Zeitschrift für interdisziplinäre Geschichte, xliii:1 (Sommer, 2012), 27–42.

DISASTER STATISTICS

Benigno E. Aguirre
Better Disaster Statistics: The Lisbon Earthquake
Major disasters are characterized by signiªcant loss of life and
Eigentum, injuries, mental distress, and the problem of numbers.
Collecting information on such events and determining their ef-
fects is of critical importance in organizing social responses and in
devising programs to ameliorate and mitigate the impact of future
disasters. Developing good statistics on such matters is important
because local, state, and federal spending is prioritized and allo-
cated according to these numbers and because life-and-death deci-
sions are based on, or inºuenced by, this information. The accu-
rate identiªcation of the effects of disasters is also valuable to
international agencies in their attempt to determine the security of
their investments in developing countries.

For these reasons, statistics and ªgures about death and
property loss from disasters, both in general and in particular,
abound in research, policy statements, professional and opera-
tional literature, and news reports. The abundance of quantitative
estimates and data documenting effects is due in part to the con-
siderable pressure from legislators and lending agencies, über-
whelmingly peopled by policy personnel keen to know how any
proposed program ªts into a cost-beneªt framework. Unglücklich-
natürlich, Jedoch, despite the obvious importance of this informa-
tion, and the call by governmental and professional committees to
optimize it, the disaster and emergency ªeld is plagued by dubious
statistical data and widespread conceptual disagreement.1

Inaccurate published statistics are often due to faulty informa-
tion about a disaster, to the different accounting methods used by

Benigno E. Aguirre is Professor of Sociology and Criminal Justice and a core faculty member
of the Disaster Research Center, University of Delaware. He is the author of “Homeland Se-
curity Warnings: Lessons Learned and Unlearned,” International Journal of Mass Emergencies and
Disasters, XXII (2004), 103–116; “Dialectics of Vulnerability and Resilience,” Georgetown
Journal on Poverty Law and Policy, XIV (2007), verfügbar unter https://articleworks.cadmus.com/
geolaw/z5800107.html.

1 Aguirre and Ernesto L. Quarantelli, “Phenomenology of Death Counts in Disasters: Der
Invisible Dead in the 9–11 WTC Attacks,” International Journal of Mass Emergencies and Disas-
ters, XXVI (2008), 19–40.

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

28 | BENIGNO E. AGUI RRE

Organisationen, and to the lack of consensus among researchers and
agencies about such central concepts as disaster, hazard, and risk.
Zum Beispiel, the European databases on hazards and disasters
maintained by a number of different organizations in different
countries show no agreement in deªnitions and protocols. Ac-
cording to Buckley, none of the existing databases provides a satis-
factory framework for the global analysis of disasters.2

Given this predicament, improving the effectiveness of disas-
ter, emergency, and crisis management faces serious challenges, es-
pecially since practitioners and the lay public alike appear unwill-
ing to consider such matters. Specialists are of two persuasions.
Alongside the signiªcant number of scholars who have called at-
tention to the problem and voiced the need for prompt solutions
in recent years are many others (sometimes even the same ones)
who are content to make claims about disaster-related demands
and the increasing frequency, severity, and etiology of disasters on
the basis of doubtful statistics. More often than not, the ªrst num-
bers and statistics issued about disasters—particularly regarding in-
jured and discomªted people and the extent of economic loss—
are hyperbolic, only to be scaled down later.3

The long-standing problem of what constitutes an injury, als
discussed in the medical disaster literature, has no simple solution.
There are major differences between injuries that require a pro-
longed hospital stay and those that need only emergency-room
treatment. Added to that uncertainty is the problem of whether
those injured in the aftermath of a disaster—for example, while
undertaking repair or recovery services—should be counted

2 Philip Buckley, Preliminary Regional Stocktaking of Natural Hazard Risk and Disaster Manage-
ment Capacity (New York, 2007).
3 On the issue of dubious statistics, see Susan K. Tubbesing (Hrsg.), Natural Hazards Data
Sources: Uses and Needs (Felsblock, 1979); Claire B. Rubin et al., Summary of Major Natural Di-
saster Incidents in the U.S. 1965–85 (Washington, D.C., 1986); Christopher Arnold, “Presenta-
tion: Loss and Damage,” in Committee on Earthquake Engineering (Hrsg.), Proceedings of a
Forum, The Economic Consequences of a Catastrophic Earthquake (Washington, D.C., 1992), 36–
48; Debarti Guha Sapir, “Health Effects of Earthquakes and Volcanoes: Epidemiological and
Policy Issues,” Disasters, XVII (1993), 255–262; Erik K. Noji, “Analysis of Medical Needs
during Disasters Caused by Tropical Cyclones: Anticipated Injury Patterns,” Journal of Tropical
Medicine and Hygiene, XCVI (1993), 1–7; Sapir and Regina Below, The Quality and Accuracy of
Disaster Data: A Comparative Analyses of Three Global Data Sets (Brussels, 2002); Dennis Mileti,
Disasters by Design: A Reassesment of Natural Hazards in the United States (Washington D.C.,
1999); National Academy of Science, Facing Hazards and Disasters: Understanding the Human
Dimensions (Washington, D.C., 2006); National Center for Environmental Health, Surveillance
and Data Collection Methods (Washington, D.C., 2003).

