Professional Documents
Culture Documents
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer
Nature Switzerland AG 2022
This work is subject to copyright. All rights are solely and exclusively licensed by the
Publisher, whether the whole or part of the material is concerned, specifically the rights
of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on
microfilms or in any other physical way, and transmission or information storage and
retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc.
in this publication does not imply, even in the absence of a specific statement, that such
names are exempt from the relevant protective laws and regulations and therefore free for
general use.
The publisher, the authors and the editors are safe to assume that the advice and informa-
tion in this book are believed to be true and accurate at the date of publication. Neither
the publisher nor the authors or the editors give a warranty, expressed or implied, with
respect to the material contained herein or for any errors or omissions that may have been
made. The publisher remains neutral with regard to jurisdictional claims in published maps
and institutional affiliations.
This Palgrave Macmillan imprint is published by the registered company Springer Nature
Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
This fifth edition of Elections and Voters in Britain has been a long time
forthcoming. The normal rhythm was to update the book every four
or five years, after each general election. This time, however, no sooner
had we updated the text to reflect one election than another was fast
approaching. The sense of trying to hit a moving target was reinforced
by the fact that more recent elections were unusual in their outcomes
as well as their frequency. Between them, the 2017 and 2019 elections
challenged a variety of received wisdoms about, for example, the role of
the economy, the limited impact of national campaigns, the advantages of
ideological moderation and the rarity of single-issue voting. Since those
received wisdoms had been carefully laid out in previous editions, the
events of the past few years called for a good deal of updating. Mean-
while, the powerful impact of the votes on Scottish independence and
EU membership has meant that referendums have been promoted from
footnote status to a chapter of their own.
Rehearsing all this explains the delayed arrival of this edition but also,
we hope, why it was worth the wait. In addition to a new chapter on
referendums, the discussions of electoral geography and electoral systems
have now been separated and expanded such that there is a second new
chapter. There is also much new material in each of the other chapters,
with results and survey data from the 2019 election fully incorporated
into the discussion as well as relevant tables and figures. There are many
v
vi PREFACE
analyses also covering the 2017 election as well as contrasts with the rather
more orthodox contest in 2015.
We used the term ‘updating’ rather than ‘rewriting’ because it is not
clear how far the most recent elections mark a temporary interruption of
normal service rather than a new electoral order. For example, part of the
very powerful Brexit effect in 2019 is likely to die away as the issue slowly
recedes, with other issues such as the economy liable to regain impor-
tance. On the other hand, part of what looked like a Brexit effect was
more a reflection of longer-term trends in the sociology and geography
of British public opinion, especially relating to the issue of immigration,
and this looks to be potentially a more enduring change. In updating the
chapters, then, we do not abandon received wisdom but acknowledge the
severity and estimate the longevity of the challenges posed to it by recent
elections.
As noted in the Preface to the first edition, this book is not primarily
for colleagues actively researching and writing about British elections.
Although there is some fresh data analysis, most specialists in the field
will already know most of what is written in the following pages. Our
hope is that they find it a useful overview which can be recommended
to their students. The broader intention is to summarise and simplify the
work of these colleagues for non-specialists and students in both schools
and universities (as well as people who just happen to be interested in
elections). One consequence of this focus is that we have tried to keep
tables as clear and uncluttered as possible. In some cases this has meant,
for instance, not showing in tables the raw numbers on which calcula-
tions are based. This would be unforgiveable in a research article, of
course, but seems, on balance, to be a sensible strategy in this case. As
before, we have also included a glossary of statistical terms, and items
appearing in the glossary are printed in bold when they first appear in
the text. The explanations and descriptions of various statistical measures
and techniques offered are intended to be non-technical. They attempt to
enable readers to understand why various techniques are used and what
the results can tell us.
Everyone writing about electoral behaviour in Britain has to acknowl-
edge a debt to the successive teams which, from 1963 onwards, have
undertaken the burden of directing the British Election Study (BES)
surveys. Their task is not getting any easier, especially in respect of the
post-election face-to-face surveys that have been the primary bases for
comparisons over time in previous editions of the book. Arranging and
PREFACE vii
ix
x CONTENTS
xiii
xiv LIST OF FIGURES
Table 1.1 Results of the 1959 and 2019 general elections in Great
Britain 2
Table 1.2 Party shares of votes in general elections, 1959–2019
(Great Britain) 8
Table 1.3 The ‘flow of the vote’ between 2017 and 2019 19
Table 2.1 Turnout in UK elections, 2015–2019 32
Table 2.2 Turnout in UK general elections, 1959–2019 34
Table 2.3 Correlates of constituency turnout (Great Britain) 1987
and 2019 37
Table 2.4 Turnout of social groups in 2019 general election (%)
and change from 2017 45
Table 2.5 Binary logistic regression analysis of voting
versus non-voting in 2019 49
Table 2.6 Comparing hypothetical choice among non-voters
with election results, 1970, 2017 and 2019 (%) 53
Table 3.1 Vote in 1964 by objective and subjective class (row
percentages) 66
Table 3.2 Party identification in 1964 by parents’ party preference
(per cent) 69
Table 3.3 Vote in 2017 and 2019 by objective and subjective class
(row percentages) 72
Table 3.4 Average net satisfaction ratings for governments,
1951–2019 81
Table 3.5 Trends in inter-election volatility (per cent) 88
xv
xvi LIST OF TABLES
Since the previous edition of this book, there have been three general
elections and a major nationwide referendum in the UK. The latter, on
the country’s membership of the European Union (EU), produced an
outcome which almost paralysed ‘normal’ politics for three years and led
directly to two further elections (2017 and 2019) following that of 2015.
Not since the 1920s had three general elections been held within less than
five years. It is clear, then, that voters—and the decisions they took—
played a central role in the dramatic political developments that occurred
in this period. Moreover, the results of at least two of those elections
and the referendum surprised most commentators and pundits, as well as
politicians, to varying degrees and in different ways.
A recurring theme of the book is that elections and voters in Britain
have become much more unpredictable and volatile than they used to be.
To illustrate this at the outset, it is useful to contrast the December 2019
general election with one held sixty years before, in October 1959. The
latter, no doubt, will seem like ancient history to many but it provides
an apt point of comparison. For one thing, it was held during an era of
relative calm. There was a broad consensus over policy among the parties
and little switching of votes between elections. In addition, it exemplified
the pattern of two-party dominance that had characterised British politics
from the end of the Second World War and had come to be accepted as
the norm which was unlikely to be threatened in the foreseeable future.
Table 1.1 Results of the 1959 and 2019 general elections in Great Britain
1959 2019
% of votes Candidates Seats % of votes Candidates Seats
Note Results exclude Northern Ireland. By convention, the Speaker is not opposed by the main
opposition parties and is counted as an ‘other’ (although in 2019 both a Green and an independent
pro-Brexit candidate stood in the Speaker’s constituency)
Source Leeke (2003); Uberoi et al. (2019)
related to one other similarity between 1959 and 2019 which conceals
a lot of change in the meantime. The old Liberal Party emerged from
the 1959 election with an unimpressive vote share and only six Members
of Parliament (MPs). By 2019, the name had changed to the Liberal
Democrats but the electoral showing was not a great deal better: 12%
of the votes and eleven seats. In the 2010 election, however, the party
had polled over 20% of the votes, won 57 seats, and become part of the
government as junior coalition partner to the Conservatives.
