2005). He abandons
the notion of emergence and instead posits a simple continuity between mundane social
situations and emergencies; both are said to be structured by pre-existing norms and roles
which guide and constrain behavior, ensuring sociality and delimiting individualized
Drury: Reactions to London bombings
panic in a crowd. Johnson’s normative approach would therefore seem to explain some of
the evidence of routine social behaviors identified in emergencies, such as queuing, men
helping women more than vice versa (i.e., gender role conformity), and the greater
assistance offered to the elderly and infirm than the able bodied (Feinberg and Johnson
2001; Johnson 1987, 1988).
While there is now an accumulated mass of evidence to support the predicted
continuity between mundane and disaster behavior suggested by Johnson’s normative
model, there are still some behavioral discontinuities which need to be explained. For
example, while it might be normative to help someone in distress in everyday
circumstances, it is surely stretching the concept of ‘normative’ to explain the risks
survivors take to help strangers, as has been found to happen in some emergencies
(Clarke 2002).
A more recent development is the affiliation approach, which is based on principles
from attachment theory (Mawson 2005). This suggests that: (i) in the face of threat, we
are motivated to seek the familiar rather than simply exit; and (ii) the presence of familiar
others (i.e. affiliates) has a calming effect, working against a ‘fight or flight’ reaction.
This theory would explain why people often prefer to remain with loved ones even at risk
of death rather than escape alone (Sime 1983).
While the affiliation approach explains the patterns of behavior when the crowd is
made up of small groups of families or friends (as in the Beverly Hills Supper Club fire;
Cornwell 2003; Johnson 1988), emergencies often involve large numbers of people who
do not know each other or become separated from their associates. Yet in these events too
there is often evidence of mutual helping and even self-sacrifice (Ripley 2005). Further, a
corollary of the affiliation approach is that where there is extreme danger and people are
amongst strangers, there will be mass panic (Mawson 2005). The research evidence that
mass panic is rare if not ‘mythical’ (Dynes 2003; Sime 1990) seems to highlight a flaw in
an otherwise well respected theory (see Aguirre 2005).
While each of these approaches to sociality in mass emergency behaviour has its
strengths, when we measure them against some features of the review evidence, we are
directed to a crucial absence. The emergent togetherness, solidarity or ‘community spirit’
observed in emergencies and disasters is surely one of the most important and striking
forms of resilience. An emergency or disaster, far from dividing people into instinct-
driven competitive individuals, can serve to create a powerful sense of unity and hence
mutual support amongst survivors both during the immediate crisis (Aguirre 2005; Clarke
2002) and in the recovery period afterwards (Fritz 1968; Fritz and Williams 1957). ENT
at least had the advantage of suggesting that there was some ‘new’ or emergent sociality
that comes out of the emergency itself. Johnson’s (1988) normative model and the
affiliation approach (Mawson 2005) have both lost this.
We suggest that a fuller explanation of togetherness and hence crowd resilience in
emergencies and disasters requires going beyond a reliance on pre-existing norms and
Drury: Reactions to London bombings
interpersonal relationships. Put differently, what is needed to complement the strengths of
existing approaches is a model of mass emergent sociality or collective resilience, i.e.
coordination and cooperation within a crowd of strangers. While the vulnerability
framework emphasizes the dissolution of social bonds, and recent normative and
affiliation models stress their maintenance, we also need to look at the possibility of the
creation of such bonds—yet without relying on empirically untenable and conceptually
individualistic notions of interpersonal interaction2.
A new model of mass emergency behavior to be explored here is derived from self-
categorization theory (SCT; Turner 1982; Turner, Hogg, Oakes, Reicher and Wetherell
1987). SCT suggests that feeling and acting with others as part of a group, crowd,
organization or even a nation, operates through self-categorizations, which may range
from personal self-categorizations (definitions of what makes us unique) to shared,
collective self-categorizations (definitions that classify us with others). Cognitively
categorizing oneself with others on some context relevant dimension (e.g. ‘we are all men
in contrast to women’) tends to heighten perceptions of similarity and unity with these
others. It also has emotional consequences; the shift from ‘me’ to ‘we’ means a greater
commitment and loyalty to the group, who are now seen more like ‘self’ than as ‘other’.
This, in turn, can mean acting in their interests in various ways, even where they are not
known personally (Drury and Reicher 1999; Levine, Prosser, Evans and Reicher 2005).
In this account, one of the bases for seeing oneself as a member of group with others
is the perception of a common fate (Campbell 1958; Turner et al. 1987). In line with this,
research on collective conflict has shown that the perception of an external threat, which
is perceived to affect everyone present indiscriminately, can transform an aggregate of
disparate individuals into a psychologically unified crowd (Reicher 1996; Stott and
Reicher 1998). This in turn would be expected to produce some of the solidarity effects
suggested by SCT outlined above.
If the SCT-based approach being suggested here is right, we would expect to find
that, even where people were mostly amongst strangers during the London bombings, if
there was a common perceived danger of death it would create a sense of shared identity.
Thus we would expect a positive association between such shared identity (as reflected in
references to enhanced ‘unity’, ‘togetherness’ and so on) and helping. Hence, finally,
rather than being concerned only with affiliates, we would expect at least some people to
help strangers even at risk to themselves personally.
Background: The London bombings of July 7th 2005
The London bombings of July 7th 2005 consisted of four explosions (three on the
London Underground and one on a London bus) which killed 56 people (including the
four bombers) and injured over 700. Those in the bombed underground trains were not
reached by the emergency services immediately, and were left in the dark, with few
announcements, and with no way of knowing whether they would be rescued, or whether
Drury: Reactions to London bombings
the rail lines were live. There were fears by both those in the trains and the emergency
services of further explosions. Triages were set up close to the explosions. Some, though
not all, those injured were ferried to various London hospitals; others made their own
way to work or home—though with some difficulty as transport in London was
massively disrupted and didn’t return to near-normal till the evening. Subsequent
research established that the events led to substantial stress in around 30 per cent of
Londoners, though most did not desire professional help (Rubin, Brewin, Greenberg,
Simpson and Wessely 2005).
Data and sample characteristics
In the months following the bombings, we sought to gather as much data as possible
on people’s experiences, perceptions, behaviors, and feelings. These can be grouped into
three main orders of data: contemporaneous newspaper material, archive personal
accounts, and primary data.
Contemporaneous newspaper accounts. One hundred and forty one accounts from
18 newspapers were collected. These data comprise the news items produced in the days
immediately after the events, which include short statements from eye witnesses and
survivors, as well as from commentators (such as journalists and public figures). Given
the limited information provided in the newspapers, there is no way of verifying how
many of the accounts are from the same people quoted more than once.
Personal (archive) accounts. Accounts from one hundred and twenty seven
witnesses and survivors were collected: 26 (transcripts and written evidence) came from
the London Assembly review hearing report (June 2006); 68 came from news websites
(mostly BBC); 22 were blogs or message board contributions; nine came from features in
newspapers; one was from an autobiographical book; and one was from a BBC radio
documentary programme.
Accounts from the London Assembly data (which was read-only) varied in length
from 1500 to 15,000 words, the mean average account being around 3500 words long.
Barring the autobiography, the other accounts (which were editable) totalled
approximately 44,000 words, and ranged from 2 lines to 3 pages with the mean length
being 380 words. One hundred and four of these personal accounts were recorded on the
day or in the immediate aftermath of the bombings, whereas the other 50 were produced
in the 12 months after the event.
Eighty one were survivors who were on one of trains or the bus that was bombed; the
rest were firsthand witnesses: 26 saw or heard one or more of the bombs, 19 were in the
area of the blast, and one was from the emergency services. Divided by location of the
explosions, 56 were at Kings Cross, 26 were at Aldgate, 27 were at Edgware Road, 13
Drury: Reactions to London bombings
were present at the bus bombing (Tavistock Square), and four more gave accounts from
more than one location.
Twelve survivors were classified as having only slight injuries (‘walking wounded’),
11 had severe injuries requiring hospitalization, 13 reported as having PTSD, and another
six had both physical and psychological trauma. There were insufficient data on the
remaining survivors. Including both witnesses and survivors, 70 of those who provided
information on their gender were male and 47 were female. All except six provided
names, and most of these provided surnames as well as first names. We were therefore
able to confirm that only one person appeared in both these data and our primary data
(discussed below).
Accounts from an additional twenty seven people who were not direct witnesses or
survivors but were people simply in London at the time were also gathered (26 from
websites and one from a newspaper). They are therefore excluded from the foregoing
analysis, except where indicated.
Primary data respondents. Advertisements were carried in newspapers and a
website3 we set up asking those caught up in the explosions to send us their personal
accounts. Links to our request were provided in the websites of both London Resilience4
and the ‘Kings Cross United’ survivors’ support group. The website contained a set of
questions, including the following:
Tell us if you were scared, how scared you felt.
Tell us how much danger you felt you were in.
Tell us if others were present and what sort of crowd it was.
Tell us whether you felt any sort of bond or any sort of unity with others
who were there.
Tell us about the reactions of others: from what you could tell, how did
they feel and how did they behave?