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

DISASTER STATISTICS | 29

among the victims of a disaster and in the documentation about
place of treatment, which is not always easy to establish. These va-
garies are also true of claims about the psychological effects of di-
sasters, especially the often-exaggerated claims about posttrau-
matic stress syndrome.

Statistics about death tend to be the most reliable, but even
these ªgures are plagued by questions about the cutoff date for fa-
talities directly attributable to a disaster (Tage, weeks, or months af-
terward?), as well as by cultural differences about how deaths are
to be counted. Arnold opines that the estimation of deaths and in-
juries caused by earthquakes is “even vaguer than the dollar-loss
aspect (for it is usually done) in a smoke-ªlled room by a small
number of people.” Sapir, in her analysis of the injuries and deaths
associated with volcanoes and earthquakes, writes, “It is rare to
ªnd systematically recorded statistics on the cause, Ort, and char-
acteristics of death and injury in disasters.” Noji states that in the
aftermath of hurricanes, the response of the medical community is
often hampered by the absence of accurate information about
deaths and injuries: “The most striking feature of data from tropi-
cal cyclone epidemiologic literature is its variable quality and na-
ture.” Systematic historical studies have shown racism to be a fac-
tor in masking the number of people from minorities who die in
disasters, as was the case for Chinese victims in the 1906 San Fran-
cisco earthquake. Darüber hinaus, until the 1960s, the Australian gov-
ernment refused to count Aborigines as persons, thereby ensuring
that their deaths in cyclones would not be recorded.4

The collection of disaster data involves at least four chal-
Längen: (1) distinguishing between direct and indirect effects,
(2) reconciling secondary data sources, (3) accessing private data-
bases, Und (4) compensating for systemic oversights in the count-
ing process. In many cases, the statistics serve the needs of the
entity collecting the data. Zum Beispiel, the information that insur-
ance agencies collect about disaster-related losses may be useful for
their needs but pose difªculties for scientiªc inquiry or general
policy development. “Numbers” incorporate the values of the
people who create them, and data collection begins with the col-

4 Arnold, “Presentation: Loss and Damage,” 43; Sapir, “Health Effects of Earthquakes and
Volcanoes,” 257–258; Noji, “Analysis of Medical Needs During Disasters,” 5. The source of
the information about the Australian government’s treatment of Aborigines was a personal
communication from Quarantelli, September 6, 2009.

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

30 | BENIGNO E. AGUI RRE

lector’s interests or concerns. Resolving inconsistencies necessi-
tates identifying such organizational interests and accommodating
their needs, even as the numbers and statistics are adjusted to suit a
broader category of interested parties. Such is the complicated
background against which this article evaluates the different claims
regarding the number of deaths caused by the Lisbon earthquake
von 1755.

the earthquake and the city The Lisbon earthquake of 1755
was centered on the Atlantic Ocean, 200 km west-southwest from
Cape St. Vincent. Most of its devastation occurred in Lisbon, Aber
other areas suffered as well. The earthquake was accompanied by a
ªre that destroyed important sections of Lisbon and by a large tsu-
nami that ravaged the riverine coast of Lisbon and Algarve in
southern Portugal as well as Seville, Cadiz, and Huelva in Spain
and Tangiers on the northern coast of Morocco. Apart from the
physical devastation, the earthquake is remembered for transform-
ing Portugal into a modern state, for helping to launch the modern
science of seismology, and for coloring the European understand-
ing of nature during the Enlightment.5

At the time of the earthquake, Lisbon was arguably the
wealthiest city in Europe. If Kendrick is to be trusted, Lisbon was
not a beautiful city, but it numbered among its admirable features
magniªcent governmental and religious structures, palaces, Und
the colorful Moorish quarters near the Castle of St. George in
Rossio—the main square of the city and its commercial and ad-
ministrative hub. Nearby were the Estatus Palace,
the Saint
Dominic Church, and the All Saint’s Royal Hospital. Lisbon’s
greatness was the result of the growth of the Portuguese Empire,
roughly from the ªfteenth to the seventeenth century, when the
Portuguese economy grew increasingly dependent on the mineral
wealth of Brazil. The ªrst decades of the long reign of King João V
(1706–1750) marked the height of its prosperity.6

By the time of the earthquake, Jedoch, the kingdom had
stagnated, and its economic dependence on Britain had become

5 Thomas Downing Kendrick, The Lisbon Earthquake (London, 1956), 29–33; Susan E.
Hough and Roger G. Bilham, After the Earth Quakes (New York, 2006), 41–51; Nicholas
Shrady, The Last Day (New York, 2008).
6 Kendrick, Lisbon Earthquake; José Augusto França, Lisboa Pombalina e o Iluminismo (Lisbon,
1987).

