33 Pages
English

Mortality in the British peerage families since 1600 - article ; n°1 ; vol.32, pg 323-352

-

Gain access to the library to view online
Learn more

Description

Population - Année 1977 - Volume 32 - Numéro 1 - Pages 323-352
Mortalité des familles nobles britanniques depuis 1600 Т.Н. HOLLINGSWORTH Les données sur la mortalité de l'aristocratie britannique depuis 1600, publiées par l'auteur en 1964, sont ici réanalysées de manière plus approfondie. Les données brutes ont d'abord été corrigées, en raison de la mauvaise répartition des âges au décès (méthode exposée en annexe) : les nombres corrigés de décès par sexe, cohorte et âge, figurent dans le tableau 1. Le tableau 2 donne les quotients de mortalité correspondants, et le tableau 3 les espérances de vie à chaque âge. On a ensuite comparé l'évolution de la mortalité avant 40 ans et après 40 ans, en calculant un taux moyen de mortalité avant 40 ans (à partir du taux de survie à 40 ans) et un taux moyen après 40 ans (à partir de l'espérance de vie à 40 ans) ; le calcul a été fait pour l'ensemble des deux sexes (tableau 4). Dans les cohortes les plus anciennes, la mortalité après 40 ans était forte, tandis que la mortalité avant 40 ans était relativement faible mais elle a augmenté après pendant environ un siècle. Le tableau 5 permet de préciser l'analyse, en distinguant trois groupes d'âges : moins de 25 ans, 25-54, 55 ans et plus. Il en ressort que la hausse de mortalité n'affectait, en fait, que le premier groupe d 'âges. Les taux moyens par grands groupes d'âges (25 ans) ont été ensuite transformés en taux du moment (tableau 6). Dans le même tableau figure, pour chaque période, le taux de mortalité générale résultant, par application des taux par âge à une structure-type. On voit ainsi que la baisse de la mortalité a été très rapide entre 1 750 et 1 775 (environ). Revenant à l'analyse par cohorte, on a calculé (tableau 7) des taux de mortalité relatifs, par sexe, âge et cohorte, en prenant pour base les taux moyens des cohortes 1550-1699. Dans le tableau 8, on a fait la moyenne des taux relatifs portant sur la même période (quels que soient les âges concernés), ce qui donne un nouvel indicateur de mortalité du moment, pour des périodes de 30 ans décalées de 5 en 5 ans. Enfin, dans le tableau 9, on a calculé par moyenne mobile un indicateur centré sur un 1er janvier de 5 en 5 ans. On trouve ainsi trois périodes de diminution rapide de la mortalité : 1860-1874 et 1800-1819, dont on avait déjà des indices pour la population totale ; et 1 745-1 764, qui n 'était pas soupçonnée. Diverses analyses des tables de mortalité relatives aux cohortes 1550-1699 sont ensuite effectuées : composantes principales (du type Ledermann), logit (Brass), comparaison avec les tables de Princeton (tableau 10). L' appendice décrit la méthode de correction des données, pour la répartition des âges non déclarés. Le tableau Al montre la répartition des âges au décès à 2 et 3 ans. Les tableaux A2, A3 et A4 décrivent la procédure de répartition de ces décès sur les âges voisins. Le cas de la mortalité adulte est traité ensuite (tableau A 5).
Mortality in the British peerage families since 1600. Т.Н. HOLLINGSWORTH The data on the death rate of British peerage since 1600, published by the author in 1964, are re-examined here in greater detail. The original figures have first been corrected on account of the poor distribution of ages at death (method shown in the appendix): the corrected death figures by sex, cohort and age are given in table 1. Table 2 shows the corresponding values of qx, and table 3, the life-expectancy at each age. The movements in death rates below and above age 40 were then compared by calculating an average death rate before age 40 (from the survivors aged 40) and an average rate above age 40 (from the life-expectancy at 40): this computation was made for all persons of both sexes (table 4). In the oldest cohorts, the death rate above age 40 was high, whilst that below age 40 was relatively low but it then increased for about a century. Table 5 permits a more detailed analysis by distinguishing three age-groups : under 25, 25-54, 55 and above. It is shown that the increase in the death rate affected only the first age-group. The average rates for large age groups (25 years) were then converted into period rates (table 6). The death rate obtained by applying the age- specific rates to a standard population is given for each period. It may thus be seen that the drop in mortality was very rapid between 1750 and 1775 approximately. Reverting to cohort analysis, in table 7 relative death rates by sex, age and cohort are calculated, with the average rates of the 1550-1699 cohorts as a base. In table 8 the average of the ratios relating to the same period (for all ages) are computed, and this gives a new current index of mortality for periods of 30 years every of 5 years. Finally, an index centered on the 1st January every five years is calculated by means of a moving average in table 9. Three periods when the death rate diminished rapidly are found: 1860- 1874 and 1800-1819, periods for which death rates for the total population were available, and 1 745-1 764, a period when no decline was suspected. Different analyses were then made of the life tables for the 1550-1699 cohorts: principal components (of the Ledermann type), logit (Brass) and a comparison with the Princeton tables (table 10). The appendix describes the method of adjusting data to allow for unstated ages. Table Al provides the distribution of ages at death from 0 to 8 years in the unweighted data; there appears an abnormal number of deaths at ages 2 and 3. Tables A2, A3 and A4 describe the method of distributing these deaths over the neighbouring ages. Adult mortality is then dealt with (table A5).
