![1
Complexity and Case
D a v i d L . H a r v e y
Case studies have long kindled the sociologi-
cal imagination. Yet their scientific status has
been contested almost from the moment soci-
ology was christened an academic discipline.
Avatars of ‘hard science’ regard case studies
as ‘too impressionistic’ in the interpretive
liberties they take and as too aggressive in the
range of subjects they presume to address.The
legitimacy of case studies has been contested
largely, therefore, on methodological and
epistemological grounds, while fundamental
issues of social ontology seldom attract the
same attention. This chapter addresses this
oversight as it bears on case-based social
research. As such, it has two goals. First,
it complements the current discussion of
case methods by exploring the ontological
requirements of case research; and second, it
uses a complex realist paradigm to illuminate
the structure of the case-object (Byrne 2002,
pp. 5–6; Byrne and Harvey 2007, pp. 64–66).
Christopher Norris (2007) defines ontol-
ogy’s intellectual niche and the type of entities
that occupy that space as follows:
Traditionally, … [ontology] … is the field of
philosophic inquiry that has to do with things … or
with real world objects and their intrinsic structures,
properties, causal powers, and so forth. That is to
say, it treats of them just in so far as they are taken
to exist and to exert such powers quite apart from
the scope and limits of human knowledge. (Norris
2007, p. 335)
Norris’s realist definition provides a feel for
the complexity and dynamic powers of the
entities with which this chapter deals. It also
anticipates the predicaments that ontological
considerations introduce into social scientific
discourse.
Three levels of ontological interest are
germane to sociological inquiry: (1) philo-
sophical ontology; (2) scientific ontology; and
(3) social ontology. Philosophical ontology
deduces from the structure of speculative
thought the fundamental nature of the entities
that constitute our everyday world. Scientific
ontologies are nested within philosophical
ontologies to the extent they flesh out the local
details of a terrain in a way philosophical
ontology cannot. They are more narrowly
focused because they deal with the existential
constitution of the entities scientists inves-
tigate. Finally, social ontologies are nested
within scientific ontologies in that they deal
with the elemental entities and dynamics](https://image.slidesharecdn.com/11843150-250520101544-6744d3fd/85/The-Sage-Handbook-Of-Casebased-Methods-David-Byrne-And-Charles-C-Ragin-38-320.jpg)
![COMPLEXITY AND CASE 19
is that they are closely and inextricably interrelated.
Obviously, all statistics ultimately come from ‘cases’.
A schedule is constructed on the basis of data
derived from the examination of individual cases.
From the examination and analysis of cases come
insights, clues, and leads, on the basis of which new
items may be introduced into the schedule and their
average significance determined. Correspondingly,
an unexpected statistical finding may require
exploration by interview to determine its meaning
and significance for [marital] adjustment. Further-
more, the analysis of ideal types of dynamic and
organic relationships derived from case studies may
suggest reorganization of predictive items under
more significant categories (Burgess 1939/1972,
pp. 386–388).
It is not untoward to suggest that Burgess
saw the relation between case studies and
survey research as a tension-ridden dialectical
unityofopposites.Hemayhaveevenregarded
the interdependence of the two techniques
as somehow paralleling the symbiotic pro-
cesses of ‘cooperative competition’ he saw
operating in his dynamic modeling of urban
ecology.
Despite the methodological hiatus separat-
ing them, when set side by side, the nomo-
thetic and the ideographic studies combined
to illuminate the complex moral texture of
Chicago’s ‘natural areas’ and ethnic neigh-
borhoods; they were also indispensable tools
for mapping the city’s centrifugal ecology and
concentric structuration (Burgess 1925/1967,
pp. 47–63). Hence, no matter how one judges
the Chicago School’s efforts at nomothetically
squaring the ideographic circle, this much is
evident: the logical contradictions between
the two methods were never resolved. This
unabated ‘benign neglect’ of methodological
schisms may have been in all likelihood a
constant source of intellectual tension during
the interwar period. At the same time, the
Chicago School’s neo-Kantianism undoubt-
edly remained a wellspring of sociological
creativity, even as it served as a leavening
source of departmental ambivalence. In any
case, it was no mean trick to negotiate
the intractable logics of social scientific
surveys and case-based narratives without
glossing their differences or submerging the
premises of the one into the other. In fact,
the genius of Chicago Sociology resided not
in the unification of these antinomies but,
rather, in their pragmatic, often piecemeal
amalgamations. It was this dialectical ambiva-
lence that sustained Chicago Sociology’s
watershed scholarship of the 1920s and
1930s.
The rise of methodological
factionalism
JusthowwellChicagoSociologymanagedthe
nomothetic/ideographic split can be gleaned
from what happened to it and to the profession
after the Second World War. A powerful
‘second Chicago School’ formed around
‘Symbolic Interactionist’ social psychology.
The Symbolic Interactionist contingent soon
set itself off from the survey researchers
(Fine1995,Gusfield1995,Abbott1999,2001,
Becker 1999).
Homologous schisms erupted elsewhere
in the profession so that by the mid-
1950s the language of the earlier neo-
Kantian antinomies would be supplanted by
an alternative oppositional vocabulary, e.g.
variable-oriented, quantitative versus qualita-
tive research and its narrative productions.At
one level, the substantive parameters of the
new controversy paralleled the old: only the
vocabularies changed. It terms of professional
consequences, though, the formal disputes
that had been held in relative abeyance now
tightened into a Gordian knot. When that
knot was finally cut, the former dialectical
ambivalence gave way to a viral, profession-
wide internecine struggle.
This mutual alienation between the two
self-interested factions hardened to such an
extent that the position of the former was
reduced to what C. Wright Mills (1959,
pp. 50–75) deprecated as ‘abstracted empiri-
cism’: fact-mongering in search of a working
hypothesis. Returning the favor, ‘positivists’
showed contempt for ‘grand theory’ (again a
Millsian epithet) and the extended narratives
associated with case research. Indeed, more
than one cutting-edge positivist suggested,
only half-mockingly, that such sociology be
banished to English Departments, American](https://image.slidesharecdn.com/11843150-250520101544-6744d3fd/85/The-Sage-Handbook-Of-Casebased-Methods-David-Byrne-And-Charles-C-Ragin-42-320.jpg)
![COMPLEXITY AND CASE 21
Conception of
cases assume
that…
Cases are empirical
units which are,
in turn…
Conceptually specific
to the single case
Conceptually subsumed
under a general
class of cases
Cell 1:
Cases as
found
Cell 2:
Cases as
objects
Cell 3:
Cases as
made
Cell 4:
Cases as
conventions
Cases are theoretical
constructions which
are, in turn…
Conceptually classed
as a single case
Conceptually subsumed
under a class of cases
Figure 1.1 Ragin’s mapping of sociological conceptions of case.
conceptual constructions whose full meaning
emerges only when placed in the context of
existing research.
Ragin concludes by noting that these four
conceptions of case could be interpreted as
pragmatic strategies available to researchers
for resolving various problems that arise
in the concrete research setting itself. The
pragmatic and eclectic tone of his conclusions
are underscored when Ragin writes:
This four-fold division of case conceptions is
not absolute. A researcher could both use con-
ventionalized empirical units, accepting them as
empirically valid (Cell 2), and try to generate new
theoretical categories or case constructs (Cell 3)
in the course of his or her research. Frustrations
with conventional case definitions and practices
(Cell 4) could lead researchers to intensify their
empirical efforts and to define cases and their
boundaries in a more inductive manner (Cell 1). In
fact, most research involves multiple uses of cases,
as specific or general theoretical categories and as
specific or general empirical units. These multiple
uses occur because research combines theoretical
and empirical analysis, and the two kinds of
analyses need not use parallel cases or units. The
point of … [Figure 1.1] … is not to establish
boundaries between different kinds of research, but
to establish a conceptual map for linking different
approaches to the question of cases. (Ragin and
Becker 1992/2005, p. 11)
In keeping with the methodological focus
of Ragin and Becker’s work, what began
ostensibly as a collation of methodological
presuppositions has become a ‘conceptual
map’ for finessing possible ‘conceptual map’
‘blockages’ encountered in field research or
later interpretive dilemmas.
Notice, however, that although What Is a
Case? deals adequately with the subjective
conceptions the researcher transports to the
casing milieu, it never directly addresses the
social ontology of the case-object itself, i.e.
the possibility that the case-object is, in and
of itself, a complex entity with autonomous
claims of its own. Instead, the act of casing
and its case-object are both approached as
if they were social constructions posited
during the research act. The typology thus
focuses on the intensive logic of various](https://image.slidesharecdn.com/11843150-250520101544-6744d3fd/85/The-Sage-Handbook-Of-Casebased-Methods-David-Byrne-And-Charles-C-Ragin-44-320.jpg)
![24 THE SAGE HANDBOOK OF CASE-BASED METHODS
ontology so difficult or even impossible? Although
the philosophical doctrines of, say, Hume or
Kant are familiar, it is important to appreciate a
phenomenological or practical basis for them. This
is in a kind of natural attitude to the world in
which we do not distinguish between knowledge
and objects, in which we don’t distinguish ontology
from epistemology … in which we just talk about
a known world. When science is doing normal
science, when things are going well, epistemically it
is natural to adopt what I call this ‘natural attitude’
in which you don’t posit ontology on the one
side and epistemology on the other: you just have
knowledge and you take it that that knowledge is
of the world. Ontology really only becomes relevant
when you are not satisfied with knowledge, when
what passes for knowledge is patently wrong or
absurd. (Bhaskar 2007, p. 192)
Bhaskar seems to have described the cur-
rent attitude of many sociologists toward
their work. Their anti-ontological stance is
manifested in the failure of social scientific
discourse systematically to ‘… distinguish
between knowledge and objects, in which
we don’t distinguish ontology from epis-
temology …’. Moreover, such systematic
glosses lead to a premature closure of self-
critical possibilities. Most immediately, this
truncation is magnified by the idea that
Humean Empiricism, to the exclusion of
more complex models of causation, forms
a sufficient basis for scientific explanations.
Consequently, when confronted with issues of
philosophical or social ontology, the scientist
too often falls into a certain obtuseness,
insisting that ‘… you just have knowledge
and you take it that [this] knowledge is
of the world’. We believe that a complex
realist framework of sociological analysis can
correct this anti-ontological bias.
As for complex realism proper, it emerged
rather haphazardly during the social and
political ferment of the 1960s. Responding to
the implausible Weltanschauung of the era,
it navigated between the exhausted claims
of Logical positivism and the exaggerated
promises of Culutral Hermeneutics. Situating
itself between an empiricism that had been
dismissive of ontology and a phenomenolog-
ical method that was verging on solipsism,
complex realism fashioned a social scientific
worldviewfromelementsofaresurgentrealist
philosophy, mathematical chaos theory, non-
equilibrium thermodynamics, and punctuated
equilibria models of evolutionary develop-
ment (Sayer 1984, Prigogine and Stengers
1984, Eldredge 1985, Gleick 1987, Gould
1989, Waldrop 1992, Arthur 1994, Collier
1994, Prigogine 1996).
The new science’s name reflected the
intellectual hurly-burly from which it grew.
That is, ‘complex realism’ was a colloquial
compression of two terms: complexity theory
and critical realism. The label critical realism
had, in turn, been formed from an elision of
two philosophical conceptions, e.g. transcen-
dental realism and critical naturalism. Four
fundamental principles subsequently charted
the development of complex realism:
• Transcendental realism: Complex realism holds
that if the sciences are to remain coherent
enterprises, their explanatory protocols must
embrace, a priori, a realist ontology, i.e. they must
of necessity assume the existence of an objective
moment of experience, some part of which stands
over and against our attempts to understand and
manipulate our world.
• Critical naturalism: Complex realism assumes the
possibility of a ‘natural science of society’ but
if, and only if, that science is properly qualified
so as to allow active human agency a role in
‘co-producing’ its everyday institutional forms.
• The ontological and hierarchical stratification of
the world: The world is ontologically stratified.
Hence, the objects of scientific inquiry (either
naturally occurring or as products of social
construction) form a loosely integrated hierarchy
of openly structured, evolving systems, which are
themselves ontologically layered and hierarchi-
cally nested.
• The asymptotic predictability of open-system
outcomes: Because of this ontological layer-
ing, actual causal processes operating in non-
laboratory settings are contingently structured
and temporally staggered. They seldom rival the
regularity of causal conjunctions elicited in con-
trolled experiments. Consequently, ‘causal laws’
in complex natural systems are tendential and
are best expressed in normic statements or in
counter-factual propositions.](https://image.slidesharecdn.com/11843150-250520101544-6744d3fd/85/The-Sage-Handbook-Of-Casebased-Methods-David-Byrne-And-Charles-C-Ragin-47-320.jpg)
![26 THE SAGE HANDBOOK OF CASE-BASED METHODS
and its hierarchical ordering is summarized by
the Nobel Laureate P.W. Anderson:
The ability to reduce everything to simple funda-
mental laws does not imply the ability to start from
those laws and reconstruct the universe. In fact,
the more the elementary particle physicists tell us
about the nature of the fundamental laws, the
less relevance they seem to have to the very real
problems of the rest of science, much less to those
of society.
The constructionist hypothesis breaks down
when confronted with the twin difficulties of scale
and complexity. The behavior of large and complex
aggregates of elementary particles, it turns out,
is not to be understood in terms of a simple
extrapolation of the properties of a few particles.
Instead, at each level of complexity entirely new
properties appear, and the understanding of the
new behaviors requires research which I think is
as fundamental in its nature as any other. That is,
it seems to me, that one may array the sciences
roughly linearly in a hierarchy …
But this hierarchy does not imply that science X
is ‘just applied Y’. At each stage entirely new laws,
concepts and generalizations are necessary, …
[thereby] … requiring inspirations and creativity
to just as great a degree as in the previous one.
Psychology is not applied biology, nor is biology
applied chemistry. (Anderson 1972, p. 393)
Using this image of the world as an open,
evolving system of ontologically stratified
entities, complex realism now moves to
distinguish between: (1) what scientists say
about the world; and (2) the world as
it exists, i.e. the constellation of obdurate
entities and dynamic mechanisms that stand
over and against human accounts of them.
Bhaskar labels these domains the tran-
sitive and the intransitive dimensions of
science.
There is a close correspondence between
the intransitive domain and the ontological
structure of science’s investigative object and
the transitive domain’s focus on scientific
method and epistemology. Of the two, the
intransitive dimension enjoys an autonomy
not allotted to the transitive domain. This
is because intransitivity is the fundamen-
tal precondition upon which the transitive
production of scientific knowledge is pred-
icated. Intransitivity thus provides scientific
understanding not only with its ontological
object of investigation, but also with its raison
d’être.
The transitive dimension, by contrast con-
sists of the historically shifting modes of
scientific understanding, viz. its theories, its
craft-based esoterica, and its technologies of
discovery. Unlike its intransitive counterpart,
science’s transitive domain is neither immune
to the distorting effects of ruling ideologies
nor free of the technological limitations
imposed upon human understanding at any
given historical moment. It is, instead, a
collectively constructed instrument of social
action that bears the material imprint of
an epoch’s social division of labor and its
intellectual mode of production.
This transitivity of scientific praxis resides
in the paramount fact that scientific problem-
atics are conditioned by the communities that
certify the investigator’s credentials and func-
tions. The productive principles of science
and its investigative ‘tricks of the trade’ are
embedded in the folklore of the profession.
Acting as a court of last resort, this community
of adepts establishes the esoteric significance
of individual programs and discoveries.
Hence, while this ‘collegium’ cannot create
science’s intransitive object, it can control
the means of legitimating the expenditure of
society’s collective labor and resources on one
project as opposed to another.
The scientific ontology of
complex systems
Critical realism’s deduction of the world’s
stratified ontology and its distinction between
the transitive and intransitive domains of
science form a suitable launching point for
discussing the social ontology of complex
social formations. What is needed, though,
is a scientific ontology that bridges the two.
Complexity theory is just such a vehicle: it is a
general systems theory, i.e. a science of open-
ended, peripatetic entities capable of material
and structural evolution (Bertalanffy 1969,
pp. 30–53, 120–154).
Complexity’s ontological foundations
beginwithRosen’sdistinctionbetween simple
systems and complex systems (Rosen 1991,](https://image.slidesharecdn.com/11843150-250520101544-6744d3fd/85/The-Sage-Handbook-Of-Casebased-Methods-David-Byrne-And-Charles-C-Ragin-49-320.jpg)
![COMPLEXITY AND CASE 31
Socialization Reproduction/
transformation
Society
Individuals
Figure 1.3 The transformational model of social action (TMSA).
a dialectically framed naturalism by sublating
the contradictory presuppositions of each
stereotype into an integrated sociological
whole. The results of his reconnaissance
are shown in Figure 1.3. The Durkheimian
stereotype consists of the triad Society →
Socialization Processes → Individual. In
accordance with Durkheimian positivism,
the downward-directed arrows represent
the coercive power of Durkheimian ‘social
facts’, which morally constrain the destructive
‘appetites’ of atomized individuals during
periods of social and cultural breakdown.
The one-sided, Weberian stereotype reads,
Individuals → Production/Transformation
→ Society; it is located on the right
side of Figure 1.3. The inverse flow of the
Weberian causal vectors signifies the weight
Weberian action theory places on human
subjectivity and its cultural productions. It
should be noted in passing that the TMSA’s
opposed triads and their dispositional flows
conceptually parallel or overlap several of
the dualisms already discussed in this paper,
viz. nomothetic versus ideographic modes of
enquiry, positivist versus social constructivist
accounts of social life, and quantitative versus
qualitative methodologies.
At this point, Peter Berger’s duality of
‘Society creates Man’ versus ‘Man creates
Society’can be added (Berger and Luckmann
1966, Bhaskar 1979/1989, pp. 31–33).
Bhaskar uses Berger’s unmediated dualism
as a foil for completing his dialectical
conception of society. Within the context of
the TMSA, man no more ‘creates’ society
than society ‘creates’ man. Instead, both are
ontologically irreducible to one another. As
different as their respective modus vivendi are,
their powers are complementary, so much so
that they form the necessary preconditions
for each other’s existence. Hence, humans
are intentional agents whose self-directed
activities:
… reproduce or transform [society]. That is, if
society is always already made, then any concrete
human praxis, or, if you like, act of objectivation
can only modify it; and the totality of such acts
sustain and change it. It is not the product of their
activity (any more, I shall argue, than human action
is completely determined by it). Society stands to
individuals, then, as something that they never
make, but that exists only in virtue of their activity.
(Bhaskar 1979/1989, p. 34)
By contrast, society, for all its power to
induce social conformity, is at best an abstract
potentiality. Its very real but slumbering
powers remain virtually inert without the
productive intercession of human agency,
individual or collective. Hence:
Society, then, is an articulated ensemble of tenden-
cies and powers which, unlike natural ones, exist
only as long as they (or at least some of them) are](https://image.slidesharecdn.com/11843150-250520101544-6744d3fd/85/The-Sage-Handbook-Of-Casebased-Methods-David-Byrne-And-Charles-C-Ragin-54-320.jpg)
![withdraw the troops and most of the public property, and to use
them, with the ordnance and ordnance stores thus saved, in
providing for the defense of Vicksburg, but especially it deprived the
enemy of any pretext for bombarding the town and sacrificing the
lives of the women and children. It appears that General Lovell
called for ten thousand volunteers from the citizens, but failed to get
them. There were many river-steamboats at the landing, and, if the
volunteers called for were intended to man these boats and board
the enemy's fleet before their land-forces could arrive, it can not be
regarded as utterly impracticable. The report of General Butler
shows that he worked his way through one of the bayous in rear of
Fort St. Philip to the Mississippi River above the forts so as to put
himself in communication with the fleet at the city, and to furnish
Commodore Farragut with ammunition. From this it is to be inferred
that the fleet was deficient in ammunition, and the fact would have
rendered boarding from river-boats the more likely to succeed. In
this connection it may be remembered that, during the war, John
Taylor Wood, Colonel and A. D. C. to the President, who had been an
officer of high repute in the old Navy, did in open boats attack
armed vessels, board and capture them, though found with nettings
up, having been warned of the probability of such an attack.[57]
Many causes have been assigned for the fall of New Orleans. Two
of them are of undeniable force: First, the failure to light up the
channel; second, the want of an obstruction which would detain the
fleet under fire of the forts. General Duncan's report and testimony
justify the conclusion that to the thick veil of darkness the enemy
was indebted for his ability to run past the forts.