Two other changes relating to party competition are clear in Table 1.1.
First, in Scotland, and to a lesser degree in Wales, the part played in elec-
tions by the nationalist parties was much more significant in 2019 than it
had been in 1959. In the 1950s the Scottish National Party (SNP) and
Plaid Cymru were generally regarded as rather peculiar parties, supported
by somewhat eccentric individuals and on the fringes of politics. Few
candidates were put forward, especially in Scotland, and those that were
won few votes. From the 1970s, however, both began to contest all seats
in their respective countries (although in 2019 Plaid stood aside in four).
Not only that, but by 2019 the SNP was the dominant party in Scotland
by some distance, winning 45% of the vote and 48 of the 59 seats. Plaid
has been much less successful but was able to take almost 10% of the
Welsh vote and hold on to its four seats.
Second, there has been a massive increase in the number and variety of
‘others’ participating as candidates in British elections. In 1959 the only
other player worth even footnoting was the Communist Party of Great
Britain (which won more than 30,000 votes). It has now disappeared
but has been more than replaced by an abundance of new parties. Some-
times, these can be highly significant participants. The United Kingdom
Independence Party (UKIP) won 13% of the vote in the 2015 general
election and, while the similarly anti-EU membership Brexit Party was
much less successful overall in 2019, it was a relevant player in many
of the constituencies in which it stood. The Greens typically now stand
in most constituencies and have held one seat, Brighton Pavilion, since
2010. In addition, in 2019 there was a flurry of candidates representing
much smaller parties. These included splinters from the major parties
(such as The Independent Group for Change), old parties that refuse
to die off (like the Social Democratic Party), new parties trying to gain
a foothold (the Women’s Equality Party, for example), local campaigners
(like the Heavy Woollen District Independents) or colourful perennials
(such as the Monster Raving Loony Party) whose primary role seems to
4 D. DENVER AND R. JOHNS
1872 the Ballot Act made voting secret (previously voters had to declare
their choice in public), so that candidates could no longer check that
the voters they had treated actually voted for them. Finally, in 1883 the
Corrupt Practices Act outlawed treating (much to the disappointment of
many voters, one suspects). Despite this, elections continued to be a form
of public entertainment until well into the twentieth century. In 1922 a
crowd of 10,000 assembled to hear the declaration of the general elec-
tion result in Dundee and was rewarded with the news that the sitting
member, Winston Churchill, had been defeated by a Prohibitionist (in a
city notorious for drunkenness).
Popular enthusiasm for elections does not reach those standards today.
Nonetheless, general elections are still major national events which precip-
itate greatly increased political activity, discussion, interest and media
coverage. On election night, the attention of much of the country is
at least partly engaged by the election. Next day, the front pages of
all newspapers worthy of the name are entirely devoted to election
news. Television channels provide round-the-clock campaign coverage
and social media platforms are inundated with election-related material.
All of this is well over the top for some people. One ‘vox pop’, Brenda
from Bristol, became a spokesperson for many with her plaintive cry of
‘not another one!’ when told about the calling of a snap election in 2017.
For others, however, general elections remain a major spectator sport:
they devour the news coverage, discuss the election online and offline,
maybe even bet on the outcome, and then stay up all night to watch the
results and drink in celebration or sorrow.
Part of the fascination is the sheer mass of information available. In
general elections there are results for 650 (UK) constituencies to be
looked at, but in addition there are elections to the Scottish Parlia-
ment, and the Welsh, Northern Ireland and London Assemblies as well
as local council and mayoral elections involving many thousands of wards
and electoral divisions. The resulting numbers can be analysed endlessly:
aggregated, averaged, correlated, graphed and used to construct maps.
Some academic studies of electoral behaviour reach such levels of statis-
tical sophistication that they are well-nigh incomprehensible to all but a
few specialists. All of these data, together with the increasing accessibility
of statistical software, partly explains the huge growth in the academic
literature focusing on elections over the past 60 years or so.
There is more to it than simply providing data for political geekery,
however. Put simply, elections are studied because they are important.
6 D. DENVER AND R. JOHNS
Studying Elections
Unsurprisingly, given their importance and fascination, there is a vast liter-
ature dealing with elections and voting in Britain. This covers almost every
conceivable aspect of the subject, including whether the weather has an
influence on turnout and whether candidates whose surnames come first
alphabetically (and hence appear at the top of the ballot paper) fare better
than others. Here we focus on the two main sources of data for studying
voting: election results and sample surveys of voters.
Table 1.2 Party shares of votes in general elections, 1959–2019 (Great Britain)
1959 1964 1966 1970 Feb Oct 1979 1983 1987 1992 1997 2001 2005 2010 2015 2017 2019
1974 1974
% % % % % % % % % % % % % % % % %
Conservative 48.8 42.9 41.4 46.2 38.8 36.7 44.9 43.5 43.3 42.8 31.5 32.7 33.2 36.9 37.7 43.4 44.7
D. DENVER AND R. JOHNS
Labour 44.6 44.8 48.8 43.9 38.0 40.2 37.8 28.3 31.5 35.2 44.3 42.0 36.2 29.7 31.2 41.0 32.9
Liberal (Lib 6.0 11.4 8.6 7.6 19.8 18.8 14.1 26 23.1 18.3 17.2 18.8 22.7 23.6 8.1 7.6 11.8
Dems)
Others 0.6 0.9 1.1 2.3 3.4 4.3 3.2 2.2 2.1 3.7 7.0 6.5 8.0 9.9 24.6 8.0 10.6
Pedersen 3.3 6.7 4.3 6.0 13.3 2.7 8.4 11.8 3.2 5.3 12.4 2.3 5.9 6.5 15.2 15.5 8.1
index
Lab-Con 1.2 −3.1 −2.8 4.9 −0.8 −2.2 5.3 4.1 −1.7 −2.1 −10.2 1.8 3.2 5.1 −0.4 −2.1 4.7
swing
Standard 2.3 2.4 1.7 2.1 2.9* 1.5 3.1 3.0* 3.2 2.8 3.4* 2.6 2.4 3.7* 4.4 4.4 3.5
deviation of
swing
Notes The figures for the Liberals in 1983 and 1987 are for the ‘Alliance’ between the Liberals and the Social Democratic Party (SDP). Standard
deviation of swing figures for elections in which there were comprehensive boundary revisions are asterisked. In these cases, constituency swings are
based on estimates of voting in the preceding election
Source Data from Rallings and Thrasher (2012) and Uberoi et al. (2019). Swing and Pedersen index statistics are calculated from the original data
1 STUDYING BRITISH ELECTIONS 9
55
50
45
40
35 Conservative
% of vote
30 Labour
25 Liberals/LD
Others
20
15
10
0
1959 1964 1966 1970 1974 1974 1979 1983 1987 1992 1997 2001 2005 2010 2015 2017 2019
Feb Oct
Fig. 1.1 Party shares of votes in general elections, 1959–2019 (Great Britain)
suggested by looking at only the first and last elections in the series.