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

DISASTER STATISTICS | 31

onerous. José I (1750–1777) who inherited these troubles had the
good fortune to be served by Minister Sebastião José de Carvalho
e Melo, Count de Oeiras and Marquês de Pombal. Pombal be-
came best known for his inspired leadership in the immediate af-
termath of the earthquake, guiding the recovery and reconstruc-
tion of the city. These effective measures earned him the trust of
the king and allowed him to act as the king’s representative. Er
created various governmental agencies to take charge of the man-
ufacture and trade of sugar, diamonds, silk, ªshing, wine, hats, Und
other items. He eventually curtailed the inºuence of Britain on
Portugal’s commerce and diminished the power of the nobility
and the Church, while enhancing the importance of individual
competence for social mobility and facilitating the economic ef-
forts of Portuguese merchants and capitalists. He laid the founda-
tion for the change from absolutism to constitutional monarchy in
1826.7

the overstatement of casualty numbers Not surprisingly,
given the important short- and long-term impacts of the Lisbon
earthquake, the literature about it is voluminous, amounting to
thousands of documents. Shrady divides his many references to it
under the subheadings of architecture and urbanism, seismology,
theological and philosophical debates, empire and commerce, Und
histories and testimonies. Signiªcant for our purpose, Jedoch, Ist
the absence in this literature of any sustained interest about the
number of people that the earthquake killed or injured—perhaps
understandable in view of its other worthy considerations. So far
as this article is concerned, Jedoch, attention to the number of
casualties is important not only for what it reveals about the earth-
quake per se but also for what it might contribute to present-day
controversies regarding the difªculties of establishing valid ªgures
about worldwide disaster-related losses.8

An examination of competing estimates of the number of
people who died in the Lisbon earthquake illustrates the lack of

7 Russell R. Dynes, “The Lisbon Earthquake of 1755: The First Modern Disaster,” in The-
odore E. D. Braun and John B. Radner (Hrsg.), The Lisbon Earthquake of 1755: Darstellungen
and Reactions (New York, 2005), 34–49; Malcolm Jack, “Destruction and Regeneration: Lis-
bon, 1755,” ebenda., 7–20.
Shrady, Last Day; Braun, “Poetic Reactions of Voltaire and Le Franc de Pompignan to the Lis-
8
bon Earthquake,” available at http://www.udel.edu/ºlt/faculty/braun/Lisbon.html (zugegriffen
Marsch 8, 2011); Aguirre and Quarantelli, “Phenomenology of Death Counts in Disasters.”

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

32 | BENIGNO E. AGUI RRE

reliability in disaster statistics. It aptly demonstrates the difªculty of
obtaining valid statistics about fatalities. Even though deaths are, In
principle, much easier to quantify in the aftermath of disasters than
injuries or psychological effects are, accuracy in Lisbon’s case is
hindered by sources that did not discriminate the deaths caused by
falling debris, ªres, tsunamis, and social unrest. The evidence indi-
cates that most of the deaths occurred in Lisbon, not in the other
cities and regions affected by the earthquake and tsunamis. Al-
though it is impossible to know how various writers came up with
their estimates of deaths in Lisbon, most of them probably used a
loose deªnition of what constituted a casualty of the earthquake.
This article follows a more restricted deªnition, basierend auf
number of dead from falling debris, with various adjustments pro-
posed to reªne this ªgure.

The ªrst casualty estimate was the ofªcial assessment by the
Marquês de Pombal that between 6,000 Und 8,000 people died—
numbers that guided the subsequent reconstruction of the city and
ultimately the transformation of the entire country. Araújo re-
cently argued that Pombal was trying to downplay the real impact
of the disaster to protect Portugal’s stature as an ally of Britain on
the eve of the Seven Years’ War (1756–1763).9

Other accounts by well-placed contemporary Portuguese
witnesses gave higher estimates, but still well within the range of-
fered by Pombal. Mendonça, Jedoch, estimated 10,000 dead;
Portal 12,000 Zu 15,000; Sacramento 18,000; and Pedegache
25,000. The Portuguese Papal Nuncio’s estimate of 40,000 Todesfälle,
probably the highest for this time period, most likely reºects the
ongoing bitter controversy between the Portuguese Catholic
Church and Pombal that culminated in the expulsion of the Jesuits
In 1759 and the subsequent drastic curtailment of the wealth and
inºuence of the Church.10

9
J. S. Brasão Farinha, “Acção dos sismos sobre as construções: Simpósio sobre a acção dos
sismos e a sua consideração no cálculo das construções,” Memória n.o 112 Boletim da Ordem dos
Engenheiros (1955); idem, “Construção da Baixa Pombalina,” Cadernos do Metropolitano de
Lisboa, VI (1997); Ana Cristina Araújo, “The Lisbon Earthquake of 1755 Public Distress and
Political Propaganda,” European Journal of Portuguese History, IV (2006), available at http://
www.brown.edu/Departments/Portuguese_Brazilian_Studies/ejph/html/issue7/pdf/aaraujo
.pdf (accessed December 16, 2011).
10 França, Lisboa Pombalina, 65–66; Joaquim Moreira de Mendonça, História Universal dos
Terremotos (Lisbon, 1758); Manuel Portal, História da ruína da cidade da Lisboa causada pelo
espantoso terramoto e incêndio (Lisbon, 1758); Antonio do Sacramento, Exhortação Consolatoria de
Jesus Christo Cruciªcado na Cruz, Ao povo Lusitano, por se ver minimamente conturbado por causa do