The argument that the guns were not of sufficiently large caliber
to stop the fleet is not convincing. If all the guns had been of the](https://image.slidesharecdn.com/11843150-250520101544-6744d3fd/85/The-Sage-Handbook-Of-Casebased-Methods-David-Byrne-And-Charles-C-Ragin-65-320.jpg)
![dashed alongside, cut our way through and over the boarder
nettings with the old navy cutlass, gained the deck, and, after a
sharp, short fight, drove the enemy below. The prizes proved to be
the gunboats Satellite and Reliance, two guns each. Landing the
prisoners, we cruised for two days in the Chesapeake Bay. A number
of vessels were captured and destroyed.]](https://image.slidesharecdn.com/11843150-250520101544-6744d3fd/85/The-Sage-Handbook-Of-Casebased-Methods-David-Byrne-And-Charles-C-Ragin-74-320.jpg)
The Sage Handbook Of Casebased Methods David Byrne And Charles C Ragin
- 1. The Sage Handbook Of Casebased Methods David Byrne And Charles C Ragin download https://ebookbell.com/product/the-sage-handbook-of-casebased- methods-david-byrne-and-charles-c-ragin-23686300 Explore and download more ebooks at ebookbell.com
- 2. Here are some recommended products that we believe you will be interested in. You can click the link to download. The Sage Handbook Of Health Care Ethics Core And Emerging Issues Ruth F Chadwick https://ebookbell.com/product/the-sage-handbook-of-health-care-ethics- core-and-emerging-issues-ruth-f-chadwick-9949814 The Sage Handbook Of Social Media Research Methods Anabel Quanhaase https://ebookbell.com/product/the-sage-handbook-of-social-media- research-methods-anabel-quanhaase-46151554 The Sage Handbook Of Drug Alcohol Studies Social Science Approaches Torsten Kolind https://ebookbell.com/product/the-sage-handbook-of-drug-alcohol- studies-social-science-approaches-torsten-kolind-46273878 The Sage Handbook Of The Digital Media Economy Terry Flew Jennifer Holt https://ebookbell.com/product/the-sage-handbook-of-the-digital-media- economy-terry-flew-jennifer-holt-46867442
- 3. The Sage Handbook Of Learning And Work Margaret Malloch Len Cairns https://ebookbell.com/product/the-sage-handbook-of-learning-and-work- margaret-malloch-len-cairns-47205188 The Sage Handbook Of Digital Society William Housley Adam Edwards https://ebookbell.com/product/the-sage-handbook-of-digital-society- william-housley-adam-edwards-47539954 The Sage Handbook Of Graduate Employability 1st Edition Tania Broadley https://ebookbell.com/product/the-sage-handbook-of-graduate- employability-1st-edition-tania-broadley-48700828 The Sage Handbook Of Leadership 2nd Edition Doris Schedlitzki https://ebookbell.com/product/the-sage-handbook-of-leadership-2nd- edition-doris-schedlitzki-49117166 The Sage Handbook Of Qualitative Research In The Asian Context 1st Edition Safary Wambaleka https://ebookbell.com/product/the-sage-handbook-of-qualitative- research-in-the-asian-context-1st-edition-safary-wambaleka-50586588
- 6. The SAGE Handbook of Case-Based Methods
- 8. The SAGE Handbook of Case-Based Methods Edited by David Byrne and Charles C. Ragin
- 9. Introduction and editorial arrangement © David Byrne and Charles C. Ragin 2009 Chapter 1 © David L. Harvey 2009 Chapter 2 © Lars Mjøset 2009 Chapter 3 © Bob Carter and Alison Sealey 2009 Chapter 4 © Malcolm Williams and Wendy Dyer 2009 Chapter 5 © David Byrne 2009 Chapter 6 © Colin Elman 2009 Chapter 7 © Emma Uprichard 2009 Chapter 8 © Dianne Phillips and John Phillips 2009 Chapter 9 © Emma Whelan 2009 Chapter 10 © Ray Kent 2009 Chapter 11 © Gisèle De Meur and Alain Gottcheiner 2009 Chapter 12 © Benoît Rihoux and Bojana Lobe 2009 Chapter 13 © Ronald L. Breiger 2009 Chapter 14 © David Byrne 2009 Chapter 15 © Nigel Fielding and Richard Warnes 2009 Chapter 16 © Seán Ó Riain 2009 Chapter 17 © Gary Goertz and James Mahoney 2009 Chapter 18 © Nick Emmel and Kahryn Hughes 2009 Chapter 19 © Fred Carden 2009 Chapter 20 © Edwin Amenta 2009 Chapter 21© John Walton 2009 Chapter 22 © Albert J. Bergesen 2009 Chapter 23 © Paul Downward and Joseph Riordan 2009 Chapter 24 © James Mahoney and P. Larkin Terrie 2009 Chapter 25 © Peer C. Fiss 2009 Chapter 26 © Frances Griffiths 2009 Chapter 27 © Sue Dopson, Ewan Ferlie, Louise Fitzgerald and Louise Locock 2009 Chapter 28 © Philippa Bevan 2009 Chapter 29 © Wendy Olsen 2009 Chapter 30 © David Byrne, Wendy Olsen and Sandra Duggan 2009 Chapter 31 © Charles C.Ragin 2009 First published 2009 Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, this publication may be reproduced, stored or transmitted in any form, or by any means, only with the prior permission in writing of the publishers, or in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency. Enquiries concerning reproduction outside those terms should be sent to the publishers. SAGE Publications Ltd 1 Oliver’s Yard 55 City Road London EC1Y 1SP SAGE Publications Inc. 2455 Teller Road Thousand Oaks, California 91320 SAGE Publications India Pvt Ltd B 1/I 1 Mohan Cooperative Industrial Area Mathura Road New Delhi 110 044 SAGE Publications Asia-Pacific Pte Ltd 33 Pekin Street #02-01 Far East Square Singapore 048763 Library of Congress Control Number: 2008931584 British Library Cataloguing in Publication data A catalogue record for this book is available from the British Library ISBN 978-1-4129-3051-2 Typeset by CEPHA Imaging Pvt. Ltd., Bangalore, India Printed in India at Replika Press Pvt Ltd Printed on paper from sustainable resources
- 10. Contents Acknowledgements viii Notes on Contributors ix INTRODUCTION Case-Based Methods: Why We Need Them; What They Are; How to Do Them 1 David Byrne PART ONE: THE METHODOLOGICAL CONTEXT OF CASE-BASED METHODS 11 1. Complexity and Case 15 David L. Harvey 2. The Contextualist Approach to Social Science Methodology 39 Lars Mjøset 3. Reflexivity, Realism and the Process of Casing 69 Bob Carter and Alison Sealey 4. Single-Case Probabilities 84 Malcolm Williams and Wendy Dyer 5. Complex Realist and Configurational Approaches to Cases: A Radical Synthesis 101 David Byrne PARTTWO: METHODS AND TECHNIQUES OF CASE-BASED RESEARCH 113 Typologies – Ways of Sorting Things Out 120 6. Explanatory Typologies in Qualitative Analysis 121 Colin Elman 7. Introducing Cluster Analysis: What Can It Teach Us about the Case? 132 Emma Uprichard 8. Visualising Types: The Potential of Correspondence Analysis 148 Dianne Phillips and John Phillips
- 11. vi CONTENTS 9. How Classification Works, or Doesn’t: The Case of Chronic Pain 169 Emma Whelan Quantitative Approaches to Case-Based Method 183 10. Case-Centred Methods and Quantitative Analysis 184 Ray Kent 11. The Logic and Assumptions of MDSO–MSDO Designs 208 Gisèle De Meur and Alain Gottcheiner 12. The Case for Qualitative Comparative Analysis (QCA): Adding Leverage for Thick Cross-Case Comparison 222 Benoît Rihoux and Bojana Lobe 13. On the Duality of Cases and Variables: Correspondence Analysis (CA) and Qualitative Comparative Analysis (QCA) 243 Ronald L. Breiger 14. Using Cluster Analysis, Qualitative Comparative Analysis and NVivo in Relation to the Establishment of Causal Configurations with Pre-existing Large-N Datasets: Machining Hermeneutics 260 David Byrne Qualitative Approaches to Case-Based Research 269 15. Computer-Based Qualitative Methods in Case Study Research 270 Nigel Fielding and Richard Warnes 16. Extending the Ethnographic Case Study 289 Seán Ó Riain 17. Scope in Case Study Research 307 Gary Goertz and James Mahoney 18. Small-N Access Cases to Refine Theories of Social Exclusion and Access to Socially Excluded Individuals and Groups 318 Nick Emmel and Kahryn Hughes 19. Using Comparative Data: A Systems Approach to a Multiple Case Study 331 Fred Carden PART THREE: CASE-BASED METHODS IN DISCIPLINES AND FIELDS 345 20. Making the Most of an Historical Case Study: Configuration, Sequence, Casing, and the US Old-age Pension Movement 351 Edwin Amenta 21. Poetry and History: The Case for Literary Evidence 367 John Walton
- 12. CONTENTS vii 22. Cultural Case Studies 383 Albert J. Bergesen 23. Social Interactions and the Demand for Sport: Cluster Analysis in Economics 392 Paul Downward and Joseph Riordan 24. The Proper Relationship of Comparative-Historical Analysis to Statistical Analysis: Subordination, Integration, or Separation? 410 James Mahoney and P. Larkin Terrie 25. Case Studies and the Configurational Analysis of Organizational Phenomena 424 Peer C. Fiss 26. The Case in Medicine 441 Frances Griffiths 27. Team-Based Aggregation of Qualitative Case Study Data in Health Care Contexts: Challenges and Learning 454 Sue Dopson, Ewan Ferlie, Louise Fitzgerald and Louise Locock 28. Working with Cases in Development Contexts: Some Insights from an Outlier 467 Philippa Bevan 29. Non-nested and Nested Cases in a Socioeconomic Village Study 494 Wendy Olsen 30. Causality and Interpretation in Qualitative Policy-Related Research 511 David Byrne, Wendy Olsen and Sandra Duggan 31. Reflections on Casing and Case-Oriented Research 522 Charles C. Ragin Author Index 535 Subject Index 539
- 13. Acknowledgements All intellectual projects are the resultant of the work of lots of individuals and indeed often of many institutions. That is particularly true of a collective effort like this Handbook. Many people have helped us in many ways. Some of them are contributors here, some are not. First, to institutions. The UK’s Economic and Social Research Council (ESRC) funded a seminar series ‘Realising the potential: realist social theory and empirical research’ in 1999–2001 organized by Caroline New of Bath Spa University and Bob Carter of the University of Warwick which played an important role in developing methodological arguments for many of our UK contributors. Second, the British Academy funded a very useful workshop organized by Wendy Olsen ‘Quantitative Social Science – Challenges and Possibilities’ in September 2004 which enabled us to bring Charles Ragin to the UK and start the dialogue which resulted in this Handbook. In 2004–2005, the ESRC under its Research Methods programme funded a series of workshops/research events under the title ‘Focusing on the Case’ which enabled us to develop the issues raised in the original BritishAcademy event. Several of the chapters in this Handbook have been written by participants in this series and the chapter on ‘Causality and Interpretation in Policy Related Research’ in Section Three is a direct product of them. We would like to acknowledge the work of Sandra Duggan and Linda Campbell at Durham in organizing these workshops and of Angela Dale, Director of the ESRC Research Methods programme and Ruth Durrell, the administrator for that programme, all of whom were always supportive and helpful in making that series work as well as it did. Wendy Olsen was the co-director of that series and she has always been a source of intellectual stimulation as well as a colleague whose administrative competence more than compensates for David Byrne’s deficiencies in that respect. The ESRC Research Methods festivals in Oxford, originally organized by Angela and Ruth, provided us with a venue for developing arguments, locating more contributors, and generally have served as a fantastic base for the development of methodology and methods in recent years. They will be an important element in the history of methodological progress in the social sciences. The suggestion that we should produce this book was originally made by Chris Rojek at SAGE and he has always been supportive and helpful. Jai Seaman at SAGE held our hands through the process of assembling chapters and calmed us down when things were a bit sticky – and to the credit of our contributors that did not happen very often. She has been an exemplary editor and an enormous help. At the personal level our families provided us with support, criticism and a sense of balance so we are grateful to them. Not only did they provide important reality checks, they also knew not to call in the medics when this project left us either staring blindly into space or working maniacally through a new argument or idea. The completion of this project provides testimony to their patience and good humour.
- 14. Notes on Contributors Edwin Amenta is Professor of Sociology and History at the University of California, Irvine. He is the author, most recently, of When Movements Matter: The Townsend Plan and the Rise of Social Security (Princeton NJ: Princeton University Press. 2006) and Professor Baseball: Searching for Redemption and the Perfect Lineup on the Softball Diamonds of Central Park (Chicago: University of Chicago Press, 2007). Albert J. Bergesen is Professor and Head of the Department of Sociology at the University of Arizona. His research on culture includes studies of popular culture (The Depth of Shallow Culture: The High Art of Shoes, Movies, Novels, Monsters and Toys. Boulder: Paradigm Publishers, 2006; paper 2007) and (God in the Movies. New Brunswick, NJ: Transaction Publications. With Andrew M. Greeley. (2000); paperback edition: Transaction Publications, 2003). His work on cultural theory includes (Cultural Analysis: The Work of Peter Berger, Mary Douglas, Michel Foucault, and Jurgen Habermas. With Robert Wuthnow, James D. Hunter and Edith Kurtzweil. London: Routledge and Kegan Paul Ltd., 1984), and more recent efforts to generate a transformational grammar of art styles can be found in, ‘Culture and Cognition’. pp. 35–47, in M. Jacobs and N. Hanrahan (Eds.) The Blackwell Companion to the Sociology of Culture. New York: Basil Blackwell, 2005; and, ‘A Linguistic Model of Art History’. Poetics, 2000, 28, 73–90. Philippa Bevan is an independent researcher in the sociology of development who has worked at the universities of Oxford and Bath. Between 2002 and 2006 she was a member of the Bath-based Research Group ‘Wellbeing in Developing Countries’ (WeD). Her main role was as Country Co-ordinator of the Ethiopia WeD study (see www.wed-ethiopia.org) and she is currently engaged in writing two books out of that experience; one on power and life quality in Ethiopia and the other on the complexity science methodology which she developed during the research programme. She and a colleague have also been using the WeD data to produce commissioned policy-related research papers and briefings on Ethiopia. Ronald L. Breiger is Professor of Sociology and a member of the Graduate Interdisciplinary Program in Statistics at the University of Arizona. He is currently Chair-Elect of the Section on Mathematical Sociology of the American Sociological Association. His interests include social networks, stratification, mathematical models, theory and measurement issues in cultural and institutional analysis. David Byrne is Professor of Sociology and Social Policy at Durham University. Previously he has worked as both an academic and in community development. His books include Complexity Theory and the Social Sciences 1998 and Interpreting Quantitative Data 2002. His major
- 15. x NOTES ON CONTRIBUTORS empirical research interest is in the consequences of deindustrialization for places and people and in the nature of social exclusion on a global basis. He is also keenly interested in the development of methodological perspectives and methods as tools which can be used in the application of social science to real social problems and sees case-based methods as having enormous potential in this context. Fred Carden is Director of Evaluation at the International Development Research Centre (Canada). He holds a PhD from the University of Montreal and was Research Fellow in Sustainability Science at Harvard University (2007–2008). He has written in the areas of evaluation, international cooperation and environmental management, and has co- published Outcome Mapping, Enhancing Organizational Performance, among others. His most recent publication is Knowledge to Policy: Making the Most of Development Research, (London: Sage, 2009). Bob Carter is Associate Professor in Sociology, University of Warwick, UK. He has published widely on the politics of racism and immigration and on social theory and sociolinguistics. He is the author of Realism and Racism: Concepts of Race in Sociological Research (New York: Routledge, 2000) and the co-author (withAlison Sealey) ofApplied Linguistics as Social Science (London: Continuum, 2004). He also co-edited (with Caroline New) Making Realism Work (New York: Routledge, 2004). Gisèle De Meur (BELGIUM) is Professor of Mathematics at the Université Libre de Bruxelles. Her research theme began with pure geometry, then mathematical anthropology and mathematics related to political science (electoral systems, theory of democracy, gender studies…). Her more recent works concern methodology of the small-N studies, and in particular ‘qulali-quantitative comparative analysis. Sue Dopson is Rhodes Trust Professor of Organisational Behaviour at Oxford University, and a Governing Body Fellow of Green Templeton College, formerly Vice-President of Templeton College. She is also Director of Research Degrees at the Oxford Said Business School. She acts as an advisor, consultant, referee and editorial board member in a variety of settings. Her research involvements lie principally in the area of health care management, built over the past 18 years. This has included the evaluation of government policy, and reviews of managerial change and organizational interventions. She has published several books and articles on changes in the management of the NHS, the changing nature of middle management, management careers and developments in public sector management. Sue worked as a Personnel Manager in the NHS before pursuing a research and academic career at the University. Paul Downward is Director of the Institute of Sport and Leisure Policy, University of Loughborough. Along with Alistair Dawson, he is the author of the textbook The Economics of Professional Team Sports (Routledge, 2000) and the author of a forthcoming book The Economics of Sport: Theory, Evidence and Policy with Alistair Dawson and Trudo Dejonghe (Butterworth Heinemann). He is the editor of a book for the Office for National Statistics on the Use of Official Data in understanding Sports, Leisure and Tourism markets and the author of numerous articles on the economics of sport. He has recently undertaken consultancy work for Sport England on participation, volunteering in sports clubs, and performance management; and UK Sport on Volunteering at the Manchester Commonwealth Games. He is a member of the International Association of Sports Economics, a founding member of the editorial board of the Journal of Sports Economics, and member of the British Philosophy of
- 16. NOTES ON CONTRIBUTORS xi Sport Association. Paul is currently playing veteran’s rugby and coaches a junior side at his local club. Sandra Duggan has experience in a wide variety of research areas with an underlying theme of effectiveness i.e ‘What works?’. She was involved in the ESRC-funded project, Focusing on the Case in Quantitative and Qualitative Research, with Dave Byrne and Wendy Olsen. The project involved a series of workshops aimed at helping researchers to develop their thinking about case-based methods using a dialogical learning approach. Sandra is now working at the University of Teesside. Wendy Dyer is Senior Lecturer in Criminology at Northumbria University. She worked with the Cleveland Diversion Team for Mentally Disordered Offenders as an Action Research Worker until 1997. She was awarded ESRC funding for her PhD at the University of Durham, Department of Sociology and Social Policy (1997–2001). Following completion of her PhD Wendy worked for two years on a project with the Centre for Public Mental Health and in 2004 she was awarded funding by the NHS National R&D Programme on Forensic Mental Health (along with Professor John Carpenter). She was then appointed Senior Research Associate, Department of Psychiatry, University of Oxford working on a project to evaluate the Dangerous and Severe Personality Disorder (DSPD) programme. In 2005 Wendy, along with two colleagues from HMP Durham and the Care Services Improvement Partnership (CSIP), set up the Prison and Offender Research in Social Care and Health – North East, Yorkshire and Humberside network (PORSCH-NEYH). The network has been awarded funding by the Home Office and Department of Health. She acts as coordinator for the networks activities. Wendy was appointed as Senior Lecturer in Criminology at Northumbria University in January 2008. Colin Elman is Associate Professor of Political Science in the Maxwell School of Syracuse University. He is currently engaged on a book project investigatingAmerica’s rise to dominance in the Western Hemisphere. He is also working on other book and article projects on realist international relations theory, qualitative methods and the diffusion of unconventional conflict practices. Elman is (with Miriam Fendius Elman) the co-editor of Progress in International Relations Theory: Appraising the Field (MIT Press, 2003); and Bridges and Boundaries: Historians, Political Scientists, and the Study of International Relations (MIT Press, 2001); and (with John Vasquez) of Realism and the Balancing of Power: A New Debate (Prentice Hall, 2003). Elman has published articles in International Studies Quarterly, International History Review, American Political Science Review, International Organization, International Security and Security Studies. He is a co-founder and Secretary-Treasurer of both the International History and Politics and the Qualitative and Multi-method Research organized sections of the American Political Science Association, and a co-founder and Executive Director of the Consortium for Qualitative Research Methods. Nick Emmel is a Senior Lecturer in the Sociology and Social Policy of Health and Development at the University of Leeds. His research interests include inequalities in health, social exclusion and poverty. Recent methodological research includes identifying methodologies for accessing socially excluded individuals and groups and Connected Lives, a qualitatively driven, mixed- method investigation of networks, neighbourhoods and communities. Currently he is using qualitative longitudinal methods to investigate the role of grandparents in supporting their grandchildren out of social exclusion in low-income communities. In all of these research projects a case-based methodology is being used to manage data and facilitate theorization and abstraction within and between cases.
- 17. xii NOTES ON CONTRIBUTORS Ewan Ferlie is Professor of Public Services Management and Head of the Department of Management at King’s College London. He has published widely in the field of organizational studies in health care, and has been co-editor of the Oxford Handbook of Public Management. Much of his work uses the comparative case method, including a co authored paper in Academy of Management Journal (best paper award, 2005). Nigel Fielding is Professor of Sociology and Associate Dean of Arts and Human Sciences at the University of Surrey. With Ray Lee, he co-directs the CAQDAS Networking Project, which provides training and support in the use of computers in qualitative data analysis, and which has latterly become a ‘node’ of the ESRC National Centre for Research Methods. His researchinterestsareinnewtechnologiesforsocialresearch,qualitativeresearchmethods,mixed method research design and criminology. He has authored or edited 20 books, over 50 journal articles and over 200 other publications. In research methodology his books include a study of methodological integration (Linking Data, 1986, Sage; with Jane Fielding), an influential book on qualitative software (Using Computers in Qualitative Research, 1991, Sage; editor and contributor, with Ray Lee), a study of the role of computer technology in qualitative research (Computer Analysis and Qualitative Research, 1998, Sage; with Ray Lee), a four- volume set, Interviewing (2002, Sage; editor and contributor), the Sage Handbook of Online Research Methods (2008; editor and contributor) and a further four-volume set on Interviewing (forthcoming). He is presently researching the application of high performance computing applications to qualitative methods. Peer C. Fiss is anAssistant Professor of Strategy at the Marshall School of Business, University of Southern California. He works in the areas of corporate governance, framing, the spread and adaptation of corporate practices and the use of set-theoretic methods, particularly in management and strategy. His research has appeared in journals such as the Academy of Management Journal, Academy of Management Review, Administrative Science Quarterly, American Sociological Review and Strategic Management Journal. He is the recipient of the 2008 Marshall Dean’s Award for Research and serves on the editorial boards of the Academy of Management Journal, Journal of International Business Studies and Business Research. Peer received his PhD jointly from the Departments of Sociology and Management & Organization at Northwestern University. LouiseFitzgerald isProfessorofOrganizationDevelopment,DepartmentofHRM,DeMontfort University. Her research centres on the management of innovation and change in professional, complex organizations and the restructuring of professional roles. She is currently working on a major, empirical project on the management of organizational networks in health care. She has published in a range of journals, such as Academy of Management Journal and Human Relations,andhasco-authoredanumberofbooks,suchasKnowledgetoAction?Evidence-Based Health Care in Context, (Oxford University Press, 2005) and The Sustainability and Spread of Organizational Change, (Routledge, 2007). She has previously worked at City University Warwick University, and Salford University. Gary Goertz teaches Political Science at the University ofArizona. He is the author or co-author of eight books and over 50 articles and chapters on issues of international conflict, institutions and methodology. His books include Social Science Concepts: A User’s Guide (Princeton 2006) Contexts of International Politics (1994), International Norms and Decision Making: A Punctuated Equilibrium Model (2003) and with Paul Diehl War and Peace in International Rivalry (2000). He is currently working on a book on the evolution of regional economic institutions and their involvement in conflict management.