La mortalidad de las f amilias de la noblesa británica a partir de 1 600 Т.Н. HOLLINGSWORTH Los datos sobre la mortalidad de la aristocracia británica a partir de 1600, publicadas por el autor en 1964, se reanalisan aqui de una manera más profunda. Primero se corrigieron los datos brutos debido a la mala distribución de las edades al morir (según el método expuesto en el anexo) : las cifras corregidas de las muertes por sexo, cohorte y edad, se exponen en la tabla 1. La tabla 2 proporciona los cuocientes de mortalidad correspondientes y la tabla 3 las esperanzas de vida a cada edad. Se compara a continuación la evolución de la mortalidad antes de 40 aňos y después de 40 aňos, calculando las tasas médias de mortalidad respectivas, mediante la tasa de sobreviviencia a 40 ano s en el primer caso y a partir de la esperanza de vida a los 40 aňos, para la tasa media correspondiente a la mortalidad sobre 40 aňos. El cálculo se realizó para ambos sexos (tabla 4). En las cohortes mas antiguas, la mortalidad sobre 40 aňos era mas alta, mientras que la mortalidad antes de los 40 aňos era relativamente baja y aumentó enseguida, durante casi todo un siglo. La tabla 5 permite precisar el análisis, en relación a très grupos de edad : menores de 25 aňos, 25-54, 55 y más. Se puede apreciar que la elevación de la mortalidad afectó sólo al primer grupo de edad. Estas tasas médias por grandes grupos de edad (25 aňos) fueron transformadas a continuanción en tasas de momento (tabla 6). En la misma tabla se expone la tasa de mortalidad general résultante, para cada periodo, mediante la aplicaciôn de las tasas por edad a una estructura-tipo . Se ve que la mortalidad expérimenta una râpida diminución entre 1750 y 1775 (approxi- madamente). Volviendo al análisis por cohorte, se calcularon (tabla 7) tasas de mortalidad relativas, por sexo, edad y cohorte, tomando со то base las tasas médias de las cohortes 1550-1699. En la tabla 8, se calcula la media de las tasas relativas correspondientes al mismo periodo (independientemente de las edades) lo que da un nuevo indicador de mortalidad del momento para periodo s de 30 aňos, desplazados de 5 en 5 aňos. Por ultimo en la tabla 9, se ha calculado, mediante promedios mo'viles, un indicador referido al primero de enero, de 5 en 5 aňos. Se encuentran asi très periodos de disminución rápida de la mortalidad : 1860-1874 y 1800-1819, para los cuales ya se tenian indices para la población total, y 1 745-1 764, para el cual no se ténia dudas. Se efectuaron a continuación diversos análisis de las tablas de mortalidad relativas a las cohortes 1550-1699 : componentes principales (del tipo Ledermann). logito (Brass), comparación con las tablas de Princeton (tabla 10). En el apéndice se describe el método utilizado para la correcciôn de los datos para la distribución de las edades no declaradas. La tabla Al muestra la distribución de las muertes entre las edades de 0 a 8 aňos en los datos brutos ; aparece un numero anormal de defunciones entre 2 y 3 anos. Las tablas A2, A3, y A4 describen el pro cedimiento de distribución de estas muertes en las edades vecinas. Después, se discute el caso de la mortalidad adulta (tabla A 5).
30 pages
Source : Persée ; Ministère de la jeunesse, de l’éducation nationale et de la recherche, Direction de l’enseignement supérieur, Sous-direction des bibliothèques et de la documentation.

Subjects

Informations

Published by
Published 01 January 1977
Reads 12
Language English
Document size 3 MB

T.h. Hollingsworth
Mortality in the British peerage families since 1600
In: Population, 32e année, n°1, 1977 pp. 323-352.
Citer ce document / Cite this document :
Hollingsworth T.h. Mortality in the British peerage families since 1600. In: Population, 32e année, n°1, 1977 pp. 323-352.
http://www.persee.fr/web/revues/home/prescript/article/pop_0032-4663_1977_hos_32_1_16483Résumé
Mortalité des familles nobles britanniques depuis 1600 Т.Н. HOLLINGSWORTH Les données sur la
mortalité de l'aristocratie britannique depuis 1600, publiées par l'auteur en 1964, sont ici réanalysées de
manière plus approfondie. Les données brutes ont d'abord été corrigées, en raison de la mauvaise
répartition des âges au décès (méthode exposée en annexe) : les nombres corrigés de décès par sexe,
cohorte et âge, figurent dans le tableau 1. Le tableau 2 donne les quotients de mortalité
correspondants, et le tableau 3 les espérances de vie à chaque âge. On a ensuite comparé l'évolution
de la mortalité avant 40 ans et après 40 ans, en calculant un taux moyen de mortalité avant 40 ans (à
partir du taux de survie à 40 ans) et un taux moyen après 40 ans (à partir de l'espérance de vie à 40
ans) ; le calcul a été fait pour l'ensemble des deux sexes (tableau 4). Dans les cohortes les plus
anciennes, la mortalité après 40 ans était forte, tandis que la mortalité avant 40 ans était relativement
faible mais elle a augmenté après pendant environ un siècle. Le tableau 5 permet de préciser l'analyse,
en distinguant trois groupes d'âges : moins de 25 ans, 25-54, 55 ans et plus. Il en ressort que la hausse
de mortalité n'affectait, en fait, que le premier groupe d 'âges. Les taux moyens par grands groupes
d'âges (25 ans) ont été ensuite transformés en taux "du moment" (tableau 6). Dans le même tableau
figure, pour chaque période, le taux de mortalité générale résultant, par application des taux par âge à
une structure-type. On voit ainsi que la baisse de la mortalité a été très rapide entre 1 750 et 1 775
(environ). Revenant à l'analyse par cohorte, on a calculé (tableau 7) des taux de mortalité relatifs, par
sexe, âge et cohorte, en prenant pour base les taux moyens des cohortes 1550-1699. Dans le tableau
8, on a fait la moyenne des taux relatifs portant sur la même période (quels que soient les âges
concernés), ce qui donne un nouvel indicateur de mortalité "du moment", pour des périodes de 30 ans
décalées de 5 en 5 ans. Enfin, dans le tableau 9, on a calculé par moyenne mobile un indicateur centré
sur un 1er janvier de 5 en 5 ans. On trouve ainsi trois périodes de diminution rapide de la mortalité :
1860-1874 et 1800-1819, dont on avait déjà des indices pour la population totale ; et 1 745-1 764, qui n
'était pas soupçonnée. Diverses analyses des tables de mortalité relatives aux cohortes 1550-1699 sont
ensuite effectuées : composantes principales (du type Ledermann), logit (Brass), comparaison avec les
tables de Princeton (tableau 10). L' appendice décrit la méthode de correction des données, pour la
répartition des âges non déclarés. Le tableau Al montre la répartition des âges au décès à 2 et 3 ans.
Les tableaux A2, A3 et A4 décrivent la procédure de répartition de ces décès sur les âges voisins. Le
cas de la mortalité adulte est traité ensuite (tableau A 5).