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

DISASTER STATISTICS | 33

From the start, the number of fatalities was part of a larger re-
ligious and cultural narrative involving Catholic and Protestant or-
thodoxies and Pombal’s’ government. Father Gabriel Malagrida
interpreted the earthquake as divine retribution for the city’s lack
of devotion, urging Lisboans to repent and renounce Pombal’s
Enlightenment philosophy. Eventually Malagrida paid with his
life; Pombal was famous for holding grudges. Dutch Calvinists and
other Protestants used the same religious logic to draw different
conclusions. According to Wesley, God was displeased and of-
fended by the cruelty of the Portuguese Inquisition and wide-
spread idolatry. Both religious narratives inºated the numbers of
the dead and the suffering of the people.11

In the twentieth century, the number of estimates prolifer-
ated. Pereira de Sousa, a Portuguese geologist, discovered material
in the National Archives convincing him that between 15,000 Und
20,000 people had perished. He found a number of questionnaires
circulated by Pombal to parish priests—probably the ªrst post-
disaster seismic questionnaires in history—to ascertain the effects
of the earthquake on their parishes. Portuguese historians made
their own calculations. Portugal
later calculated 12,000 dead.
França, in his study of Lisbon’s reconstruction and recovery after
the earthquake, accepted the 10,000-dead ªgure of Mendonça.
Oliveira accepted Pombal’s high estimate of 8,000. Marques is
known to have surmised that 5,000 people died in the city imme-
diately from falling debris and 5,000 later from heart attacks and
injuries. Other well-informed scholarly books are those by
Kendrick, Maxwell, and Leidner. The ofªcial position of the Por-
tuguese government is that between 7,000 Zu 10,000 people died,
Und 20,000 were injured.12

Terremoto de primeiro de Novembro de 1755 (Lisbon, 1757); Miguel Tibério Pedegache, Nova e ªel
relação do terramoto que experimentou Lisboa, e todo o Portugal no 1 de Novembro de 1755 com algunas
observações curiosas, e a explicação das suas causas (Lisbon, 1756); Dynes, “Lisbon Earthquake,”
34–43.
11 Gabriel Malagrida, juízo da verdadera causa do terramoto (Lisbon, 1756); John Wesley, Seri-
ous Thoughts Ocassioned by the Great Earthquake at Lisbon (London, 1755); Shrady, Last Day,
113–146; Kendrick, Lisbon Earthquake, 72–118.
12 Francisco Luis Pereira de Sousa, “O terramoto do 1 de novembro de 1755 en portugal,”
Serviços Geol’ogicos, ICH (1919), 1–277 (a special issue of the journal devoted to Sousa’s ªndings);
Fernando Portugal, Lisboa em 1758: memórias paroquiais de Lisboa (Lisbon, 1974); França, Lisboa
Pombalina, 66; Carlos Sousa Oliveira, “Descrição do Terramoto de 1755, Sua extensão, causas
e efeitos: O sismo, O tsunami. O incêndio,” in Fundacão Luso-Americana para o Desen-
volvimento (Hrsg.), O Grande Terremoto de Lisboa. ICH. Descrições. (Lisbon, 2005), 23–86; A. H.

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

34 | BENIGNO E. AGUI RRE

The majority of writings today reproduces, or even embel-
lishes, the most outlandish claims of the eighteenth and nineteenth
centuries. Zum Beispiel, Pereira states that deaths in Lisbon alone
numbered 30,000 Zu 40,000. Braun mentions that 30,000 gestorben, Und
100,000 were injured. Continuing with the inºation of numbers,
the Paciªc Disaster Center, Levin, Rée (writing for the Times Lit-
erary Supplement), and the Encyclopædia Britannica set the number of
fatalities at 60,000; the Institute of Computational Mathematics
and Mathematical Geophysics of the Siberian Division of the Rus-
sian Academy of Sciences at 80,000; Warnes at 90,000; and noaa’s
National Geophysical Data Center (ngdc) bei 60,000 Zu 100,000.
Most recently, Hough and Bilham, unter anderen, in their recent
and otherwise excellent monograph on earthquakes, estimate
close to 100,000 dead. The inºation in the number of people who
are said to have perished in Lisbon because of the quake is impos-
sible to ignore.13

reasons for overstating, and doubting, the casualty claims
The present-day tendency is to assume a much greater number of
dead in the Lisbon earthquake than the relevant facts can accom-
modate. Wenn, as Rodrigues argues, the population of Lisbon at the