Tell us if you saw any examples of people helping others or else simply
looking out for themselves and ignoring others.
Tell us if you saw any examples of cowardice or heroism.
Tell us if your feelings or perceptions towards other people around you
changed over the course of the event. If so in what way?
Fifteen email accounts were received, averaging about 1410 words each. Such email
responses were necessarily brief. Respondents were therefore asked to be interviewed so
that their responses could be probed in detail. Eight agreed to this request. Four further
interviewees were recruited via advertisements in the press, approaching support
organizations and official bodies, and through snowballing personal contacts.
Both the website questionnaire and the interview schedule were structured through
standard interview techniques to try to identify processes of interest yet to avoid
Drury: Reactions to London bombings
interviewer bias. Thus each new interview theme began with open questions (e.g., ‘How
did people respond?’) before being followed up with closed questions about specific
behaviors (e.g. ‘Did you see anybody helping? Did you see any selfish behavior?’)
There was a potential hazard of causing distress through the interview as speakers
relived the trauma. The interviews took place at least two months after the events, which
is when clinicians screen for PTSD; survivors would have had a diagnosis by then and
therefore could be warned that the interview could be distressing. We also took steps to
minimize the risk by ensuring that (a) the interviewer had a clear and graded set of steps
of take if there were any signs of distress during the interview; and (b) the nature of the
interview (i.e., research not therapy) was clear throughout the session. Ethical approval
was granted to the project by the University of Sussex School of Life Sciences Ethics
Committee in September 2005. Each interviewee provided informed consent for their
interview responses to be analysed and published anonymously.
The twelve interviews each lasted between 45 minutes and an hour, and produced a
mean of 6875 words per verbatim transcript. Six of the interviewees were men and six
women. Of the seven email-only respondents, six were women. Ages were not recorded.
Five of the interviewees and four of the email-only respondents were survivors; the rest
were witnesses. Of the survivors, three suffered minor injuries, one of these requiring
hospital treatment; one of them was also diagnosed as suffering from PTSD.
In summary, excluding the contemporaneous newspaper accounts and allowing for
some unverified overlap between the personal accounts and the primary data, the data
comprise accounts from over 146 survivors and witnesses, most of whom (90) were
actually caught up in the explosions. Based on the London Assembly Report (2006, p.
73) estimate that 4000 people were ‘directly affected’ by the four explosions, this means
the sample was around five per cent of this total population.5
Data analysis principles
Triangulation. A first methodological principle was that of triangulation.
Widespread agreement between accounts and across different sources would give us
some confidence in any claims about the main contours of behavior and perception
amongst those caught up in the explosions—for example, that mutual aid was or was not
Hierarchical quality of data. A second methodological principle was that the
different sources of data were each of different value and quality. While each could
contribute to the analysis, they would not do so equally.
The newspaper report dataset was the largest and had the advantage of being
contemporaneous. At the same time, however, it was both the most superficial and the
most partial. Media agendas circumscribe the questions that witnesses and survivors are
asked, and determine which features of such accounts finally end up on the printed page.
For example, the common sense image of ‘mass panic’ could operate as a frame through
Drury: Reactions to London bombings
which the media filters information on the events, excluding some elements and
emphasizing or adding others. Thus, while these data provided some evidence of the
types of behaviors, emotions and perceptions observed during the events, there was little
indication of the processes behind these observations.
The corpus of 127 archive personal accounts went beyond these snippets of
observations and included more reflections on and reasons for behavior. (Some accounts
were also contemporaneous, and so cannot be criticized for being simply post hoc
reconstructions.) However, in the production of these accounts, the agenda and aims are
again different than ours as researchers. For example, people were not asked and tended
not to volunteer whether or not their behavior during the events was motivated by
affiliation, and whether there was a strong sense of common identity (or a sense of
Hence there was a need for primary data—in other words, to ask systematically not
only what people saw and did, but about their various possible motivations. This dataset,
then, while the smallest, is the most informative, detailed and elaborate.
Coding and thematic analysis
The data were subject to content and thematic analyses, each of which entailed coding
material in relation questions of interest. For example, for each source we checked for
and grouped all references to the term ‘panic’. For the contemporaneous newspaper
material, this comprised a count of positive or negative references per article, whereas for
the archive personal accounts and primary data this comprised a ‘yes’ or ‘no’ as to
whether each person used the term.
Personally ‘selfish’ or competitive behaviors were operationalized as any behaviors in
which someone acted at the expense of another when they had a choice to do otherwise.
Examples included someone elbowing another out of the way in order to get out, or
someone ignoring another’s request for help when they had a choice to do otherwise. The
same procedure was carried out for ‘helping behavior’, which was defined as anything
done or said with the purpose of assisting another. This included comforting others,
offering them bottles of water, physically helping people up or along, giving people
information or directions, giving first aid, tying tourniquets, and applying makeshift
bandages. (Where possible, ‘help’ was broken down into three categories: ‘number I
helped’, ‘number who helped me’, ‘number I saw being helped by others’.)
The other categories coded and counted included perceived threat (i.e. references to
how much danger people felt they were in); fear (own and others); ability to help
(references to physical or other limits to the help people could give each other);
references to calm and orderliness; concern for others (friends and family versus
strangers). Common identity was operationalized in terms of references to unity (e.g.,
solidarity, togetherness).
Drury: Reactions to London bombings
In each case, the size of a piece of coded text within this scheme varied from a
sentence to a multisentence chunk. Sentences or chunks were coded according to the rule
of thumb: assign the single most appropriate code in the scheme (Miles and Huberman
1994: 65).
Are the events characterized as ‘panic’?
In the contemporaneous newspaper data, 57 eyewitness and survivor accounts used
the term ‘panic’. (See Table 1, below, for summaries of data on all measures.) However,
there were 20 such accounts that explicitly denied there was panic, while 37 referred to
‘calm’ amongst those affected by the bombs, and 58 described the response as an ‘orderly
In the archive personal accounts, 46 people described the events as ‘panic’ or referred
to people ‘panicking’; but 53 of them (and indeed 17 of the same people) also
characterized the evacuation as ‘orderly’6.
Interviewees and email respondents were much less ready than those in the
newspapers and personal accounts to use the term ‘panic’. In describing the behavior of
the rest of the crowd, only two respondents endorsed the term ‘panic’. But even here,
when asked what she meant, one of these respondents replied simply that people were
Int: Do you think anybody panicked?
LB12: In our carriage no, or if they did they panicked inwardly, they
didn’t express their panic. I mean there was no screaming in our carriage I
mean people were trying to get out the door but they weren’t trying to get
out of the door stupidly.
The same respondent also described others’ behavior as overwhelmingly ‘calm’. Five
other respondents straightforwardly denied that people panicked, said that people ‘started
to panic’ but were ultimately calm or controlled (two), said that they didn’t see any panic
(two), or reserved the term for one individual or a small minority in the crowd
‘hyperventilating’, ‘screaming’ or becoming ‘hysterical’ (four).
Drury: Reactions to London bombings
Table 1. Summary statistics for all measures
newspaper accounts
Archive personal
Primary data:
Interviews and emails
Total 141 127 19
‘Crowd panic’ 57 33 2
‘No panic’ 20 - 5
‘Crowd calm’ 37 - 13
‘Crowd orderly’ 58 52 3
‘I helped’ 57 42 13
‘I was helped’ 17 29 7
‘I saw help’ 140 50 17
‘Selfish’ behaviors 3 11 4
With strangers - 57 15
With affiliates - 8 4
Fear 31 89 6
Possibility of death 70 68 12
Not going to die - 2 1
Common fate 0 11 5
Unity 7 19 11
Disunity 0 0 1
Unity / ‘I helped’ 7 / 3 19 / 11 11 / 10
Unity / ‘I was helped’ 7 / 2 19 / 8 11 / 6
Unity / ‘I saw help’ 7 / 5 19 / 12 11 / 10
Unity / total ‘selfish
behaviors’ 7 / 2 19 / 0 11 / 2
Concern for affiliates 12 24 3
Concern for strangers 21 24 9
No concern for affiliates - 7 4
Risks to help strangers - 12 5
In describing their own behavior, only one said s/he ‘panicked’, and four others said
they ‘felt’ panicky. However, these were again references to feelings (of fear) rather than
overt behavior. The rest of our respondents were either explicit that they did not panic (2)
or simply did not refer to panic in describing their own behavior. Thirteen respondents
explicitly said that people in the crowd were mostly calm:
Drury: Reactions to London bombings
It took about twenty twenty five minutes before we got out … and some
people were really itching to get off the train so more people the more
agitated people were not being shaken up they felt they were, even though
they wanted to get off at the same time so it was quite a calm calm evenly
dispersed evacuation there wasn’t people running down the train
screaming their heads [off]. It was very calm and obviously there was
people crying [ ] 7
but generally most sort of people were really calm in
that situation, which I found amazing. (LB 1)
Seven said that they themselves felt calm, and three described the crowd as orderly.