The period might be divided into four main phases. The first, which ran
until the two elections in 1974, featured a series of close races between
the Conservatives and Labour with just small fluctuations in their vote
shares. Understandably, these fluctuations captured most attention at the
time. In hindsight, however, the bigger story was the gains made by other
parties at the expense of the main two. The 1959 election would mark
the last time that they would win 90% of the British vote between them.
By October 1974, not only were the Liberals close to 20% but the nation-
alist parties had surged in Scotland and Wales, the SNP winning 30.4% of
Scottish votes.
The second and third phases saw dominance by the Conservatives and
then Labour. While the Liberals (in various incarnations) maintained a
healthy vote share throughout, they and the other parties were side-lined
by the big victories won first by Margaret Thatcher’s Conservatives and
then Tony Blair’s Labour. The pattern was one of big majorities, at first
eroded only slowly but then suddenly overturned as voters finally lost
patience with the governing party. The 1997 and 2010 elections thus
marked painful ends to long periods in office for the Conservatives and
Labour, respectively.
10 D. DENVER AND R. JOHNS
a reasonable estimate of the national swing could be made, and the final
result of the election could be predicted with some confidence, even to
the extent of identifying which seats would change hands.
Again, things have changed. This is confirmed by the final statistic in
Table 1.2: the standard deviation of constituency swings. This gives an
indication of the extent to which swing varied across constituencies in
each election: whether it was broadly similar (lower score) or varied a
good deal (higher score). Amid some short-term fluctuations, the long-
term pattern is one of greater variability in more recent elections than
in the early elections reported in Table 1.2. Overall shifts in support
between the major parties have become less uniform across the country.
In 2019, for example, despite the big overall swing to the Conserva-
tives, 31 constituencies recorded a swing to Labour. Every single one
of these is estimated as having had a ‘Remain’ majority in the 2016 refer-
endum, confirming the cross-currents of Brexit as a major reason for
recent variations in swing.
Another reason why swings vary more, and why national swing is a
less useful notion in the context of more recent elections, is the growing
vote for the Liberal Democrats and other parties. Swing was developed
in a situation in which elections were essentially contests between two
parties only. It cannot tell us about relative change between three (or
more) national parties, plus strong regional contenders. There have been
attempts to devise three-way swing figures (Miller, 1981) but these are
complicated to calculate and lack the elegance and simplicity of tradi-
tional swing. The commonest way of measuring aggregate or net electoral
change nowadays is simply to use the changes in each party’s percentage
share of the vote from one election to another.
sampling theory, we simply assert that if the sampling methods are appro-
priate this can be done with a fair amount of accuracy and that surveys
are, therefore, among the most important tools of empirical social science.
At the same time, the experience of polling errors at recent elections—
covered in Chapter 6—teaches us that we must be careful consumers of
survey results.
Academic survey studies of voting behaviour were pioneered in the
1940s in the United States. They arrived in Britain in the 1950s but were
initially local studies, surveying individual constituencies like Greenwich
(Benney et al., 1956) or Bristol North East (Milne & Mackenzie, 1958).
The first national survey study of the British electorate was undertaken in
1963 under the direction of David Butler and Donald Stokes. Although
1963 was not an election year (the survey was one of two focusing on the
1964 election), this proved to be the first of a series of national surveys
carried out at every general election since 1964 under the auspices of the
British Election Study (BES). Butler and Stokes themselves covered the
1964, 1966 and 1970 elections. A team from the University of Essex took
over for the elections of 1974 (two) and 1979. The study returned to
Oxford where, from 1983 until 1997, Anthony Heath, Roger Jowell and
John Curtice were the principal investigators. A new Essex-based team ran
the study from 2001 to 2010 and then the duopoly was finally broken:
the current BES team, with five principal investigators, is based at the
University of Manchester (see www.britishelectionstudy.com). Over the
years, the BES surveys have formed the bases of a number of major works
on voting in Britain and hundreds, possibly thousands, of book chapters,
scholarly articles and papers.
One core component of the BES has remained constant over the
decades: a post-election survey based on face-to-face interviews with a
representative sample of British voters, designed not only to record but
also to explain whether, how and why they voted. Nonetheless, the scale
and complexity of the studies have increased considerably. From 2001 the
BES has commissioned the online survey agency YouGov to collect daily
survey data during the ‘short’ campaign. The 2015–2017–2019 edition of
the study involved almost 100,000 respondents, interviewed face-to-face
or online, and over 20 ‘waves’ of data collection beginning a year before
the 2015 election and still running at the time of writing. Beyond the
BES, there are individual election studies for the Scottish Parliament and
the Wales and Northern Ireland Assembly elections, and in 2010 there
was an ethnic minority add-on to the BES, providing much larger samples
14 D. DENVER AND R. JOHNS
technical term, the ecological fallacy, for the error of drawing conclu-
sions about individual voters from the patterns in aggregate data. A vivid
example comes from the rise of the Nazi Party in 1930s Germany. This
went hand in hand with soaring national unemployment rates and the
party also polled especially heavily in districts where there was high unem-
ployment. Most observers concluded that the Nazis must have appealed
in particular to the angry or desperate jobless. Yet that conclusion goes
beyond the data. Later and more finely-grained analysis indicated instead
that unemployed voters turned to the Communists while the employed
middle classes, seeing in the lengthening dole queues the threat of unrest
and revolution, turned to a Nazi Party promising order, security and a
steadfast opposition to communism.
The other crucial advantage of survey data is that they can begin to
tell us why people vote the way they do. Aggregate statistics refer only to
the objective characteristics and behaviour of a population; voting deci-
sions are driven by people’s beliefs, attitudes and judgements. Consider
the hypothesis discussed just above—that unemployment drives support
for parties on the far right of politics—which remains a live debate in
studies of voting in Britain and beyond (Rooduijn, 2018). Underlying
that hypothesis are various ideas about why the jobless might turn to a
radical right-wing party like UKIP or the Brexit Party. One is that unem-
ployed people are prone to blame immigrants for their plight and so
are attracted by an anti-immigration party. Another is that unemploy-
ment leads to frustration and anger with the system and, in turn, to
sympathy with the anti-establishment stances of radical right parties. Atti-
tudes like hostility to immigration and anger with the system can only be
measured (and, even then, only imprecisely) using surveys. Thus, if we
want to know why certain characteristics are associated with party choice,
individual-level data from surveys are crucial. Without them, theories
about why people vote the way they do would remain highly speculative.