Oliveira Marques, Breve História de Portugal (Lisbon, 2006), 391; Kenneth Maxwell, Pombal,
Paradox of the Enlightenment (New York, 1995); Allison Leidner, “The Lisbon Earthquake of
1755,” available at http://water.stanford.edu/nur/GP50/allison.pdf (abgerufen im Juli 16, 2009).
Serviço Nacional de Protecção Civil, Lisboa, Programa de acções para minimização do risco sísmico:
Fase I: Primeiro Relatório ªnal Plano Nacional de Emergência. Presidência do Conselho de Ministros
(Lisbon, 1983).
13
S. Álvaro Pereira, “The Opportunity of a Disaster: The Economic Impact of the 1755
Lisbon Earthquake,” available at http://www.york.ac.uk/res/cherry/docs/Alvaro3.pdf (ac-
cessed July 16, 2009); Braun, “Poetic Reactions of Voltaire”; Paciªc Disaster Center, Geschichte
of Great Earthquakes, available at http://74.125.47.132/suchen?Q(cid:2)cache:K2Lev0Wg9VcJ
:www.pdc.org/iweb/earthquake_history.jsp(cid:3)lisbon(cid:3)earthquake(cid:3)Paciªc(cid:3)Disaster(cid:3)Center
&cd(cid:2)1&hl(cid:2)In&ct(cid:2)clnk&gl(cid:2)us&client(cid:2)ªrefox-a (abgerufen im Juli 17, 2009); Boris Levin,
“Tsunamis: Causes, Consequences, Prediction, and Response,” available at www.undl.org/
unldoc/EOLSS/E4-06-01-06-TXT.aspx (accessed March 8, 2011); Jonathan Rée, “A Mean
and Rootless Fungus,” Times Literary Supplement, XVIII, 18 Oct. 2002, available at http://
www.susan-neiman.de/docs/book05.html (abgerufen im Juli 17, 2009). Another high estimate
came from the Institute of Computational Mathematics and Mathematical Geophysics of the
Siberian Division Russian Academy of Sciences. See also Kathy Warnes, “The 1755 Lisbon
Earthquake—Pombal Uses Science to Rebuild,” Suite 101, available at http://www.
suite101.com/content/the-1755-lisbon-earthquake—-pombal—uses-science-to-rebuild-
a259781) (accessed March 7, 2010); ngdc, “November 1, 1775, Lisboa, Portugal,” available at
http://www.ngdc.noaa.gov/hazard/tsu_travel_time_events.stml (accessed June 12, 2008);
Hough and Bilham, After the Earth Quakes, 265.

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

DISASTER STATISTICS | 35

time of the earthquake was approximately 191,000 persons, Die
higher estimates of 40,000 Zu 100,000 dead would mean that ap-
proximately one-quarter to more than one-half of the city’s popu-
lation perished, a level of mortality that ªnds no support in the
contemporary record.14

An important question derived from this case study, and other
similar ones, is why the number of dead and injured victims, eco-
nomic losses, and psychological impairments caused by disasters
have a tendency to increase over time. One reason is pecuniary.
Albala-Bertrand writes that distortions are often introduced by
“local or national government, reºecting their vested interest in
proªting from outside assistance, by magnifying all categories of
loss, especially death tolls, so as to elicit a higher inºow of response
resources, less pressure from international creditors, and better
longer-term borrowing conditions.” Many of the ªgures ad-
vanced, especially for technological disasters and increasingly for
natural disasters, derive from interest and advocacy groups that
usually have a vested concern to make situations look worse than
they actually are. Most of the quantitative ªgures produced in the
United States overstate deaths, injuries, economic losses, and other
presumed consequences of disasters, such as looting and criminal
behavior, volunteering, longer-run divorce rates, and persistence
of psychological problems.15

In certain cases, ofªcials from governments and nongovern-
mental organizations inºate the numbers for their own advantage.
But another contributing factor is that people like big numbers,
and news outlets, bureaucrats, Wissenschaftler, and academicians have all
been known to enjoy delivering the “big” story. The emphasis of
the mass media on extraordinary and unusual circumstances and
reporters’ general ignorance of what actually happens in commu-
nities in the aftermath of disasters is well documented. Noch, diese
popular outlets are often responsible for sustaining disaster myths.
Alexander documents many of these misrepresentations in his de-
tailed examination of the Haitian earthquake of January 2010. In-

14 Teresa Rodrigues, “Lisboa: Das longas permanências demográªcas à diversidade social,”
available at http://geo.cm-lisboa.pt/ªleadmin/GEO/Imagens/GEO/Demograªa_historica/
Jornadas_demograªa_2008/Demograªa_doc1.pdf (accessed December 16, 2011); see also
idem, Cinco Séculos de Quotidiano: A vida em Lisboa de Século XV aos nossos dias (Lisbon, 1997).
15
1993), 41.

José M. Albala-Bertrand, The Political Economy of Large Natural Disasters (New York,

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

36 | BENIGNO E. AGUI RRE

cluded in the news about this disaster was erroneous information
about the health hazard of bodies left unburied, the occurrence of
disease epidemics, survivors emerging dazed and apathetic from
long periods under rubble, and the need for military intervention
to reduce the likelihood of looting and social chaos. Alexander
also discusses the ways in which the mass media misunderstood the
need for humanitarian aid, medicines, and ªeld hospitals, more in-
tent on creating villains and heroes.16

The exaggerations with which survivors and their communi-
ties often describe their ordeals reºect a certain “pride of place”
and an inºated sense of their ability to handle adversity, born of
immense relief. In vielen Instanzen, as the Lisbon earthquake dem-
onstrates ideology and pre-judgment play a part in distorting the
effects of disasters. The overstatements that attended the Lisbon
earthquake are an example of sedimentation—the creation, repeti-
tion, and accumulation of unveriªed claims over time that at some
point become self-sustaining facts. Although the goal of scientiªc
discovery is to avoid the dangers of replication, cumulative error is
always a possibility.

a methodology for calculating reliable casualty estimates
Given the importance of the effects of the earthquake on Lisbon’s
Entwicklung, Portuguese engineers and seismic scientists in the
twentieth century studied the earthquake from the perspective of
modern plate tectonics and the characteristics of Lisbon’s housing
stock when the disaster struck. Coelho examined the engineering
principles employed in the reconstruction of the city, and Oliveira