Abstract
Mortality in the British peerage families since 1600. Т.Н. HOLLINGSWORTH The data on the death rate
of British peerage since 1600, published by the author in 1964, are re-examined here in greater detail.
The original figures have first been corrected on account of the poor distribution of ages at death
(method shown in the appendix): the death figures by sex, cohort and age are given in table
1. Table 2 shows the corresponding values of qx, and table 3, the life-expectancy at each age. The
movements in death rates below and above age 40 were then compared by calculating an average
death rate before age 40 (from the survivors aged 40) and an average rate above age 40 (from the life-
expectancy at 40): this computation was made for all persons of both sexes (table 4). In the oldest
cohorts, the death rate above age 40 was high, whilst that below age 40 was relatively low but it then
increased for about a century. Table 5 permits a more detailed analysis by distinguishing three age-
groups : under 25, 25-54, 55 and above. It is shown that the increase in the death rate affected only the
first age-group. The average rates for large age groups (25 years) were then converted into "period
rates" (table 6). The death rate obtained by applying the age- specific rates to a standard population is
given for each period. It may thus be seen that the drop in mortality was very rapid between 1750 and
1775 approximately. Reverting to cohort analysis, in table 7 relative death rates by sex, age and cohort
are calculated, with the average rates of the 1550-1699 cohorts as a base. In table 8 the average of the
ratios relating to the same period (for all ages) are computed, and this gives a new "current index" of
mortality for periods of 30 years every of 5 years. Finally, an index centered on the 1st January every
five years is calculated by means of a moving average in table 9. Three periods when the death rate
diminished rapidly are found: 1860- 1874 and 1800-1819, periods for which death rates for the total
population were available, and 1 745-1 764, a period when no decline was suspected. Different
analyses were then made of the life tables for the 1550-1699 cohorts: principal components (of theLedermann type), logit (Brass) and a comparison with the Princeton tables (table 10). The appendix
describes the method of adjusting data to allow for unstated ages. Table Al provides the distribution of
ages at death from 0 to 8 years in the unweighted data; there appears an abnormal number of deaths at
ages 2 and 3. Tables A2, A3 and A4 describe the method of distributing these deaths over the
neighbouring ages. Adult mortality is then dealt with (table A5).
Resumen
La mortalidad de las f amilias de la noblesa británica a partir de 1 600 Т.Н. HOLLINGSWORTH Los
datos sobre la mortalidad de la aristocracia a partir de 1600, publicadas por el autor en 1964,
se reanalisan aqui de una manera más profunda. Primero se corrigieron los datos brutos debido a la
mala distribución de las edades al morir (según el método expuesto en el anexo) : las cifras corregidas
de las muertes por sexo, cohorte y edad, se exponen en la tabla 1. La tabla 2 proporciona los
cuocientes de mortalidad correspondientes y la tabla 3 las esperanzas de vida a cada edad. Se
compara a continuación la evolución de la mortalidad antes de 40 aňos y después de 40 aňos,
calculando las tasas médias de mortalidad respectivas, mediante la tasa de sobreviviencia a 40 ano s
en el primer caso y a partir de la esperanza de vida a los 40 aňos, para la tasa media correspondiente a
la mortalidad sobre 40 aňos. El cálculo se realizó para ambos sexos (tabla 4). En las cohortes mas
antiguas, la mortalidad sobre 40 aňos era mas alta, mientras que la mortalidad antes de los 40 aňos era
relativamente baja y aumentó enseguida, durante casi todo un siglo. La tabla 5 permite precisar el
análisis, en relación a très grupos de edad : menores de 25 aňos, 25-54, 55 y más. Se puede apreciar
que la elevación de la mortalidad afectó sólo al primer grupo de edad. Estas tasas médias por grandes
grupos de edad (25 aňos) fueron transformadas a continuanción en tasas "de momento" (tabla 6). En la
misma tabla se expone la tasa de mortalidad general résultante, para cada periodo, mediante la
aplicaciôn de las tasas por edad a una estructura-tipo . Se ve que la mortalidad expérimenta una râpida
diminución entre 1750 y 1775 (approxi- madamente). Volviendo al análisis por cohorte, se calcularon
(tabla 7) tasas de mortalidad relativas, por sexo, edad y cohorte, tomando со то base las tasas médias
de las cohortes 1550-1699. En la tabla 8, se calcula la media de las tasas relativas correspondientes al
mismo periodo (independientemente de las edades) lo que da un nuevo indicador de mortalidad "del
momento" para periodo s de 30 aňos, desplazados de 5 en 5 aňos. Por ultimo en la tabla 9, se ha
calculado, mediante promedios mo'viles, un indicador referido al primero de enero, de 5 en 5 aňos. Se
encuentran asi très periodos de disminución rápida de la mortalidad : 1860-1874 y 1800-1819, para los
cuales ya se tenian indices para la población total, y 1 745-1 764, para el cual no se ténia dudas. Se
efectuaron a continuación diversos análisis de las tablas de mortalidad relativas a las cohortes 1550-
1699 : componentes principales (del tipo Ledermann). logito (Brass), comparación con las tablas de
Princeton (tabla 10). En el apéndice se describe el método utilizado para la correcciôn de los datos
para la distribución de las edades no declaradas. La tabla Al muestra la distribución de las muertes
entre las edades de 0 a 8 aňos en los datos brutos ; aparece un numero anormal de defunciones entre
2 y 3 anos. Las tablas A2, A3, y A4 describen el pro cedimiento de distribución de estas muertes en las
edades vecinas. Después, se discute el caso de la mortalidad adulta (tabla A 5).MORTALITY IN THE
BRITISH PEERAGE FAMILIES
SINCE 1600 (I)
Т.Н. HOLIJNGSWORTH
University of Glasgow
Introduction Our knowledge of mortality levels in the past might be
regarded as fairly good, in the sense that we know that these
levels must have been high in all early periods. Before modern methods of
hygiene, nutrition and medicine had appeared, it is scarcely possible to imagine
that any population would achieve and sustain an expectation of life at birth of
more than forty years or so. It is only when we attempt to discover the pattern
of this heavy mortality that difficulties arise.