Helping versus personal ‘selfishness’
In the contemporaneous newspaper articles, there were 57 reports from people who
said they helped others, 17 accounts from people who were helped by others, and 140
further witness observations of help between survivors. This help included people
reassuring each other (by hugging or talking), pulling people from the wreckage, and
holding people up as they evacuated. There were only three eyewitness reports of
personally selfish behaviors in the newspapers. An example is at the bus bombing where
a witness described people elbowing each other aside in their efforts to get away.
Forty two of those providing archive personal accounts reported helping others (most
of them helping more than one person), 29 reported being helped by others, and 50
reported witnessing others affected by the explosions helping others (most of these again,
including the train drivers, helping more than one person).8
There were only 11 personal accounts of observed behavior that could be described as
personally selfish (such as the case of someone described as ‘selfish’ for phoning work to
cancel his meetings rather than call the emergency services); six of these were cases
where the speaker suggested that another survivor behaved ‘selfishly’ to them or to
someone else. Four people glossed their own behavior as ‘selfish’. However three of
these were coded as being unable to help others, usually because of some physical
impediment, even where they wanted to (cf. Cornwell, Harmon, Mason, Merz and Lampe
2001). Therefore we might suggest that such self-reports of ‘selfishness’ could be cases
of survivor guilt (Titcher and Frederick 1976).9
Of our 19 respondents, 13 reported at least one instance of themselves helping
another—ranging from comforting them to giving them water. Those that didn’t report
helping were not themselves in a position to help: two were not near any survivors, one
was in plaster, and the other attributed his behavior to shock. Seven of our eight survivors
reported being helped by others; the eighth didn’t give enough information. All of our
respondents (except two who didn’t give enough information) reported witnessing people
helping others, and in most cases this helping was described as widespread, despite
difficult conditions (such as darkness, injury and pain):
Drury: Reactions to London bombings
I remember walking towards the stairs and at the top of the stairs there was a
guy coming from the other direction. I remember him kind of gesturing;
kind of politely that I should go in front—‘you first’ that. And I was struck I
thought God even in a situation like this someone has kind of got manners
really. Little thing but I remember it. (LB 11)
Seven of our respondents said they felt ‘selfish’ or guilty for being overly concerned for
their own personal safety. However, again in line with the suggestion that this may be no
more than survivor guilt, only one of these described actually neglecting someone when
in a position to help. Seven respondents were explicit that they had witnessed no selfish,
competitive, or similarly morally reprehensible behavior from other people:
I didn’t see any uncooperative activity, I just saw some people who were so
caught up in their own feelings that they were kind of more focused on
themselves but I didn’t see anyone who was uncooperative. I didn’t see any
bad behavior” (LB 4)
However, two respondents described one individual being concerned with his mobile
phone when they thought he could have been helping, and one described people ‘ignoring
others, walking past’. This makes a total of four selfish acts witnessed or carried out by
three of our respondents.
Thus, while, as we noted earlier, some people used the language of panic, their
account of their actual behavior did not match the classic description. The most that could
be said is that some people expressed the fear, but not the behavioral responses, usually
associated with ‘panic’.
Were people amongst strangers?
The affiliation model would suggest that the widespread helping noted above
occurred because people were amongst family members or other people they knew. The
model also predicts panic if people are with strangers in a situation of extreme danger.
Therefore, to the extent that the helping noted above occurred even though people were
in fact with strangers rather than affiliates, affiliation cannot be the major explanation,
and self-categorization is a possible antecedent of the collective resilience observed.
In the contemporaneous newspaper material, there are no figures on the number of
survivors who were with people they knew, although many of the reports describe those
affected as ‘commuters’, with the implication that most people were with strangers.
Of those providing archive personal accounts who gave information on who they
were with, only eight people reported being with friends or family when the bombs
Drury: Reactions to London bombings
exploded, while 57 reported being amongst strangers. This includes 48 people who were
actually on the trains or bus that exploded.
Among our 19 respondents, only four people (one interviewee and three email
respondents) were with friends or family.
Was there a perceived danger of death?
Conceivably, the evidence shown here of people helping strangers could be
explicable simply in terms of the danger being (perceived to be) passed. In other words,
once the bombs had exploded, perhaps people felt that there was no longer a threat of
death; hence the help they gave was personally risk- or cost-free. If this is the case, the
present evidence could easily be accommodated by the ‘mass panic’ model. If on the
other hand there was still a perceived danger after the explosion, the widespread helping
behavior noted above is more consistent with a resilience approach to mass emergency
behavior, and with self-categorization theory in particular.
In the contemporaneous newspaper accounts, 31 eyewitnesses reported experiencing
fear or observing it in others; and 70 of them reported thinking they might die.10 This
figure of around 50% perceiving a threat of death is noteworthy prima facie evidence that
people close to the explosions still felt in danger even after the bombs had gone off.
Indeed this makes sense; for an unexpected explosion is likely to make people feel less
safe immediately afterwards as it renders the world much more dangerous and
In the archive personal accounts, there were 89 total reports of fear: 39 self-reports of
own fear (and only four denials of fear), and 50 reports of observed fear in others (and
only one denial). There were in addition a total of 68 reports of anticipations of death: 44
said (and only one denied) they thought they might die; and 24 said (and only one denied)
that they could see others thought they might die. The personal accounts are also useful
for the details they provide of why people felt in danger even after the bombs had gone
off. Possible sources of death mentioned by survivors included the tunnels collapsing,
collision with an oncoming train, smoke and fire, electrocution on live rails, and
secondary explosions.
Among our respondents, there were six reports of fear (five of own fear, and one an
observation of others), though it seems that this fear was not constant, and that other
concerns may have been greater:
My initial feelings of anxiety did turn to being scared early on but when it
became obvious that I would have to ensure my colleague got home the
challenge of that overtook and feelings of worry or fear I had. (LB 16)
Twelve of our respondents were explicit that they thought they might die, and nine of
these (plus one who didn’t feel in danger himself, LB 6) said that others appeared to think
Drury: Reactions to London bombings
they might die. Again, possible sources of death cited included smoke (two respondents),
more bombs going off (six), suffocation (one) and fire (two). It is also interesting and
important to note that there was little difference in proportions between survivors (seven
out of nine thought they might die) and witnesses (five out of ten thought they might die)
in the expectation of death, making the point that objective proximity to the bombings
wasn’t necessarily the best predictor of subjective danger.
Common fate and unity
Evidence that there were feelings of common fate (reflecting the shared danger) and
hence of unity in the crowd would be consistent with the self-categorization approach
proposed here, according to which survivors helped strangers due to a common identity.
In the contemporaneous newspaper accounts, there were seven references to unity
from witnesses and survivors. However, while some witnesses described isolated
individuals who behaved in ways apart from the rest of the crowd (e.g., ignoring others),
there were no statements from survivors or witnesses referring to crowd disunity,
individualism or fragmentation.11
In the archive personal accounts, eleven people (ten survivors and one witness)
describe feeling a common fate with others caught up in the bombing. Nineteen (five
‘common fate’ people plus 15 others) said they felt a sense of unity with others during the
event. (There were also a number of references to the ‘Blitz spirit’, a cliché referring to
unity and resilience among those surviving the air attacks on Britain by German bombers
during the Second World War.) However, while these numbers are again small, it is
important to note that no survivors or witnesses contradicted them by describing disunity,
conflict or individualism in the crowd, or otherwise denied that there was a common fate
or sense of unity.
It was only in the primary data of respondents’ accounts that survivors and witnesses
were actually asked about feelings of unity; this was not a topic on the agenda of those
gathering the other data. Among respondents, references to unity in the crowd were not
only typical but also highly elaborate.
Thus eleven respondents (nine of our twelve interviewees plus two of our additional
seven email-only respondents) were explicit that there was a strong sense of unity in the
crowd; i.e. that they felt it themselves (nine of them) and/or saw it in others (nine of
them). Indeed respondents sometimes mentioned this before the topic was introduced by
the interviewer. They also used a variety of their own terms to describe the experience—
‘unity’, ‘together’, ‘similarity’, ‘affinity’, ‘part of a group’, ‘everybody, didn’t matter
what colour or nationality’, ‘you thought these people knew each other’, ‘teamness’[sic].
These were in turn associated with emotional references to others in the group, e.g.
‘warmness’, ‘vague solidity’, ‘empathy’. Such rich descriptions were sometimes
complemented by numerical ratings that some of them were able to provide for the
strength of this feeling of common identity (8/10, 9/10, 100%, 10/10). Moreover, some
Drury: Reactions to London bombings
speakers explicitly contrasted this positive feeling of unity in the emergency with the
unpleasant sense of competition and atomization in relation to other individual public
transport users they experienced ordinarily:
Int: Can you say how much unity there was on a scale of one to ten?
LB 1: I’d say it was very high I’d say it was seven or eight out of ten.
Int: Ok and comparing to before the blast happened what do you think the
unity was like before?
LB 1: I’d say very low—three out of ten, I mean you don’t really think
about unity in a normal train journey, it just doesn’t happen you just want
to get from A to B, get a seat maybe.