- 18. NOTES ON CONTRIBUTORS xiii Alain Gottcheiner is Assistant Lecturer in Applied Mathematics and Computer Science at the Université Libre de Bruxelles, Brussels, Belgium. His main areas of interest are mathematical models as applied to decision-making, to human sciences and to linguistics. He has also studied the mathematical methods (operational research, combinatorial mathematics, groups) and computer science methods which allow the data modeling in administrative sciences, social anthropology, cultural anthropology, comparative linguistics, and phonology, with particular emphasis on the transfer of these methods to non-mathematicians. Alain has reviewed current methods of research in mathematics, physics, astronomy and cosmologony, linguistics, etc. all in light of theories of proof of Popper and Lakatos, and positivists’ theories of truth. Frances Griffiths trained in medicine at University of Cambridge and Kings College Hospital, London and went on to become a general practitioner in Stockton-on-Tees. While working as a GP she undertook her PhD at the University of Durham, Department of Sociology and Social Policy and was a founder member of the Northern Primary Care Research Network. Frances joined the University of Warwick in 1998 and became founding Director of the Warwick West- Midlands Primary Care Research Network and developed her research interest in the impact of technology on perceptions of health. In 2003 she was awarded a Department of Health National Career Scientist Award for developing a programme of research on Complexity and Health. Frances undertakes her research with interdisciplinary research teams both within the University and internationally. Although her research mainly links social science and medicine, her research programme related to complexity now includes research collaboration with physical sciences and the humanities. David L. Harvey is Professor Emeritus of Sociology at the University of Nevada, Reno. He has written on such topics as: poverty and the subculture of poverty; social alienation and the socialpsychologicalmeasurementofreifiedconsciousness;andCriticalMarxistsociology.More recently, he has attempted to apply the Critical Realist perspective and the Chaos/Complexity paradigm of non-equilibrium systems to sociological subject matters. His essay in this Handbook reflects these dual interests as they pertain to the understanding of qualitative case studies. Finally, Professor Harvey is currently completing a social history of a poor, Southern–white community located in the American Midwest. It is entitled Potter Addition: The Social History of a ‘Hillbilly Slum’. When completed, it will be a companion volume to an earlier ethnographic investigation, Potter Addition: Poverty, Family and Kinship in a Heartland Community. Kahryn Hughes is a Senior Research Fellow, and Co-Director of Families Lifecourse and Generations Research Centre (FlaG), at the University of Leeds. Her main theoretical interests are theorizing time, space and relational identity constitution. Thematically, she is interested in exploring inter-generational experiences and meanings of social exclusion; practices of addiction; and theorizing social networks in the context of low-income communities. Ray Kent is currently Senior Lecturer in the Department of Marketing at the University of Stirling. He has published several books in sociology, media audience measurement, data analysis and marketing research. He is currently exploring alternative methods of data analysis including the use of fuzzy-set analysis, chaos theory and neural network analysis. Bojana Lobe completed her PhD in Social Sciences Methodology at the University of Ljubljana. She is a researcher at Centre for Methodology and Informatics, and teaching methodological courses at Faculty of Social Sciences. She is a member of editorial board of International Journal ofMultipleResearchApproaches.SheisinvolvedinEuKidsOnlineproject.Hercurrentresearch activities and interests are: methodological issues in mixed methods research, new technologies
- 19. xiv NOTES ON CONTRIBUTORS in social science data collection (virtual ethnography, online focus groups, online qualitative interviews), systematic comparative methods, methodological aspects of researching children’s experience. Louise Locock is a Qualitative Researcher interested in personal experiences of health and illness, evidence-based medicine, quality improvement and organizational change in health care. Since 2003 she has been a senior researcher with the DIPEx Health Experiences Research Group at the University of Oxford. James Mahoney (PhD 1997, University of California, Berkeley) is a Professor of political science and sociology at Northwestern University. He is the author ofThe Legacies of Liberalism: Path Dependence and Political Regimes in Central America (Johns Hopkins University Press, 2001) and co-editor of Comparative Historical Analysis in the Social Sciences (Cambridge University Press, 2003). His work also includes articles on political and socioeconomic development in Latin America, path dependence in historical sociology and causal inference in small-N analysis. Mahoney’s most recent book is Colonialism and Development: Spanish America in Comparative Perspective (Cambridge University Press, forthcoming). Lars Mjøset (b. 1954) is Professor at the Department of Sociology and Human Geography, and Director of The Oslo Summer School for Comparative Social Science Studies, both at the University of Oslo, Norway. His main fields of interest are the philosophy of the social sciences and comparative political economy with special reference to the Nordic countries and the integration of small countries in the world economy. Besides books in Norwegian, he published The Irish Economy in a Comparative Institutional Perspective (Dublin: NESC 1992), and recently co-edited and contributed to two special issues of Comparative Social Research: Vol. 20 with Stephen van Holde, The Comparative Study of Conscription in the Armed Forces (Amsterdam: Elsevier Science, 2002), and Vol. 24 with Tommy H. Clausen, Capitalisms Compared (Amsterdam, London, New York: Elsevier Science, 2007). Wendy Olsen works as a Senior Lecturer in Socioeconomic Research at University of Manchester. Her research covers UK labour markets, global economic development and research methods. Her specialist areas include the regulation of labour markets; feminist analyses of labour relations; gendered labour force participation in India; global patterns of gendered human development; and methodological pluralism (especially across the qualitative–quantitative ‘divide’). Her publications include Rural Indian Social Relations (Delhi: Oxford University Press, 1996); The Politics of Money (2002); Modelling Gender Pay Gaps (with S. Walby, Equal Opportunities Commission, 2004). She is currently working on fuzzy-set causal analysis, causes of gender pay gaps, habitus and moral reasoning strategies about economic policy. Seán Ó Riain is Professor of Sociology at the National University of Ireland, Maynooth. He has carried out a range of case study research, including ethnographic studies of software workplaces and transnational studies of high technology growth and development. Publications include Global Ethnography (2000, co-edited) and The Politics of High Tech Growth (2004). Current projects include a study of Silicon Valley and Ireland as a case study of a transnational system of work and employment and a life history study of social change in twentieth century Ireland that mixes qualitative and quantitative methods. Dianne Phillips (b. 1940), BA (Econ) MA (Econ), and PhD, in Sociology at the University of Manchester. Postgraduate diploma in Mathematical Statistics, University of Manchester. Former Principal Lecturer in Social Research Methodology and Social Statistics in the Faculty of
- 20. NOTES ON CONTRIBUTORS xv Humanities and Social Science at the Manchester Metropolitan University in the UK. Currently independent consultant, providing support for social research and evaluation. In addition to commissioned research and evaluation reports, publications include contributions to a number of books on methodological topics and to other social research oriented publications. Also an occasional reviewer for Sociology, Social Science Computer Review, Sociological Research Online and the Royal Statistical Society (Social Statistics). Current interests in evaluation methodologies and ‘measurement’ of soft outcomes. John Phillips (b. 1939), Degrees Oxford, Psychology, Philosophy; London, Economics; Salford, Sociology. Lectured in Social Theory, Manchester Polytechnic, 1969–1992. After retirement, varied consultancy work. Publications in Social Theory. Current interests inArtificial Intelligence, Dialectics and Sociology. Charles C. Ragin holds a joint appointment as Professor of Sociology and Political Science at the University ofArizona. In 2000–2001 he was a Fellow at the Center forAdvanced Study in the Behavioral Sciences at Stanford University, and before that he was Professor of Sociology and Political Science at Northwestern University. His substantive research interests include such topics as the welfare state, ethnic political mobilization and international political economy. However, for the past two decades his work has focused primarily on broad issues in social science methodology, especially the challenge of bringing some of the logic and spirit of small- N case-oriented research to the study of medium-sized and large Ns. Central to this effort is the use of set-theoretic methods of analysis and the treatment of cases of bundles of interpretable combinations of aspects and not simply as raw material for the assessment of the net effects of ‘independent’ variables. His most recent books are Redesigning Social Inquiry: Fuzzy Sets and Beyond, published by University of Chicago Press in 2008 and Configurational Comparative Methods: Qualitative Comparative Analysis and Related Techniques (co-edited with Benoît Rihoux), also published in 2008 (Sage). He is the author of more than 100 articles in research journals and edited books, and he has developed two software packages for set-theoretic analysis of social data, Qualitative Comparative Analysis (QCA) and Fuzzy-Set/Qualitative Comparative Analysis (fsQCA). Benoît Rihoux is Professor of Political Science at the Centre de Politique Comparée of the Université catholique de Louvain (Belgium). His substantive research interests include political parties, new social movements, organizational studies, political change, environmental politics and policy processes. He is Co-ordinator of the COMPASSS international research group (http://www.compasss.org) and oversees the management of its linked Web pages, databases andarchives.Heisalsojointconvenerofinternationalinitiativesaroundmethodsmoregenerally, such as the ECPR Standing Group on Political Methodology, the ECPR Summer School in Methods and Techniques and the ECPR Research Methods book series (Palgrave; co-editor with B. Kittel). He has recently published Innovative Comparative Methods for Policy Analysis: Beyond the Quantitative–Qualitative Divide (Springer/Kluwer, co-editor with H. Grimm, 2006) and Configurational Comparative Methods: Qualitative Comparative Analysis (QCA) and Related Techniques (Sage, co-editor with Charles C. Ragin, 2008). Joseph Riordan is a Senior Lecturer in Economics at the University of Hertfordshire. He is a member of the International Association of Sports Economics (IASE) and the European Association of Sports Management (EASM) and recently presented a paper on the effect of participation in sport on people’s self-reported health at the EASM conference in 2008. He has acted as a referee for academic journals on sports and economics including the Journal of Sports Economics and European Sport Management Quarterly. With Dr Paul Downward, he is currently
- 21. xvi NOTES ON CONTRIBUTORS examining sports participation in the UK using a variety of official datasets. Joe is a keen runner and regularly participates in the London Marathon, despite his weight! Alison Sealey is a Senior Lecturer in Modern English Language, University of Birmingham, UK. She has published extensively on various aspects of sociolinguistics, applied linguistics and discourse analysis. She is the author of Childly Language: Children, Language and the Social World (Longman, 2000) and the co-author (with Bob Carter) of Applied Linguistics as Social Science (Continuum, 2004). P. Larkin Terrie is a Doctoral candidate in Political Science at Northwestern University. His dissertation project focuses on state development in Latin America. His interests also include qualitative and quantitative methodology. Emma Uprichard is a Lecturer in Social Research methods at the Department of Sociology, University ofYork. Her research interests revolve around the methodological challenge involved in studying the complex social world, and include the sociology of children and childhood, cities and urban change, time and space, and complexity theory. Her current research extends her previous doctoral study of cities as complex systems by focusing specifically on the changing patterns of food and eating throughout the life course in York and the UK since the post-war period. John Walton is Research Professor of Sociology at the University of California, Davis. He was Distinguished Professor of Sociology at UCD, 1978–2004, and previously Professor of Sociology and Urban Affairs at Northwestern University, 1966–1978. He has been a fellow at the Center for Urban Affairs at Northwestern University and the Woodrow Wilson International Center for Scholars inWashington, DC. In 2001 he was a LeverhulmeTrust Fellow in the UK. He has held offices in both theAmerican SociologicalAssociation and the International Sociological Association. In 2005 he received the Robert and Helen Lynn Award for Distinguished Career Achievement in Community and Urban Sociology from theAmerican SociologicalAssociation. John is author and editor of twelve books and he has also written some two hundred articles, chapters, essays and reviews in the areas of historical sociology, international development, social movements and collective action, and environmental sociology. Richard Warnes is an Analyst at RAND Europe, Cambridge UK and his interests lie in the fields of Policing and Counter-terrorism. He joined RAND in early 2007 after serving for nine years as an Officer with the Metropolitan Police Service in London and earlier service in the British Army. Between completing his first degree and military service he worked for seven years in international relief and human rights, travelling extensively to Eastern Europe, the Middle East and South East Asia. He is currently a part time PhD candidate with the University of Surrey where he is carrying out doctoral research on counter-terrorist responses in seven countries. This research is supported by both an Airey Neave Trust Fellowship, and a Bramshill (National Police Staff College) Fellowship. He holds an MA in Criminal Justice Studies from Brunel University and a BSc (Honours) in International Politics from the University of Portsmouth. Emma Whelan is an Associate Professor of Sociology in the Department of Sociology and Social Anthropology, Dalhousie University, Halifax, Canada. Her research and publications are concerned with claims-making, standardization and credibility struggles in health, illness and medicine. She has particular interests in pain medicine, endometriosis, expert-lay relations and disease classification.
- 22. NOTES ON CONTRIBUTORS xvii MalcolmWilliams is Professor of Social Research Methodology and Head of the School of Law and Social Science at the University of Plymouth. His main area of interest is methodological issues in social science, in particular probability, causality and objectivity, and he has published extensively in these areas. His empirical research is in the area of housing need, counter- urbanization and household change. Malcolm is the author/editor of six books/edited collections including Philosophical Foundations of Social Research (Sage, four volumes 2006); Making Sense of Social Research (Sage, 2003); Science and Social Science (Routledge, 2000); Knowing the Social World (with Tim May, Open University Press, 1998).
- 24. Introduction Case-Based Methods: Why We Need Them; What They Are; How to Do Them D a v i d B y r n e Science studies cases. It examines – to use one of the more than ten dictionary definitions of the multi-faceted word ‘case’ – instances of a particular situation or set of circumstances. A plethora of methods use the term ‘case’ in some way to describe approach and focus. The purpose of this collection is to present an account of those methods in their range and variety of application. However, and this qualification has the most emphatic of emphases, the focus on cases here is much more than a way of demarcating a set of methods understood merely as techniques of investigation. The ‘turn to the case’ represents a fundamental break with a traditionofexplanationthathasdominatedthe physical sciences, been understood (although notwithoutchallenge)asrepresentingamodel for the biological sciences and has been much contested in the social sciences. We – the editorsofthisbook–startfromtheproposition that the central project of any science is the elucidation of causes that extend beyond the unique specific instance. Danermark et al. (2002, p. 1) have firmly stated: … first, that science should have generalizing claims. Second, the explanation of social phenom- ena by revealing the causal mechanisms which produce them is the fundamental task of research. We agree, and have considerable sympathy withtheturntocriticalrealismasabasicmeta- theoretical foundation for social research practice that is proposed by those authors. However, we want to qualify and elaborate somewhat on these fundamental premises. First, we want to make it clear that for us generalizing is not the same as universalizing. It is important to be able to develop an understanding of causation that goes beyond the unique instance – the object of ideographic inquiry. However, it is just as important to be able to specify the limits of that generalization. We cannot establish universal laws in the social sciences. There is no valid nomothetic project. We hope to demonstrate
- 25. 2 THE SAGE HANDBOOK OF CASE-BASED METHODS here that case-based methods help us both to elucidate causation and to specify the range of applicability of our account of causal mechanisms. The emphasis on the plural form of mechanism in the preceding sentence is very important. It is not just that different mechanisms operate to produce, contingently in context, different outcomes. It is rather that different mechanisms may produce the same outcome – the complete antithesis of the form of understanding that is implicit in, and foundational to, traditional statistical modelling’s search for the – that is to say the universal, always and everywhere, nomothetic – model that fits the data. The traditional mode through which sci- ence has sought to determine causality has centred on the notion of the variable as the causal agent in the world with real entities – cases – understood as the determined consequences of the operation of relationships among variables. Another way of saying this is that the conventional view is to see cases simply as sites for observing and measuring variables. Cross-case analysis of these variables, once measured via ‘cases’ (as instances of generic processes) is then seen as the primary ‘source’ of scientific knowledge established in the form of causal models derived from the description of associ- ations among variables.This analytic strategy, derived from the programme of Newtonian physics,hascometodominatethequantitative programme across science as a whole and has a far greater influence on qualitative social science than is often recognized. Against the method of explanation that centres on vari- ables and works through analysis has been set a tradition operating under a range of related labels – interpretation, thick description, the phenomenological programme. Insofar as this alternative project – perhaps best understood as that of the human sciences as defined by Dilthey – has worked with cases, it has tended to understand them as specific and unique instances to be understood in their own terms – the ideographic project – rather than explained by general covering laws – the nomothetic project. We might see this approach as analogous to explaining each individuallandslideintermsofuniquecontext and conditions, without reference to general geological principles. We want to get beyond this. As always in science, this can be achieved not only by a radical set of technical innovations but also by recognizing that many of our existing scientific practices are implicitly, and to an increasing extent explicitly, founded around a ‘focus on the case’. So the purpose of this collection will be to present the methods that can be employed by case-focused researchers, certainly in such a way as to enable people to employ them in practice, but also as the foundation for a practical social science – and indeed for a practical biological and particularly biomedical science – which gets beyond the dichotomies of quantitative/qualitative – explanation/interpretation. A number of streams come together in this project. In particular, the synthesis of complexity theory and critical realism proposed by Reed and Harvey (1992, 1996) resonates very strongly with the idea that cases are central to proper social scientific understanding. One answer, although by no means the only one, to the important question: ‘What is a case?’ (Ragin and Becker 1992) is that cases are complex systems – a position that transcends the holistic/analytic dichotomy by recognizing that in complex systems (far from equilbric systems), trajectories and transformations depend on all of the whole, the parts, the interactions among parts and whole, and the interactions of any system with other complex systems among which it is nested and with which it intersects. Moreover, complexity theory’s mode of understanding the causal determination of the trajectories of cases and ensembles of cases resonates with critical realism’s understanding of causality as complex and contingent, and both correspond in essence with the configurational conception of causality that underpins Ragin’s (2000) assertion that what matters for social scientists are set-theoretic relationships rather than causal models couched in terms of descriptions of associations among abstracted variables.