We encounter a great problem in measuring infant mortality at all, owing
to the gaps in the data, or (what is just as serious) suspected gaps that may or
may not be real. Mortality in early childhood has similar inconveniences, so
much so that many historical studies, and Anciennes Familles Genevoises(2) is
one of them, are obliged to consider mortality before adulthood only in a very
summary way. This is better than asserting that one's figures are accurate when
they cannot be accurate at all, but it leaves a large question mark over our knowl
edge of the pattern of mortality. When we recall that nearly half the population
might by dying before age twenty, the absence of reliable information about
them is especially serious.
Such considerations apply particularly to genealogically -based research,
for which adult mortality will usually be more reliable. Parish register studies,
on the other hand, have a contrary problem. Migration from the parish, and
(1) An earlier version of this paper was given during a seminar at the University of
Oslo in March 1976, and a revised version was read to the Demographic Section of the
Polish Academy of Sciences at Warsaw in October 1976.
(2) By Louis Henry, Paris, PUF, Cahiers Travaux et Documents n° 26, 1956.
Population, numéro spécial, 1977. MORTALITY IN THE BRITISH 324
temporary absences of some years, lead to a degree of uncertainty about
mortality rates that must become greater with advancing age, quite apart from
any omissions or inaccuracies in the register itself. Thus again we find that our
knowledge of mortality is worst at just the part of the lifespan when the
incidence of is highest. When we recall that the tendency of
monumental inscriptions, rather like popular newspapers, is to announce the
unusual and ignore the typical(3), leading to highly erroneous notions of the
mortality rates by age(4), we must be highly suspicious of all data that might
lead to estimates of mortality patterns in the past. Accuracy is possible, but
human tendencies are not favourable to it except when mortality was consciousl
y being measured.
At present, several attempts have been made to construct life tables that
would apply to populations undergoing heavy mortality, despite the almost
complete absence of detailed knowledge of the levels of mortality that might
then be appropriate to each age. The approaches vary, but the main difficulty
is that the most reliable information on mortality patterns comes from popula
tions with relatively light mortality. The method by which one extrapolates
from the known to the unknown is then crucial, but there is no way of verify
ing whether or not any particular method gives especially good results, because
the correct answers are unknown. The logit method developed by Brass(5) does
not depend on more than one assumed schedule of mortality to generate a set of
life tables, but, for the same reason as before, it is not obvious that the logit
method has any particular advantages for the purpose.
For historical purposes, moreover, we are interested in the progress of the
mortality pattern for the same population and not in geographical differences
between populations at the same period. It is therefore discouraging to recall
that when Coale and Demeny established a "North" pattern of mortality(6),
they were unable to keep the Scandinavian countries in it for the whole period
for which data were available. This is worse in that Scandinavia has the lon
gest series of reliable mortality data, so that we may fear that other countries
might also have strayed from one group to another if there had been more data
for them.
We are therefore attempting in this paper to determine as accurately as
possible what were the rates of mortality by age and sex for the British peerage
(3) L. Henry, La mortalité d'après les inscriptions funéraires, Population, 12, 1, 1957,
pp. 149-152.
(4) L. Henry, L'âge du décès d'après les inscriptions funéraires, Population, 14, 2
1959, pp. 327-329.
(5) W. Brass et al. The Demography of Tropical Africa, Princeton, 1968, pp. 127-135.
(6)A.J. Coale and P. Demeny, Regional Life Tables and Stable Populations,
Princeton, 1966, pp. 13-14. PEERAGE FAMILIES SINCE 1600 325
families since 1600. It is a reworking of our old data(7) on more subtle lines,
and takes the matter almost as far as the limitations of that material allow. At
the same time, many problems of English demographic history before civil
registration began in 1837 are still unresolved, and further light may be shed on
the precise nature of the mortality decline that must have occurred between,
say, 1665 and 1837.
Comparisons with other historical results are largely omitted, since we are
here attempting to study the details of mortality, which can scarcely be done
without a good body of data. We also do not discuss the causes of death, but we
can assume some high incidence of deaths by violence for the younger males of
the aristocracy at most historical periods. The data on known causes of death
are, in fact, very incomplete.
Results For convenience, the details of the methods of refining our data
are given in the appendix. The main problem is to correct the raw
data for the heaping effects at certain ages at death, which are inevitable under
the system for filling gaps in the historical records of dates of birth and death.
The data are compiled to the nearest tenth of an individual (a convenient
reminder that they are not data actually observed, but have been subject to
corrections), and are arranged in the customary quinquennial age-groups.
Because the size of the population of any one cohort is rather small, it is very
important to know the precise number of people dying in each age-group. This
is provided in table 1, which is the only place where we present data rather
than calculations from data. Further calculations than those given here should
always take table 1 as a starting point. Whole numbers with no decimal indicate
figures requiring no corrections.
It is clear from table 1 that the mortality rate for a single quinquennium
of life cannot be known very accurately, since the size of the cohorts is rather
small, even at birth. It is not therefore possible to make a detailed analysis of the
development of mortality over the centuries, and we must adopt some means of
grouping or smoothing in order to reduce the amount of sheer sampling error.