Where respondents offered an explanation for the feeling of unity, they attributed it to
their shared experience of threat and danger. Thus five of the interviewees who described
unity linked this to the common experience of the bombing, as illustrated in the following
I felt that we’re all in the same boat together [ ] and then for the feelings
that I was feeling could well have been felt by them as well ‘cos I don’t
think any normal human being could just calmly sat there going oh yeah
this is great [ ] it was a stressful situation and we were all in it together and
the best way to get out of it was to help each other … yeah so I felt exactly
I felt quite close to the people near me. (LB 1)
Only one interviewee described not feeling unity with others.
Is unity associated with helping (or personal selfishness)?
The self-categorization approach predicts that evidence of unity should at least to
some extent be associated with helping behavior. In the contemporaneous newspaper
accounts, three of the seven eyewitnesses who reported feeling unity also reported
helping (each more than once), two reported being helped and five reported seeing others
help. Only one who reported unity did not report help given, received or observed.
Moreover, of the seven reporting unity, only two also reported receiving, participating in
or witnessing personally selfish acts.
In the archive personal accounts, eleven of the 19 survivors or witnesses who
described feeling unity with others said they helped someone else; eight of the 19 were
helped by others; and twelve of the 19 observed help. While there are more reports of
help than this (i.e. more people reported help than reported unity), the pattern is
important. In short, where there was unity there tended to be help (but help didn’t
Drury: Reactions to London bombings
necessarily predict unity). None of those 19 who reported seeing or experiencing unity
reported any personally selfish acts.
Of our eleven respondents who described unity (felt or observed), ten described
helping others, six described being helped, and ten described seeing others help each
Int: And was there this kind of sense of unity generally with people there
who were walking as opposed to just…
LB 2: Yeah I think people were yeah I think people were just helping each
other out giving directions and stuff.
Only two of the 11 who saw or felt unity described seeing others engage in selfish
acts (the person who used his phone apparently rather then help others). Thus, for all
three datasets, those who reported unity reported fewer total personally selfish acts than
acts of help seen, given or received (see Table 1).
Concern for affiliates versus taking risks to help strangers
We have seen (i) that helping was commonplace among survivors, (ii) that most survivors
were amongst strangers and, (iii) that the perceived threat of death was present even after
the bombs had exploded. Hence we can infer that, rather than panicking when faced with
danger and the unfamiliar, at least some survivors helped strangers. This is more in line
with the suggestion of SCT that the emergency brought people together rather than with
the predictions of the other models.
However, the case for SCT against the affiliation model in particular would be further
strengthened by (i) evidence that affiliates were not survivors’ only or overriding
concern; and (ii) any explicit examples of people putting themselves at further risk to
help strangers.
In the contemporaneous newspaper data there were 12 reports of people showing
concern for their friends and loved ones, but 21 reports of concern for others (strangers).
In the archive personal accounts, 24 people could be classed as expressing concern
for affiliates: 22 reported trying to contact their family members as soon as possible and a
further two reported emotional concern for family who were not in danger. Seven
explicitly reported no such concern for affiliates. Twenty four expressed concern for
There were 12 examples of people being observed or reporting risking their own
safety to help strangers. An example is people staying to help others when they were
themselves were able to move away from what they perceived as a likely site of a further
explosion or tunnel collapse.
Among both our witness and survivor respondents, there were three expressions of
concern for family members not present – although two of there were cases of the
Drury: Reactions to London bombings
survivor wanting to reassure others that he was okay rather than reassuring himself that
they were okay. Four more were explicit that they were not concerned for their families.
Nine respondents expressed concern for strangers:
LB 7: I felt a lot of concern really, I felt really sorry for this poor guy that
I saw sitting on the seats, this guy that had just lost his leg, I don’t know
it’s hard to put it on a scale, I guess probably 10 cos that’s the worst, you
know that’s the most sorry I have ever felt for anyone, so yeah.
Int: Ok and also did you feel concern for people who weren’t there, like
family and loved ones, thinking, ‘oh what’s happened to them I hope they
are all right?’
LB 7: Um don’t think so no.
Three of our respondents referred to the helping behavior of other survivors as
‘brave’ or ‘risky’:
This woman that came and talked to me, I think she was quite brave, she
had been on the platform when it happened and they were just evacuating
the tube station and she said she started seeing people walk out and she
stayed in the tube station that I was there, for ages, and she was first aid
trained so she was kind of running around trying to do what she could.
(LB 7)
One witness described himself carrying on helping others despite his awareness of the
possibility of secondary explosions. Another witness gave examples of where he had
helped other people despite his perceived danger of death. (None of this includes
accounts of the behavior of the emergency services which was also described as ‘heroic’
because of the risk of death—usually attributed to possible secondary devices.)
In describing the London bombings of July 2005, the term ‘panic’ was used by a
number of witnesses and survivors—and, indeed, more so by commentators who did not
witness events directly. Yet the concrete and detailed descriptions of survivors’ behaviors
tell the opposite story. Rather than personal selfishness and competition prevailing,
mutual helping and concern was predominant amongst survivors, despite the fact that
most people were amongst strangers rather than affiliates. There is also evidence that this
helping behavior took place in spite of perceived danger rather than because people felt
that they were now out of danger.
Drury: Reactions to London bombings
In many ways what has been described here reflects a familiar pattern. Comparisons
with the World Trade Centre disaster of September 11th 2001 are obvious, not least
because, like the London bombs of 2005, this was a terrorist attack with a large civilian
casualty count. Analyses of 9-11 refer to the relative absence of panic (Blake, Galea,
Westeng and Dixon 2004), the calm and orderliness of the evacuation (Proulx and Fahy
2003), and the frequency of helping and acts of ‘mundane heroism’ amongst strangers
(Connell 2001; Tierney 2002).
Resilience in individuals, organizations and crowds
In psychology, there is a long established developmental, clinical and psychiatric
literature on resilience, individual differences in which are explained in terms of a
combination of both genetic and acquired characteristics, including early experiences and
attachments, repertoires of knowledge, as well as ongoing family, peer, school and work
relationships (Williams 2008). This kind of framework has also been employed in
organizational studies to explain how people adapt to stress (Haslam 2004).
In relation to disasters and terrorist attacks, there has until now been no group-level
model of resilience to fill the gap between, on the one hand, accounts of individual
resilience (e.g. Noppe, Noppe and Bartell 2006) and, on the other, accounts of the ability
of organizations to improvise and function in the face of attack (e.g., Dynes 2003;
Tierney 2002; Tierney and Trainor 2004). The analysis presented here thus offers a prima
facie case for a new conceptualization of resilience in unstructured crowds in
emergencies, which we have termed collective resilience. In the London bombings,
survivor behavior was characterized by adaptive features, such as order, solidarity and
mutual aid rather than the dysfunctional individualism and panic that characterizes
psychosocial vulnerability. Importantly, it was the crowd itself that was the basis of the
resilience displayed by survivors. In this account, then, the crowd is a psychosocial
resource: a sense of psychological unity with others during emergencies is the basis of
being able to give and accept support, act together with a shared understanding of what is
practically and morally necessary, and see others’ plight as linked to our own rather than
The concept of collective resilience also offers a new way of thinking about aspects
of personal resilience and recovery in mass emergencies. Being part of a psychological
crowd increases individuals’ chances of physical survival and psychological recovery,
since the crowd enables them practically to realise goals they cannot achieve alone,
including organizing the world around them to minimize the risks of being exposed to
further trauma (see Williams and Drury 2009).
Collective resilience can be derived from the principles of self-categorization theory,
and at least some of the data fits very well with the theory, and none actually contradicts
it. While positive evidence for a common identity (in the form of references to unity in
the crowd) is weak in the secondary data, there is no counterevidence (e.g. statements
Drury: Reactions to London bombings
about disunity, conflict or individualism). Moreover, here people were not asked about
unity or for their reasons for helping others (or, at least, they are not reported). In the
primary data, by contrast, there is clear evidence that the perception of common danger or
fate created a strong sense of unity, at least for some people. The suggestion that a shared
social identity arose from the common experience threat makes sense of the evidence of
inclusive solidarity, including the risks that some people took to help strangers. The
analysis of these data thus turns around one of the basic tenets of the mass panic
approach—i.e. that threat of death in an emergency serves to divide people against each
The foregoing explanation for resilience amongst survivors of the London bombs is
not meant to suggest that there will be equally enhanced unity and high levels of mutual
aid amongst all participants in every emergency (Drury, Cocking, and Reicher 2009). But
where there is mutual aid and other indicators of resilience in the crowd, we would argue
that self-categorization processes are part of the explanation. In suggesting that self-
categorization processes explained collective behavior following the London bombings, it
is also necessary to consider the contribution of other possible psychological bases for
mutual aid and resilience. It might be argued that there is some evidence here, albeit
weak, for other models of crowd functionality, sociality and adaptive behavior in
First, then, could the collective resilience displayed be explicable in terms of
emergent norm theory? While survivors obviously communicated with each other, there
is no evidence that extended milling was necessary before collective action could take
place. Many people seemed to know what to do (‘morally’ and practically) without
debate: help others and try to get out. Moreover, the data suggests the importance
specifically of ‘feeling part of a group’ rather than ‘shared vision’ per se (cf. Turner
1996) in the subjective sense of unity.