- 26. CASE-BASED METHODS: WHY WE NEED THEM; WHAT THEY ARE; HOW TO DO THEM 3 Issues of methodology and social research practice informed by methodological argu- ment have a particular contemporary saliency. This is not just a matter of arguments in the academy about the constitution of knowledge, although those arguments do matter. It is also about knowledge as it is deployed in more or less democratic societies where political differences founded on overt differences of material and cultural interest have tended to be replaced by competing claims to competence in political administration. This is especially true of post-ideological and post-interest politics in that most scientized and administered of societies, the United Kingdom, where programme and policy outcomes as measured by statistical indicators are the basis of political claims to competence, and hence electability. Simultaneously, the translation of such indices into performance indicators sustains an audit culture that has largely displaced effective local democracy and replaced it with an apparently neutral and technical political administration that is target driven. However, although most developed in the UK, these are general tendencies in democratic capitalist societies where voters are perceived – it may well be wrongly – essentially as consumers, and ‘evidence’ is supposed to be the basis of policy and practice. We will return to the issues of the rela- tionship between social research and political administration – the domain of applied social research – but let us begin in the academy itself. Hedström (2005, p. 1) has remarked, in our view correctly, that: … much sociological theory has evolved into a form of metatheory without any specific empirical referent and … much sociological research has developed into a rather shallow form of variable analysis with only limited explanatory power. Again we agree, although our solution to unlocking this logjam is diametrically opposed to Hedström’s turn to analytical sociology coupled with agent-based mod- elling. Likewise, we have some sympathy with Flyvbjerg’s desire to make ‘social science matter’ (2001), although rather less with his assertion of phronesis – virtuous judgement we might say – as a practice that can be distinguished from both episteme and techne – from abstract knowledge and applied skill. When Flyvbjerg (2001, p. 57) asserts that: Phronesis is that intellectual activity most relevant to praxis. It focuses on what is variable, on that which cannot be encapsulated by universal rules, on specific cases. Phronesis requires an interaction between the general and the concrete; it requires consideration, judgment and choice. we see the second sentence as in fact a general statement to science as a whole, and the third as the first part of the application of science in social practice. The second part of that application in practice, of social science as praxis, is the requirement that social scientists as experts, for expertise we do indeed possess, have to deploy that expertise in dialogue with other people in the social contexts in which that knowledge is applied.1 Flyvberg does turn to the case study in arguing for ‘the power of example’ and we agree with his endorsement of the explanatory power of case-based narratives, especially in relation to the formulation of policy interventions. However, we want to go further here because we are firmly asserting what Flyvberg’s rather tentatively suggests – that case studies of different kinds, and in particular multiple case studies founded on systematic comparison – are the foundations of useful theoretical descriptions of the social world. This position has a good deal in common with Pawson’s argument after Merton for the saliency of ‘theories of the middle range’. PawsondeployedtwoquotationsfromMerton to illustrate what this term means and we draw on his usage here. Theories of the middle range are: Theories that lie between the minor but necessary working hypotheses that evolve in abundance in day to day research and the all inclusive systematic efforts to develop a unified theory that will explain all the observed uniformities of social behaviour, social organization and social change. (Merton 1968, p. 39)
- 27. 4 THE SAGE HANDBOOK OF CASE-BASED METHODS Moreover: The middle range orientation involves the spec- ification of ignorance. Rather than pretend to knowledge where in fact it is absent, it expressly recognizes what still must be learned in order to lay the foundations for more knowledge. (Merton 1968, p. 68) Perhaps we go further than Merton because our position really rejects the notion of any grand unifying social theory, seeing this as not only impossible in terms of difficulty but in fact as quite incommensurate with the nature and character of the emergent world. Indeed, the use of that word – emergence – implies that our view of the impossibility/undesirability of grand unifying theory also applies to much of the natural world, and with particular force to intersections between the natural and the social. Something needs to be said now, and to be said with a good deal of force. The turn to case methods is predicated on an explicit rejection of the utility of causal modelling based on variables. Abbott has delivered an intellectual coup-de-grace to that approach in his imagining of the reflections of a scholar of the future, which bears repeating here: The people who called themselves sociologists believed that society looked the way it did because social forces and properties did things to other social forces and properties. Sometimes the forces and properties were individual characteristics like race and gender, sometimes they were truly social properties like population density or social disorganization. Sociologists called these forces and properties “variables”. Hypothesizing which of these variables affected which others was called “causal analysis”. The relation between the variables (what these sociologists called the “model”) was taken as forcible, determining. In this view, narratives of human actions might provide “mechanisms” that justified proposing a model, but what made social science science was the discovering of these “casual relationships”. (Abbott 2001, p. 97) As Abbott remarks, can that really be what sociologists thought? Well, some did, some still do, and more importantly other social scientists, and in particular social statisticians, quantitative political scientists and economists, not only believe this and carry out academic work on that basis, but have a major influence on policy through the magic of numerically based accounts of the working of reality. That said, we want to make it absolutely clear that our conception of how case-based methods work explicitly rejects any fundamental distinction between the quantitative and the qualitative. We think Hedström is flat wrong when he asserts that: ‘Although qualitative research can be important for the development of explanatory theory, it lacks the reliability and generalisability of quantitative research ….’ (2005, p. 101) whilst agreeing absolutely with his preceding statement that: ‘… quantitative research is essential for sociology’ (2005, p. 101). Our rejection is of the disembodied variable, not of quantitative measurement, description, and systematic exploration. In the social sciences it is often contended that quantitative work is concerned with cause whereas qualitative work is concerned with meaning. This is characteristically accom- panied by an explicit citation of Weber’s dictum that explanation must be adequate at both the levels of cause and meaning. However, the separation of qualitative work from the investigation of causes is miscon- ceived. Indeed, Weber was to a considerable degree engaged with the interpretation of meaning precisely because he understood meanings as causal to social actions. Pawson and Tilley (1997, p. 21) correctly identify the traditional interpretative programme of social research – what they call hermeneutics I – as concerned with the establishment of a true representation, which must include an account of the causal processes in the context and actions being interpreted. We regard Geertz’s (1975) commitment to ‘thick description’, with its emphasis on the specific and the contextual, as wholly compatible with this traditional conception of the hermeneutic programme and with complexity science’s emphasis on the inherent localism but nonetheless achievable reality of scientific knowledge. This has to be distinguished from the radical relativism of contemporary post- modernism,which,incontrastwiththeproject
- 28. CASE-BASED METHODS: WHY WE NEED THEM; WHAT THEY ARE; HOW TO DO THEM 5 represented by the different contributions to this handbook, denies the possibility of any knowledge. Of course, knowledge is socially constructed, as is the social itself, but reality has a say in that construction as well as the acts, agency and social character of the knowledge makers. We make a bold contention here – it is by thinking about cases that a proper and explicitly dialectical synthesis can be achieved between cause and meaning/interpretation in order to achieve explanation. When we think about cases we have to think not just conceptually but also in relation to the actual tools we use for describing, classifying, explaining and understanding cases. Hayles (1999) and Cilliers (1998) have in different ways commented on how the availability of computing technology has given us access to domains of reality in a fashion that is radically different from that of the traditional conception of scientific practice. We have become cyborgs – able to do with technology what we cannot do with ‘science’. This has much in common with actor-network theory’s notion of ‘actants’ but differs in that the tool and user become one.2 We do have very important tools here: the numerical taxonomy techniques of cluster analysis, the configurational approaches developed on the basis of Boolean algebra by Ragin and others, the potential of neural networks in social research and the availability of computer- based methods of what is usually miscalled ‘qualitative analysis’ (although there are analytic components) but would be better described as systematically structured qualita- tive interpretation. All of these are important for case-based work and will be dealt with here. We want to integrate the discussion of modes of thought and modes of practice as research as a process necessarily depends on this synthesis. At this point we want to make something explicit that has thus far been implicit. Although we will include consideration of individual cases and methods for under- standing them, the comparative method in its different forms is central to case-based understanding. We see this in terms both of classification on the basis of comparison and in the exploration of complex and multiple causality. Classification compares to distinguish like from not like. Systematic comparison based on interpretation of a range of cases – with its origins in eighteenth century German Kameralwissenschaft – seeks to establish distinctive characteristics of particular cases or sets (ensembles) of cases and to explore how those characteristics taken together are causal to the current condition of cases. There is always at least an implicit and usually explicit process of categorization in which cases are grouped into categories and there is a qualitative examination of historical trajectories in order to ascertain which trajectories produced which outcomes. Classification and comparison are central to any case-based method that attempts any sort of generalization. To reinforce this point, let us refer first to Linnaeus on classification and then to my co-editor on comparison: All the real knowledge which we possess, depends on methods by which we distinguish the similar from the dissimilar. The greater number of natural distinctions this method comprehends, the clearer becomes our idea of things. The more numerous the objects which employ our attention, the more difficult it becomes to form such a method and the more necessary. (General Plantarum quoted by Everitt 1993, p. 2) Implicit in most social scientific notions of case analysis is the idea that the objects of investigation are similar enough and separate enough to permit treating them as instances of the same general phenomenon. At a minimum, most social scientists believe that their methods are powerful enough to overwhelm the uniqueness inherent in objects and events in the social world. … The audiences for social science expect the results of social scientific investigation to be based on systematic appraisal of empirical evidence. Use of evidence that is repetitious and extensive in form, as when based on observations of many cases or of varied cases, has proved to be a dependable way for social scientists to substantiate their arguments. (Ragin 1992, p. 2) A central theme of the collection is the importance of demonstrating how case-based methods work in practice. Here, by prac- tice we mean both disciplinary application and application to the resolution of cen- tral problems of social practice. We note
- 29. 6 THE SAGE HANDBOOK OF CASE-BASED METHODS that case-focused methods are of particular importance for interdisciplinary studies at a time when social science is increasingly ‘post-disciplinary’ both in conceptualization and in practical application. They are also a central part of the repertoire of applied social research. These two areas of interest intersect, particularly in the domains of the applied area academic fields exemplified by health studies, urban studies, education, policy studies, social policy, environmental studies, development studies, science studies and others. It may well be that it is in the applied areas that the social sciences will ‘open’ as the Gulbenkian Commission (Wallerstein 1996) on their future suggested they should. Case- focused methods and understanding will be a crucial tool in this process. One final methodological point has to be made now. Ragin and Becker (1992) asked ‘What is a case?’, and in his introduction to that seminal collection Ragin began precisely by questioning the very idea of case itself and noting that all the contributors to the collection and participants in the workshops fromwhichitderivedagreedthatcaseanalysis is fundamental to social science and: ‘… has a special, unexamined status’ (1992, p. 8). Ragin’s solution to the dilemmas this posed has been to propose that we focus on ‘casing’ and the process of social inquiry – to quote the title of his concluding chapter to What is a Case? He urged us to examine precisely the tactic that social scientists engage in when they delimit or declare cases, and gave this a name – casing. As we shall see, several of the contributions to this Handbook address this idea, and it resonates very much with Cilliers’ (2001) discussion of the ways in which we might specify complex systems. Here it is important to make some preliminary comments about how casing works in applied social research and what issues emerge from this. Typically, applied social researchers work with what appear to be given cases. This is particularly the case when they start, as I (Byrne) have in many research projects, from a statistical array of data arranged in typical spreadsheet style with the cases forming the rows in the matrix and ‘variate traces’forming the columns. In other words, we start with the products of surveys and it is worth remem- bering that Marsh (1982) defined the social survey precisely as something that generated just such a case/variable matrix. Here, we should remember that much of the data that we might use comes not from dedicated research exercises, including official exercises such as the census, but from administrative processes. However, once it is arranged as a case–variate trace matrix, we have a survey. So statistical work takes its cases as given – as the entities presented in the data array. Note that the phrase here is statistical work, not quantitative work. Quantitative data construction involves specification of cases. A very simple example is provided by the question ‘Is the case for a social survey, the individual or the household, composed of varying numbers of individuals and having emergent properties beyond those of the individuals constituting it?’ Actually, multi-level modelling techniques designed to pursue causality across levels (see Goldstein 1995) do address the issue of casing by trying to take account of the ways in which multiple levels in a hierarchical dataset (as Goldstein correctly remarks, such datasets have the form they do because that is the form of the reality from which they are constructed) work together to produce outcomes. In general, when we engage in causal modelling, the data have been cased for us. Sometimes this makes good sense. How- ever, even as simple a social phenomenon as the household has a distinct fuzziness, which we can appreciate very well when we pursue the trajectories of households through time using longitudinal panel datasets such as the British Household Panel Survey. In other words, policy researchers have to pay as much attention to the constitution of their cases as they do to processes of operationalization in the construction of measures. This applies with particular force to the evaluation of social intervention programmes but it has general application. An example is appropriate. One of my graduate students, Katie Dunstan, is working on and around an evaluation of an intervention programme in ‘depressed’
- 30. CASE-BASED METHODS: WHY WE NEED THEM; WHAT THEY ARE; HOW TO DO THEM 7 (the depression is contestable) local housing markets, with the ‘local’ defined as relatively small areas within a city region. It rapidly became apparent that sense could only be made of the trajectories of these localities if they were understood in relation first to the city region of which they were part and the dynamics of housing markets throughout that region.Andthencamethesub-primeissue,the collapse of Northern Rock and a global or at least British–US collapse in housing markets generally. What is the case here for which causes can be established? Indeed, the problem of boundaries in spatial studies illustrates the issues of casing in relation to applied research very well. My own work on the trajectories of former industrial city regions into a post-industrial future requires me to think very carefully about the changing spatial boundaries of the regions, and indeed to consider equally carefully temporal boundaries in relation to the explo- ration of major qualitative changes. This is a relatively pure intellectual problem, although it can have profound policy consequences at a local level when municipal boundaries that do have a reality bear very little relationship to patterns of living and working and revenue raisingincomplexmetropolises.Itcantakeon a more extreme character when administrative boundaries taken as cases form the basis of applied policy research and consequent policy implementation with very significant outcomes for local populations. Ruane’s (2007) examination of the reconfiguration of emergency and maternity services in Greater Manchester demonstrated that by taking the administrative health region as the case, there was a marked tendency for policy research to argue for services to be moved towards the centre of the region. As this was also happening in adjacent regions, the peripheries were losing key services as it were in both directions – in the case of the Pennine towns towards both Leeds and Manchester. The case was given as the city region but that was not the significant health-care system for people in the periphery. We have here entered into discussion of the ‘boundaries between nested and overlapping cases’, in which it might be regarded as a great abyss (Charles Ragin’s suggested term) of social science. We are not claiming to have an immediate practical solution to the issues posed by this abyss, although the content of this Handbook is intended to offer us approaches to dealing with it. Rather, our intention is to problematize in a way that shows the knowledge and theory dependence of social research, and the consequent need for close connection to/understanding of empirical cases, however defined or constructed.The reality of the abyss might be what drives many researchers to go micro and focus on the individual as the one true case. We are arguing very firmly that we can do more than that. So the methodological principles and issues have been laid out before you. Let us now turn to the objectives and organization of the Handbook as a whole. Our intention in assembling these chapters was to cover three interrelated domains of knowledge: (1) the methodological status of case-based methods; (2) the sets of techniques that can be deployed in case-based investigation; and (3) actual examples of the use of the methods in the fields and disciplines that constitute the territory of the social sciences, both in the academy and applied in practice. Of course, these three domains are remarkably fuzzy sets. Contributions in all three sections of this book draw on concepts, methodological programmes and actual examples of empirical research in practice. This is exactly as it should be but there is a heuristic value in the sectioning we have deployed. Each section is introduced by a guide to that section, which essentially presents a brief editor’s exposition of the contents of the chapters within that section. First, we thought it essential to have a set of arguments that relate to the actual methodological status of case-based methods. We wanted to set them in the historical context of the development of social science as an interwovenprojectofunderstandingthesocial that requires both the empirical investigation of the world as it is, as it has become what it is and as it represents a potential for the future;
- 31. 8 THE SAGE HANDBOOK OF CASE-BASED METHODS and the formulation of coherent accounts – theories – of the causal processes that generate becoming, presence and future potential. To a considerable degree we return to some of the central issues raised when Ragin and Becker asked ‘What is a case?’ but do so more explicitly in relation to arguments about the nature of social science’s methodological programme as a whole. The first half of this introduction started this account. It is developed in the first section of the book through a set of five chapters that engage with the way in which contemporary social science handles causality. These pieces develop a set of overlapping themes, which include the historical development of causal reasoning in social explanation, the resonances among complexity theory, critical realism and config- urational approaches to identifying complex and contingent causes, and the ways in which method, methodology and meta-theoretical reasoning about the nature of the world and how it may be known to interact with each otherinformingthecharacterofscience–note no apologetic confining of this to the social sciences. Implicit in this discussion is a theme that emerges in the succeeding sections: how does how we know engage with how we act in shaping the social world itself? The second section of this Handbook includes a set of chapters that deal with the methods and techniques of case based research. We begin with a subset that addresses the range of issues associated with classification. The four papers here combine in varying degrees the outlining of methods for classifying with methodologically focused discussions of just what we are doing when we classify and how classification shapes our ways of understanding. Two explicit developed accounts emerge here. One is essentially a classification located repetition of the ontological position which is common to the chapters in section one. The other represents the major coherent rival to the concern with causality that underpins that style of work and draws on actor-network theory (ANT). ANT represents the most sophisticated version of a conventionalist account of the nature of scientific knowledge and, moreover, one that is firmly grounded in careful consideration of scientific practices themselves. It is important that this different way of thinking is represented here. The next subset comprises a set of chapters that deal in different ways with quantitative, and especially configurational, approaches. These contain elements of tech- nical discussion and demonstration but are primarily concerned with the underlying logic of explanation that informs configurational techniques. Again, it is absolutely essential to reiterate that the relationship between logic of explanation and actual technique is not unidirectional. It is true that the general principle of a shift from variable-centred views of causality to the proposition that what matters is set-theoretic relationships is fundamental and foundational to qualitative comparative analysis and related configura- tional techniques, but we see here that the actual use of these techniques in practice leads to the development of methodological arguments themselves. The final subset of section 2 takes a turn to the qualitative. The classification and configurational approaches reviewed in the preceding subsections have a strong but not exclusively quantitative character. Here again we see how the actual processes of doing research and the development of techniques and procedures for that ‘doing’ interact with our approaches to describing and explaining how the world works. What these chapters demonstrate is that the issues of explanation transcend the false and – let it be bluntly said now – really rather silly false dichotomy between quantitative and qualitative approaches to understanding the social world. Rather, we have a set of unifying issues that include those of generalization, the need to consider relationships as just as real and socially significant as entities and implications for science as description, explanation and practice that flow from the agentic potential of human beings. It is not that these issues do not emerge in relation to quantitative modes of work, and indeed several of the chapters in the two preceding sections address them specifically.
- 32. CASE-BASED METHODS: WHY WE NEED THEM; WHAT THEY ARE; HOW TO DO THEM 9 It is just that they are absolutely forefronted in qualitative modes of social research and we get a sharp reminder of their significance here. Part 2 of the book is by no means just a set of demonstration of case-based methods, although it does include very substantial demonstration of case-based methods. Rather, it mixes demonstration, empirical example and methodological discussion. It is fuzzy but focused – the central principle underlying case-based understanding and consequent methods themselves. This approach continues in part 3, where eleven chapters address the issues of deploying case-based methods in disciplines and fields. Again, we find a constant engagement with methodological issues and with the interrelationship between those arguments and the actual deployment of methods in practice. The areas covered in these chapters include historical account, political science, organizational studies, eco- nomics,medicineasbothmedicalpracticeand biomedical science, health policy, develop- ment studies, cultural studies and the general deployment of case methods in applied social research as understood by applied social researchers. The focus of these chapters is on how methods can be understood and used, with the argument often illustrated by a specific empirical example or examples of research, explanation and interpretation in the discipline or field. So we begin with a justification of case-based methods and move through an account of the character of such methods into a discussion of how they work in the practice of knowledge construction both in the academy and in applied research, which is intended to inform policy development and practice. The Handbook concludes with an overview of the themes emerging in it by my co-editor, but even here it is worth mentioning some points of agreement and disagreement that emerge from reading over the contributions as a whole: • Agreement: case-based methods are useful and represent, among other things, a way of moving beyond a useless and destructive tradition in the social sciences that have set quantitative and qualitative modes of exploration, interpretation and explanation against each other. • Agreement: generalization matters, but that generalization is best understood as involving careful attention to the setting of scope. In other words, we cannot generate nomothetic laws that are applicable always and everywhere in the social sciences. No method can construct theories with that character. However, comparative case-based methods do enable us to give accounts that have more than unique ideographic range and individual case studies can contribute to this. The scoping requirement is to pay careful attention to the limitations of our knowledge claims in time and space. • Disagreement: some contributors consider that the whole logic of case-based methods requires an ontological position that necessarily rejects the value of abstracted variable-based analyses in explanation in the social sciences. Others do not go so far and simply see case-based methods, and in particular set-theoretic methods, as representing an important alternative approach whilst allowing variable-based modelling a role in social explanation. • Case-based methods are a way of making social science useful: they always have been and they continue to be so. And so to the meat of the matter … NOTES 1. The ideas and practice of Paolo Freire in relation to participatory research and dialogical learning are of enormous importance for us here. 2. We must also note that our endorsement of critical realism’s deep ontology means that we accept that the real in the form of generative mechanisms and the actual as the contingent consequence of those generative mechanisms have form before we shape that form as the empirical in our construction of knowledge from the materials of reality. Actor- network theory, in contrast, has a shallow ontology in which form is wholly constructed by the practices of science. In essence, we allow reality a role in representations of reality. Representations are made from something, not reified from nothing. REFERENCES Abbott, A. 2001 Time Matters. Chicago: University of Chicago Press.
- 33. 10 THE SAGE HANDBOOK OF CASE-BASED METHODS Cilliers, p. 1998 Complexity and Postmodernism. London: Routledge. Cilliers, p. 2001 ‘Boundaries, hierarchies and networks in complex systems’ International Journal of Innova- tion Management 5(2): 135–147. Danermark, B., Ekström, M., Jakobsen, L. and Karlson, J.C. 2002 Explaining Society. London: Routledge. Everitt, B.S. 1993 Cluster Analysis. London: Edward Arnold. Flyvbjerg, B. 2001 Making Social Science Matter. Cambridge: Cambridge University Press. Geertz, C. 1975 The Interpretation of Cultures. London: Hutchinson. Goldstein, H. 1995 Multilevel Statistical Models. London: Edward Arnold. Hayles, K. 1999 How We Became Posthuman. Chicago: University of Chicago Press. Hedström, P. 2005 Dissecting the Social. Cambridge: Cambridge University Press. Marsh, C. 1982 The Survey Method. London: Allen and Unwin. Merton, R. 1968 Social Theory and Social Structure (enlarged edition). New York: Free Press. Pawson, R. ‘Middle range theory and programme theory evaluation: from provenance to practice’ in F. Leuuw (Ed.) Evaluation and the Disciplines (in press). Pawson, R. and Tilley, N. 1997 Realistic Evaluation. London: Sage. Ragin, C. 1992 ‘Introduction’ in C. Ragin and H. Becker (Eds) What Is a Case? Cambridge: Cambridge University Press, pp. 1–19. Ragin, C. 2000 Fuzzy Set Social Science. Chicago: University of Chicago Press. Ragin, C. and Becker, H. 1992 What Is a Case? Cambridge: Cambridge University Press. Reed, M. and Harvey, D.L. 1992 ‘The new science and the old: complexity and realism in the social sciences’ Journal for the Theory of Social Behaviour 22: 356–379. Reed, M. and Harvey, D.L. 1996 ‘Social science as the study of complex systems’ in L.D. Kiel, and E. Elliott (Eds) Chaos Theory in the Social Sciences. Ann Arbor, MI: University of Michigan Press, pp. 295–324. Ruane, S. 2007 Report Commissioned by Rochdale Borough Council Re: The Healthy Futures and Making It Better Proposals for Health Services in Rochdale Leicester: Health Policy Research Unit, De Montfort University. Wallerstein, I. (Chair of Gulbenkian Commission) 1996 Open the Social Sciences. Stanford, CA: Stanford University Press.