Nevertheless, as a beginning, we do calculate the mortality rates for the data of
table 1. It is easy to do so, of course, because the cohorts are real ones, and the
number at risk at the beginning of any age-period is simply the number of
survivors up to that age. The mortality rate is then the number of deaths in the
age-group divided by the number at risk. The figures are given in table 2, which
naturally shows some large variations and apparent anomalies. These are only to
be used as a basis for more detailed investigation, however, since many of them
(7) Т.Н. Hollingsworth, The Demography of the British Peerage, Supplement to
Population Studies, 18, 2, November 1964. i
i
1
1
i
i
I
i
1
i
MORTALITY IN THE BRITISH 326
Total 1027 162 222 251 451 293 83 042 053 197 329 343 1462 40 ON 68 480 1006 1032 066 000 930 758 1040 559 040 019 042 806 034 Tř О •-< Tt O ro 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 7 3 rt 5 8 2 0 5 3 9 2 9 9 40 ЧО Tř 4 4 4 ГО r— 00 rt Tt O CO 40 rt 40 Tř Г~ rt О O 00 CO rt 287 O ON Tt ГО 34 00 40 o ГО rt rt ГО VO ON •-< ~< 00 ^ rt es rs 00 rt rt Tt VO ON rt es
Ol vo O O es oo es rt 00 rt O СО О Tt 40 40 67. VO Tt 40 О CO Tt 80-8 ^^ t-~ O t— f- ГО 40 00 rt O O ON 40 CO 40 00 123 102 123 О ГО СО 40 197 75 t— i es es es es es Tt ro VO ON O 40
r^ on о es ro i-H ro oo es 4o vo ON rt 00 O VO O VO 40 ON ON CO 75-7 60) CO rt Tt ON Tt Tt rt rt Ol rt 00 ЧО O t — Г^ ЧО 107 110 135 160 vooor-o 175 o 86 *— * es •—» vo r— es Tt ro ro чо ro es es es es ro Tt Tt VO t_i
es со 4o Tt 40 O 40 40 O eo co ro eo co ГО ON 40 O ЧО O Tt 9 Tt 70-7 r~ es r-i os o 40 rt ON Tř 149 Tt Г~ O ON 40 CO VO 40 77 O Tt O ЧО ro Tt О Tt O rt со es ges rH Tt CO es li ro es 40 Tř Tt Tt Tt Tt Tt 40 <D я1 я es 40 >-i *— i es r- r~ oo co O Tř О Tt O CO rt ON On CO rs O all al 65-6 го vo r~ rs 70 89 91 79 106 rS 00 40 40 38 55 60 55 98 104 127 120 84 rs ro Tt rs 4o Tř 40 Tt 40 Tt at at
oo on oo t— 1 co es i— co 1 8 5 4 0 ТГ 0 4 9 6 4 5 60-6 40 O t~~ ON r~ r~ eo es 139 Tř vo Tř 4o O Tt VO 40 ГО Tt Tt CO ГО ЧО Tt 40 t— i^ оосочо чо oo oo On CO VO 40 40 00 VO Tt ON VO Tt ЧО 40 Tt 40 Tt IS)
— i Tt 00 ГО CO ON Tt ON Tř vo rt Tt t~- rt 40 Tt 40 CO ГО 40 ЧО ON 55-5 On es ЧО О On on ^j. vo 00 40 00 O VO ГО ON Tt 40 со VO Г- 40 Tt 40 Tt Tt Tt 40 CO Tt oo O ro oo vo Tř чо чо vo 00 Tt VO 40 Tt
Tt O rt Tt ON <— 1 VO ГО Г~ rt O 40 vo ro ro O VO rS rs ro rs oo 50-. О Г- t— CO 40 Г— •— ГО ЧО co vo ro O vo 4o es чо 00 CO 40 40 ЧО 40 40 40 Tt ro 40 OS f- Tt 40 Tt Ol Tt Tt 40 CO es i — CO 40 VO Tt Tt Г~ 40 40 Tt Tt 40 Tt Tt t-4 45-49 t-~ vo 40 vo CO C~~ O O rt 40 40 — < 00 00 ГО oo Tt on es Tt Tt R, t e-Grou ON 00 VO 40 ГО ГО О ЧО r~ Tt 38 39 43 38 co Tř on on Tt es 49 33 42 47 58 60 50 54 ages ages ЧО ГО 4o го со со es es Ol VO 40 es Tř чо
Tř ГО O O rt ro O t- Tt 3 1 8 8 5 1 9 5 4 5 4 4 îales rt Tř С- ГО rS O <— 40 40 VO ГО ГО Tř 40 a 40 od es rt oo Tt es oo 40 00 00 O ГО я les го го чо чо Tt Tt ГО Ol Tt Tt Tř ro es es at < co vo vo CO CO 40 rS ГО Tt 40 40 '-0f 15 r— (245 C- (— vo o r~ o ГО ЧО ГО Tt t— 00 rt Tt Tt CO Tt 40 Tt O 00 o> ON 35- ЧО r- os ro es ON 40 ГО ON ON ГО U. rs vo ro oo Tt c- Os O 00 O es 40 O >— ГО Tt O Tt ON rs чо Tř Tt Tt ro es ro co Tř ro oi es ГО 40 Tt 40 Tt Tt rS ГО ГО 40 40 Tt 40
Tt r^ t- Tř Г— ГО 40 VO 40 rS 40 ON СО Tt ON VO 40 ГО 40 OS CO vo 30-: Tt Г- 40 t~~ 47 40 О ГО ГО г— чо О чо os со со чо o го го rs ON Tt O Tt r- го чо eo es 40 vo 40 VO Tt ГО rs Tř ro Tt es es чо чо Tt 40 40 rS 40
OS On co i— ro ro co co ЧО \О ЧО es ro oo O Tt 00 Г~ CO rs со чо 25-; r— .— es vo oo es чо rt vo ro •— oo 00 rt O Tt 40 Tt ГО ro co 4o eN О о о Tt Tt СО ГО es ГО 40 40 40 - ^ ЧО ■<> 40 40 Tt ГО ГО ЧО VO чо чо Tt es Tt 40
Tt rt г- Г- О 40 ON VO es es Tt •H vo es N On O ro es Tt VO O g > г- о 20- 0 Tř •— 1 ЧО 40 40 O t^ 00 es r~ es oo r^ ON 00 O 00 Tt чо es oo rs ro Tt Tt ГО rS es es vo Tt 4o 40 чо es ro ГО 00 rt es со чо 40 ro es ro ЧО 40 Tt «i sa»j ON я я On r- es -H o 40 Tt ON 40 O es es on es чо eo ro ro Tt 4O al 15- — i r- es o Tt rt Tt rt ГО О Г- ЧО t"H O rt Tt rS ro O t~- O 27 rt es es 40 Tt О я es co ro rs es го со rs rt CO Tt Tř 40 ГО CO Tt ГО Tř rs oi ro rs it