Second, it seems at least some of the data on helping behavior described in the
present paper might be explicable in terms of everyday societal norms. Yet in everyday
circumstances it is perhaps more ‘normative’ in fact to compete with one’s fellow
passengers for space on the London Underground rather than to help them (cf. Johnson
1987); indeed, as we have seen, this was the comment of some of our interviewees. On its
own, the concept of norm merely redescribes behaviors rather than explains them, since it
does not suggest when and how such normative behaviors might be instantiated in
particular contexts.
Relatedly, there is also some evidence here consistent with the suggestion that people
in emergencies act in terms of their given social roles (cf. Donald and Canter 1992;
Johnson 1988). For example, male survivors and witnesses helped others more than
women did. Arguably the actions of the train drivers in taking a lead in the evacuation is
also evidence for the maintenance of social roles. However, the social role of most
survivors of the bombs is unclear. Most were ‘commuters’; but what does this mean for
Drury: Reactions to London bombings
practice in these circumstances? Like the concept of norms, the notion of social roles may
describe some behaviors but does not explain why one set of rules for behavior and not
another will be operating at a given time.
Finally, data in support of the affiliation approach is patchy at most. Most people
were amongst strangers; hence preexisting relationships could not have been the
motivation behind most of the helping behavior. The evidence that people displayed at
least as much concern for the strangers around them as they did about affiliates, and the
lack of panic that people displayed despite being amongst strangers in a situation of
extreme fear, both suggest that Mawson’s (2005) account is incomplete.
Limitations of the data
Part of the value of the present study is that it is to our knowledge the first social
scientific attempt to analyse collective behavior, perceptions and motivations among
survivors in the July 7th 2005 London bombings, the most serious bombing attack on
mainland Britain since the Second World War. Yet the nature of the event—the
impossibility of gathering contemporaneous data and the obvious sensitivity of the topic
afterwards—means that the data are less than ideal. A more systematic examination of
the ideas proposed here could be the subject of a further study, such as a comparison in
solidarity (versus ‘selfishness’) between different crowd events varying in shared fate and
unity (Drury, Cocking and Reicher 2009). Any such further research would need to
address the particular methodological limitations of the present study, however. These
fall into three areas.
A first set of problems has to do with self-selection in the sample. With such a painful
topic as a terrorist attack, there is inevitably selectivity in who speaks and gets reported in
the media and the courts, and in who comes forward to be interviewed for the research.
Obviously, it is impossible to control for this, and it is possible that other stories of the
event could be told if different people were interviewed.
Second, and relatedly, broader social pressures may have led to certain biases in the
dataset as a whole. For example it could be argued that, in the context of a terrorist attack,
there is a political imperative to tell a positive story of unity and heroism, and that other
voices and versions of events have been marginalized. Yet it is clearly not the case that
one homogenous version of events prevailed in the data set. There were numerous
examples of the pathologizing discourse of ‘mass panic’ in the accounts of both survivors
and witnesses, which contradicted their own accounts of their behavior. Common sense
discourse, as has been observed, is a repository of contradictory ready-made explanations
(Billig 1987).
A third possible limitation of the present analysis is that the reliance on self-reports
means the data may reflect the operation of self-presentational biases. Clearly behavior in
an emergency is a highly charged topic. To ask survivors whether they helped others or
simply looked after themselves when others were in need is perhaps to invite them to
Drury: Reactions to London bombings
defend themselves. What tells against an explanation of this data solely in terms of
survivors’ desires to present themselves favourably, however, is that the analysis does not
rely on self-reports of own behavior alone: witnesses’ observations of others’ behavior
are consistent with the story that mutual aid was more prevalent than personal selfishness.
On top of this, the fact that, in each of the different datasets analysed, people reported
more instances of helping by others than helping by themselves tells against the idea that
this data is simply an artefact of a self-presentational bias.
There is a final, general, point to be made on the quality of these data and the
plausibility of the theoretical claims we have made. It is obvious to researchers of real
world collective behavior such as mass emergencies that data in this field, which excel in
ecological validity, will never achieve the ‘completeness’ and reliability of studies using
controlled designs and measures, such as laboratory experiments or ‘attitude’ surveys
with student samples. Yet what tells in favour of the present analysis is that this pattern of
findings and its interpretation is fully in line with a growing body of well controlled
laboratory and survey studies that likewise link group formation, identity, support and
responses to stress (e.g. Haslam, Jetten and Waghorn 2009).
The high levels of mutual aid amongst survivors and witnesses of the bombings
supports the view that the public in general and crowds specifically are more resilient
than they are given credit in the influential ‘vulnerability’ framework with its emphasis
on ‘risk’ prevention and inevitable mental health problems (Wessely 2005a, b). The
present study suggests that resilience within an emergency is not restricted to
spontaneous coordination amongst the emergency services (cf. Rodriguez, Trainor and
Quarantelli 2006; Tierney 2002), and that what has been called a ‘therapeutic’ or
‘altruistic’ community (Barton 1969; Fritz 1968; see Furedi, 2007, for a more recent
discussion) can develop very quickly and without existing social ties beyond the common
human capacity to categorize others with self (Turner et al. 1987). As such, these data
make the case (i) for those in authority to encourage a common identity amongst the
public in emergencies (instead of promoting a message of ‘stranger danger’) and (ii) for
the inclusion of the public in the policymaking and practice of their own defence—not
least through keeping them properly and practically informed, rather than excluded for
fear of ‘mass panic’ (Dynes 2003; Wessely 2005a, b).
If informal collective resources are the basis of resilience within an emergency, it is
possible they may operate in the clinical aftermath too. A number of survivors in this
study mentioned that their membership of the group of ‘survivors’ continued to be
psychologically important to them: ‘I've been left most of all with this huge sense of
solidarity’ (Personal account 12)
Some of these survivors therefore reported seeking out other survivors; they
suggested that sharing their feelings with each other in support groups helped in their
Drury: Reactions to London bombings
recovery. The finding that mutual support groups have some subjective wellbeing
function for survivors is in line with recent studies demonstrating the reduction of stress
through social identification-based social support (Haslam, O’Brien, Jetten, Vormedal
and Penna 2005; Haslam and Reicher 2006). What is needed in future research is some
test of the objective effects of such mutual support groups. The implications for aftercare,
where, at the moment, there is an exclusive but controversial reliance on professional
experts (Wessely 2005a, b), could be significant.
Taken to an extreme, it needs to be acknowledged that the rhetoric of ‘resilience’ can
be used politically not only to boost (e.g., national) morale but also to minimize
government and corporate responsibility by downplaying real hardships (Furedi 2007).
The data analysed in this paper do not suggest an absence of distress, suffering or
symptoms; nor do these findings serve to reify the collective resilience of the survivors as
peculiar to Londoners or the British, as has been suggested in populist accounts (e.g.
Elms 2005). (For example, the fact that Tierney (2002) found that New Yorkers also
demonstrated such resilience is in line with the implication of SCT that the processes of
mutual aid in adversity identified here are to some extent universal.) However, despite its
potential dangers, it is argued here that a notion of resilience in unstructured crowds is
necessary to counter the currently dominant vulnerability framework which not only
neglects the human capacity for collective survival in the face of disasters but threatens to
undermine it.
1 The term ‘vulnerability’ is also used in the field of disaster research to refer to the
structural or geographical (i.e., physical) susceptibility of some populations to such
disastrous events as floods, hurricanes and earthquakes (e.g., Comfort 1990). In the
present analysis, however, we are concerned with the psychosocial conception of
vulnerability—as used in US based disaster research (Dynes 2003), UK disaster
mental health strategy (Wessely 2005a, b), and critical sociology (Furedi 2007).
2. Given the timeframe of community (re)construction in the postdisaster recovery
period (i.e., days and months rather than the minutes or hours it takes for the
occurrence of the disaster itself), it might be thought that research on unity and
support in this field would find evidence for the role of extended interaction in the
development of such sociality. Yet, so far as we are aware, there is little reference in
this literature to ENT or its principles (e.g. Comfort, 1990; Kaniesty and Norris 1997;
Paez, Basabe, Ubillos and Gonzalez-Castro 2007; see Drury and Winter 2004 for a
Drury: Reactions to London bombings
5. As mentioned, there is undoubtedly duplication within the newspaper data. There is
probably some overlap between survivors and witnesses in the contemporaneous
newspaper accounts and the other data. However, the newspaper data are likely to
include accounts from people not captured in the other datasets. Thus we can
plausibly conclude that 146 is an underestimate of the number of people giving
accounts in this study.
6. The category ‘no panic’ was not coded in this dataset, and ‘orderly’ was coded
instead of ‘calm’.
7. [ ] denotes material edited from the transcript for reasons of space.
8. Helping given to others was the only variable in the personal accounts for which a
statistically significant gender difference was identified: men were more likely to
report instances where they had helped others (M = 3.50 people helped) than were
women (M = 1.95), t38 = 2.57, p = 0.01.
9. See also National Institute for Clinical Excellence NICE Guidelines on PTSD. (2006)
Available at
0bombings.pdf, last accessed 6th September 2007.