The limitations of aggregate data are well illustrated by returning to
the topic of swing. As explained above, swing is a summary measure of
net electoral change based on aggregate data (namely, election results).
It cannot tell us how individuals behaved. The five per cent Labour-to-
Conservative swing between 2017 and 2019, for example, does not mean
that five per cent of the electorate switched their vote from Conserva-
tive to Labour. It is possible that more than five per cent did so; it is
also possible that no one did. This can be easily understood if we distin-
guish the different components of electoral change. Whether over the
18 D. DENVER AND R. JOHNS
Measures like swing simply summarise the effects of all of these ebbs
and flows. Only if there is imbalance in one or more type of switching—
one major party gaining more from other parties, or losing less through
abstention or physical replacement—will that individual-level switching
translate into aggregate-level change. For more detailed information
about the various components of change we have to turn to individual-
level data produced by surveys.
Ideally, a survey study designed to analyse electoral change would inter-
view a sample of voters after one election and then re-interview the same
people after a second election. This is a panel survey. It is well estab-
lished that many voters do not accurately recall which party they voted for
some years previously and the use of a panel reduces this problem. Panel
surveys have their own problems, however. They suffer from ‘attrition’ as
many respondents—especially those less interested in politics—drop out
and the sample becomes unrepresentative; and even this approach strug-
gles to measure the impact of the physical replacement of the electorate
(since it is impossible to re-interview people who have died or emigrated
between the two elections). Nonetheless, the panel design is an extremely
1 STUDYING BRITISH ELECTIONS 19
powerful tool for those analysing electoral change. The usual method is
to construct a two-way table showing exactly what people did at the two
elections concerned. A table of this kind is sometimes called a ‘flow of
the vote’ table and Table 1.3 is an example of this based on BES Internet
panel respondents who were asked about their votes in the 2017 and
2019 elections. The table shows how survey data can provide informa-
tion about the various elements of electoral change which could not be
derived from election results alone.
In most British elections, switching between the two major parties
is relatively rare. That was less true in 2019 (and in 2017) than usual
because the cross-pressures of Brexit provided an unusual powerful
inducement for Labour Leavers and Conservative Remainers to cross to
the other side. However, since a larger proportion of former Labour
voters switched to the Conservatives (10%) than vice versa (4%), this
already accounts for part of the overall swing between the two. Labour
also lost more voters to the other parties than did the Conservatives,
although it picked up more of those who had voted Liberal Democrat
in 2017. On the other hand, the Conservatives made much greater gains
from ‘others’ (mainly former UKIP voters). Finally, Labour also lost out
from the switching between voting and abstention. Since Conservative
supporters tend to turn out more dependably, the general pattern is that
Labour loses more of its former voters to abstention but also gains more
of those who return to the polls. Only the first of these happened in 2019,
however, the Conservatives proving slightly more popular among erst-
while abstainers. The likely reason is explored in the next two chapters:
Table 1.3 The ‘flow of the vote’ between 2017 and 2019
2017 vote
2019 vote Did not vote Conservative Labour Lib Dem Other
% % % % %
Theories of Voting
The study of elections and voting behaviour is not just about collecting
facts and ‘number crunching’ on the basis of complicated statistical anal-
ysis. As in all social sciences, facts have to be selected, collected, ordered,
analysed and interpreted. For that, theory is required. We have never yet
seen a survey study of voting which asked respondents what colour their
eyes are, for the simple reason that there is no theory suggesting that
eye colour may affect a person’s choice of party. Theories give guidance
as to which facts should be collected and how they should be organised,
interpreted and explained (see Sanders, 2017).
1 STUDYING BRITISH ELECTIONS 21
The first condition requires that the social groups concerned must have
differing material or symbolic interests which are affected by government
policy. Thus council tenants and owner-occupiers might have different
interests with regard to housing policy. Policy in areas such as abortion
or embryo research might not directly affect many people but are of
symbolic importance to some groups (such as Roman Catholics). The
second and third conditions relate to the processes by which social and
political divisions are maintained and reinforced. In Chapter 3, we show
how these conditions used to be fulfilled in the context of social class in
Britain—but are much less clearly in evidence today.
We have here, then, what might be termed an ‘interests plus sociali-
sation’ theory or model. Different social groups have different interests
and hence different policy preferences. They tend, therefore, to vote
for different parties which they perceive as representing these inter-
ests. Awareness of a group’s distinctiveness and of group-party links is
sustained by regular contact with fellow group members in the family,
among peers and in the community.
This model is appealingly simple and has played a large part in
voting research in Britain. However, it is not without difficulties. One of
these is the issue of overlapping group memberships. People belonging
to the same social class may differ in terms of religion, ethnicity or
income bracket, calling into question the idea that they all share the
same interests. This might explain why the sociological approach is often
confounded by ‘deviants’: that is, the (often large) minorities who do not
conform to group voting norms. A case in point is British Indians. In
the 2017 and 2019 elections, around 40 per cent of that group voted
Conservative (Martin & Khan, 2019). This looks surprisingly high if
we assume that ethnicity was their primary group identity, given that the
1 STUDYING BRITISH ELECTIONS 23
Conservative vote share among BAME voters as a whole was only 20 per
cent. However, if social class had driven the decisions of these predomi-
nantly affluent, suburban voters, we might have expected an even larger
Conservative vote share. The theory offers few clues about which group
membership will be decisive in determining an individual’s party support.
That is partly because it risks giving the impression that voters do all the
work, automatically calculating their interests and finding the party that
best serves them. This ignores the active role that political parties play in
mobilising and structuring the electorate—that is, in encouraging voters
to feel certain identities while downplaying others. Third, the theory also
neglects the key part played by national identities. Voting in Scotland has
long been partly a competition between Scottish and British identities
which largely transcend the kinds of social divisions discussed by Berelson
and his colleagues (Carman et al., 2014, ch. 2). Meanwhile, an assertion
of English—as opposed to British—identity was part of the UKIP surge
in 2015 that attracted votes across social classes (Hayton, 2016).
Some of these difficulties were addressed—at least in part—by an
important extension to the sociological model. This, too, derives from
the study of voting in the United States, and specifically from a famous
book—The American Voter (Campbell et al., 1960). Campbell and his
colleagues developed what has come to be known as the ‘Michigan
model’ (since the original research was directed from the University of
Michigan). It shared the same foundations as the sociological model in
its suggestion that long-term factors are most important in determining
party choice. However, in the Michigan model there is no simple step
from social location to voting behaviour. Rather, the social position that
an individual occupies affects the kinds of influences that he or she will
encounter in interacting with family, friends, neighbours, work colleagues
and so on. As a consequence of these interactions, especially those within
the family, the individual acquires a party identification. This means a
sense of attachment to a party, a feeling of commitment to it, being a
supporter of the party and not just someone who happens to vote for
it from time to time. It has something in common with national iden-
tity: just as most of us think of ourselves as English, Scottish, Welsh,
British or whatever, many also think of themselves as Conservative or
Labour. Other analogies might be support for a football team or the kind
of ‘brand loyalty’ that some consumers have for a particular make of car
or chocolate bar.