16 David E. Alexander, “News Reporting of the January 12, 2010, Haiti Earthquake: Der
Role of Common Misconceptions,” Journal of Emergency Management, VIII (2010), 15–27. Der
estimate of the number of people who died in the 2010 Haiti earthquake went from 40,000 Zu
50,000 in the ªrst days after the quake to approximately 100,000 by the end of the ªrst two or
three weeks, und zu 230,000 months later. Im Januar 2011, the estimate reached 300,000, als
given by Robert Maguire in “Rebuild Haiti, Not Just Its Capital,” Contemporary History, CX
(2011), 81–82. Among the many examples of unveriªed claims that are often considered fac-
tual is the number of people who died in the ªrestorm that destroyed large parts of Dresden,
Deutschland, on February 13 Und 14, 1945. At the time of the air raid, the British estimated that
the attack killed around 16,000. Von 1948, Jedoch, two German generals had raised the num-
ber to 250,000; in the 1950s and 1960s, scholars settled on numbers near 100,000. Apparently,
the true number is 25,000 Zu 30,000 victims; the number of people buried after the attack in
municipal cemeteries was 21,271. Details are presented by Rebecca Grant in “The Dresden
Legend,” Airforce Magazine (Oktober 2004), available at http://www.airforce-magazine.
com/MagazineArchive/Pages/2004/October%202004/1004dresden.aspx.

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

DISASTER STATISTICS | 37

studied the source and spread of the seismic forces and its differen-
tial effects on the built environment. This important line of
scientiªc investigation culminated in the work of Sousa, welche
borrowed engineering formulas proposed by The Federal Emer-
gency Management Agency (fema), Coburn, Spence, and Pomonis
and Tiedemann, to estimate ranges of numbers of fatalities.17

Sousa adopted a range for the total population for Lisbon at
the time of the earthquake that agrees with the best historical writ-
ings on the city, from a low of 150,000 to a high of 260,000. Wenn
Rodrigues’ well-regarded analysis of Portugal during the eigh-
teenth century is correct, Lisbon at the time of the earthquake had
a population of 191,052; the city would not reach that size again
for half a century. Sousa’s computer simulations created a seismic
scenario similar to that of 1755. For different housing damage and
casualty models based on various assumptions, the seismic death
toll varies between 0.3 percent and 3.3 percent of the population.
For the low estimate of 150,000 Menschen, the number of fatalities
reicht von 450 Zu 4,950; for the high estimate of 260,000 Menschen,
it ranges from 780 Zu 8,580.18

Given these comparatively low estimates in the face of the
much higher ones already noted, this article reproduces Sousa’s
ªndings via a different approach—applying standard demographic
methods to the Catholic Church’s death rate in Lisbon to deter-
mine the total number of people who died in the city on the day
of the earthquake. Although the requisite information about the
Catholic Church in Lisbon is not complete at this time, the discus-
sion clearly displays the reasoning behind the results permitted by
the current data, laying the foundation for future reªnement.

Calculating death rates in an afºicted population requires
knowing the number of dead and the population at risk. Wann
the information does not cover the entire population, as in this

17 A. Gomes Coelho, “Do ‘inquérito do Marquês de Pombal’Ao Estudo de Pereira de
Sousa sobre o Terramoto de 1 de Novembro de 1775,” in Fundacão Luso-Americana para o
Desenvolvimento (Hrsg.), A Grande Terramoto de Lisboa. ICH. Descrições (Lisbon, 1985),143–187;
Sousa Oliveira, “Descrição do Terramoto de 1755, 41–66; Maria Luísa Sousa, “Risco Sísmico
em Portugal Continental,” unpub. Ph.d. diss. (University of Lisbon, 2008), 236; Andrew W.
Coburn, Robin J. S. Spence, and Antonios Pomonis, “Factors Determining Human Casualty
Levels in Earthquakes: Mortality Prediction in Building Collapse,” Proceedings of the Eleventh
World Conference on Earthquake Engineering, Madrid, X (1992), 5989–5994; HerbertTiedemann,
Earthquakes and Volcanic Eruptions: A Handbook on Risk Assessment (Zurich, 1992).
18 Rodrigues, Lisboa: Das longas permanencias.

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

38 | BENIGNO E. AGUI RRE

Fall, there is always some question concerning the extent to
which the death rate of a sub-population can represent that of the
general population. Portugal’s list of the 266 ofªcials of the Catho-
lic Church who died in Lisbon as a result of the earthquake is
credible, corroborated by other sources. Many of the churches and
monasteries in Lisbon were destroyed. França and Sousa Oliveira
document that only eleven of the sixty-ªve convents in the city
survived: fewer than ten of the forty churches were able to escape
severe damage. Leonor Ferrao listed the hundreds of large reli-
gious buildings in Lisbon at the time of the earthquake, sowie
the many that were destroyed. The earthquake happened on All
Saints’ Day, November 1, während 9 Bin. mass—a time when a
large proportion of church personnel would have been present at
religious services.19

The membership numbers of Lisbon’s Catholic Church in
1755 would be difªcult to obtain without extensive research in the
Arquivo Histórico do Patriarcado de Lisboa. The Perilongue Cen-
sus of 1765, Jedoch, established the number of parishioners in
the national Church at 42,200. Higgs estimated its size to be half of
the Spanish Catholic Church, oder 91,389 persons. Whichever of
these two numbers is more accurate, the exact percentage of
Church ofªcials in Lisbon on the day of the earthquake remains a
mystery, but certain educated guesses are possible.20