10. Fear and expectation of death are coded separately, since it was clear that some
people thought they might die yet were not actually afraid (or, at least, did not report
11. In the accounts of journalists, politicians and other commentators, on the other hand,
there were 83 references to unity as well as numerous references to the resilience of
the UK or London identity. Given the political context, such talk might be understood
as a call to unity rather than a description of observed unity, however.
The research described in this report was made possible by a grant from the Economic
and Social Research Council awarded to J Drury, S Reicher (University of St Andrews),
P Langston (University of Nottingham) and D Schofield (RMIT University, Melbourne),
Ref. no: RES-000-23-0446. The authors thank the following: for their assistance in
transcription, data input and other clerical assistance in this project: T Maxwell and R
Marlow; and for initial website development: D Schofield.
Drury: Reactions to London bombings
Aguirre, B.E. 2005. “Commentary on ‘Understanding Mass Panic and Other Collective
Responses to Threat and Disaster’: Emergency Evacuations, Panic, and Social
Psychology.” Psychiatry 68: 121-129.
Aguirre, B.E.. 2006. On the Concept of Resilience. University of Delaware, Disaster
Research Center. Preliminary paper #356.
Aguirre, B.E., Wenger, D. and Vigo, G. 1998. “A Test of the Emergent Norm Theory of
Collective Behavior.” Sociological Forum 13: 301-320.
Aylwin C.J., König, T.C, Brennan, N.W, Shirley, P.J, Davies G., Walsh, M.S. and Brohi,
K. 2006. “Reduction in Critical Mortality in Urban Mass Casualty Incidents:
Analysis of Triage, Surge, and Resource Use After the London Bombings on July
7, 2005.” The Lancet 368: 2219 -25
Barton, A.H. 1969. Communities in Disaster: A Sociological Analysis of Collective Stress
Situations. New York: Doubleday.
Billig, M. 1987. Arguing and Thinking. Cambridge, UK: Cambridge University Press.
Blake, S.J., Galea, E.R., Westeng, H., and Dixon, A.J.P. 2004. “An Analysis of Human
Behavior During the World Trade Center Disaster of 11 September 2001 Based
on Published Survivor Accounts.” Proceedings of Third International Symposium
on Human Behavior in Fire, Belfast, September.
Campbell, D.T. 1958. “Common Fate, Similarity and Other Indices of the Status of
Aggregates of Persons as Social Entities.” Behavioral Science 3: 14-25.
Clarke L. 2002. “Panic: Myth or Reality?” Contexts 1: 21-26.
Comfort, L. 1990. “Turning Conflict Into Cooperation: Organizational Designs for
Community Responses in Disasters”. International Journal of Mental Health 19:
Connell, R. 2001. Collective Behavior in the September 11, 2001 Evacuation of the
World Trade Center. University of Delaware, Disaster Research Center.
Preliminary paper #313.
Cornwell, B. 2003. “Bonded Fatalities: Relational and Ecological Dimensions of a Fire
Evacuation.” The Sociological Quarterly 44: 617-638.
Cornwell, B., Harmon, W., Mason, M., Merz, B., and Lampe, M. 2001. “Panic or
Situational Constraints? The Case of the M/V Estonia.” International Journal of
Mass Emergencies and Disasters 19: 5-25. Available at
Donald, I. and Canter, D. 1992. “Intentionality and Fatality During the Kings Cross
Underground Fire.’ European Journal of Social Psychology 22: 203-218.
Drury J. and Reicher, S. 1999. “The Intergroup Dynamics of Collective Empowerment:
Substantiating the Social Identity Model.” Group Processes and Intergroup
Relations 2: 381-402.
Drury: Reactions to London bombings
Drury, J., Cocking, C. and Reicher, S. 2009. “Everyone for Themselves? A Comparative
Study of Crowd Solidarity Among Emergency Survivors.” British Journal of
Social Psychology 48:
Drury, J., and Winter, G. 2004. Social Identity as a Source of Strength in Mass
Emergencies and Other Crowd Events. International Journal of Mental Health
special issue on ‘Coping with Disasters: The Mental Health Component’ 32: 77-
Durodié, B. and Wessley, S. 2002. “Resilience or Panic? The Public and Terrorist
Attack.” The Lancet 360: 1901-02.
Dynes, R.R. 2003. “Finding Order in Disorder: Continuities in the 9-11 Response.”
International Journal of Mass Emergencies and Disasters 21: 9-23. Available at
Elms R. 2005. “There Was No Panic, No Hysteria. This Is a Hard-Nosed, Battle-Scarred
City. And we Londoners are Hard to Scare.” The Independent on Sunday 10 July.
Feinberg, W., and Johnson, N.R. 2001. “The Ties That Bind: A Macro-Level Approach
to Panic.” International Journal of Mass Emergencies and Disasters 19: 269-295.
Available at
Freud, S. 1921. “Group Psychology and the Analysis of the Ego” Pp. 91-178 In
Civilization, Society and Religion edited by A. Dickson, translated by J. Strachey,
1985). Harmondsworth, UK: Penguin.
Fritz, C.E. 1968. “Disasters.” Pp. 200-207 In International Encyclopedia of the Social
Sciences Vol 4, edited by D. Sills. New York: Macmillan and Free Press,
Fritz, C.E., and Williams, H.B. 1957. “The Human Being in Disasters: A Research
Perspective.” The ANNALS of the American Academy of Political and Social
Science 309: 42-51.
Furedi, F. 2007. Invitation to Terror. London: Continuum Press.
Glass, T.A. and Schoch-Spana, M. 2002. “Bioterrorism and the People: How to
Vaccinate a City Against Panic.” Clinical Infectious Diseases 34: 217-223.
Haslam, S.A. 2004. Social Identity in Organizations: The Social Identity Approach.
London: Sage.
Haslam, S.A., Jetten, J., and Waghorn, C. 2009. “Social Identification, Stress, and
Citizenship in Teams: A Five-Phase Longitudinal Study.” Stress and Health 25:
Haslam, S.A., O’Brien, A., Jetten, J., Vormedal, K., and Penna, S. 2005. “Taking the
Strain: Social Identity, Social Support, and the Experience of Stress.” British
Journal of Social Psychology 44: 355-370.
Haslam, S.A., and Reicher, S.D. 2006. “Stressing the Group: Social Identity and the
Unfolding Dynamics of Responses to Stress.” Journal of Applied Psychology 91:
Drury: Reactions to London bombings
Helbing, D., Farkas, I., and Vicsek, T. 2000. “Simulating Dynamical Features of Escape
Panic.” Nature 407: 487-490.
Johnson, N.R. 1987. “Panic at ‘The Who Concert Stampede’: An Empirical Assessment.”
Social Problems 34: 362-373.
Johnson, N.R. 1988. “Fire in a Crowded Theatre: A Descriptive Investigation of the
Emergence of Panic.” International Journal of Mass Emergencies and Disasters
6: 7-26. Available at
Jones, E., Woolven, R., Durodié, B., and Wessely, S. 2006. “Public Panic and Morale:
Second World War Civilian Responses Re-Examined in the Light of the Current
Anti-Terrorist Campaign.” Journal of Risk Research 9: 57-93.
Kaniasty, K., and Norris, F. (1997). Social Support Dynamics in Adjustment to Disasters.
Pp 595-619 In Handbook of Personal Relationships, 2
nd Edition. Edited by S.
Duck, London, UK: Wiley.
Kendra, J. and Wachtendorf, T. 2001. Elements of Community Resilience in the World
Trade Center Attack. Newark DE: University of Delaware Disaster Research
Center. Preliminary paper #318.
Le Bon, G. 1895. The Crowd: A Study of the Popular Mind. Dunwoody, GA: Norman S.
Levine, M., Prosser, A., Evans, D., and Reicher S. 2005. “Identity and Emergency
Intervention. How Social Group Membership and Inclusiveness of Group
Boundaries Shape Helping Behavior.” Personality and Social Psychology Bulletin
31: 443-453.
Mawson, A.R. 2005. “Understanding Mass Panic and Other Collective Responses to
Threat and Disaster.” Psychiatry 68: 95-113.
McDougall, W. 1920. The Group Mind. New York: GP Putnam's Sons.
McPhail, C. 1991. The Myth of the Madding Crowd. New York: Aldine de Gruyter.
Miles, M.B., and Huberman, A.M. 1994. Qualitative Data Analysis: An Expanded
Sourcebook. Thousand Oaks, CA: Sage.
Noppe, I. C., Noppe, L. D., and Bartell, D. 2006. “Terrorism and Resilience:
Adolescents’ and Teachers’ Responses to September 11, 2001”. Death Studies 30:
Paez, D., Basabe, N., Ubillos, S., and Gonzalez-Castro, J.L. 2007. “Social Sharing,
Participation in Demonstrations, Emotional Climate, and Coping With Collective
Violence After the March 11th Madrid Bombings.” Journal of Social Issues 63:
Proulx, G., and Fahy, R.F. 2003. “Evacuation of the World Trade Center: What Went
Right?” Pp. 27-34 in Proceedings of the CIB-CTBUH International Conference
on Tall Buildings, Oct. 20-23, Malaysia.