24 D. DENVER AND R. JOHNS
which they think will bring them the greatest benefit. This apparently
simple suggestion has importantly influenced recent electoral analysis,
spawning a variety of more specific approaches to understanding party
choice such as issue voting, spatial or proximity voting, economic voting,
and valence or ‘performance’ politics—the notion that voters choose the
party that looks most capable of governing effectively. Admittedly, it is a
broad definition of rational choice that includes all of these approaches.
For example, the valence model of voting has acquired an importance in
recent studies of British voting such that it could reasonably have been
singled out as a third model in itself. The different approaches that we
have collected under the heading of ‘rational choice’ represent different
ways of specifying how voters evaluate the ‘benefits’ available from the
competing parties; otherwise, saying that voters choose the party that will
bring them the greatest benefit is trivial or even tautological. The reason
for collecting them together is that, in this book, we use the term ‘rational
choice’ largely as a contrast with the sociological or Michigan approach
and the word ‘choice’ is just as important as the word ‘rational’. The point
is that, rather than voting more or less automatically in line with long-
standing attachments, voters instead make a choice anew at each election
on the basis of what the parties are offering at that time. It is also worth
noting that ‘rational choice’ does not imply arduous and complex calcu-
lations of the costs and benefits associated with each party. It may well
be more rational, especially for those less engaged with politics, to find a
quick and simple way of judging the parties.
To summarise, then, theories or models are required to provide frame-
works within which appropriate data can be collected and understood
and also to provide explanations for the empirical relationships that are
discovered. Whether the data are at the aggregate or individual level,
theory guides us about which variables are likely to matter—which socioe-
conomic characteristics of a constituency to investigate, which survey
questions to ask. Having outlined the most influential general theories
of voting behaviour here, we will discuss them again, offering some eval-
uation of their relative usefulness, in the final chapter. Before examining
patterns of party choice, however, it is important to discuss another deci-
sion that electors have to make—whether to vote at all—which logically
must precede a decision about which party to vote for. The next chapter,
therefore, is concerned with turnout in elections.
1 STUDYING BRITISH ELECTIONS 27
References
Benney, M., Gray, A. P., & Pear, R. H. (1956). How people vote. Routledge &
Kegan Paul.
Berelson, B., Lazarsfeld, P., & McPhee, W. (1954). Voting. University of Chicago
Press.
Butler, D. (1998). Reflections on British elections and their study. Annual
Review of Political Science, 1, 451–464.
Butler, D., & Stokes, D. (1969). Political change in Britain (1st ed.). Macmillan.
Butler, D., & Stokes, D. (1974). Political change in Britain (2nd ed.).
Macmillan.
Campbell, A., Converse, P., Miller, W., & Stokes, D. (1960). The American
Voter. Wiley.
Carman, C., Johns, R., & Mitchell, J. (2014). More Scottish than British: The
2011 Scottish Parliament Election. Palgrave Macmillan.
Crewe, I. (1985). Great Britain. In I. Crewe & D. Denver (Eds.), Electoral
change in Western democracies (pp. 100–150). Croom Helm.
Edgeworth, F. (1905). The law of error. Transactions of the Cambridge Philo-
sophical Society, 20, 35–65 and 113–44.
Eriksson, L. (2011). Rational choice theory: Potential and limits. Palgrave
Macmillan.
Hayton, R. (2016). The UK independence party and the politics of Englishness.
Political Studies Review, 14(3), 400–410.
Heath, O., & Johns, R. (2010). Measuring political behaviour. In M. Bulmer, J.
Gibbs, & L. Hyman (Eds.), Social measurement through social surveys (pp. 47–
68). Ashgate.
Lazarsfeld, P., Berelson, B., & Gaudet, H. (1944). The people’s choice. Duell,
Sloan and Pearce.
Leeke, M. (2003). UK election statistics: 1945–2003. House of Commons
Library.
Martin, N., & Khan, O. (2019). Ethnic minorities at the 2017 British general
election. Runnymede Trust.
Mellon, J., Evans, G., Fieldhouse, E., Green, J., & Prosser, C. (2018). Brexit or
Corbyn? Campaign and inter-election vote switching in the 2017 UK General
Election. Parliamentary Affairs, 71(4), 719–737.
Miller, W. (1981). The end of British politics. Clarendon Press.
Milne, R. S., & Mackenzie, H. C. (1958). Marginal seat. Hansard Society.
Pedersen, M. N. (1979). The dynamics of European party systems: Changing
patterns of electoral volatility. European Journal of Political Research, 7 , 1–26.
Rallings, C., & Thrasher, M. (2012). British electoral facts 1832–2012. Biteback.
Rooduijn, M. (2018). What unites the voter bases of populist parties? Comparing
the electorates of 15 populist parties. European Political Science Review, 10(3),
351–368.
28 D. DENVER AND R. JOHNS
It is now one o’clock P.M. and Mr. Taft is at home, where this personage stands
out more boldly before us, since the trials through which the country is passing do
not permit us yet to judge him politically.
The President of the Commission, in his private life, has many points of
similarity with Count de Caspe, that stainless gentleman the Filipinos still recall
with veneration. Excepting the brilliancy of those splendid entertainments with
which he endeavoured to blot out all racial differences by mingling in fraternal
embrace Filipinos and Spaniards at the Malacañan villa, there ordinarily reigned in
the governor’s mansion the placid silence of the home of a well-to-do retired
merchant. The Countess, who on Thursdays did the honours of her salon with
exquisite tact, was during the other days of the week a housekeeper who did not
disdain to go to a grocery store to make purchases, or to look over the laundry list.
The same thing happens in the elegant chalet at Malate where Mr. Taft lives.
This is a quiet and peaceful home, a temple erected to the affections, under whose
roof Mr. Taft rests some hours after the efforts which his political work demands.
His table reflects his modest character. Four courses, two kinds of fruit, a dessert
and sauterne compose the menu of the luncheon where Mr. Taft is always
accompanied by some guest, either Filipino, American or Spanish. During the meal
politics are banished; if the guest is a Filipino who speaks French Mrs. Taft
interrogates him on the customs of the archipelago; if he is Spanish, as to the
toilettes worn in Manila by the ladies at the most brilliant receptions held here; as
to the favourite musical composer of the Hispano-Filipino society; and this
conversation increases in attraction when Miss Herron, sister-in-law of Mr. Taft
and the incarnation of the modern woman’s education, takes part therein. Miss
Herron speaks French correctly, has travelled much, and journeyed through Spain
like an intelligent tourist. The architectural lace-work of the Alhambra charmed
her, and she went into ecstasies over the orange blossoms growing along the banks
of the Guadalquivir. With what Miss Herron was not in harmony, and she berates
them like an unsubsidised journalist, were the Spanish railroads. Miss Herron is
right.