According to Higgs, the Pino Manique Census published in
1789 found 477 religious houses in the country for 1739. Estrema-
dura, the region of Portugal in which Lisbon is located, accounted
für 38.2 percent of them. A good many of these buildings were
probably located in Lisbon, which at the time of the earthquake
had thirty-seven parishes, thirty-two monasteries, and eighteen
nunneries. Assuming that the proportion of these houses in the
city is a good proxy for the size of the Portuguese Church in Lis-
bon, and given the absence of any other viable information, Das
article establishes that 19 Zu 38 percent of Estremadura’s Church
personnel resided in Lisbon. Although the highest percentage is

19 Portugal, Lisboa em 1758, 307, 310; França, Lisboa Pombalina, 67; Sousa Oliveira,
“Descrição do Terramoto de 1755,” 56, 60; Leonor Ferrão, “Desenvolvimento urbanístico:
Os palácios e os conventos,” in Irisalva Moita (Hrsg.), O Livro de Lisboa (Lisbon, 1994), 239–282.
20 Nicholas Atkin and Frank Tallett, Priests, Prelates & People: A History of European Catholi-
cism since 1750 (New York, 2003); David Higgs, “The Portuguese Church,” in William J.
Callahan and idem (Hrsg.), Church and Society in Catholic Europe of the Eighteenth Century (Neu
York, 1979), 51–65.

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

DISASTER STATISTICS | 39

highly unlikely, it provides an effective top range nonetheless. Noch,
the numbers of Church personnel must have been unusually large.
Foreigners visiting the city often remarked about the size and
growth of the Catholic Church in Lisbon, and the Portuguese
Crown even tried to limit the swelling number of Church person-
nel in Lisbon by denying their requests for more religious houses
throughout the seventeenth and eighteenth centuries.21

the death rates of

Results Table 1 zeigt an

the Catholic
Church in Lisbon for various sets of assumptions. Wenn 38 percent of
the Catholic Church’s personnel resided in Lisbon at the time of
the earthquake, the death rate for the two Church population esti-
mates are .016667 Und .007660, jeweils. Assuming the more
probable distribution, according to which only half of the Catho-
lic Church’s personnel in Estremadura resided in Lisbon, the death
rates for the two Church population estimates are .033333 Und
.015319, jeweils. Using these death rates and the two popula-
tion estimates of the city, the table presents the estimated number
of people who died in Lisbon. If half, oder 19 Prozent, of the reli-
gious personnel in Estremadura were inside Lisbon’s city walls,
deaths for the low population ªgure numbered 2,298 Und 5,000.
For the high population ªgure, the corresponding numbers are
3,983 Und 8,867. Assuming that all religious personnel in Estrema-
dura were inside the city walls, the corresponding numbers of
dead people are 1,149 Zu 2,500 when the population size of the
city is 150,000, Und 1,992 Zu 4,334 when it is 260,000.

It is reassuring to ªnd that the results of this study conform in
almost all respects with Sousa’s earlier ªndings. Although her
lower estimations of 450 Und 780 are artifacts of her methodology
without historical validity, her upper numbers are close to the
ªndings of this study given the assumption that 19 Prozent der
Portuguese Catholic Church personnel resided in Lisbon. Her es-
timate of 4,950 fatalities falls inside the range of 2,298 Zu 5,000
dead found in this study for a population of 150,000, and her esti-
mate of 8,580 fatalities falls within the range of 3,983 Zu 8,667
deaths for a population of 260,000. The combined ªndings of both
studies indicate that falling debris during the Lisbon earthquake
probably killed between 5,000 Und 8,700 people in Lisbon.

Shrady, Last Day, 9; Braga et al., Lisboa Setecentista Vista por Estrangeiros (Lisbon, 1992),

21
38–82; Fernando Castelo Branco, Lisboa Seiscentista (Lisbon, 1990), 67–69.

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

40 | BENIGNO E. AGUI RRE

Tisch 1 Number of Dead from Earthquake, as Calculated by Death Rate,
Proportion of Catholic Church Personnel in Lisbon, and the Size of
Lisbon’s Population

personnel in portugal’s church (upper and lower estimates)

42,000

91,389

proportion of personnel in lisbon (upper and lower estimates)

19%

38%

19%

38%

death rate (based on 266 Todesfälle)

.03333

.01667

.01532

.00766

total deaths in lisbon based on population of 150,000

5,000

2,500

2,298

1,149

total deaths in lisbon based on population of 260,000

8,667

4,334

3,983

1,992

notes and sources The upper and lower estimates of Portugal’s Church personnel derive,
jeweils, from Nicholas Atkin and Frank Tallett, Priests, Prelates & People: A History of Eu-
ropean Catholicism since 1750 (New York, 2003); David Higgs, “The Portuguese Church," In
William J. Callahan and idem (Hrsg.), Church and Society in Catholic Europe of the Eighteenth Cen-
tury (New York, 1979), 51–65. The upper and lower estimates of the total population in Lis-
bon derive from Maria Luísa Sousa, Risco Sísmico em Portugal Continental (Lisbon, 2008). Der
number of Church ofªcials who died in the earthquake, 266, used to calculate the death rate,
derives from Fernando Portugal, Lisboa em 1758: memórias paroquiais de Lisboa (Lisbon, 1974),
307, 310.