Proulx, G., and Sime, J.D. 1991. “To Prevent ‘Panic’ in an Underground Emergency:
Why Not Tell People the Truth?” Pp. 843-852 in Fire Safety Science:
Drury: Reactions to London bombings
Proceedings of the Third International Symposium, edited by G. Cox and B.
Langford. London: Elsevier Applied Science.
Quarantelli, E. L. (1960). “Images of Withdrawal Behavior in Disasters: Some Basic
Misconceptions.” Social Problems 8: 68-79.
Reicher, S. D. 1984. “The St Pauls Riot: An Explanation of the Limits of Crowd Action
in Terms of a Social Identity Model.” European Journal of Social Psychology 14:
Reicher S. 1996. “‘The Battle of Westminster’: Developing the Social Identity Model of
Crowd Behavior in Order to Explain The Initiation and Development of
Collective Conflict.” European Journal of Social Psychology 26: 115-34.
Ripley, A. 2005. How to Get Out Alive. Time, 16th May: 48-52.
Rodriguez, H., Trainor, J., and Quarantelli, E.L. 2006. “Rising to the Challenges of a
Catastrophe: The Emergent and Prosocial Behavior Following Hurricane
Katrina.” ANNALS of the American Academy of Political and Social Science 604:
Rubin, G.J., Brewin, C.R., Greenberg, N., Simpson, J., and Wessely, S. 2005.
“Psychological and Behavioral Reactions to the Bombings in London on 7 July
2005: Cross Sectional Survey of a Representative Sample of Londoners.” British
Medical Journal 331: 606-611.
Schultz, D.P. 1964. Panic Behavior: Discussions and Readings. New York: Random
Sherif, M. 1936. The Psychology of Social Norms. New York: Octagon books.
Sime, J.D. 1983. “Affiliative Behavior During Escape to Building Exits.” Journal of
Environmental Psychology 3: 21-41.
Sime, J.D. 1990. “The Concept of ‘Panic’.” Pp. 63-81 In Fires and Human Behavior 2nd
ed., edited by D. Canter. London: David Fulton.
Sime, J.D. 1995. “Crowd Psychology and Engineering.” Safety Science 21: 1-14.
Smelser, N.J. 1962. Theory of Collective Behavior. London: Routledge and Kegan Paul.
Strauss, A.L. 1944. “The Literature on Panic.” Journal of Abnormal and Social
Psychology 39: 317-328.
Stott, C., and Reicher, S. 1998. “Crowd Action as Inter-Group Process: Introducing the
Police Perspective.” European Journal of Social Psychology 28: 509-529.
Tierney, K.J. 2002 . Strength of a City: A Disaster Research Perspective on the World
Trade Center Attack. Available at Last accessed Sept
6th 2007.
Tierney, K., Bevc, C. and Kuligowski, E. 2006. “Metaphors Matter: Disaster Myths,
Media Frames and Their Consequences in Hurricane Katrina.” ANNALS of the
American Academy of Political and Social Science 604: 57-81.
Tierney, K. and Trainor, J. 2004. “Networks and Resilience in the World Trade Center
Disaster.” Pp. 157-172 in MCEER: Research Progress and Accomplishments
Drury: Reactions to London bombings
2003-2004. Buffalo NY: Multidisciplinary Center for Earthquake Engineering
Titcher, J.L., and Frederick, T.K. 1976. “Family and Character Change at Buffalo
Creek.” American Journal of Psychiatry 133: 295-299.
Turner, J.C. 1982. “Towards a Cognitive Redefinition of the Social Group.” Pp. 15-40 in
Social Identity and Intergroup Relations edited by H. Tajfel. Cambridge, UK:
Cambridge University Press.
Turner, J.C., Hogg, M.A., Oakes, P.J., Reicher, S.D., and Wetherell, M.S. 1987.
Rediscovering the Social Group: A Self-Categorisation Theory. Blackwell:
Turner, R.H. 1964. “Collective Behavior.” Pp. 382-425 in Handbook of Modern
Sociology edited by R.E.L. Faris. Chicago: Rand McNally.
Turner, R.H. 1994a. “Ideology and Utopia after Socialism.” Pp. 79-100 in New Social
Movements: From Ideology to Identity edited by E. Laraña, H. Johnston, and J.R.
Gusfield. Philadelphia, PA: Temple University Press.
Turner, R.H. 1994b. “Rumor as Intensified Information Seeking: Earthquake Rumors in
China and the United States.” Pp. 244-260 in Disasters, Collective Behavior and
Social Organization edited by R.R. Dynes and K.J. Tierney. Newwark: University
of Delaware Press.
Turner, R.H. 1996. “The Moral Issue in Collective Behavior and Collective Action.”
Mobilization 1: 1–15.
Turner, R.H. and Killian, L.M. 1957. Collective Behavior. Englewood Cliffs, NJ:
Turner, R.H. and Killian, L.M. 1972. Collective Behavior, 2nd ed. Englewood Cliffs, NJ:
Turner, R.H. and Killian, L.M. 1972. Collective Behavior, 3rd ed. Englewood Cliffs, NJ:
Weller, J., and Quarantelli, E.L. 1973. “Neglected Characteristics of Collective
Behavior”. American Journal of Sociology 79: 665-685.
Wessely, S. 2005a. “Editorial: Don’t Panic! Short and Long Term Psychological
Reactions to the New Terrorism: The Role of Information and the Authorities.”
Journal of Mental Health 14: 1-6.
Wessely, S. 2005b. “Victimhood and Resilience”. New England Journal of Medicine
353: 548-550.
Williams R. 2008. “Resilience and Risk in Children and Young People.” Pp. 107-118 in
Child and Adolescent Mental Health: A Handbook edited by C. Jackson, K. Hill,
and P. Lavis. Brighton, UK: Pavilion Publishing.
Williams, R.J.W., and Drury, J. in press. “Psychosocial Resilience and Its Influence on
Managing Mass Emergencies and Disasters.” Psychiatry.
Wright, S. 1978. Crowds and Riots. London: Sage.
... Similar patterns have been observed during the COVID-19 pandemic, where people join mutual aid groups to help their communities with for example dog-walking, shopping, and other things to facilitate reduction in virus transmission (see Fernandes-Jesus et al., 2021;Mao et al., 2021). Research on collective action in the form of cooperation and mutual aid during and after disasters demonstrates that in disasters a shared fate often develops which facilitates a shared identity to emerge (e.g., Cocking et al., 2009;Drury et al., 2009). Through shared fate, people come to see themselves as part of the same group or community, which functions as a base for solidarity, cohesion, and empowerment to emerge (Drury andReicher, 1999, 2009). ...
... We argue that disasters can offset or increase already existing perceived group-based injustice (e.g., Templeton et al., 2020), create a shared social identity (e.g., Cocking et al., 2009;Drury et al., 2009) efficacy (e.g., Drury et al., 2016) through a perception that "we" can be effectful in toppling the power or changing power relations (e.g., changing policy around cladding, increasing transparency in state communication, or shifting the political governance). Increased identification from shared injustice has been emphasized elsewhere both broadly in terms of sharing minority membership or victimhood (see e.g., rejectionidentification model by Branscombe et al., 1999) as well as in the disaster literature in terms of common fate (e.g., Drury et al., 2009). ...
... As disasters often contains the prerequisites for collective action participation-maybe the question that should be asked next is "why do protests not occur after disasters?". One potential factor to the lack of disaster-related protests could be the need to prioritize the cooperation route (see e.g., Cocking et al., 2009;Drury et al., 2009) to ensure one self 's and others' survival. However, we argue that due to the likelihood of an increase in number and severity of disasters due to the climate crisis (e.g., Van Aalst, 2006; Benevolenza and DeRigne, 2019), we need ...
Full-text available
People around the globe are affected by disasters far beyond the disaster properties. Given that certain social groups are affected disproportionately, disasters need to be considered as political events which may cause political actions. Therefore, we aim to discuss, from a social psychological perspective, how and why protests might occur during or after a disaster. We argue for an elaborated model of collective action participation suggesting that disasters enhance the predictors of protest mobilization and participation though emerged or enhanced social injustice. We also suggest that disaster properties can be used to delegitimise protests and social movements, limiting the mobilization and collective resilience during and after a disaster. Finally, we discuss the gaps in current research and emphasize the need for more attention to the disaster-protest link as we can expect more disasters due to climate crisis, likely to lead to more protests and political collective action.
... Here, frontline workers have found that their previous understanding of support and solidarity is no longer prevalent from leadership, and is being weakened from the public, and so they have recalibrated their source of solidarity to those with whom they have shared experience and identity. Research from the parallel field of collective resilience suggests that identity-relevant social norms can confer social support at the collective level (Drury et al., 2009b;Ntontis et al., 2021), and so recentring the collective from which that resilience can be drawn would seem to be a healthy and -perhaps -inevitable consequence to weakened broader solidarity. ...