The children, Robert, about eleven years old; Helen, a girl of nine, and Charles, a
baby of three, who is the king of the household:—the McKinley, as it were, of this
patriarchal republic—do not come to the table; they eat with the governess.
After the meal, in the fine gallery overlooking the sea, sipping the coffee, Mr.
Taft talks of the education of his children, of the difficulties met in the Philippines
in the solution of so interesting a problem; and his wife converses of the charitable
work she expects to undertake when she shall have assumed a more permanent
place in the Archipelago, which Magellan discovered for Spain, and which, through
a horrible fatality, is no longer ours. Politics are also eschewed on the gallery.
Needless to say this extraordinary editorial afforded us all
boundless amusement; we began to caution Mr. Taft frequently
about the careful preservation of his “aureole” and Maria and I
decided that we would have to walk warily indeed, if we were
destined to be so minutely reported.
CHAPTER VI
A STRANGE ENVIRONMENT
I had a pair of ponies and a Victoria; Mr. Taft had his two little
brown horses and a Victoria; besides which there was an extra horse
to be used in case of accident to one of the others, as well as a pony
and calesa for the children. This rather formidable array was
necessary because we found it impossible to take a horse out more
than twice a day, and usually not more than once, on account of the
sun. My ponies were taken out only in the early morning or the late
evening, and those of Mr. Taft had all they could do to take him to
the office and bring him home twice a day. Distances were long and
there were no street cars which ran where anybody wanted to go.
This number of conveyances made a good many stablemen
necessary and all of them, with their families, lived in quarters
attached to the stables. These families consisted of fathers, mothers,
sisters, brothers, uncles, aunts, cousins near and far removed, wives,
children, grandchildren, and a few intimate and needy friends with
their family ramifications. Besides our three cocheros and the stable
boys, there was a gardener with his parientes, so it is no wonder that
on my first inspection of the lower premises I should have thought
that some sort of festivity was in progress. I might have lived in
Manila twenty years without being able to straighten out the
relationships in this servant colony; it was not possible to learn who
had and who had not a right to live on the place; and my protest was
met with the simple statement that it was el costumbre del pais, so I,
perforce, accepted the situation.
Filipino servants never live in the master’s residence; they never
want to; they want the freedom of a house of their own, and these
houses are, as a rule, built on the outer edges of the garden, or
compound. I believe Americans now are learning to meet the
pariente habit by having room for just as many people as they need,
and no more. But those who live in the old places, with their ample
quarters, still gather the clans and are permitted to enjoy a most
expansive and patriarchal sensation.
My horses, when I first saw them, were a source of the greatest
pride. A beautifully matched pair of coal-black, stylishly-paced and
glossy little stallions, hardly larger than Shetland ponies, they looked
as if they had been washed in some sort of shrinking soap and had
come out in perfect condition except that they were several sizes
smaller than they ought to have been. These Philippine ponies are
doubtless descendants of the Arabian horses brought over by the
Spaniards and have been reduced to their present size by the change
of climate and the difference in food and environment, but they still
have the fine lines and the general characteristics of their
progenitors.
Mr. Taft secured mine from Batangas, where all the best ponies
come from, through the kindness of Mr. Benito Legarda, the
staunchest of Americanistas. Batangas was a most unquiet province,
the last, in fact, to become pacified, and Mr. Legarda had to pay an
insurrecto for bringing the horses through the insurgent lines and
delivering them at Calamba, near Manila. Although he did not know
their exact origin when he bought them, Mr. Taft said that if the facts
became known he would be accused, in certain quarters, of giving
indirect aid to the revolutionists; but he wanted the ponies so he did
not return them.
When they were hitched to the shining little Victoria which had
been built for them, they were as pretty as a picture and, as I did not
propose to have such a turn-out ruined by a couple of Filipinos on
the box in untidy camisas hanging outside of as untidy white
trousers, I had made for my cochero and boy, or coachman and
footman, a livery of white and green in which they took such
inordinate pride that they seemed to grow in stature and dignity.
Maria and I felt a sense of the utmost satisfaction the first time we
stepped into this carriage for a drive down to the Luneta where we
were sure to see everybody we knew and hundreds of people besides;
but our vanity was destined to be brought to a sudden termination.
As we were driving along with much satisfaction, a bit of paper
floated down alongside the blinkers of the little ebony steed on the
right and he made one wild leap into the air. His companion gave
him an angry nip, and then the fight was on. Maria and I jumped out,
which was not difficult in a low-built Victoria, and no sooner had we
done so than we saw the complete wreck of all our grandeur. With all
the leaping and plunging and biting and kicking, in the vicinity of a
handy lamp-post, the smash-up was fairly complete. Neither of the
ponies was hurt, except by the lash of the whip, and I must say the
little wretches looked rather funny; like very pretty and very bad
children, sorry for what they had done. But their characters were
established and they proceeded after that to live up to them. We
never could have any confidence in them and my coachman was the
only person who could do anything with them. He was a most
unsatisfactory man in many ways and used often to call for us at
dinner parties in a state of gay inebriety, but we didn’t dare discharge
him because everybody else in the stables stood in awe of the blacks
while he seemed greatly to enjoy his constant and spectacular
struggles with them.
The Filipinos are a most temperate people; there is no such thing
as drunkenness among them; but coachmen seem to be an exception
in that they allow themselves a sufficient stimulation of the fiery vino
to make them drive with courage and dash, sometimes minus all care
and discretion. The drivers of public vehicles seem to love their little
horses in a way; they are inordinately proud of a fast paced or
stylish-looking pony; yet they are, as a rule, quite harsh to them.
They overload them and overdrive them, and under all conditions
they lash them continuously.
No Filipino cochero likes to have another cochero pass him, and
the result is constant, indiscriminate racing, on any kind of street,
under any circumstances,—and never mind the horse.
My children were driving with their governess to the Luneta one
evening, when two caromatas came tearing down behind them, each
driver hurling imprecations at the other and paying no attention to
what was ahead of him. The result was a violent collision. The two
caromatas went plunging on, the cocheros not stopping to see what
damage they might have done—which was very characteristic—and
the children narrowly escaped a serious accident. Charlie was hurled
out and fell under the children’s calesa and Robert and Helen both
declare they felt a sickening jolt as a wheel passed over him. The
baby, too, vowed that the calesa “went wight over me, wight dere,”
indicating a vital spot; but upon the closest examination we could
discover nothing more serious than a few bruises. However, it made
us very much afraid to trust the children out alone.
The gardener had two little boys, José and Capito, who were a few
years older than Charlie, but about his size, and he took a
tremendous fancy to them. They were clad, simply, in thin gauze—or
jusi—shirts which came down a little below their waists, and I think
Charlie envied them this informal attire. He used to order them
around in a strange mixture of Spanish, Tagalog and English which
made me wonder at my wholly American child; but it was an
effective combination since he seemed to have them completely
under his thumb and, as he revelled in his sense of power, he never
tired of playing with them.