For a complete picture of the devastation, we can add to these
deaths the victims of subsequent ªres, tsunamis, and civil unrest,
whom neither Sousa’s work nor this study included in its esti-
Kumpels. Personnel of the Catholic Church (the only group about
which demographic analysis about the earthquake can be done at
this point) probably comprised one of the most decimated sub-
populations in Lisbon during the earthquake. Other groups also at
high risk but without documentation were dock workers and
other laborers in the port area of the River Tagus, as well as chil-
dren and the elderly, who usually suffer severely in earthquakes,
and the poor. Due to the lack of reliable information, we cannot
even speculate about their plight.

Antunes recently discovered that young people predominated
among the mortal remains of victims found in the graves at the
Convento de Jesus in Lisbon. The injuries, deep cuts, and burns

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

DISASTER STATISTICS | 41

evident in their bones reveal the difªculty of separating the effects
of the falling debris from the ensuing tumult that engulfed the city
and its environs in those fateful days. By way of contrast, few
members of the nobility and the more well-to- do families died in
the earthquake; these people generally left the city on Sundays and
attended religious services in the chapels of their mansions.22

Although no data are available to substantiate the number of
people who died as a result of the ªres, tsunamis, and civil unrest
that accompanied the earthquake, roughly the same number of
people might well have died due to these causes as from falling de-
bris, bringing the total fatalities to a range of 10,000 Zu 17,400,
close to Pereira de Sousa’s estimate of 15,000 Zu 20,000.

Whether or not the Marquês de Pombal underreported the effects
of the tragedy depends on whether the deaths caused by catastro-
phes after the quake are included in the estimate. Our ªndings
support his initial claims only if they count deaths from falling
debris. Jedoch, Pombal’s early estimates might not have been
high enough because the full scale of the disaster was not yet evi-
dent. Hopefully, future historical research will shed light on these
matters.

Much more attention needs to be placed on the problem of
faulty disaster statistics to facilitate initiatives for mitigation and re-
silience in the aftermath of disasters. The creation of a worldwide
network of national centers for disaster statistics would help to
ameliorate, if not solve, dieses Problem. In der Tat, as far back as 1966, A
committee of the Division of Medical Sciences of the National
Academy of the Sciences (nas) recommended the development of
a national center in the United States to “document and analyze
types and numbers of casualties in disasters, to identify by on-site
medical observation the problems encountered in caring for disas-
ter victims, and to serve as a national educational and advisory
Körper . . . to meet the needs imposed by disaster or national emer-
gency.”23

Certainly historians should become more interested in the

22 Miguel Telles Antunes, “Victims of the Great 1755 Earthquake at the Convento de
Jesus,” E-Terra, III (2006), available at http://e-terra.geopor.pt; Shrady, Last Day, 52.
23 Committee on Trauma and Committee on Shock, Division of Medical Sciences, Bereits-
tional Research Council, Accidental Death and Disability, The Neglected Disease of Modern Society
(Washington. D.C., 1966), 29. See also the more recent effort to create a national center in
Quarantelli, Possibilities and Problems in Developing a Global Disaster Information Network (GDIN)
(Newark, 1998).

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3

42 | BENIGNO E. AGUI RRE

study of disasters and the lessons that can be learned from them.
Zum Beispiel, a little attention to history could have helped to an-
ticipate, and even obviate, the problems created by the recent
militarization of emergency management in the United States (als
reºected in the downgrading of fema and the ascendance of the
Department of Homeland Security [dhs] after the terrorist attacks
von 9/11). After all, emergency management replaced the military
intervention that had been in vogue during the Cold War for
good reason. During the decades after the war, the military proved
incapable of handling the demands that disasters generated, wie zum Beispiel
early warning, evacuation, sheltering, reconstruction, and recov-
ery. Nor could it foster the necessary change in culture to mitigate
the risks associated with disasters and increase the resilience of so-
cial institutions in their wake.24

While we await a new era of disaster preparation and re-
sponse, we should learn to take with a grain of salt the reports of
news organizations, the popular press, and even scientiªc publica-
tions that either do not have the technical proªciency to make
claims or the fact-checking capacity to conªrm them. As the ex-
perience of the Lisbon earthquake of 1755 attests, the effects of di-
sasters are frequently inºated to mythic proportions.

24 For an example of recent historical research about disaster, see Gerrit Jasper Schenk and
Jens Ivo Engels (Hrsg.), “Historical Disaster Research: Concepts, Methods and Case Studies,”
introduction to a special issue of Historical Social Research, XXXII (2007), 9–31; Monica Juneja
and Franz Mauelshagen (Hrsg.), Coping with Natural Disasters in Pre-Modern Societies (Neu
Delhi, 2007).

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u

/
J
ich

N
H
A
R
T
ich
C
e
–
P
D

F
/

4
3
1
2
7
1
7
0
5
7
6
2

/
J
ich

N
H
_
A
_
0
0
3
3
7
P
D

B
j
G
u
e
S
T

Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3
PDF Herunterladen