t rt О ГО Tt vo r- es ^, O ^ O 40 ГО co es vo oo Tt Tt rs O 10- r— rs «— on rt rt r~ es Г- ГО 00 O es On O 00 чо vo es O vo VO ГО О ГО On Tř 00 ЧО w es co rt es rt rt rt es es es Tt rt ■^ es es es rs rt
P~ OS t~- 00 40 00 On ro со ЧО Os 40 чо ЧО co vo o vo es vo VO Tt O Tt Tt o rt vo rs On es os es Tt Tt O Os O ro О ~* о rt rt ЧО Tt Os 40 es co 00 40 00 Tt ЧО Tt СО о 40 ro es Tt es eo co es ro es rs ro co Tt ro co co es es
rt ro Tt rs O t— 00 Tt O 40 ON CO 40 co О 00 го 0О Os vo O co Tt ON 7. »—l 53. 105. 00 00 131. 125. 111. 117. 81. 1-4 41 t— C— 40 >— 1 Tt 71 ON ON 83 73 66 58 ro 130 102 95 83 83 69 49 Tt о Tt O O 40
1 7 3 8 2 0 7 es О 13. VO i— 88. t CO ON 9 3 6 5 0 0 6 4 4 ■""* t— 40 O rt oo r~- rt o oo o VO VO 161 207 158 123 104 VO 00 00 106 О ro es o о 169 170 о Os rt ЧО Tt r- on c- Tt es t~~ oo On 40 - Tt
ON Tt ON Tt ON Tt ON Tt On Tt On Tt OS Tř Os On Tt Os On Tt On 774 Os Tt Os 874 Tt Tt Tt r- ON On es Tt Tt ON r- rS Tt ON rS Tt r» on es Tt r— on es Tt t- ON CN Tt Os rS Tř r- f- Г- on es 00 00 Tř vo vo Г— Г- Г— 00 00 00 ON ON 00 00 ON ON 40 40 ЧО VO 40 40 ЧО ЧО 1750-1 1775-1 1800-] 1825-1 Bor с 1550-1 1575-1 1600-1 1625-1 1650-1 1675-1 1700-1 1725-] 1850-1 1875-1 1900- 1925-1 1550- 1575-1 1600-1 1625- 1650- 1675-1 Cohc 1700- 1725- 1750- 1775- 1800- 1825- 1850- 1875- 1900- 1925- <
i
I
i
I
i
i
I
i
с
i
I
PEERAGE FAMILIES SINCE 1600 327
00 ON 00 ^- « П О> 1~ ОO ЧО СЧ О00ON ON Г— .-н ГО 40 Tt Г- О О г~ со О О ел Tt ■ч- Tt rf to ЧО 1- р- ЧО г- ЧО "О
О — < 00 СЧ OriTtO ОООНО чо to
О tO Tt ГО сч сч
го to On оо О со m оо to Tt to ЧО Tt tO ЧО Г- м m ví О ■— ГО СЧ * р- оо ON OO Tt to Tj- ГО ГО ГО to го го го ГО ГО ГО ГО ГО ГЧ ГЧ ^н to ГО Tt Tt СЧ СО СЧ СЧ
О ON •ч 00 P- Tt to Г- Tt ЧО го ON On чо го О О О ГО ГО Tt СЧ чо г- с- >— < ON — < Г- ГО ЧО ■Ч" 00 ЧО ON ГО ГЧ О to ГО ЧО ON О rt О ГО Г- ГО ГЧ-н ГО О -ч to On On чо ON СЧ О ^О ^н ГО СЧ ГО сч го го сч ГЧ СЧ СЧ ГО Tt 't ГО сч сч сч сч
чо сч —4 ~- г-- СЧ чо О г~ ЧО Tt ГО Г- ГО ГЧ on оо оо сч сч сч го сч to го оо >— Г-^ Tt O tO СЧ ^н Tt •— чо чо го го to ГО 00 СЧ О ЧО ГО ГО ГО СЧ го го го го
00 I— О On —4 ЧО О ЧО О ON ГО Р- О Ю О 00 ЧО 00 ГО ЧО ON ON ЧО Г~ оо сч г~ 1 го го сч ON ЧО ГО 1О ЧО О ■* СЧ СЧ Г~- >— оо чо О ГО 00 ГО On О О 00 О СЧ On r- СЧ СЧ г-1 ^н СЧ СЧ гЧ ,-н Tt ГО СЧ ГО
Tt г~ ЧО .— сч го го О ГО О ON ^ч го On чо ГО ON О to to oo O t-4 р- ON Г-. о ^н ^н ON О О ON »-н го чо го ON I/I ON 00 р- г- оо г- г- тг ГО 0О г- О to чо сч го СЧ ■* г-4 ЧО ГО ГО ЧО ЧО Tt го — ~ч сч сч сч сч сч
ГО ■— I СЧ ON Tř тГ ON Tt 00 tO 00 со сч чо сч чо <ř оо чо го оо i— < ГО ГО ON О t-~ чо сч г^- сч чо ЧО СЧ On чо 00 ^н СЧ 00 О ЧО ГО Р- i— О СЧ ГЧ ГЧ 00 On чо >О чо СЧ 00 ON О ЧО «О VI СЧ ЧО Tt
О «о тг О чо ^S- t- го го ■* чо Tt 00 On O On OO VO Г- ЧО V) ГО
чо t— О г- чо сч r~ [ — ЧО ЧО ГО ЧО ■— ON in ЧО ON Tt О ON CO 00 ON ГО 00 ON V) чо О v-) сч о г— го о -d- о .— О го -H ЧО 00 О ГО .4 ON ГО СЧ СЧ СЧ СЧ СЧ ТГ 00 ЧО ЧО Tt Р- tO tO On —t ON On On 00 to to to го сч сч сч О О О On СЧ ОО 00 ЧО чо ■* "о чо
Г~ О ЧО Г- СЧ t~- ON г~ 00 ON ЧО HNO ча оо чо to сч Tj- r~ t го to On Г- г~ — О 00 00 tO -н ГО со го сч чо ON « on v~> Tt On Tt to оо оо i— ON О 00 00 00 ON On
t-- 00 ч ЧО ON Tt 00 ГО O ЧО ^н О •— i On СЧ to ю оо о Tt О Tt C— ON 00 t- Г- СЧ tO ON ON ЧО >O ГО ГО
О t— ГО <— I fN Tt -н О 00 00 СЧ СЧ О Tt О ИГЧЛ r ON Tt r~- tO — О ЧО to ГО Г- ON О Р- ЧО О О го О P- Tt TÍ- ЧО го I — On г^ ГО 00 Tt 00 ЧО to 00 On On 00 го
ЧО СЧ Г-- ЧО сч to сч to ЧО 00 ^Н ГО сч сч сч О СЧ Tt O ЧО Tt ГО чо г- >о to —4 Г) Tt СЧ ГО Tt V> сч го го сч
Tt ГЧ Tř <-ч го СЧ О О оо го ^^ >о О СЧ ЧО О 00 О О Г- О 00 ЧО Tt CO t-~ О to ON 00 О to сч го О СЧ О Tt — i Tf ГО ГО ГО ГЧ ГО -н СЧ Ч гНГ>| j СЧ ГО ГО
го сч г- чо Tt СЧ г- 00 чо го О О ON ON ГО i—t to сч On to p- СЧ ON чо •— i чо сч ON -н Г-- 00 Tt 00 Г~ ON Tt 00 ЧО ЧО Tt ГО ГЧ СЧ СЧ -* Tt CO to тг Tt чо СЧ ГО СЧ -ч СЧ СЧ Tt tO tO Tt tO tO