Maria and I soon adopted the universal habit of driving down to
the Escolta in the early morning to do such shopping as was
necessary. We found a variety of interesting shops, but with very
little in them to meet the ordinary demands of an American woman.
There were delightful Indian bazaars and Chinese tiendas where all
manner of gaudy fabrics and strange oriental articles were on sale,
while the Spanish shops upon which everybody had to depend in
those days, and which had such grandly European names as Paris-
Manila and La Puerta del Sol, catered largely to the Filipino taste for
bright colours.
The Escolta at that time was full of saloons, established by the
inevitable followers of a large army, and the street being very narrow
and the old, rickety, wooden buildings being very wide open, the
“beery” odour which pervaded the atmosphere at all hours was really
dreadful. Mr. Taft decided that as long as this was the only street in
town where women could go shopping, the saloons would have to be
removed. There was opposition on the Commission to the bill which
provided for their banishment, and it was fought from the outside
with great vigour and bitterness, but a majority were in favour of it,
so it passed, and the saloons had to move. There has not been a
saloon on the Escolta from that day to this and, indeed, they have
ever since been under such satisfactory regulation that there is little
evidence left of their existence in the city.
I am afraid it is going to be very difficult to convey an adequate
picture of Manila society during the first years of American
occupation. There had been, in the old days, a really fine Spanish and
rich mestizo society, but all, or nearly all, of the Spaniards had left
the Islands, and the mestizos had not yet decided just which way to
“lean,” or just how to meet the American control of the situation. I
may say here that most of the educated, high-class Filipinos are
mestizo; that is, of mixed blood. They may be Spanish mestizo or
Chinese mestizo, but they have in them a strong strain of foreign
blood. Besides the Spanish- and Chinese-Filipinos, there are a
number of British mestizos who are very interesting people. Mr.
Legarda, Chief Justice Arellano, Dr. Pardo de Tavera and Mr.
Quezon, the Filipino delegate to the United States Congress, are
Spanish mestizos, while Mr. Arañeta, the Secretary of Finance and
Justice, as well as the Speaker of the Philippine Assembly and many
able lawyers and successful business men are of Chinese descent.
The mestizos control practically all the wealth of the Philippines, and
their education, intelligence and social standing are unquestioned. It
is the only country in the world that I know about—certainly the only
country in the Orient—where the man or woman of mixed blood
seems to be regarded as superior to the pure blooded native.
Dating back also to the Spanish days was quite a numerous foreign
society consisting of a few consuls, some professional men, the
managers of banks and large British and European mercantile firms,
and their families. The leaders of the British colony were Mr. and
Mrs. Jones—Mr. Jones being the manager of the Manila branch of
the Hongkong and Shanghai Banking Corporation. Mrs. Jones, a
very beautiful and charming woman, gave some very elaborate
parties during that first winter. Bank House, the residence
maintained by the bank for its manager in Manila, is a beautiful
place in Uli-Uli, a district on the picturesque banks of the upper
Pásig, and it is finely adapted for balls and large receptions. Then
there were several German families who also entertained quite
lavishly, and I remember, especially, one Austrian exile; indeed, I
shall never be able to forget him because my husband took such joy
in pronouncing his name. He was Baron von Bosch.
This was the “set” which entertained the Commission most
cordially during our first season in Manila, while the Army officers,
following the lead of their Commanding General, held themselves
somewhat aloof. I kept up a constant round of parties of different
kinds in my house, and gave a dinner at least once a week at which
were gathered companies of a most interestingly cosmopolitan
character. And we did not fail to observe all the desirable forms. Both
Filipinos and Europeans expect a certain amount of ceremony from
the representatives of government and are not at all impressed by
“democratic simplicity”; so believing in the adage about Rome and
the Romans, we did what we could. Beside the spic and span guard at
the outer gate of the illuminated garden, we always, on dinner party
nights, stationed coachmen, or other stable boys disguised as liveried
footmen, on either side of the entrance, to receive guests and conduct
them to the dressing-rooms, and up the stairs to the reception room.
Our house was nicely adapted for a dinner of twelve and I usually
tried to confine myself to that number. We always had an orchestra,
orchestras being very plentiful in Manila where nearly every native
plays some sort of instrument, and the music added greatly to the
festive air of things, which was enhanced, too, by a certain oriental
atmosphere, with many Japanese lanterns and a profusion of potted
plants and great, hanging, natural ferneries and orchids which were
brought in from the forests by the Filipinos and sold on the streets.
My husband is supposed to be the author of the phrase: “our little
brown brothers”—and perhaps he is. It did not meet the approval of
the army, and the soldiers used to have a song which they sang with
great gusto and frequency and which ended with the conciliating
sentiment: “He may be a brother of William H. Taft, but he ain’t no
friend of mine!”
We insisted upon complete racial equality for the Filipinos, and
from the beginning there were a great many of them among our
callers and guests. Their manners are models of real courtesy, and,
while their customs are not always like ours, wherever they are able
they manifest a great willingness to be conforme,—to adapt
themselves,—and their hospitality is unbounded.
I shall never forget my first call from a Filipino family. They
arrived shortly after six in the evening: el señor, la señora and four
señoritas. We went through a solemn and ceremonious handshaking
all around. I received them first, then passed them on to my husband
who, in turn, passed them on with a genial introduction to my sister
Maria. We had been sitting on the verandah, and when a semi-circle
of chairs had been arranged, the six of them sat down; el señor
noisily cleared his throat a couple of times while the ladies calmly
folded their little hands in their laps and assumed an air of great
repose. It was as if they had no intention of taking any part whatever
in the conversation.
El señor explained in Spanish that they were our near neighbours
and that they had called merely to pay their respects. Mr. Taft had
been studying Spanish diligently ever since he left the United States,
but he is not conspicuously gifted as a linguist, and he had not yet
waked up—as he so often expressed a wish that he might—to find
himself a true Castilian. However, his ready laugh and the cordiality
of his manners have always had a peculiar charm for the Filipinos,
and he was able on this occasion, as he was on many future ones, to
carry off the situation very well. We all nodded and smiled and said,
“Si Señor” and “Si Señora,” to long and no telling what kind of
speeches from our guests; then Maria and I complimented the ladies
on their beautifully embroidered camisas, which started things off
properly. They praised everything in sight, and what we didn’t get
through the little Spanish we knew, we got from gesture and facial
expression. They got up and wandered all around, feeling of my
Japanese tapestries and embroideries, breathing long “ahs!” of
admiration over my gold screens and pictures and curios, and acting
generally like callers who were being very well entertained. Then the
children came in and they broke out afresh in voluble praise of them.
I assumed the proper deprecatory mien in response to their
laudation of my children, and altogether I felt that we were
acquitting ourselves rather well in this first inter-racial social
experience.
But at the end of half an hour the strain was getting a little severe
and I was wondering what to do next, when our six callers arose and
said they must be going. I breathed an inward sigh of relief and was
making ready to escort them to the top of the stairs, when my
husband cordially exclaimed:
“Why, no! Porque? Tenemos bastante tiempo. Why hurry?” And—
they—all—sat—down!