- 34. PART ONE The Methodological Context of Case-Based Methods This Part of the book comprises five papers that locate case-based methods in their methodological context. The first, by David Harvey, sets case-based approaches in relation to the methodological programme of the his- tory of social science in general, but sociology in particular. Harvey’s core argument is that debates about the scientific status of case stud- ies have been conducted in methodological and epistemological terms, whereas issues of ontology have either scarcely been considered or have been conflated with ‘epistemological presuppositions’. He develops a critique founded in the synthesis of critical realism and complexity frames of reference, which he has proposed in a range of earlier work and proceeds by identifying first ‘the anti- ontological bias’ of the empirical social sciences and then proposing a complex realist alternative position. He then proceeds to delineate a complex realist standpoint – a position adopted explicitly by several other contributors to the Handbook, and delivers a historical narrative of the ‘ambivalent sociological status’ case studies have held in American sociology through the twentieth century, concluding with an elaboration of a complex realist model of the case object. Lars Mjøset takes up Harvey’s issue in contemporary terms. He presents an account of‘thecontextualistapproachtosocialscience methodology’ beginning with a specification of the role of methodologists as mediators between professional philosophy of science and the actual technical apparatuses of the practice of social research. As he puts it, for methodologists the: ‘… task is to produce practical philosophies of social science, methodologies that provide concrete guidelines in research practice, and criteria as to what knowledge shall count as science’ (see p. 40). There is an interesting argument to be had with the notion that the professional philosophy of science should have any privileged position in adjudicating on the adequacy of practice here, but it is generally held and Mjøset does articulate clearly that methodologies are the product of a two-way dialogue between philosophical orientation and the methods themselves in practice. He develops his argument by identifying three practical philosophies of social science. First is the standard attitude that emerges from some sort of synthesis of logical positivism, the work of Popper and Lakatos, and analytical philosophy. This tends to what we might call scientism – privileging quantification, experiments including thought experimentsandstatisticalanalyses.Secondis the social–philosophical practical philosophy
- 35. 12 THE SAGE HANDBOOK OF CASE-BASED METHODS of social science for which the leading question is ‘how is social science possible?’ In very crude summary, we might say that the positions here are those that tend to hold that the ‘human sciences’are different. Mjøset shows that this position is itself bifurcated with, on the one hand, a transcendental or reconstructionist programme of establishing the general basis for social theory and, on the other, a deconstructionist tradition asserting that no such programme is possible. Finally, we have the contextualist practical philosophy of social science – in effect an approach founded on and displayed in the actual distinctively social methods of social science research as practised. Mjøset demonstrates the relationship of this programme with pragmatism and emphasizes that in contrast with the two preceding positions, those who endorse/practise it, see social science as concerned with intervention rather than representation. Plainly, case-based methods resonate particularly with this style, a style of work as much as a style of thinking, or perhaps it is best to say a style of thinking while working and working while thinking. Mjøset’s chapter proceeds through a five- fold specification of the essentials of each attitude in terms of explanatory logic, popularization of fundamental metaphysical questions, explanatory priorities given to the autonomy of social science, implied soci- ology of knowledge and assumptions about relationships between the sciences. Mjøset demonstrates how the positions become visible in episodic struggles on methodology but also notes that methodological camps generally ‘keep to themselves’ and that this is the case between ‘mechanism scholars’ and ‘case scholars’. Essentially, his chapter demarcates the issues and develops a full account of their implications for both the construction of social theory and the practice of social research. Bob Carter and Alison Sealey develop themes introduced by Harvey and Mjøset with specific reference to the central issues of defining and using cases. Drawing on realist social theory, they follow Ragin in emphasizing the importance of ‘casing’ and suggest that: ‘because of its explicit social ontology, realism provides the means for making the process of casing a more fastidious one’ (see p. 69). These authors put considerable emphasis on the significance of reflexivity in the practices of researching the social in terms of which could be described using the Freirean conception of dialogue, but also insist that social reality is not exhausted by either actors’ or researchers’ accounts and that we must pay serious attention to realism’s stratified social ontology, which has important implications for methodological research. Malcolm Williams and Wendy Dyer take the idea of the case into engagement with probability as the basis of causal claims in social science. They outline the three different interpretations of probability that have been variously deployed, viz. the frequency interpretations that underpin con- ventional statistical modelling, the propensity interpretation following on from the work of Popper and subjectivist interpretation, noting that this last is not generally associated with causal explanation but focuses more on personal beliefs about the probability of an event. By taking up Popper’s discussion of single case probability, these authors seek to work out a technical approach to the causal exploration of large datasets describing the trajectories of cases through time and interpreting these trajectories in terms of the propensities inherent in individual cases as described in the dataset. The example employed is that of cases moving through experience in a large ‘custody diversion’ project (a project intended to divert offenders and potential offenders with mental health problems from imprisonment). This chapter combines all of a methodological synthesis of realism, complexity and a propensity view of probability; a developed demonstration of a set of practical techniques using databases, cluster analysis approaches and the mapping of clusters through time; and an account of the experience of an actual research project. For Williams and Dyer: ‘What is important and useful is to be able to talk about the probabilities of antecedent conditions that led tospecificoutcomesandtousethisknowledge
- 36. THE METHODOLOGICAL CONTEXT OF CASE-BASED METHODS 13 to produce explanations (and thus predictions) of phenomena’ (see p. 98). This Part concludes with a piece by David Byrne attempting to synthesize complex realist (very much after Harvey as modified by engagement with Cilliers) with configura- tional approaches to cases.This was originally identified by the author as a rant – and to a considerable extent it continues to be one, but it is a rant to a purpose. That purpose is to respond to the resonance between complex realism as a meta theoretical account of social reality and case-based methods in general, but configurational approaches in particular, as modes of understanding complex causal processes in social reality. By paying particu- lar attention to the idea of control parameters and seeking to understand how such control parameters might be identified using case- based methods, the chapter seeks to initiate methods that can be applied in practice to social purposes.
- 38. 1 Complexity and Case D a v i d L . H a r v e y Case studies have long kindled the sociologi- cal imagination. Yet their scientific status has been contested almost from the moment soci- ology was christened an academic discipline. Avatars of ‘hard science’ regard case studies as ‘too impressionistic’ in the interpretive liberties they take and as too aggressive in the range of subjects they presume to address.The legitimacy of case studies has been contested largely, therefore, on methodological and epistemological grounds, while fundamental issues of social ontology seldom attract the same attention. This chapter addresses this oversight as it bears on case-based social research. As such, it has two goals. First, it complements the current discussion of case methods by exploring the ontological requirements of case research; and second, it uses a complex realist paradigm to illuminate the structure of the case-object (Byrne 2002, pp. 5–6; Byrne and Harvey 2007, pp. 64–66). Christopher Norris (2007) defines ontol- ogy’s intellectual niche and the type of entities that occupy that space as follows: Traditionally, … [ontology] … is the field of philosophic inquiry that has to do with things … or with real world objects and their intrinsic structures, properties, causal powers, and so forth. That is to say, it treats of them just in so far as they are taken to exist and to exert such powers quite apart from the scope and limits of human knowledge. (Norris 2007, p. 335) Norris’s realist definition provides a feel for the complexity and dynamic powers of the entities with which this chapter deals. It also anticipates the predicaments that ontological considerations introduce into social scientific discourse. Three levels of ontological interest are germane to sociological inquiry: (1) philo- sophical ontology; (2) scientific ontology; and (3) social ontology. Philosophical ontology deduces from the structure of speculative thought the fundamental nature of the entities that constitute our everyday world. Scientific ontologies are nested within philosophical ontologies to the extent they flesh out the local details of a terrain in a way philosophical ontology cannot. They are more narrowly focused because they deal with the existential constitution of the entities scientists inves- tigate. Finally, social ontologies are nested within scientific ontologies in that they deal with the elemental entities and dynamics
- 39. 16 THE SAGE HANDBOOK OF CASE-BASED METHODS sociohistorical formations must exhibit if they are to sustain themselves over time. A clear understanding of ‘casing’, to use Ragin’s (1992, pp. 219–220) suggestive term, requires a full explication of a case-object’s ontology at all three of these levels. Moreover, in keeping with the complex realist paradigm, we assume that case-objects are ontologically real, i.e. they exist prior to the research act itself. Of course, the conceptual imagery of the case-object is undoubtedly a product of operationist research strategies. The case- object, however, is not a wholly nominalist invention. Casing does not create the case- object, but only a construct of the case-object. Consequently, there is always an aspect of the case-object that eludes or stands over and against casing and its epistemology. For this reason, the ontological autonomy of the case-object imbues social inquiry with a self–referential dialectical tension, akin to T.W. Adorno’s (1973, p. 5) ‘negative dialectic’. Paraphrasing one ofAdorno’s most powerful maxims, we can say the relation between casing and its object is such that whenever a casing construct goes into its case-object, some aspect of the object is left over. When not trivialized, this ‘material remainder’ prevents premature closures of inquiry. Hence, the natural attitude associated with complex realism insists that some part of the case-object’s ontology resides in its obdurate properties. Accordingly, if the ontological autonomy of the case-object offers a fresh set of checks and counter-checks for critically evaluating case research in general, it also forces a re-examination of those casing strategies that epistemologically gloss the case-object’s autonomy. As we will see, this glossing of ontological issues has been justi- fied by an axiomatic aversion on the part of social scientists to metaphysical speculation. Today, however, there is a growing realization that any science deserving of the name must eventually attend to the ontological underpin- nings of its subject matter. Indeed, there is a growing awareness among social scientists of the need to explicate all three ontological moments in framing their research. In what follows, we first examine how and why it is that sociology gives such short shrift to ontological considerations. In the latter portion of this chapter, we follow up that critique by constructing an ideal typical model of the case-object and its ontological structure. THE ANTINOMIES OF CASE-BASED RESEARCH The neo-Kantians and the social sciences The current casing debate is due in large part to American sociology’s long-standing debt to neo-Kantian philosophy. Even now, neo- Kantianism, with its axiomatic aversion to metaphysicsanditsavoidanceofobjectivistor realist ontologies limits the forms sociological praxis takes. Although few still use its argot, its canonical reduction of ontology to problems of method are at the heart of what we will call sociology sans social ontology. The neo-Kantian movement emerged among German intellectuals in the last half of the nineteenth century, a period of industrial crises and class realignments. These crises were accompanied (in some instances, preceded) by the ascent of the neo- Hegelian and the Positivist worldviews in the academy (Iggers 1968, Ringer 1969, 1997, 2004, Burger 1976). A century be- fore, Kant had developed a watershed subjective idealist philosophy with which he deflected David Hume’s withering empirical attack on Newtonian science and its constitutive categories. Under the duress of Modernist thought, the neo-Kantians sought to resuscitate Kantian humanism and, in doing so, to protect nineteenth century historiography from the disenchantment of an aggressive utilitarianism bent upon ‘reducing cattle to tallow and men to ‘profits’. This defense was accomplished by a self- conscious ‘return to Kant’. That is, the neo- Kantians attempted to revive philosophically the formative role individual subjectivity and
- 40. COMPLEXITY AND CASE 17 morality played in shaping human history and culture.Hence,theneo-Kantiansmomentarily repelled the various fin-de-sicle challenges to Kant’s subjective idealism. Ironically, their triumph over Positivism, on the one hand, and neo-Hegelianism on the other, was compromised on two important fronts. First, they made a fateful concession to Kantian orthodoxybyretainingKant’sinsurmountable firewall that separated the sensate world of phenomenal experience from the noumenal world of ideas and values. The same partition also segregated the constructive logics of scientific discovery from metaphysical spec- ulation, and, finally, the deterministic world of nature from human agency’s willfully constructed social and moral orders. Their second point of compromise was more serious. In ‘vanquishing’ neo- Hegelianism, a great many neo-Kantians inadvertently incorporated elements of Hegelian realism into their own philosophical framework. Consequently, as Lewis White Beck (1967, pp. 468–473) observes, the ‘neo-Kantian’ label actually designated: … a group of somewhat similar movements that prevailed in Germany between 1870 and 1920 but had little in common beyond a strong reaction against irrationalism and speculative materialism and a conviction that philosophy could be a ‘science’ only if it returned to the method and spirit of Kant. … Because of the complexity and internal tensions in Kant’s philosophy, not all the neo-Kantians brought the same message from the Sage of Königsberg, and the diversity of their teachings was as great as their quarrels were notorious. At the end of the nineteenth century the neo-Kantians were as widely separated as the first generation Kantians had been at its beginning, and the various neo- Kantian movements developed in directions further characterized by such terms as neo-Hegelian and neo-Fichtean. (Beck 1967, p. 468) Sociologists are familiar with the Marburg School of neo-Kantianism led by Paul Natrop and Hermann Cohen, and the Baden or South- western School led by Wilhelm Windelband and Heinrich Rickert. To these, Beck adds a third ‘Sociological School’, which centered on the works of Wilhelm Dilthey and Georg Simmel. Dilthey’s so-called Lebensphilosophie established what would soon become a standard distinction between Naturwissen- schaft and Geisteswissenschaft. This division was grounded in a substantive partition of experience into objects of nature and historically situated human productions (Tapper 1925, Holborn 1950, Masur 1952, Rickman 1960). These two ontological domains also prescribed two radically different approaches to knowledge. Naturwissenschaft remained, in essence, the system Kant had defended. Geisteswis- senschaft was a science of the human mind, a moral science based on the scholar’s imaginative reconstruction of past Weltanschauungen. Hence, the Geisteswis- senschaften could plumb human history and motives with a depth the Naturwissenschaften could not. Windelband (1894/1980) accepted the wisdom of Dilthey’s ‘two sciences’ but offered an alternative theory of what divided them from each other. Dilthey’s reply to Lebensphilosophie sought to purge the dis- tinction between the two sciences of any ontological residues. Windelband’s partition was predicated not in diverging ontologies but in the contrasting logics of procedures the two deployed. Man and society could be understood using both the generalizing methods of the natural science and the particu- larizing modes of understanding practiced by historians. Hence, nomothetic inquiries could seek universal laws of human nature just as readily as ideographic inquiry might pursue and grasp the historical particularity of a distant era and its culture. All that was required was to keep the two logics of investigation and their respective operational lines of thought separated. Moreover, by ‘de-ontologizing’ the differences between the two, metaphysical entanglements could be avoided. There was, of course, a price to be paid for Windelband’s ‘epistemological end around’, and that price would become evident in Chicago Sociology’s attempt to construct a coherent sociology using Windelband’s philosophical distinctions.
- 41. 18 THE SAGE HANDBOOK OF CASE-BASED METHODS The paradox of methodological complementarity Both Simmel and Windelband had profoundly influenced Robert Ezra Park during his doctoral studies in Germany – Simmel at Berlin and Windelband at Strasbourg and Hei- delberg (Baker 1973, pp. 256–257). Hence, it should be of little surprise that a neo- Kantian worldview dominated the Chicago School’s sociological outlook during the interwar decades of the 1920s and 1930s. This neo-Kantian dualism is most evident in the fact that both nomothetic and ideographic studies were pursued during this era, although at times unevenly so. Much of this unevenness was related to the unequal value the Progressive Move- ment placed on nomothetic and ideographic research. During the Progressive Era, his- torians and social scientists alike were increasingly expected to produce instru- mental knowledge that could facilitate the ‘Americanizing’ of immigrants and the ‘dan- gerous classes’. This emphasis on scien- tific prediction and programmatic control undoubtedly tended to marginalize ideo- graphic research. Whereas case studies were still to be valued for their anecdotal insights and aesthetic immediacies, their emotion- laden ambience relegated them in some Progressive circles to a status similar to the ethereal programs of Christian uplift, the pleadings of muck-raking novelists, and the sensationalist huckstering of yellow jour- nalism (Bulmer 1984, pp. 1–45, Matthews 1977, pp. 85–120). This predicament was soon translated into a methodological conundrum for Park and Burgess (1921/1969), as they labored to construct an ostensibly disinterested scientific sociology, while simultaneously preserving the dual heritages of the nomothetic and ideographic traditions. Their task was made incrementally more difficult by the fact that the public reputation of Chicago Sociology’s scientific surveys often hinged upon the benchmarkparticipantobserverstudies,urban ethnographies, and autobiographies it was producing – works whose power flowed directly from their inner sense of verisimili- tude (Bulmer 1984, pp. 151–171). By then, Thomas and Znaniecki’s The Polish Peasant in Europe and America (1918/1958) had provided a venerated tem- plate for Chicago Sociology’s case-base studies. Park and Burgess (1921/1969) drew on this masterpiece with stunning effect, although in later years – especially after Park’s retirement – it fell to Burgess and his co-workers at the Local Community Research Committee (LCRC) to work out the everyday nuts and bolts of designing a comparative framework for conducting case- based researches and integrating them into the Committee’s overall agenda (Burgess1941, Bulmer 1984, pp. 129–150). Aware of the encroaching power of the new statistical techniques then sweeping the social sciences, Burgess nonetheless continued to navigate between the quantitative social sciences and the lived substance of the field studies being gathered under his supervision (Burgess 1945, Burgess 1974, pp. 367–373, 386–392). Some notion of the practical difficulty of reconciling these methodological comple- mentarities can be garnered from Burgess’s 1927 article, Statistics and Case Studies as Methods of Sociological Research. Here we find Burgess looking forward to ‘the emancipation of case studies from the dom- ination of statistics’ (Burgess 1974, p. 370) even as he underscored the intractable tensions between the two. Hence, Burgess traces the role this methodological tension played in predicting marital adjustment outcomes. It is in revealing the dynamic factors in human relationships that the distinctive value of case studies exists. An intimate personal document discloses the organic relationship between factors, which, more than the mere presence of the factors, is often significant for marital adjustment. In order, therefore, to improve prediction, case- study analysis should be employed in combination with statistical procedure. In larger part, case study and statistical methods can be profitably used in close interrelation, but in considerable part the case-study method must be employed in relative independence of statistical procedure in order to derive from it its highest value for prevision. The fact
- 42. COMPLEXITY AND CASE 19 is that they are closely and inextricably interrelated. Obviously, all statistics ultimately come from ‘cases’. A schedule is constructed on the basis of data derived from the examination of individual cases. From the examination and analysis of cases come insights, clues, and leads, on the basis of which new items may be introduced into the schedule and their average significance determined. Correspondingly, an unexpected statistical finding may require exploration by interview to determine its meaning and significance for [marital] adjustment. Further- more, the analysis of ideal types of dynamic and organic relationships derived from case studies may suggest reorganization of predictive items under more significant categories (Burgess 1939/1972, pp. 386–388). It is not untoward to suggest that Burgess saw the relation between case studies and survey research as a tension-ridden dialectical unityofopposites.Hemayhaveevenregarded the interdependence of the two techniques as somehow paralleling the symbiotic pro- cesses of ‘cooperative competition’ he saw operating in his dynamic modeling of urban ecology. Despite the methodological hiatus separat- ing them, when set side by side, the nomo- thetic and the ideographic studies combined to illuminate the complex moral texture of Chicago’s ‘natural areas’ and ethnic neigh- borhoods; they were also indispensable tools for mapping the city’s centrifugal ecology and concentric structuration (Burgess 1925/1967, pp. 47–63). Hence, no matter how one judges the Chicago School’s efforts at nomothetically squaring the ideographic circle, this much is evident: the logical contradictions between the two methods were never resolved. This unabated ‘benign neglect’ of methodological schisms may have been in all likelihood a constant source of intellectual tension during the interwar period. At the same time, the Chicago School’s neo-Kantianism undoubt- edly remained a wellspring of sociological creativity, even as it served as a leavening source of departmental ambivalence. In any case, it was no mean trick to negotiate the intractable logics of social scientific surveys and case-based narratives without glossing their differences or submerging the premises of the one into the other. In fact, the genius of Chicago Sociology resided not in the unification of these antinomies but, rather, in their pragmatic, often piecemeal amalgamations. It was this dialectical ambiva- lence that sustained Chicago Sociology’s watershed scholarship of the 1920s and 1930s. The rise of methodological factionalism JusthowwellChicagoSociologymanagedthe nomothetic/ideographic split can be gleaned from what happened to it and to the profession after the Second World War. A powerful ‘second Chicago School’ formed around ‘Symbolic Interactionist’ social psychology. The Symbolic Interactionist contingent soon set itself off from the survey researchers (Fine1995,Gusfield1995,Abbott1999,2001, Becker 1999). Homologous schisms erupted elsewhere in the profession so that by the mid- 1950s the language of the earlier neo- Kantian antinomies would be supplanted by an alternative oppositional vocabulary, e.g. variable-oriented, quantitative versus qualita- tive research and its narrative productions.At one level, the substantive parameters of the new controversy paralleled the old: only the vocabularies changed. It terms of professional consequences, though, the formal disputes that had been held in relative abeyance now tightened into a Gordian knot. When that knot was finally cut, the former dialectical ambivalence gave way to a viral, profession- wide internecine struggle. This mutual alienation between the two self-interested factions hardened to such an extent that the position of the former was reduced to what C. Wright Mills (1959, pp. 50–75) deprecated as ‘abstracted empiri- cism’: fact-mongering in search of a working hypothesis. Returning the favor, ‘positivists’ showed contempt for ‘grand theory’ (again a Millsian epithet) and the extended narratives associated with case research. Indeed, more than one cutting-edge positivist suggested, only half-mockingly, that such sociology be banished to English Departments, American
- 43. 20 THE SAGE HANDBOOK OF CASE-BASED METHODS Studies Programs, or other academic dens of antiquity. Renegotiating old faits accomplis Given the progressive anomie that ensued, it should not be surprising that the renewed debate over casing appeared upon first reading to be little more than another counter- hegemonic call for a radical makeover of the discipline. A second reading, however, revealed a series of restorative motifs under- lying these polemical appearances. Moreover, this call for restorative self-clarification came from both sides of the quantitative/qualitative divide, resembling what mathematicians call a ‘squeezing theorem’ tactic. That is, the re-vindication of the proper place of case studies in the social sciences was not ana- lytically deduced from first principles but, instead, was pragmatically pieced together by the practical interests and experiences of both camps. This ameliorative strategy is at the heart of Ragin and Becker’s synoptic work What Is a Case? Exploring the Foundations of Social Inquiry (1992). The volume is a wide-ranging ‘taking of stock’, a momentary pause designed to assess the current status of case research. It reports the proceedings of a symposium that attempted to codify the methodological bounds of case studies. As an exercise in methodological self-clarification, the symposium accomplished two tasks: (1) it established the methodological bounds between casing and survey research tech- niques; and (2) it constructed an interpretive typology for classifying the manifold forms casing strategies might take. As to the first goal, symposium partici- pants established a rough consensus on the proscriptive limits of casing methods: • Methodological individualism, in and of itself, did not provide an adequate framework for grasping case processes, ‘… nor the idea that social life can be understood only from the perspective of individual actors’ (Ragin and Becker 1992, p. 7). • Cases were not purely given as social entities to be defined just as they were, nor were they solely the product of the interpretations the researcher brought to the problem being investigated (Ragin and Becker 1992, p. 7). • Whereas theoretical or interpretively selective samplings of case elements were deemed permis- sible, case analysis could seldom, if ever, ground itself in random samplings of a population (Ragin and Becker 1992, p. 7). • Finally,inacarefullystatedadumbrationoftheuse of multivariate techniques in directly delineating the domain of case studies, it was concluded that the ‘… concept of “the case” is logically prior both to the concept of “the population” and the concept of “the variable”’. In a context where the concept of ‘the case was made problematic, the other concepts appear impenetrable’ (Ragin and Becker 1992, p. 7). The second goal of the symposium was more formidable. Using as his ‘sample’ the eight invited papers that provided the focus of the symposium’s discussions’, Ragin developed a four-fold typology of conceptions of case to categorize different styles of casing. Ragin’s typology was generated from two dichotomies. The first pertained to the researcher’s assumptions concerning the nature of casing: did the researcher con- ceptualize the case as being an empirical unit or was it approached as a theoretical construct. The second dichotomy centered on whether the case in question was treated as a ‘one-of-a-kind’ sociohistorical interpretation or if it was seen as belonging to an already existing class of cases. When Ragin cross- referenced the two dichotomies, a four-fold table (stochastically represented in Figure 1.1) resulted. His synthetic classification gen- erated four cells: (1) ‘Cases as Found’, presumptively real, social entities not unlike other ‘natural phenomena’ given to experi- ence; (2) ‘Cases as Objects’, social entities posited to be both phenomenally bounded and conforming empirically to conceptually validated categories already in use; (3) ‘Cases as Made’, a conceptually mediated category in which the case in question is defined as a product of the specific unfolding of field research, one that has no parallel or comparative equivalent in the literature; and (4) ‘Cases as Conventions’, specific