ON i-H го On ЧО Г>> CTv Tt СЧ 00 ГО ГО 00 -H ГО СЧ ON to ^ О о to сч сч Tt ГО ГО СЧ Tt СЧ СЧ ГО On ЧО чо
Tt Tt Г^ -ч -н СЧ 00 ON ЧО — "-H Tt ON ON ГО -H СЧ 00 to о сч чо оо Tt ЧО <О СЧ сч о to о •-h О 00 to p- ЧО ^н ГО СО 00 СЧ 00 Tt to ON ЧО О 00 00 О 00 ЧО ГО Tt Г~ ГО Tt tO 00 Г- «О ЧО О оо оо чо О го го чо ON 00 ЧО ЧО чо to to чо ON ЧО tO
Tt On тг ON Tt ON Tt On Tt ON Tt ON Tt ON Tt O\ Tt ON Tt ON Tt Г~ ON СЧ Tt Tt ON ЧО ЧО Г— t— p- p- oo oo 00 00 ON ON to to чо чо to to чо чо
O ví O O о to о to О to О to ó >A ó to Ó v\ o v~> о to о to tO Р- О СЧ ifl г- N to г~ О сч to r~ о сч •О P- О ГЧ p- P- О00 00 СЧ чо чо р~ Р- r~ p~ oo oo 00 00 ON On ÍO «О ЧО ЧО to to чо чо i
i
i
i
MORTALITY IN THE BRITISH 328
го го го го О Г- 00 1-н eo oo oo о О О О О о о го о с* мл r— «о so p- rr es -ч rr 00 P- P- 00 m г- t— 00 OO 00 00 OONO eo Os eo Os rr -h О <П m v) in in гГ Os СО Os О ГО Г4! О V) 00 ГГ V5 es' <N rô rô rô ro ró го го О чо rř rř i/o ró rř ■* iň rř rř rr rř rř rř -н rř ГО го eš еч го rr
SO Os 00 i— о t- es so i-i es os ro oo oo oo oo p~ p~ so Os Os Os Os m p- 00 О О es oo es О OS in -н О so О —I SO, ГО OS oo oo ro in SO vo VO VO oç О es rr О 00 >O SO vo ro vo гГ rr rr rr V5
t— oo oo oo О ~* ro uo es os г— -н vo m m m Os © -н 00 oo in es -h — so -и so г~ es es es os os t— oo in vo so so p- rr i— m O Tř so V) О rr ro io гГ SO SO eS ГО Г— f- Г- uo r- so os Os ro О О es ro ro го 00 eS p~ 00 нМОМ vOKIvOvO vo* P- 00 00 OS es: r-í so К Os Os Os Os vo V) in oo so so oo
vo oo vo m 00 ГО ГО ГО Os O ro i— 00 00 00 00 ro p- p- h rr oo O< rr О P- P~ P~ irj CM Tfr in o' es* 00- Os" r— t— p- p- Г- p- p~ OO 00 00 Os Os О Os Os Os Os ОС? oš o es es es es
P- SO Г- -h so rr -h ro ro О ~* -* Os О О О О ■* -н os O г- O es —i es «M in Os in -h о SO OS ^н -Н О ro ro ro so rr OS Os p- 00 Os <N »ч SO V) OS 00 О Os r~ rs es es »o es so ro es p~ ro rr О О О — i О Os Os Os Os OS Os O fN ГО ГО ГО Р-ОМЯО in so vo so
г- es es p- ro p- p- — -h in Г- V> so О О О eo о Os in vo es es es ГО OS OS SO p- rr es os P- Os OtS Os ГГ SO 1П t— r- so о os О го го го rr Os О ro oo' <rî — ! es rr 00 Os Os os ro ro io "^t in SO SO кО rř И vo
p~ ГО OS Г~ ro oo es V) in rO Os oo ro ro so esp- vo vo vo Os гГ гГ vi P- V) ГГ 00 rr г- -и го es oo rr ro 00 ГО О t— p~ о О г- VO Os OS t— vo ro os so -^ oo es SD О О О p- oo os es es es -н es is Н 00 OS Os Os Os O i-i >-i rr ro m so es rr rř rř es es es es
—i 00 00 00 ro r— ro о ro го го о SO ГГ Os rS Os тТ го Р-- es' — 5 es rô C5 es ro so rî rr ГГ rř es es es es
rr -h in in 00 SO i— i V) Г- Г— tS О ro rr ro ro rr О n oo О О vo Os 1-н Р- ГО SO 00 ГО Г- О SO Os Tř 00 rs oo ro ro гГ P- ro VO 00 O\ OS ГО ГГ V) V) ~ О г- -< ro in oo oo es го О 00 rr oo es in m р-; го -э rř uó so so
О -н m Os so О 1-н so vo Ю so Г- 00 VO V> P- ГГ SO Tt ГО ro so CN Г-- rr rr о о t— i ГО -н in P- Os SO vo ГО OS Tf — < in oo oo о co oo О rr Г-Н00 1Л eo -h ro es i-H es m in ro r~- о oo SO ro 00 SO О es in es eS Щ Os ГО oo' Р-; ГО О О г-' os oo oš oó oš oš
-н rr O Os OS _< Os -h vo rr p- p- H Tt О Os 00 ГО -Ч Tfr OS es ro os os es oo os ro es 00 rr P- 00 ГО SO VO ГО ГО 1П Os OOt^HH ro oo m os О vo го О <N ГО Г- CM p- oo ro so es rr vo vo w> Ю Os O О so ТГ ■* so rr «л os vo V) vo О ы 00* Ó ri — i eS O eš ro oó OŠ О О О rî TÍ tÍ го го го го ro rr rr rr N N M S MM(N(N СУ ш OS 00 fS -H es — i rr es -h in m Os m oo O os rr ro oo oo •— oo es r- es o es —i -h ro ro oo so oo os -н os es oo oo SO ^1- ГО SO p- in >n es Os Os 00 Os -н so О -н tri 00 -H 1-н
ví rř oó oo í ř ro ro ro ro ro ro ro m ro ro ro rr
rs p- стч о p- vo es p- os oo o Tř es in so oo rr О 00 rr SO Os ГГ 00 00 rs| Tf OS es 1-н oo os m oo ro in os es oo so — m rr © OS in SD 00 i— • со о p- 00 P- P- V) VO Os vo -h Г- V) V) Os t— Os <N IO -н 00 O so os rr es in
só об Os rř <N fN <S О
О i-i О го 00 P- SO О vo rr vj es «1ЛООО eo Os in Os in in ro oo NOt P- rr rr P~ -н -н SO SO m p~ p^ oo SO ГО -н О 00 SO Tt О SO 1О -H Os О rr so es 00 00 О С— rr 00 г- О 00 V) ГГ ^н гГ Os Р~ 00 Tt ГО -Н р- О rr ro in os о es -" oš ^ rř oó ó 5 © 00 OŠ го ro es es ro rr rr rr