The phrase ‘in it together’ has been used liberally since the outbreak of COVID‐19, but the extent that frontline workers felt ‘in it together’ is not well understood. Here, we consider the factors that built (or eroded) solidarity while working through the pandemic, and how frontline workers navigated their lives through periods of disconnection. Semi‐structured interviews with 21 frontline workers, across all sectors, were conducted in the United Kingdom and Ireland. The qualitative data were analysed systematically using reflexive thematic analysis. The three themes identified in the data were: (1) Solidarity as central to frontline experiences; (2) Leadership as absent, shallow and divisive: highlighting ‘us‐them’ distinctions and (3) The rise of ‘us’ and ‘we’ among colleagues. Our research offers insights into how frontline workers make sense of their experiences of solidarity and discordance during the first year of the COVID‐19 pandemic, with relevance for government and organizational policy‐makers shaping future conditions for frontline workers.
... Of relevance to the current research, the pathway from social connection to wellbeing has been shown to emerge under conditions of collective threat, such as public emergencies or natural disasters. Here we see that the perception of a common fate allows for the establishment of a shared identity, and that this emergent social identity leads to mutual support and subsequently to enhanced individual and collective health (82)(83)(84). Furthermore, research into formal support provision has demonstrated that rates of volunteerism are associated with increased feelings of personal self-efficacy and empowerment, as well as improved mental and physical health (85,86), and that increases in shared identity are associated with higher levels of wellbeing for volunteers (87). ...
Full-text available
While the relationship between loneliness and psychological distress is well documented, the mechanisms underlying this relationship are less clear. One factor known to be related to loneliness as well as psychological distress, is social support, with some studies suggesting that support–both received and provided–can serve as a mechanism to reduce the distress associated with loneliness. In this paper we examine the mediating role of both aspects of support in the relationship between loneliness and psychological distress in the COVID-19 context. We used a multi-country dataset collected at two timepoints during the pandemic; the first during the early stages ( N = 6,842, 11 countries) and the second collected for a subset of countries ( N = 1,299, 3 countries) 3 months later. Across all eleven countries, results revealed significant positive associations between loneliness and distress. Furthermore, using longitudinal data, we investigated the directionality of this relationship and found that increased loneliness over time was associated with increased psychological distress. The data also showed that both feeling unsupported and feeling unable to provide support to others mediated this relationship. These findings point to the need to facilitate people's ability to draw effective social support and help others–particularly at times when social connectedness is threatened–as a way of alleviating the psychological distress that commonly presents with loneliness.
... Instead, the focus has shifted to the factors that explain why people cooperate rather than compete in response to a crisis 35 . One of these factors is an emerging sense of shared identity and concern for others, which arises from the shared experience of being in a disaster 40 . This feeling can be harnessed by addressing the public in collective terms and by urging 'us' to act for the common good 41 . ...
يعد تغير المناخ أكبر تهديد يواجه البشرية بأكملها، وأحد أكثر القضايا إلحاحًا التي تواجه المجتمعات، ويهدد خطط التنمية والتقدم والصحة؛ بل ويهدد كوكب الأرض بأكمله. غالبًا ما يُوصف تغير المناخ بأنه مشكلة بيئية، ولكن من الأفضل فهمه على أنه مشكلة بشرية؛ إذ أن السلوك البشري مسؤول عن تغير المناخ إلى حد كبير. سيتأثر البشر بهذه التغيرات المناخية، وسيتطلب مواجهتها ضرورة تغيير السلوك البشري. ولا شك أن تغير المناخ يغير عالمنا، وبالتالي سيؤدي بالضرورة إلى تغيير المجتمعات. ولهذا؛ يمثل تغير المناخ حاليًا تهديدًا وجوديًا وجسديًا ونفسيًا. ومن ثم، أصبح تطبيق إجراءات وتدابير التخفيف والتكيف والقيادة الخضراء والحوكمة الرشيدة والتنمية المستدامة والتنمية المتجددة ضرورية بشكل متزايد؛ للحفاظ على رفاهية الفرد والمجتمع وبقاء كوكب الأرض. تم تأليف هذا الكتاب سعيًا مني لتأصيل قضية التغير المناخي وتأثيراته على الانسان من الناحية النفسية من جهة، ولمسح البحوث والدراسات النفسية الحديثة حول مجموعة من الموضوعات ذات الصلة بالتغير المناخي من جهة ثانية. ولإثارة تفكير الباحثين والمختصين في علم النفس بالوطن العربي من أجل إجراء دراسات وبحوث مستقبلية حول كيفية إدراك البشر أو تصوراتهم واستجاباتهم للتغيرات المناخية التي تحدث في العالم من حولهم وتأثير ذلك على الجانب النفسي من جهة ثالثة. وكذلك لوضع إطار إجرائي للخدمات والرعاية النفسية وقت الأزمات والكوارث المناخية، وللمساهمة في تعزيز قيم المواطنة المناخية مع مراعاة حقوق الإنسان وحرياته والعدالة المناخية لجميع المجتمعات والأفراد بمختلف فئاتهم من أجل استدامة الصحة واستدامة البيئة من جهة رابعة. يتكون هذا الكتاب الذي بين أيديكم؛ من تسع فصول.
The present paper explores psychological processes that underpin the success of community change in the context of urban regeneration schemes. We adopt a social identity approach to develop an understanding of the ways in which social identity dynamics may impact upon peoples' experiences of regeneration, and what influence these identity processes have on the creation of new communities. Qualitative interviews, using thematic analysis as an analytic technique, were conducted with community members (n = 14) in a recently (2001–2011) regenerated area in the South‐West of England. Three overarching themes were identified: Patterns of identification, willingness to engage, and the notion of regeneration as an event. The research overall highlights the central role of group‐based identity in understanding the processes of regeneration and how this is experienced by different community members. Findings are discussed in relation to the impact regeneration schemes have on community members' sense of collective self, unity, and engagement. The research highlights the pivotal role of social identity processes in delivering successful and sustainable strategies of urban regeneration.
Panic buying (PB) during disasters is a global phenomenon, though underresearched. However, with the COVID-19 pandemic, it has gained prominence because of its adverse consequences. It has a reciprocal relationship with crises, i.e., it occurs during crises and can lead to the worsening of problems. PB can have different characteristics according to the type of disaster. The proposed models to understand the underlying mechanism of the PB include the causal model, cognitive model, model-driven causal factors, stimulus-organism-response and dual-system theory model, ABC model concerning social media information and compulsive PB, cultural dimension model, etc. Understanding the determinants of this behavior can help predict, prevent, and manage this condition, thus having public health and welfare implications during crises. This chapter intends to highlight the significant disasters and associated PB behavior, how PB behavior can affect disaster management, and the underlying mechanism of PB and its determinants.KeywordsPanic buyingDisasterEmergencyCrisesDeterminantsPredictors mechanismModel
Collective victimhood and collective resilience are two sides of the same coin. However, most literature to date has focused on the experiences and consequences of collective victimhood. In the present research, we focused on the experiences of Black Americans, a group that has a legacy of victimization and resilience. As a part of Black Americans' collective memory, we explored the nature of historical collective resilience and examined its role in explaining collective responses to present‐day oppression, over and above any effect of historical collective victimhood. When they were asked to reflect on their group's history, across Studies 1 (N = 272) and 2 (N = 294), we found that Black Americans generated narratives of collective resilience. In both studies, we also found evidence that perceived historical collective resilience was linked to a greater sense of collective continuity, which, in turn, explained greater support for the ongoing Black Lives Matter movement. Our findings underscore the importance of considering narratives of resilience in a group's history and point to the way such collective resilience narratives can serve as a resource for the group in the present.
Full-text available
This paper evaluates behavior among individuals during the sinking of the M/V Estonia in 1994, which caused the deaths of 851 people. Survival rates of those on board indicate a drastically higher proportion of men surviving than women (.22 vs .05), and a higher proportion of crew members surviving than regular passengers (.23 vs .12). These patterns suggest that intense competition and panic may have ensued during the escape, since it appears that individuals with (socially-defined) role obligations (e.g., crew members) disregarded others’ (e.g., regular passengers) need for aid during the emergency. However, we can believe that the unusually tumultuous physical constraints effectuated by the disaster may have made it difficult for these individuals to help one another escape from the ship. Thus, unlike previous research on disasters, this paper treats the situational environment of the disaster as a factor that can influence survival rates differentially across groups. The usefulness of reaching conclusions about the existence of panic based solely on observations of overt action and/or covert emotional states (especially those based on quantitative analyses) is thus called into question.
Emergent norm theory is clarified and expanded through analysis of the processes and circumstances by which the normative element emerges and persists in general and in collective behavior and collective action. An interdependency in risk principle is suggested, coupled with triggering conditions to evoke a specifically moral sense. The paper concludes with brief discussions of the sources of normative content and social structural conditions that are conducive to normative emergence.
Alex Haslam has thoroughly revised and updated his ground-breaking original text with this new edition. While still retaining the highly readable and engaging style of the best-selling First Edition, the author presents extensive reviews and critiques of major topics in organizational psychology - including leadership, motivation, communication, decision making, negotiation, power, productivity and collective action - in this thoroughly revised edition. New to the Second Edition: An entirely new chapter on organizational stress which deals with highly topical issues of stress appraisal, social support, coping and burnout.; New, wider textbook format and design making the entire book much more accessible for students.; A wide range of pedagogical features are included - suggestions for further reading at the end of each chapter and comprehensive glossaries of social identity, social psychological and organizational terms