- 44. COMPLEXITY AND CASE 21 Conception of cases assume that… Cases are empirical units which are, in turn… Conceptually specific to the single case Conceptually subsumed under a general class of cases Cell 1: Cases as found Cell 2: Cases as objects Cell 3: Cases as made Cell 4: Cases as conventions Cases are theoretical constructions which are, in turn… Conceptually classed as a single case Conceptually subsumed under a class of cases Figure 1.1 Ragin’s mapping of sociological conceptions of case. conceptual constructions whose full meaning emerges only when placed in the context of existing research. Ragin concludes by noting that these four conceptions of case could be interpreted as pragmatic strategies available to researchers for resolving various problems that arise in the concrete research setting itself. The pragmatic and eclectic tone of his conclusions are underscored when Ragin writes: This four-fold division of case conceptions is not absolute. A researcher could both use con- ventionalized empirical units, accepting them as empirically valid (Cell 2), and try to generate new theoretical categories or case constructs (Cell 3) in the course of his or her research. Frustrations with conventional case definitions and practices (Cell 4) could lead researchers to intensify their empirical efforts and to define cases and their boundaries in a more inductive manner (Cell 1). In fact, most research involves multiple uses of cases, as specific or general theoretical categories and as specific or general empirical units. These multiple uses occur because research combines theoretical and empirical analysis, and the two kinds of analyses need not use parallel cases or units. The point of … [Figure 1.1] … is not to establish boundaries between different kinds of research, but to establish a conceptual map for linking different approaches to the question of cases. (Ragin and Becker 1992/2005, p. 11) In keeping with the methodological focus of Ragin and Becker’s work, what began ostensibly as a collation of methodological presuppositions has become a ‘conceptual map’ for finessing possible ‘conceptual map’ ‘blockages’ encountered in field research or later interpretive dilemmas. Notice, however, that although What Is a Case? deals adequately with the subjective conceptions the researcher transports to the casing milieu, it never directly addresses the social ontology of the case-object itself, i.e. the possibility that the case-object is, in and of itself, a complex entity with autonomous claims of its own. Instead, the act of casing and its case-object are both approached as if they were social constructions posited during the research act. The typology thus focuses on the intensive logic of various
- 45. 22 THE SAGE HANDBOOK OF CASE-BASED METHODS conceptions of casing, while only tangentially touching on the case-objects’s extensive ontology. Hence, What Is a Case?, for all its theoretical sophistication, never breaks free of the neo-Kantian circle of epistemological concerns. METHODOLOGICAL WARS OF POSITION The results reported in What Is a Case? give us a fair index of the current thinking of those involved in case-based research. Furthermore, its conceptual tone and con- clusions are firmly fixed in the profession’s post-war schismatics. Although reasoned in its claims and moderate in its goals, What Is a Case? none the less bears the marks of those virulent wars of position that have marked the profession for more than six decades. Waged in a blunt, no-quarters-given style, these seemingly endless struggles have been carried out in the name of building a ‘unified social science’. Fought to the point of stalemate, both sides of the controversy have been reduced to encroaching upon the other’s ‘turf’ and attempting to supplant it methodologically. Despite this impasse, there are still those who believe that these ‘border raids’ will eventually produce closure on the issue of what is or what is not ‘scientific sociology’. Andrew Abbott (1999, 2001), by contrast, promotes a contrarian view concerning the inevitability of methodological and episte- mological closure. Although working from differing initial premises than those employed here, he similarly argues that sociology’s methodological dualism is deeply rooted in the history of the profession, if not its thought, and is likely to remain so for some time. Abbott’s stance is exemplified by his deconstruction of the molar oppositions of quantitative versus qualitative methodologies into a series of paired oppositional ‘affinities’ (Figure 1.2). When combined, this ensemble of opposed pairings spans what Abbott calls a methodological manifold that ultimately sets quantitative sociology and qualitative sociology at loggerheads with one another: Perhaps the strongest of these affinities is what we might call the methodological manifold: an affiliation of four or five separate distinctions generally labeled by the distinction of qualitative versus quantitative. At the heart of this manifold is the nearly absolute association of positivism with analysis and of narrative with interpretation. So strongly linked are these two dichotomies that the other way of lining them up – narrative with positivism and analysis with interpretation – is nearly nonsensical to most sociologists. … Analytic positivism is nearly always realist rather than constructivist in its epistemology, generally concerns social structure rather than culture, and usually has a strongly individual rather than emergent cast. Most of its proponents are strong believers in transcendent social knowledge. By contrast, narrative interpretation usually invokes culture (perhaps social structure as well), is willingly emergent, and nearly always follows a construction- ist epistemology. Most members of this school – from Blumer to Foucault – believe social knowledge ultimately to be situated, not transcendent. (Abbott 2001, p. 28) Abbott’s manifold is schematized in the lower portion of Figure 1.2. The lower rows are taken directly from Abbott’s work (2001, p. 28). The reversible arrows have been added to represent the dynamics of paradigmatic interaction and the fact that the paired elements of each affinity are antinomies. The upper two rows of Figure 1.2, Natur- wissenschaft versus Geisteswissenschaft and nomothetic methods versus ideographic meth- ods have been superimposed to underscore our contention that the quantitative versus qualitative methodological dispute is a finely grained duplication of the original neo- Kantian antinomies. Consequently, the repro- ductive dynamics of these various ‘wars of methodological position’ parallel those of so-called ‘stationary waves’, e.g. they are configurations that sustain their relative position and structural patterning vis-à-vis their immediate milieu, even as new materials and media traverse it. By our account, then, the continuing century-long imbroglio over the scientific legitimacy of case research finds a com- mon root in a permutating neo-Kantianism
- 46. COMPLEXITY AND CASE 23 Quantitative method versus Qualitative method Posiivism Interpretation Analysis Narrative Realism Constructionism Social structure Culture Individual level Emergent level Transcendent knowledge Situated knowledge Naturwissenschaft versus Geisteswissenschaft Nomothetic science versus Ideographic science Figure 1.2 Historical continuity of methodological affinities in sociology. This figure is a reconstruction of Abbott, i.e. lots added to Abbott’s original. that, even now, shapes the methodological controversies between ‘positivist’ sociology and its ‘humanist’ counterpart. According to our reconstruction, the history of that controversy began with Lebensphilosophie and Dilthey’s seminal distinction between the Naturwissenschaften and the Geisteswis- senschaften. This problematic was countered with Windelband’s division between the nomothetic sciences and the ideographic sciences. Chicago Sociology’s singular philo- sophical achievement was that it internalized these elements without resolving the anti- nomical ambiguities inherent inWindelband’s and Dilthey’s work. After the Second World War, the neo-Kantian hegemony remained intact, but was externally reconfigured to flow along lines of intense professional factionalism. The newly defined struggle emerged in an ideological space in which quantitative research commitments occupied one methodological pole, while the qualitative commitments of a ‘humanist sociology’ and its narrative traditions defended the other. Although a powerful force for the next five decades, Logical positivism’s arrival in the United States did not become, as first thought, the triggering mechanism of a Kuhnian revolution. Quite the contrary, while profoundly altering social science practice positivism never transcended its own neo- Kantian horizon and the former’s tendency to ‘discard’ontological questions as unscientific (Freidman 2000). COMPLEX REALISM AND SOCIAL ONTOLOGY It is neither unfair nor alarmist to say that the current methodological debate over the scientific status of case research is at an impasse. Nor is it out of order to suggest that the source of that impasse resides in the anti-ontological stance in which not only sociology, but modern science in general, is steeped. The British philosopher Roy Bhaskar has suggested as much when he asks: How did it come about that ontology was a virtually taboo subject? What was it that made
- 47. 24 THE SAGE HANDBOOK OF CASE-BASED METHODS ontology so difficult or even impossible? Although the philosophical doctrines of, say, Hume or Kant are familiar, it is important to appreciate a phenomenological or practical basis for them. This is in a kind of natural attitude to the world in which we do not distinguish between knowledge and objects, in which we don’t distinguish ontology from epistemology … in which we just talk about a known world. When science is doing normal science, when things are going well, epistemically it is natural to adopt what I call this ‘natural attitude’ in which you don’t posit ontology on the one side and epistemology on the other: you just have knowledge and you take it that that knowledge is of the world. Ontology really only becomes relevant when you are not satisfied with knowledge, when what passes for knowledge is patently wrong or absurd. (Bhaskar 2007, p. 192) Bhaskar seems to have described the cur- rent attitude of many sociologists toward their work. Their anti-ontological stance is manifested in the failure of social scientific discourse systematically to ‘… distinguish between knowledge and objects, in which we don’t distinguish ontology from epis- temology …’. Moreover, such systematic glosses lead to a premature closure of self- critical possibilities. Most immediately, this truncation is magnified by the idea that Humean Empiricism, to the exclusion of more complex models of causation, forms a sufficient basis for scientific explanations. Consequently, when confronted with issues of philosophical or social ontology, the scientist too often falls into a certain obtuseness, insisting that ‘… you just have knowledge and you take it that [this] knowledge is of the world’. We believe that a complex realist framework of sociological analysis can correct this anti-ontological bias. As for complex realism proper, it emerged rather haphazardly during the social and political ferment of the 1960s. Responding to the implausible Weltanschauung of the era, it navigated between the exhausted claims of Logical positivism and the exaggerated promises of Culutral Hermeneutics. Situating itself between an empiricism that had been dismissive of ontology and a phenomenolog- ical method that was verging on solipsism, complex realism fashioned a social scientific worldviewfromelementsofaresurgentrealist philosophy, mathematical chaos theory, non- equilibrium thermodynamics, and punctuated equilibria models of evolutionary develop- ment (Sayer 1984, Prigogine and Stengers 1984, Eldredge 1985, Gleick 1987, Gould 1989, Waldrop 1992, Arthur 1994, Collier 1994, Prigogine 1996). The new science’s name reflected the intellectual hurly-burly from which it grew. That is, ‘complex realism’ was a colloquial compression of two terms: complexity theory and critical realism. The label critical realism had, in turn, been formed from an elision of two philosophical conceptions, e.g. transcen- dental realism and critical naturalism. Four fundamental principles subsequently charted the development of complex realism: • Transcendental realism: Complex realism holds that if the sciences are to remain coherent enterprises, their explanatory protocols must embrace, a priori, a realist ontology, i.e. they must of necessity assume the existence of an objective moment of experience, some part of which stands over and against our attempts to understand and manipulate our world. • Critical naturalism: Complex realism assumes the possibility of a ‘natural science of society’ but if, and only if, that science is properly qualified so as to allow active human agency a role in ‘co-producing’ its everyday institutional forms. • The ontological and hierarchical stratification of the world: The world is ontologically stratified. Hence, the objects of scientific inquiry (either naturally occurring or as products of social construction) form a loosely integrated hierarchy of openly structured, evolving systems, which are themselves ontologically layered and hierarchi- cally nested. • The asymptotic predictability of open-system outcomes: Because of this ontological layer- ing, actual causal processes operating in non- laboratory settings are contingently structured and temporally staggered. They seldom rival the regularity of causal conjunctions elicited in con- trolled experiments. Consequently, ‘causal laws’ in complex natural systems are tendential and are best expressed in normic statements or in counter-factual propositions.
- 48. COMPLEXITY AND CASE 25 Critical realism The philosophical core of the complex realist perspective originates in the works of Roy Bhaskar (1975/1997, 1979/1989, 1993). Bhaskar’s transcendental realist deduction of the world forms the Archimedean point of his critical philosophy. The best way to grasp the logic of his deduction is to note that Bhaskar takes his lead from the critical method of Immanuel Kant. Just as Kant stood Humean Empiricism on its head, Bhaskar stands Kant’s philosophy on its head to arrive at his ontologically real conception of the world. Recall that Kant’s ‘Copernican revolution’ hinged on an ‘epis- temological turn’. That is, Kant ironically accepted Hume’s empiricist epistemology, but only as a preparatory ploy for reconstructing his subjective idealist defense of Newtonian science and its categorical infrastructure. By contrast, Bhaskar’s Transcendental Realism begins with an ontological turn. While Kant began his demonstration of the validity of scientific knowledge with an ‘epistemological feint’, Bhaskar’s ontological turn accepts science’s account of the world in order to deduce from it the world’s ontologi- cal structure. Hence, he asks, ‘Given the unparalleled success of the physical sci- ences in describing and explaining events in the natural world (the social sciences being another matter altogether), what do the patterns and historical sequencing of significant scientific discoveries tell us about the structure and dynamics of nature itself?’ By accepting science’s successful methods and epistemological presuppositions of how best to study the world, Bhaskar attempts to tease from method the ontological structure of reality. Having posed the question, Bhaskar moves post-haste to critically examine the paragon of scientific methodology, e.g. the modern laboratory experiment. In doing so, he demonstrates that if experimentally derived descriptions of natural laws are to remain logically coherent, then they must be justified by appealing to the existence of a reality outside the laboratory. It is here, however, that a paradox emerges: the ‘world outside’ seldom comports itself with the same causal regularity and predictability reproduced in the laboratory. In nature’s open milieu, the constancy of causal sequences – the empiricist’s guarantor of nature’s ‘iron-clad laws’ – breaks down. If the scientific method is to be vindicated, the difference between replicating natural laws in the laboratory and their erratic actualization in everyday life demands a new accounting. For critical realism, that vindication begins with the recognition that the constant conjunction of cause and effect engineered in the laboratory is an ideal approximation of natural processes. This is because natural causation is the result of a complex knitting together of a plurality of entities and their powers. This joint production is further complicated by the fact that these causal powers are often ‘temporallystaggered’intheir‘causalfirings’. Hence, the ontological complexity of joint causal inputs and their actual timing seldom match the consistency of outcomes produced in the laboratory. A cursory inspection of the historical patterns of significant scientific discovery suggests that the historical sequencing of significant scientific discoveries involves a ‘deepening’ across time of already existing knowledge. This deepening points to the possibility that these complex causal entities are arranged in a series of hierarchically arranged ontological layerings. Hence, the ontological stratification of natural entities is both nested and hierarchically ordered vis-à- vis one another. Moreover, this ‘onion-like layering’ is structured so that more complex orders tend to emerge from more simply composed strata. In their turn, these newly emerged complexities form the seedbed from which even more complex unities are built.As importantly, despite this common generative chain, each successive stratum possesses an ‘organizational surplus’ that makes it and its powers ontologically autonomous of the strata from which it has emerged. The generality of this ontological complexity
- 49. 26 THE SAGE HANDBOOK OF CASE-BASED METHODS and its hierarchical ordering is summarized by the Nobel Laureate P.W. Anderson: The ability to reduce everything to simple funda- mental laws does not imply the ability to start from those laws and reconstruct the universe. In fact, the more the elementary particle physicists tell us about the nature of the fundamental laws, the less relevance they seem to have to the very real problems of the rest of science, much less to those of society. The constructionist hypothesis breaks down when confronted with the twin difficulties of scale and complexity. The behavior of large and complex aggregates of elementary particles, it turns out, is not to be understood in terms of a simple extrapolation of the properties of a few particles. Instead, at each level of complexity entirely new properties appear, and the understanding of the new behaviors requires research which I think is as fundamental in its nature as any other. That is, it seems to me, that one may array the sciences roughly linearly in a hierarchy … But this hierarchy does not imply that science X is ‘just applied Y’. At each stage entirely new laws, concepts and generalizations are necessary, … [thereby] … requiring inspirations and creativity to just as great a degree as in the previous one. Psychology is not applied biology, nor is biology applied chemistry. (Anderson 1972, p. 393) Using this image of the world as an open, evolving system of ontologically stratified entities, complex realism now moves to distinguish between: (1) what scientists say about the world; and (2) the world as it exists, i.e. the constellation of obdurate entities and dynamic mechanisms that stand over and against human accounts of them. Bhaskar labels these domains the tran- sitive and the intransitive dimensions of science. There is a close correspondence between the intransitive domain and the ontological structure of science’s investigative object and the transitive domain’s focus on scientific method and epistemology. Of the two, the intransitive dimension enjoys an autonomy not allotted to the transitive domain. This is because intransitivity is the fundamen- tal precondition upon which the transitive production of scientific knowledge is pred- icated. Intransitivity thus provides scientific understanding not only with its ontological object of investigation, but also with its raison d’être. The transitive dimension, by contrast con- sists of the historically shifting modes of scientific understanding, viz. its theories, its craft-based esoterica, and its technologies of discovery. Unlike its intransitive counterpart, science’s transitive domain is neither immune to the distorting effects of ruling ideologies nor free of the technological limitations imposed upon human understanding at any given historical moment. It is, instead, a collectively constructed instrument of social action that bears the material imprint of an epoch’s social division of labor and its intellectual mode of production. This transitivity of scientific praxis resides in the paramount fact that scientific problem- atics are conditioned by the communities that certify the investigator’s credentials and func- tions. The productive principles of science and its investigative ‘tricks of the trade’ are embedded in the folklore of the profession. Acting as a court of last resort, this community of adepts establishes the esoteric significance of individual programs and discoveries. Hence, while this ‘collegium’ cannot create science’s intransitive object, it can control the means of legitimating the expenditure of society’s collective labor and resources on one project as opposed to another. The scientific ontology of complex systems Critical realism’s deduction of the world’s stratified ontology and its distinction between the transitive and intransitive domains of science form a suitable launching point for discussing the social ontology of complex social formations. What is needed, though, is a scientific ontology that bridges the two. Complexity theory is just such a vehicle: it is a general systems theory, i.e. a science of open- ended, peripatetic entities capable of material and structural evolution (Bertalanffy 1969, pp. 30–53, 120–154). Complexity’s ontological foundations beginwithRosen’sdistinctionbetween simple systems and complex systems (Rosen 1991,
- 50. COMPLEXITY AND CASE 27 pp. 314–327). Western science has long been drawn to mechanist metatheories and it seems still to be the hegemonic preference in sociology. Classical Newtonian mechanics is prototypical of simple system problematics. Such a science is predicated upon five assumptions: (1) Classical systems are closed systems in the sense that exogenous forces are either con- trolled or are analytically excluded from causal consideration. (2) Once the initial conditions of a system’s location relative to a fixed set of spatio-temporal coordinates are known; and (3) Once the researcher has access to the universal laws governing such systems; then (4) Precise predictions of future system states can be made, and these predictions can be tested empirically. (5) Finally, this explanatory template is sustainable if and only if the combined causal forces driving the developmental dynamics are mechanically related, viz. they do not interact significantly among themselves. Given these constraints, the dynamics of simple systems can be expressed in terms of linear and reductive formats. They are ‘linear’ in the sense that the system’s dynamics can be causally expressed by a simple summing over the independent vectors driving the system. Sometimes called ‘superposition’, the resulting linear combination of causal factors entails the reductive assumption that simple systems as a class are analytically equal to the sum of their parts. When these conditions are met, simple systems are said to be ‘conservative’. Underpinning this conservative designation, though, is the axiomaticassumptionthatthespatio-temporal nexus framing the system’s movements are fixed, viz. the referential framework cannot, in-and-of-itself, evolve during the analytic span of the system. In contrast to the linear dynamics of simple systems, complex systems roughly conform to realism’s ontological deductions. Complex systems are composed of internally stratified elements whose causal powers reciprocally reinforce one another so as to generate a structurally coherent skein. Moreover, the actualsequencingoftheircausalcapacitiesare often staggered and uneven. Whether natural entities or sociohistorical formations (case- objects not excepted), this asynchronous sequencing of causal inputs confounds most linear explanations. Little wonder, then, that when sociologists confront complex social formations, they resort to ‘super-organic’ analogies, rather than to the classical, clock- work allusions celebrating Nature’s inherent parsimony. Differing sharply from simple systems, the organization of complexity is so tightly knotted by reciprocal causation and stabilizing redundancies that their adaptive integration approximates what Abbott (1999, pp. 196–205) has called a spatio-temporal interaction field. The most significant difference between simple and complex systems, however, is found in the latter’s evolutionary capacities. Under propitious circumstances, complex systems can undergo irreversible phase tran- sitions. Such sudden perturbations not only induce changes in the material and structural composition of the system itself but, on occasion, such changes actually entail the qualitative transformations of the evolution- ary phase space itself. Put more succinctly, the conditions under which complexity’s evolution occurs often requires the evolution of evolution itself. Finally, complexity’s evolution can be radically disjunctive. The crux of this dis- continuity resides in complexity’s sensitive dependence on initial conditions. Such sensi- tivity expresses itself when incremental shifts in the system’s initial conditions give rise to disproportionately large differences in its final state. Moreover, these small differences and their non-linear amplifications often originate as simple errors in iterative replication or some other series of accidental occurrences. Dissipative social systems and dissipative social theory Dissipative systems are complexly structured, material constellations whose growth and
- 51. 28 THE SAGE HANDBOOK OF CASE-BASED METHODS evolution are regulated by non-linear pro- cesses of evolution. Complex social systems (including case-objects) are considered a subset of dissipative systems because of the open interchange of matter and energy with their environment, as well as the role material production and ecological adaptation play in sustaining their livelihood. As it presently stands, dissipative systems theory provides the social scientist with a valuable heuristic for grasping the dynamics and long-term fate of social formations. It can aid sociologists in their understanding of social atrophy and the anomic disorgani- zation of social institutions, as well as their subsequent renewal. Dissipative systems theory was originally grounded in the Second Law of Thermo- dynamics and was the provenance of the relatively new discipline of non-equilibrium thermodynamics. Unlike the First Law of Thermodynamics, which posits the con- servation of matter and energy and their continual transformation of one into the other, the Second Law makes no such conservative assumption. Quite the contrary, it posits that, under certain conditions, thermodynamic systems inevitably descend into a state of structural disorder (entropy). For example, closed systems, by defini- tion, have no access to environments in which free energy or assimilable matter is available. With no resources by which they could renew and extend themselves, they will undergo organizational degrada- tion. In the jargon of thermodynamics, this progressive disorganization is due to an irreversible ‘accumulation of positive entropy’. One need only touch a hot tea-kettle to appreciate the point of the Second Law.As we know, only a fraction of the heat summoned from the flame of a gas stove to heat the water in the kettle is transferred to the kettle; and only a portion of that energy actually heats the water. The energy not working to heat the water is radiated into the ambient air as ‘wasted energy’: the heat is ‘dissipated’. But more than wasted heat and working efficiency are lost in the process. The complexity of the molecular structure of the natural gas being used to heat the water is itself structurally degraded into the combustible by- products of oxidation. Hence, if the stove/ tea-kettle/ambient air were a ‘closed system’ then, along with the mechanical loss due to heat transfer, positive entropy would also have accumulated via oxidation. And as the system was ‘closed’ – bereft of renewable resources – the growth of positive entropy would be irreversible. That is, the system would have moved itself a step closer toward being a homogeneous milieu devoid of struc- tural form. In such instances, that spatially homogeneous distribution of energies would be called ‘thermal equilibrium’. When dissipative systems are transported to an ‘open milieu’, the implications of the Second Law shift dramatically. In open- system settings, dissipative organizations have hypothetical access to environmental stores of assimilable matter and free energy: these materials can be used as potential resources for fostering dissipative system development.As they increase their structural complexity, they secure, ceteris paribus, the means for staving off their immanent descent into thermal equilibrium. And, in special cases, certain open system milieus may be rich enough and robust enough to set into motion a countervailing dynamic that can blunt, or even temporarily reverse, the system’s descent into entropic equilibrium. Evolution is just such a counter-tendency. Its ascendant tendency toward increased com- plexification represents the natural possibility that new levels of order and novelty can offset extant levels of immanent disorder. Positive entropy’s nemesis, so to speak, takes the form of an increasing accumu- lation of negative entropy or ‘negantropy’. Whether as organisms, symbiotically inte- grated ecologies, or sociocultural structures, dissipative systems are material embodi- ments of organizational complexity. Taken as systemic totalities, dissipative forma- tions are concrete, spatio-temporal sites in which the dialectics of entropic decay and negantropic structuration constantly play off one another.