CS Г- eS 00 Os OS ro rr гГ О es О oo oo v) h rr О t- es es oo p~ -h Os oo es тг es in VI os 00 ro es vo -h 00 oo vo о es rr О ГО Os ГГ es -h so rr es rf го с; SO ко Tt so r~-' г- es os го SO rř eN CÍ in m ro ro SO VO 00 О rô v5 es ГО CS ГО ГО
г- го о es eS ГГ ГО t— Tt P~ О О P~ rr Os so го so rr О so О es —i VO eS v> 00 ГО <N t~- <— i ro es o os es o o> -h rr ro О Os rr ■* О t- ro es >n oo es rr v) so in vo oo p» vo Os SO ГГ io го го rř es' o" Ô © eš rô о rô rr rr rr rr vo vo vo vo
m -h — < rs oo es es Os
Г~ П- Г- О rr oo es 00 SO ГО 00 vo rT Os о Г-; rr- -h' ГО* 1/î r-'< ОЙ -h ro Ô in es" m eô v) vo О in >o m so
00 -h Os SO os о -h es so es о о rr rr oo oo ■^- ro p- -h i—i os m v» OS О Os OS Os Os in ГО es p- ro Г— OS 00 Г- О uo es so os SO Os О 00 О 00 in rr OS V) V) 00 Os ГО ГГ V) OS i-l OŠ VO VO P~ VO
p- p- os es m « по so ro oo es SO OS 00 ГГ in oo ro p- rr 00 гГ гГ ro 00 >O 00 ro rs oo --и i—i oo es oo 00 ГО 00 Os 00 t- ГО -h V) OS Os О ГО ГО V) i— I ГО 00 00 UO О -ч 00 Г- Г— SO i-t SO -h ГО 00 t-i so es in rr 00 i-i 00 ГГ V) P^ 00 P~ so os es es " o' -- iň oš — rř <A oó eš vb os о vo vo vo r- rr ■rr in m
rT OS гГ Os ■* OS ■* OS rr Os rT OS rr Os rr Os rr Os rr Os Os rř OS rr Os
vo SO P- P~ so so г~ г- 00 00 CJS Os in in so vo 00 00 Os Os V) V) SO SO
© v> © m o «о о ю Ó IO Ó Ю о </•> о in O V) O V) m г~ О es in г- es v) г- о es VI Р-~ О ÍS SO SO Г- t— 00 00 £У\ Os V) V) SO SO m in so чэ 00 00 Os OS PEERAGE FAMILIES SINCE 1600 329
must stem from sampling errors, to say nothing of the procedures of correction
of the data as described in the appendix. The last cohorts contain some data
based on only part of their mortality experience, because 1959 was the last
year included in the study.
We also supply a table of expectations of life as table 3. The method of
compilation was somewhat complex. The expectation of life at age 85 was
taken from the original (uncorrected) tabulations by single years of age, and
for all age-groups between 5 and 85 the years lived by the cohort within an
age-group was taken simply as 2.5 times the sum of the numbers at the begin
ning and at the end of the age-group. Under age 5, following Coale and
Demeny(8) but with some simplification, the formulae used were (in the
ordinary notation):
= 0.34 /0 + 0.66 and + 2.64 /5
These separation factors have the effect of reducing the expectation of
life at birth by about 0.1 year, an amount which, although small, must give
a better approximation than the simpler formulae:
= 0.5 /0 + 0.5 I, and = 2 ly + 2 /5
We can only complete the expectation of life for the last four cohorts
by making some assumption about mortality at the higher ages. In table 3
we assume that the most recent mortality will continue. In this way, we can
project the whole life table even for the most recent cohort.
Appropriate comparisons with table 2 are found as tables 40 and 41
in our earlier work(9) and with table 3 as tables 42 and 43 of the same(10).
In table 2, it is obvious that the grosser distortions due to heaping of ages
at death have disappeared. Table 3 results in slightly reducing the expectation
of life at birth when it is high and slightly increasing it when it is low. At some
higher ages, however, the changes are quite large. For example, e5Q for the
female cohort born 1625-1649 had previously been given as 10.8 years, but
it is now 13.208. One might study the amendments in detail, but all we shall
say here is that the revised series have more verisimilitude.
Table 3 provides us with a simple way of summarizing the mortality
experience of each cohort if we concentrate upon the expectation of life at
birth. The first two cohorts and the last two cohorts are smaller than normal,
and therefore particularly liable to sampling error, and in fact the first and last
(8) op. cit., p. 20
(9) op. cit., pp. 54-55.
(10) op. cit., pp. 56-57.