I regretted then even the little Spanish Mr. Taft had learned,
though, of course, he didn’t expect them to heed his polite protest.
He knew nothing at all about Filipino manners; he didn’t know they
expected to receive some sign from him when it was time to go and
that they would consider it discourteous to go while he was urging
them to stay. He kept up, without much assistance, a brave if
laboured conversation, and the minutes slowly passed. Our dinner
hour approached and I darted warning glances at him, for I had a
horrible fear that he just might ask them to remain and dine. But at
the end of another hour a strained expression began to spread itself
over even his face, and there was not a word of protest from him
when, at a quarter past eight, our little brown neighbours once more
indicated an intention of going home. We entertained Filipino callers
nearly every day after that, but never again did we urge them to
reconsider their sometimes tardy decision to depart.
With regard to Filipino manners and customs; I am reminded that
we were nonplussed, though greatly amused by the costumbre del
pais which decreed that some return be made by a Filipino for any
and all favours bestowed upon him. We grew accustomed to this
before we left the Islands, and came to expect a few offerings of sorts
almost any day in the week, but in the beginning it was usually most
embarrassing.
One time, soon after our arrival, a very loyal Americanista was
shot down in the street, during the peaceful discharge of his duty, by
an insurrecto. His widow, with her children, came into Manila in a
state of utter destitution, to secure some recompense from the
government for her husband’s services, and while her case was
pending Mr. Taft, in great pity for her, sent her money enough to live
on. The next day the whole family, from the wide-eyed boy to the
babe carried astride the mother’s hip, came to call on their
benefactor, bringing with them as a gift a basket containing a few
eggs, some strange Philippine fruits and a lot of sea-shells. Mr. Taft
was deeply touched, and with the brusqueness of a man who is
touched, he told her he had given her the money to buy food for
herself and her children and not for him, and he refused her offering.
I know, by the light of a fuller knowledge of the character of the lowly
Filipino, that she went away feeling very much cast down.
But in connection with such gifts there were always more laughs
than sighs. We invited to luncheon one day a dashing Filipino named
Tomaso del Rosario. Señor Rosario, a man of wealth and prominence
who had a fine Spanish education and was well dressed in the high-
collared, patent-leathered and immaculate-linened Spanish style,
was quite self-confident and enjoyed himself very much. He seemed
attracted to Maria and she, being linguistic, was able to talk to him in
a mixture of many languages. The next day she received from Señor
Rosario, not a floral offering, but a basket filled with nuts, a canned
plum-pudding, some canned chocolates and preserved fruits. This
attention did not seem so remarkable, however, when we learned, to
our amusement, that he had sent exactly the same present to Alice
Worcester, then five years old.
Our life, on the whole, was intensely interesting in its unusual
atmosphere and curious complications, but throughout everything
we were made to feel the deep significance of our presence in the
Islands; and the work of the Commission was first, last and always to
us the subject of the greatest moment. Even in our daily round of
social affairs we dealt with tremendous problems whose correct
solution meant the restoration of peace and prosperity to what then
should have been, and what we knew could be made, a great country.
That for which the American flag had always stood began to assume,
for many of us, a broader and a finer meaning; and being so much a
part of our flag’s mission in a strange field a certain zest was added to
our patriotism which we had never felt before. I believe, and I think
all those who know the truth believe, that Americanism, in its highest
conception, has never been more finely demonstrated than in the
work done by the United States in the Philippine Islands; work, the
broad foundation for which the Commission was engaged in
constructing during the period of which I write.
So many were the problems to be met and dealt with that in the
beginning the Commissioners were each given a set of subjects for
investigation and study, their findings being submitted for debate
and consideration in the general meetings.
Taxation, civil service, provincial and municipal organisations,
currency and finance, police, harbour improvements, roads and
railways, customs, postal service, education, health, public lands, an
honest judiciary and the revision of the code of laws; these were
some of the vital problems, but underlying them all was the
immediate necessity for the establishment of tranquillity and
confidence throughout the archipelago.
In order to make clear, in any degree, the Philippine situation as
we found it, it is essential that, briefly, the position of the Catholic
Church and its representatives, the Friars, be explained. For the first
time in its history the American government found itself compelled
to adjust a seemingly insurmountable difficulty between a church
and its people.
With us the Church is so completely separate from the State that it
is difficult to imagine cases in which the policy of a church in the
selection of its ministers, and the assignment of them to duty could
be regarded as of political moment, or as a proper subject of
comment in the report of a public officer, but in the first reports of
the Philippine Commission to Washington this subject had to be
introduced with emphasis.
The Spanish government of the Philippine Islands was a
government by the Church through its monastic orders, nothing less.
In the words of the Provincial of the Augustinians, the Friars were
the “pedestal or foundation of the sovereignty of Spain” which being
removed “the whole structure would topple over.” The Philippine
people, with the exception of the Mohammedan Moros and the non-
Christian tribes, belonged, during the Spanish dominion, to the
Roman Catholic Church, and the Church registry of 1898 showed a
total membership of 6,559,998. The parishes and missions, with few
exceptions, were administered by Spanish Friars of the Dominican,
Augustinian and Franciscan orders, and it was to the nature of this
administration that Spain owed the insurrections of 1896 and 1898,
the latter of which terminated only upon our assuming control of the
islands.
In 1896 there were in the Philippines 1,124 monks of the
Augustinian, Dominican and Franciscan orders, which body included
a company of Recolletos, who are merely an offshoot of the order of
St. Augustine and differ from the Augustinians only in that they are
unshod. In addition to these there were a few Jesuits, Capuchins,
Benedictines and Paulists, but they engaged in mission and
educational work only and did not share with the other orders the
resentment and hatred of the people. Filipinos were not admitted to
any of the orders, but they were made friar curates and served as
parish priests in some of the smaller places.
When a Spanish Friar curate was once settled in a parish he
remained there for life, or until he was too old for service, and
because of this fact he was able to establish and maintain an
absolutism which is difficult to explain in a few words. He was simply
everything in his parish. As a rule he was the only man of education
who knew both Spanish and the native dialect of his district, and in
many parishes he was the only Spanish representative of the
government. In the beginning, through his position as spiritual
guide, he acted as intermediary in secular matters between his
people and the rest of the world, and eventually, by law, he came to
discharge many civil functions and to supervise, correct or veto
everything which was done, or was sought to be done in his pueblo.
He was Inspector of Primary Schools, President of the Board of
Health and the Board of Charities, President of the Board of Urban
Taxation, Inspector of Taxation, President of the Board of Public
Works, Member of the Provincial Council, Member of the Board for
Partitioning Crown Lands, Censor of Municipal Budgets, and Censor
of plays, comedies or dramas in the dialect of his parish, deciding
whether or not these were against the public peace or morals. In a
word, he was the government of his parish; and in addition to all
things else, it was he who, once a year, went to the parish register,