- 52. COMPLEXITY AND CASE 29 Recognizing the analytic potential of this dialectical process, Ilya Prigogine and his collaborators in the Brussells School (Prigogine and Stingers 1984, Nicolis and Prigogine 1989) constructed a general, four- part paradigm for investigating the dissipative dialectics inherent in complex systems. The modelposits:(1)aboundeddissipativesystem whose developmental complexity is driven by the dialectically tensed accumulations of positive entropy and negantropy; (2) the system’s immediate environment, from which matter and energy transfers flow back and forth across the system’s boundaries; (3) the systemic ‘metabolism’of that ‘free-energy’to generate structural growth and, occasionally, give way to evolutionary complexity; and (4) the capacity of the dissipative system to slough off the ‘waste by-products’ (positive entropy) of system complexification into its immediate environment. Under the assumptions of this model, open, complex systems are able to appropriate environmental resources and, in varying degrees, to convert them into increased struc- tural complexity (negantropic accumulation). When sufficient accumulations of negantropic complexity outweigh the dissipative system’s ability to slough off its positive entropy, a possible take-off point toward emergent evolutionary development is possible. Such dissipative entities are then said to be in a far-from-equilibrate state. When, on the other hand, positive entropy accumulations outstrip negantropic growth, then the dissi- pative system tends toward a steady-state thermal equilibrium. A third possibility, near- to-equilibrate systems, enables a dissipative system to sustain itself in a state of home- ostatic equilibrium. Unable to evolve to a far-from-equilibrate regime, they nonetheless are able to stave off descent into thermal equilibrium. The hallmark of these near-to- equilibrate systems is their ability to settle into cycles in which they regularly oscillate between two or more fixed reference states. We now possess a heuristic for modeling the dissipative dialectics of structural order and disorder in complex social systems. More to the point, we have a plausible scientific ontology for reconnoitering the presumed structure and dynamics of case-objects. Of the three dissipative possibilities just listed, far-from-equilibrate, dissipative systems are of greatest interest to sociology. Case-objects may be provisionally likened to ontologically complex entities that live off their material and social environments and thereby risk entropically degrading their relations with each. As dissipative entities, case-objects are ‘unstable’ in the sense that they are driven by boundary-testing engines that perpetually explore their evolutionary options. When the case-object evolves, this shift may be precipitated as much from within as from without. Consequently, much of the anomic disorder encountered in social formations, case-objects included, is the price exacted for continually seeking transitory adaptations. This anomie is itself a sign of institutional entropy. It exacerbates the case-object’s far- from-equilibrate tendencies and, as such, sets in motion new adaptive quests. TRANSITIVE AND INTRANSITIVE CASE Two implications concerning the social ontol- ogy of case-objects flow from the foregoing discussion of dissipative systems and their scientific ontology. First, case-objects are best approached as if they were complex, dissipative systems, as opposed to simple mechanical ensembles. Second, the transfor- mational dynamics of case-objects follow the regimen associated with far-from-equilibrate, dissipative systems. They are, in short, evolving, sociohistorical entities and should be treated as such. Ontologically, case-objects are minimally bounded, complex, dissipative systems that interact with their material and social environments. Moreover, as we will argue below, they are the joint product of the intentional activities of historical actors as well as of inertial institutions. Taken together, the contradictory ‘push and pull’ of agency and social structure form the sociological source of the case-object’s proactive innovations.
- 53. 30 THE SAGE HANDBOOK OF CASE-BASED METHODS The social ontology of the case-object When approached from the combined per- spectives of complex realism’s assumptions concerning the ontology of the case-object and the epistemological bracketing of casing developed in What Is a Case?, four aspects of casing and the case-object can be used to map the intransitive domain of case research. They are: (1) The case-object must support a self-standing narrative: The case-object must be a sufficiently self-contained entity that it can provide, prima facie, meaningful analytic closure. The case- object must also be ontologically complex enough to sustain a relatively well-formed explanatory narrative. (2) The case-object must be a minimally integrated social system: The case-object should be a well- bounded, internally nested set of ontologically layered structures that form a minimally integrated constellation. This social integration usually occurs via structural redundancies and ‘backups’ or functioning reciprocities. It need not, of necessity, be consensually integrated. Indeed, such stabilization could just as readily result from systematic class coercion, a tran- sitory truce or cooptation between competing ethnic or racial factions, or a hegemonic stalemate between symmetrically empowered social or cultural elites. (3) Case-objects are open, historically evolving systems: As a nested, ontologically stratified, entity, the case-object must be in continual exchange with its material habitat and its var- iegated social environments. As an emergently organized entity, case-objects are historically constituted and reconstituted in social space and social time. They have the self-directed capacity to evolve, albeit at uneven rates of transformation. (4) Case-objects are dialectically reproduced by human intentionality and by institutional stasis: If the sociological investigation of the case- object is to remain scientifically viable, the intentionality of human agency must be granted equal recognition to that given to the ability of social institutions to coercively channel or exploit the self-directed motives of human agents. Moreover, the chaos that inheres in everyday life is often due to as much as to burgled efforts at institutional control as it does because of the anarchic ‘perversity’ authored by human agency. In the final analysis, though, just as these four characteristics of case-objects are irreducible to one another, so the case-object as an emergent totality is itself an irreducible entity in its own right. A complex realist paradigm At this point, Bhaskar’s critical naturalism provides a way for plumbing the ontological complexity of the case-object. In this context, the term ‘naturalism’ denotes the positivist presupposition that social facts may be investigated using the same methods as those employed in the physical sciences. This naive naturalism becomes critical naturalism when the constitutive paradox of sociology as a science is confronted. That is, any science of society worthy of the name must reconcile the fact that the coercive powers of historically situated social institutions are always effectuated by human agents and their discretionary conduct. Within limits, agents intentionally interpret (or misinterpret) institutional missions and thereby filter soci- ety’s ‘objective powers’through their situated understandings. Add to the contingencies the serendipitous errors of applying institutional edictsorofanomiethatisinducedbylingering institutions whose powers no longer produce taken-for-granted outcomes give a a fair sense of the ontological complexity social scientists face when pursuing their craft. In sorting out these complications, com- plex realists follow Bhaskar’s (1979/1989, pp. 31–37) lead. He begins by selecting two classic, but diametrically opposed, accounts of social life, what he calls the ‘Durkheimian’ and the ‘Weberian stereotypes’. Playing off their inverted but mutually exclusive accounts, he develops his own realist concep- tion of sociology’s intransitive domain. In this analytic context, the term ‘stereotype’ denotes the partial nature of the scientific truths the Durkheimian and Weberian paradigms express. From these one-sidedabstractions,Bhaskarthendevelops
- 54. COMPLEXITY AND CASE 31 Socialization Reproduction/ transformation Society Individuals Figure 1.3 The transformational model of social action (TMSA). a dialectically framed naturalism by sublating the contradictory presuppositions of each stereotype into an integrated sociological whole. The results of his reconnaissance are shown in Figure 1.3. The Durkheimian stereotype consists of the triad Society → Socialization Processes → Individual. In accordance with Durkheimian positivism, the downward-directed arrows represent the coercive power of Durkheimian ‘social facts’, which morally constrain the destructive ‘appetites’ of atomized individuals during periods of social and cultural breakdown. The one-sided, Weberian stereotype reads, Individuals → Production/Transformation → Society; it is located on the right side of Figure 1.3. The inverse flow of the Weberian causal vectors signifies the weight Weberian action theory places on human subjectivity and its cultural productions. It should be noted in passing that the TMSA’s opposed triads and their dispositional flows conceptually parallel or overlap several of the dualisms already discussed in this paper, viz. nomothetic versus ideographic modes of enquiry, positivist versus social constructivist accounts of social life, and quantitative versus qualitative methodologies. At this point, Peter Berger’s duality of ‘Society creates Man’ versus ‘Man creates Society’can be added (Berger and Luckmann 1966, Bhaskar 1979/1989, pp. 31–33). Bhaskar uses Berger’s unmediated dualism as a foil for completing his dialectical conception of society. Within the context of the TMSA, man no more ‘creates’ society than society ‘creates’ man. Instead, both are ontologically irreducible to one another. As different as their respective modus vivendi are, their powers are complementary, so much so that they form the necessary preconditions for each other’s existence. Hence, humans are intentional agents whose self-directed activities: … reproduce or transform [society]. That is, if society is always already made, then any concrete human praxis, or, if you like, act of objectivation can only modify it; and the totality of such acts sustain and change it. It is not the product of their activity (any more, I shall argue, than human action is completely determined by it). Society stands to individuals, then, as something that they never make, but that exists only in virtue of their activity. (Bhaskar 1979/1989, p. 34) By contrast, society, for all its power to induce social conformity, is at best an abstract potentiality. Its very real but slumbering powers remain virtually inert without the productive intercession of human agency, individual or collective. Hence: Society, then, is an articulated ensemble of tenden- cies and powers which, unlike natural ones, exist only as long as they (or at least some of them) are
- 55. Other documents randomly have different content
- 56. high stage of the river gave to its surface an elevation above that of the natural bank; but a continuous levee to protect the land from inundation existed on both sides of the river. When the ascending fleet approached these batteries, a cross-fire, which drove two of the vessels back, was opened upon it, and continued until all the ammunition was exhausted. The garrison was then withdrawn- casualties, one killed and one wounded. The regret which would naturally arise from the fact of these batteries not having a sufficient supply of ammunition is modified, if not removed, by the statement of the highly accomplished and gallant officer, Major-General M. L. Smith, who was then in command of them. He reported: Had the fall of New Orleans depended upon the enemy's first taking Forts Jackson and Philip, I think the city would have been safe from an attack from the Gulf. The forts, in my judgment, were impregnable as long as they were in free and open communication with the city. This communication was not endangered while the obstruction existed. The conclusion, then, is briefly this: While the obstruction existed, the city was safe; when it was swept away, as the defenses then existed, it was within the enemy's power. On the other hand, General Duncan, whose protracted, skillful, and gallant defense of the forts is above all praise, closes his official report with the following sentence: Except for the cover afforded by the obscurity of the darkness, I shall always remain satisfied that the enemy would never have succeeded in passing Forts Jackson and St. Philip. The darkness to which he referred was not only that of night, but also the absence of the use of the means prepared to light up the river. As further proof of the intensity of the darkness, and the absence of that intelligent design and execution which had been
- 57. claimed, I will quote a sentence from the report of Commodore Farragut: At length the fire slackened, the smoke cleared off, and we saw to our surprise that we were above the forts. On the 25th of April the enemy's gunboats and ships of war anchored in front of the city and demanded its surrender. Major- General M. Lovell, then in command, refused to comply with the summons, but, believing himself unable to make a successful defense, and in order to avoid a bombardment, agreed to withdraw his forces, and turn it over to the civil authorities. Accordingly, the city was evacuated on the same day. The forts still continued defiantly to hold their position. By assiduous exertion the damage done to the works was repaired, and the garrisons valiantly responded to the resolute determination of General Duncan and Colonel Higgins to defend the forts against the fleet still below, as well as against that which had passed and was now above. On the 26th Commodore Porter, commanding the mortar-fleet below, sent a flag-of-truce boat to demand the surrender of the forts, saying that the city of New Orleans had surrendered. To this Colonel Higgins replied, April 27th, that he had no official information that New Orleans had been evacuated, and until such notice was received he would not entertain for a moment a proposition to surrender the forts. On the same day General Duncan, commanding the coast- defenses, issued the following address: SOLDIERS OF FORTS JACKSON AND ST. PHILIP: You have nobly, gallantly, and heroically sustained with courage and fortitude the terrible ordeals of fire, water, and a hail of shot and shell wholly unsurpassed during the present war. But more remains to be done. The safety of New Orleans and the cause of the Southern Confederacy—our homes, families, and
- 58. everything dear to man—yet depend upon our exertions. We are just as capable of repelling the enemy to-day as we were before the bombardment. Twice has the enemy demanded your surrender, and twice has he been refused. Your officers have every confidence in your courage and patriotism, and feel every assurance that you will cheerfully and with alacrity obey all orders, and do your whole duty as men and as becomes the well-tried garrisons of Forts Jackson and St. Philip. Be vigilant, therefore, stand by your guns, and all will yet be well. J. K. DUNCAN, Brigadier-General, commanding coast-defenses. Not less lofty and devoted was the spirit evinced by Colonel Higgins. His naval experience had been energetically applied in the attempts to preserve and repair the raft. As immediate commander of Fort St. Philip he had done all which skill and gallantry could achieve, and, though for forty-eight hours during the bombardment he never left the rampart, yet, with commendable care for his men, he kept them so under cover that, notwithstanding the long and furious assault to which the fort was subjected, the total of casualties in it was two killed and four wounded. Their conduct was such as was to be anticipated, for, had these officers been actuated by a lower motive than patriotism, had they been seeking the rewards which power confers, they would not have taken service with the weaker party. Their meed was the consciousness of duty well done in a righteous cause, and the enduring admiration and esteem of a people who had only these to confer.
- 59. During the 25th, 26th, and 27th, there had been an abatement of fire on the forts, and with it had subsided the excitement which imminent danger creates in the brave. A rumor became current that the city had surrendered, and no reply had been received to inquiries sent on the 24th and 25th. About midnight on the 27th the garrison of Fort Jackson revolted en masse, seized upon the guard, and commenced to spike the guns. Captain S. O. Comay's company, the Louisiana Cannoneers of St. Mary's Parish, and a few others remained true to their cause and country. The mutiny was so general that the officers were powerless to control it, and therefore decided to let those go who wished to leave, and after daybreak to communicate with the fleet below and negotiate for the terms which had been previously offered and declined. Under the incessant fire to which the forts had been exposed, and the rise of the water in the casemates and lower part of the works, the men had been not only deprived of sleep, but of the opportunity to prepare their food. Heroically they had braved alike dangers and discomfort; had labored constantly to repair damages; to extinguish fires caused by exploding shells; to preserve their ammunition by bailing out the water which threatened to submerge the magazine: yet, in a period of comparative repose, these men, who had been cheerful and obedient, as suddenly as unexpectedly, broke out into open mutiny. Under the circumstances which surrounded him, General Duncan had no alternative. It only remained for him to accept the proposition which had been made for a surrender of the forts. As this mutiny became known about midnight of the 27th, soon after daylight of the 28th a small boat was procured, and notice of the event was sent to Captain Mitchell, on the Louisiana, and also to Fort St. Philip. The officers of that fort concurred in the
- 60. propriety of the surrender, though none of their men had openly revolted. A flag of truce was sent to Commodore Porter to notify him of a willingness to negotiate for the surrender of the forts. The gallantry with which the defense had been conducted was recognized by the enemy, and the terms were as liberal as had been offered on former occasions. The garrisons were paroled, the officers were to retain their side- arms, and the Confederate flags were left flying over the forts until after our forces had withdrawn. If this was done as a generous recognition of the gallantry with which the forts had been defended, it claims acknowledgment as an instance of martial courtesy—the flower that blooms fairest amid the desolations of war. Captain Mitchell, commanding the Confederate States naval forces, had been notified by General Duncan of the mutiny in the forts and of the fact that the enemy had passed through a channel in rear of Fort St. Philip and had landed a force at the quarantine, some six miles above, and that, under the circumstances, it was deemed necessary to surrender the forts. As the naval forces were not under the orders of the general commanding the coast-defenses, it was optional with the naval commander to do likewise or not as to his fleet. After consultation with his officers. Captain Mitchell decided to destroy his flagship, the Louisiana, the only formidable vessel he had, rather than allow her to fall into the hands of the enemy. The crew was accordingly withdrawn, and the vessel set on fire. Commodore Porter, commanding the fleet below, came up under a flag of truce to Fort Jackson, and, while negotiations were
- 61. progressing for the surrender, the Louisiana, in flames, drifted down the river, and, when close under Fort St. Philip, exploded and sank. The defenses afloat, except the Louisiana, consisted of tugs and river-steamers, which had been converted to war purposes by protecting their bows with iron so as to make them rams, and putting on them such armament as boats of that class would bear; and these were again divided into such as were subject to control as naval vessels, and others which, in compliance with the wish of the Governor of Louisiana and many influential citizens, were fitted out to a great extent by State and private sources, with the condition that they should be commanded by river-steamboat captains, and should not be under the control of the naval commander. This, of course, impaired the unity requisite in battle. For many other purposes they might have been used without experiencing the inconvenience felt when they were brought together to act as one force against the enemy. The courts of inquiry and the investigation by a committee of Congress have brought out all the facts of the case, but with such conflicting opinions as render it very difficult, in reviewing the matter, to reach a definite and satisfactory conclusion. This much it may be proper to say, that expectations, founded upon the supposition that these improvised means could do all which might fairly be expected from war-vessels, were unreasonable, and a judgment based upon them is unjust to the parties involved. The machinery of the Louisiana was so incomplete as to deprive her of locomotion, but she had been so well constructed as to possess very satisfactory resisting powers, as was shown by the fact that the broadsides of the enemy's vessels, fired at very close quarters, had little or no effect upon her shield. Without power of locomotion, her usefulness was limited to employment as a floating battery. The question as to whether she was in the right position, or whether, in
- 62. her unfinished condition, she should have been sent from the city, is one, for an answer to which I must refer the inquirer to the testimony of naval men, who were certainly most competent to decide the issue. One of the little river-boats, the Governor Moore, commanded by lieutenant Beverly Kennon, like the others, imperfectly protected at the bow, struck and sunk the Varuna, in close proximity to other vessels of the enemy's fleet. Such daring resulted in his losing, in killed and wounded, seventy-four out of a crew of ninety-three. Then finding that he must destroy his ship to prevent her from falling into the hands of the enemy, he set her on fire, and testified as follows: I ordered the wounded to be placed in a boat, and all the men who could to save themselves by swimming to the shore and hiding themselves in the marshes. I remained to set the ship on fire. After doing so, I went on deck with the intention of leaving her, but found the wounded had been left with no one to take care of them. I remained and lowered them into a boat, and got through just in time to be made a prisoner. The wounded were afterward attended by the surgeons of the Oneida and Eureka. This, he says, was the only foundation for the accusation of having burned his wounded with his ship. Another, the Manassas, lieutenant-commanding Warley, though merely an altered tug- boat, stoutly fought the large ships; but, being wholly unprotected, except at her bow, was perforated in many places, as soon as the guns were brought to bear upon her sides, and floated down the river a burning wreck. Another of the same class is thus referred to by Colonel Higgins:
- 63. At daylight, I observed the McRae, gallantly fighting at terrible odds, contending at close quarters with two of the enemy's powerful ships. Her gallant commander, Lieutenant Thomas B. Huger, fell during the conflict, severely, but I trust not mortally, wounded. This little vessel, after her unequal conflict, was still afloat, and, with permission of the enemy, went up to New Orleans to convey the wounded as well from our forts as from the fleet. On the 23d of April, 1862, General Lovell, commanding the military department, had gone down to Fort Jackson, where General Duncan, commanding the coast-defenses, then made his headquarters. The presence of the department commander did not avail to secure the full coöperation between the defenses afloat and the land-defenses, which was then of most pressing and immediate necessity. When the enemy's fleet passed the forts, he hastened back to New Orleans, his headquarters. The confusion which prevailed in the city, when the news arrived that the forts had been passed by the enemy's fleet, shows how little it was expected. There was nothing to obstruct the ascent of the river between Forts Jackson and St. Philip and the batteries on the river where the interior line of defense rested on its right and left banks, about four miles below the city. The guns were not sufficiently numerous in these batteries to inspire much confidence; they were nevertheless well served until the ammunition was exhausted, after which the garrisons withdrew, and made their way by different routes to join the forces withdrawn from New Orleans.
- 64. Under the supposition entertained by the generals nearest to the operations, the greatest danger to New Orleans was from above, not from below, the city; therefore, most of the troops had been sent from the city to Tennessee, and Captain Hollins, with the greater part of the river-fleet, had gone up to check the descent of the enemy's gunboats. Batteries like those immediately below the city had been constructed where the interior line touched the river above, and armed to resist an attack from that direction. Doubtful as to the direction from which, and the manner in which, an attempt might be made to capture the city, such preparations as circumstances suggested were made against many supposable dangers by the many possible routes of approach. To defend the city from the land, against a bombardment by a powerful fleet in the river before it, had not been contemplated. All the defensive preparations were properly, I think, directed to the prevention of a near approach by the enemy. To have subjected the city to bombardment by a direct or plunging fire, as the surface of the river was then higher than the land, would have been exceptionally destructive. Had the city been filled with soldiers whose families had been sent to a place of safety, instead of being filled with women and children whose natural protectors were generally in the army and far away, the attempt might have been justified to line the levee with all the effective guns and open fire on the fleet, at the expense of whatever property might be destroyed before the enemy should be driven away. The case was the reverse of the hypothesis, and nothing could have been more unjust than to censure the commanding General for withdrawing a force large enough to induce a bombardment, but insufficient to repel it. His answer to the demand for the surrender showed clearly enough the motives by which he was influenced. His refusal enabled him to
- 65. withdraw the troops and most of the public property, and to use them, with the ordnance and ordnance stores thus saved, in providing for the defense of Vicksburg, but especially it deprived the enemy of any pretext for bombarding the town and sacrificing the lives of the women and children. It appears that General Lovell called for ten thousand volunteers from the citizens, but failed to get them. There were many river-steamboats at the landing, and, if the volunteers called for were intended to man these boats and board the enemy's fleet before their land-forces could arrive, it can not be regarded as utterly impracticable. The report of General Butler shows that he worked his way through one of the bayous in rear of Fort St. Philip to the Mississippi River above the forts so as to put himself in communication with the fleet at the city, and to furnish Commodore Farragut with ammunition. From this it is to be inferred that the fleet was deficient in ammunition, and the fact would have rendered boarding from river-boats the more likely to succeed. In this connection it may be remembered that, during the war, John Taylor Wood, Colonel and A. D. C. to the President, who had been an officer of high repute in the old Navy, did in open boats attack armed vessels, board and capture them, though found with nettings up, having been warned of the probability of such an attack.[57] Many causes have been assigned for the fall of New Orleans. Two of them are of undeniable force: First, the failure to light up the channel; second, the want of an obstruction which would detain the fleet under fire of the forts. General Duncan's report and testimony justify the conclusion that to the thick veil of darkness the enemy was indebted for his ability to run past the forts. The argument that the guns were not of sufficiently large caliber to stop the fleet is not convincing. If all the guns had been of the
- 66. largest size, that would not have increased the accuracy but would have diminished the rapidity of the fire, and therefore in the same degree would have lessened the chances of hitting objects in the dark. Further, it appears that the forts always crippled or repulsed any vessels which came up in daylight. The forts would have been better able to resist bombardment if they had been heavily plated with iron; but that would not have prevented the fleet from passing them as they did. Torpedoes might have been placed on the bar at the mouth of the river before the enemy got possession of it, and subsequently, if attached to buoys, they might have been used in the deep channel above. Many other things which were omitted might and probably would have been done had attention been earlier concentrated on the danger which at last proved fatal. If the volunteer river-defense fleet was ineffective, as alleged, because it was not subject to the orders of the naval commander, that was an evil without a remedy. The Governor of Louisiana had arranged with the projectors that they should not be subject to the naval commander, and the alternative of not accepting them with that condition was that they would not agree to convert their steamers into war-vessels. Unless, therefore, it can be shown that they were worse than none, their presence can not be properly enumerated among the causes of the failure. The fall of New Orleans was a great disaster, over which there was general lamentation, mingled with no little indignation. The excited feeling demanded a victim, and conflicting testimony of many witnesses most nearly concerned made it convenient to select for censure those most removed and least active in their own justification. Thus the naval constructors of the Mississippi and the Secretary of the Navy became the special objects of attack. The
- 67. selection of these had little of justice in it, and could not serve to relieve others of their responsibility, as did the old-time doom of the scapegoat. New Orleans had never been a ship-building port, and when the Messrs. Tift, the agents to build the iron-dad steamer Mississippi, arrived there, they had to prepare a ship-yard, procure lumber from a distance, have the foundries and rolling-mills adapted to such iron-work as could be done in the city, and contract elsewhere for the balance. They were ingenious, well informed in matters of ship-building, and were held in high esteem in Georgia and Florida, where they had long resided. They submitted a proposition to the Secretary of the Navy to build a vessel on a new model. The proposition was accepted after full examination of the plan proposed, the novelty of which made it necessary that they should have full control of the work of construction. To the embarrassments above mentioned were added interruptions by calling off the workmen occasionally for exercise and instruction as militiamen, the city being threatened by the enemy. From these causes, unexpected delay in the completion of the ship resulted, regret for which increased as her most formidable character was realized. These constructors—the brothers Tift—hoped to gain much reputation by the ship which they designed, and, from this motive, agreed to give their full service and unremitted attention in its construction without compensation or other allowance than their current expenses. It would, therefore, on the face of it, seem to have been a most absurd suspicion that they willingly delayed the completion of the vessel, and at last wantonly destroyed it. Mr. E. C. Murray, who was the contractor for building the Louisiana, in his testimony before a committee of the Confederate
- 68. Congress, testified that he had been a practical ship-builder for twenty years and a contractor for the preceding eighteen years, having built about a hundred and twenty boats, steamers, and sailing-vessels. There was only a fence between his shipyard and that where the Mississippi was constructed. Of this latter vessel he said: I think the vessel was built in less time than any vessel of her tonnage, character, and requiring the same amount of work and materials, on this continent. That vessel required no less than two million feet of lumber, and, I suppose, about one thousand tons of iron, including the false works, blockways, etc. I do not think that amount of materials was ever put together on this continent within the time occupied in her construction. I know many of our naval vessels, requiring much less materials than were employed in the Mississippi, that took about six or twelve months in their construction. She was built with rapidity, and had at all times as many men at work upon her as could work to advantage— she had, in fact, many times more men at work upon her than could conveniently work. They worked on nights and Sundays upon her, as I did upon the Louisiana, at least for a large portion of the time. The Secretary of the Navy knew both of the Tifts, but had no near personal relations or family connection with either, as was recklessly alleged. He, in accepting their proposition, connected with it the detail of officers of the navy to supervise expenditures and aid in procuring materials. Assisted by the chief engineer and constructor of the
- 69. navy, minute instructions were given as to the manner in which the work was to be conducted. As early as the 19th of September he sent twenty ship-carpenters from Richmond to New Orleans to aid in the construction of the Mississippi. On the 7th of October authority was given to have guns of heaviest caliber made in New Orleans for the ship. Frequent telegrams were sent in November, December, and January, showing great earnestness about the work on the ship. In February and March notice was given of the forwarding from Richmond of capstan and main-shaft, which could not be made in New Orleans. On March 22d the Secretary, by telegraph, directed the constructors to strain every nerve to finish the ship, and added, work day and night. April 5th he again wrote: Spare neither men nor money to complete her at the earliest moment. Can not you hire night-gangs for triple wages? April 10th the Secretary again says: Enemy's boats have passed Island 10. Work day and night with all the force you can command to get the Mississippi ready. Spare neither men nor money. April 11th he asks, When will you launch, and when will she be ready for action? These inquiries indicate the prevalent opinion, at that time, that the danger to New Orleans was from the ironclad fleet above, and not the vessels at the mouth of the river; but the anxiety of the Secretary of the Navy and the efforts made by him were of a character applicable to either or both the sources of danger. Thus we find as early as the 24th of February, 1862, that he instructed Commander Mitchell to make all proper exertions to have guns and carriages ready for both the iron-clad vessels the Mississippi and the Louisiana. Reports having reached him that the work on the latter vessel was not pushed with sufficient energy, on the 15th of March he authorized Commander Mitchell to consult with General Lovell, and, if the contractors were not doing everything practicable to complete her at the earliest moment, that
- 70. he should take her out of their hands, and, with the aid of General Lovell, go on to complete her himself. On the 5th of April, 1862, Secretary Mallory instructed Commander Sinclair, who had been assigned to the command of the Mississippi, to urge on by night and day the completion of the ship. In March, 1861, the Navy Department sent from Montgomery officers to New Orleans, with instructions to purchase steamers and fit them for war purposes. Officers were also sent to the North to purchase vessels suited to such uses, and in the ensuing May an agent was dispatched to Canada and another to Europe for like objects; and in April, 1861, contracts were made with foundries at Richmond and New Orleans to make guns for the defense of New Orleans. On the 8th of May, 1861, the Secretary of the Navy communicated at some length to the Committee on Naval Affairs of the Confederate Congress his views in favor of iron-clad vessels, arguing as sell for their efficiency as the economy in building them, believing that one such vessel could successfully engage a fleet of the wooden vessels which constituted the enemy's navy. His further view was that we could not hope to build wooden fleets equal to those with which the enemy were supplied. The committee, if it should be deemed expedient to construct an iron-clad ship, was urged to prompt action by the forcible declaration, Not a moment should be lost. Commander George Minor, Confederate States Navy, Chief of the Bureau of Ordnance, reported the number of guns sent by the Navy Department to New Orleans, between July 1, 1861, and the fall of the city, to have been one hundred and ninety-seven, and that before July twenty-three guns had been sent there from Norfolk, being a total of two hundred and twenty guns, of which forty-five were of large caliber, supplied by the Navy Department for the defense of New Orleans.
- 71. Very soon after the Government was removed to Richmond, the Secretary of the Navy, with the aid of Commander Brooke, designed a plan for converting the sunken frigate Merrimac into an iron-clad vessel. She became the famous Virginia, the brilliant career of which silenced all the criticisms which had been made upon the plan adopted. On May 20, 1861, the Secretary of the Navy instructed Captain Ingraham, Confederate States Navy, to ascertain the practicability of obtaining wrought-iron plates suited for ships' armor. After some disappointment and delay, the owners of the mills at Atlanta were induced to make the necessary changes in the machinery, and undertake the work. Efforts at other places in the West had been unsuccessful, and this was one of the difficulties which an inefficient department would not have overcome. The iron- clad gunboats Arkansas and Tennessee were commenced at Memphis, but the difficulty in obtaining mechanics so interfered with their construction, that the Secretary of the Navy was compelled, December 24, 1861, to write to General Polk, who was commanding at Columbus, Kentucky, asking that mechanics might be detached from his forces, so as to insure the early completion of the vessels. So promptly had the iron-clad boats been put under contract, that the arrangements had all been made in anticipation of the appropriation, and the contract was signed on the very day the law was passed. On December 25, 1861, Lieutenant Isaac N. Brown, Confederate States Navy, a gallant and competent officer, well and favorably known in his subsequent service as commander of the ram Arkansas, was sent to Nashville. Information had been received that four river-boats were there, and for sale, which were suited for river defense. Lieutenant Brown was instructed to purchase such as
- 72. should be adaptable to the required service, and to proceed forthwith with the necessary alteration and armament. In the latter part of 1861, it having been found impossible with the means in Richmond and Norfolk to answer the requisitions for ordnance and ordnance stores required for the naval defenses of the Mississippi, a laboratory was established in New Orleans, and authority given for the casting of heavy cannon, construction of gun- carriages, and the manufacture of projectiles and ordnance equipments of all kinds. On December 12, 1861, the Secretary of the Navy submitted an estimate for an appropriation to meet the expenses incurred for ordnance and ordnance stores for the defense of the Mississippi River. Secretary Mallory, in answer to inquiries of a joint committee of Congress, in 1863, replied that he had sent a telegram to Captain Whittle, April 17, 1862, as follows: Is the boom, or raft, below the forts in order to resist the enemy, or has any part of it given way? State condition. On the next day the following answer was sent: I hear the raft below the forts is not in best condition; they are strengthening it by additional lines. I have furnished anchors. To further inquiry about the raft by the Committee, the Secretary answered: The commanding General at New Orleans had exclusive charge of the construction of the raft, or obstruction, in
- 73. question, and his correspondence with the War Department induced confidence in the security of New Orleans from the enemy. I was aware that this raft had been injured, but did not doubt that the commanding General would renew it, and place an effectual barrier across the river, and I was anxious that the navy should afford all possible aid. . . . A large number of anchors were sent to New Orleans from Norfolk for the raft. Though much more might be added, it is hoped that what has been given above will sufficiently attest the zeal and capacity of the Secretary of the Navy, and his anxiety, in particular, to protect the city of New Orleans, whether assailed by fleets descending or ascending the river. Having thus reviewed at length the events, immediate and remote, which were connected with the great catastrophe, the fall of our chief commercial city, and the destruction of the naval vessels on which our hopes most rested for the protection of the lower Mississippi and the harbors of the Gulf, the narrative is resumed of affairs at the city of New Orleans. [Footnote 57: Captain Wood had a number of light row-boats built, holding each about twenty men. They were fitted with cradles to wagons, and could be quickly moved to any point by road or rail. He writes: In August, 1863, I left Richmond with four boats and sixty men for the Rappahannock, to look after one or two gunboats that had been operating in that river. Finding always two cruising together, I determined to attempt the capture of both at once. About midnight, with muffled oars, we pulled for them at anchor near the mouth of the river. They discovered us two hundred yards off. We
- 74. dashed alongside, cut our way through and over the boarder nettings with the old navy cutlass, gained the deck, and, after a sharp, short fight, drove the enemy below. The prizes proved to be the gunboats Satellite and Reliance, two guns each. Landing the prisoners, we cruised for two days in the Chesapeake Bay. A number of vessels were captured and destroyed.]
- 75. CHAPTER XXIX. Naval Affairs (continued).—Farragut demands the Surrender of New Orleans.—Reply of the Mayor.—United States Flag hoisted.— Advent of General Butler.—Barbarities.—Antecedents of the People.— Galveston.—Its Surrender demanded.—The Reply.—Another visit of the Enemy's Fleet.—The Port occupied.—Appointment of General Magruder.—Recapture of the Port.—Capture of the Harriet Lane.— Report of General Magruder.—Position and Importance of Sabine Pass.—Fleet of the Enemy.—Repulse by Forty-four Irishmen.— Vessels captured.—Naval Destitution of the Confederacy at first.— Terror of Gunboats on the Western Rivers.—Their Capture.—The most Illustrious Example.—The Indianola.—Her Capture.—The Ram Arkansas.—Descent of the Yazoo River.—Report of her Commander.— Runs through the Enemy's Fleet.—Description of the Vessel.— Attack on Baton Rouge.—Address of General Breckinridge.—Burning of
- 76. the Arkansas. Sad though the memory of the fall of New Orleans must be, the heroism, the fortitude, and the patriotic self-sacrifice exhibited in the eventful struggle at the forts must ever remain the source of pride and of such consolation as misfortune gathers from the remembrance of duties well performed. After the troops had been withdrawn and the city restored to the administration of the civil authorities, Commodore Farragut, on April 26, 1862, addressed the Mayor, repeating his demand for the surrender of the city. In his letter he said: It is not within the province of a naval officer to assume the duties of a military commandant, and added, The rights of persons and property shall be secured. He proceeded then to demand that the emblem of sovereignty of the United States be hoisted over the City Hall, Mint, and Custom-House by meridian this day. All flags and other emblems of sovereignty other than those of the United States must be removed from all the public buildings by that hour. To this the Mayor replied, and the following extracts convey the general purport of his letter: The city is without the means of defense, and is utterly destitute of the force and material that might enable it to resist an overpowering armament displayed in sight of it. . . . To surrender such a place were an idle and unmeaning ceremony. . . . As to hoisting any flag other than the flag of our own adoption and allegiance, let me say to you that the man lives not in our midst whose hand and heart would not be paralyzed at the mere thought of such an act; nor could I
- 77. find in my entire constituency so wretched and desperate a renegade as would dare to profane with his hand the sacred emblem of our aspirations. . . . Peace and order may be preserved without resort to measures which I could not at this moment prevent. Your occupying the city does not transfer allegiance from the government of their choice to one which they have deliberately repudiated, and they yield the obedience which the conqueror is entitled to extort from the conquered. Respectfully, JOHN T. MONROE, Mayor. On the 29th of April Admiral Farragut adopted the alternative presented by the answer of the Mayor, and sent a detachment of marines to hoist the United States flag over the Custom-House, and to pull down the Confederate flag from the staff on the City Hall. An officer and some marines remained at the Custom-House to guard the United States flag hoisted over it until the land-forces under General Butler arrived. On the 1st of May General Butler took possession of the defenseless City; then followed the reign of terror, pillage, and a long train of infamies, too disgraceful to be remembered without a sense of shame by any one who is proud of the name American. Had the population of New Orleans been vagrant and riotous, the harsh measures adopted might have been excused, though nothing could have justified the barbarities which were practiced; but, notable as the city had always been for freedom from tumult, and occupied as it then was mainly by women and children, nothing can extenuate the wanton insults and outrages heaped upon them. That
- 78. those not informed of the character of the citizens may the better comprehend it, a brief reference is made to its history. When Canada, then a French colony, was conquered by Great Britain, many of the inhabitants of greatest influence and highest cultivation, in a spirit of loyalty to their flag, migrated to the wilds of Louisiana. Some of them established themselves in and about New Orleans, and their numerous descendants formed, down to a late period, the controlling element in the body-politic. Even after they had ceased, because of large immigration, to control in the commercial and political affairs of the city, their social standard was still the rule. No people were more characterized by refinement, courtesy, and chivalry. Of their keen susceptibility the Mayor informed Commodore Farragut in his correspondence with that officer. When the needy barbarians of the upper plains of Asia descended upon the classic fields of Italy, their atrocities were such as shocked the common-sense of humanity; but, if any one shall inquire minutely into the conduct of Butler and his followers at New Orleans, he will find there a history yet more revolting. Soon thereafter, on May 17, 1862, Captain Eagle, United States Navy, commanding the naval forces before Galveston, summoned it to surrender, to prevent the effusion of blood and the destruction of property which would result from the bombardment of the town, adding that the land and naval forces would appear in a few days. The reply was that, when the land and naval forces made their appearance, the demand would be answered. The harbor and town of Galveston were not prepared to resist a bombardment, and, under the advice of General Herbert, the citizens remained quiet,
- 79. resolved, when the enemy should attempt to penetrate the interior, to resist his march at every point. This condition remained without any material change until the 8th of the following October, when Commander Renshaw with a fleet of gunboats, consisting of the Westfield, Harriet Lane, Owasco, Clifton, and some transports, approached so near the city as to command it with his guns. Upon a signal, the Mayor pro tem, came off to the flag-ship and informed Commander Renshaw that the military and civil authorities had withdrawn from the town, and that he had been appointed by a meeting of citizens to act as mayor, and had come for the purpose of learning the intentions of the naval commander. In reply he was informed that there was no purpose to interfere with the municipal affairs of the city; that he did not intend to occupy it before the arrival of a military commander, but that he intended to hoist the United States flag upon the public buildings, and claim that it should be respected. The acting Mayor informed him that persons over whom he had no control might take down the flag, and he could not guarantee that it should be respected. Commander Renshaw replied that, to avoid any difficulty like that which occurred in New Orleans, he would send with the flag a sufficient force to protect it, and would not keep the flag flying for more than a quarter or half an hour. The vessels of the fleet were assigned to positions commanding the town and the bridge which connected the island with the mainland, and a battalion of Massachusetts volunteers was posted on one of the wharves. Late in 1862 General John B. Magruder, a skillful and knightly soldier, who had at an earlier period of the year rendered distinguished service by his defense of the peninsula between the
- 80. James and York Rivers, Virginia, was assigned to the command of the Department of Texas. On his arrival, he found the enemy in possession of the principal port, Galveston, and other points upon the coast. He promptly collected the scattered arms and field artillery, had a couple of ordinary high-pressure steamboats used in the transportation of cotton on Buffalo Bayou protected with cotton- bales piled from the main deck to and above the hurricane-roof, and these, under the command of Captain Leon Smith, of the Texas Navy, in coöperation with the volunteers, were relied upon to recapture the harbor and island of Galveston. Between night and morning on the 1st of January, 1863, the land-forces entered the town, and the steamboats came into the bay, manned by Texas cavalry and volunteer artillery. The field artillery was ran down to the shore, and opened fire upon the boats. The battalion of the enemy having torn up the plank of the wharf, our infantry could only approach them by wading through the water, and climbing upon the wharf. The two steamboats attacked the Harriet Lane, the gunboat lying farthest up the bay. They were both so frail in their construction that their only chance was to close and board. One of them was soon disabled by collision with the strong vessel, and in a sinking condition ran into shoal water. The other closed with the Harriet Lane, boarded and captured the vessel. The flag-ship Westfield got aground and could not be got off, though assisted by one of the fleet for that purpose. General Magruder then sent a demand that the enemy's vessels should surrender, except one, on which the crews of all should leave the harbor, giving until ten o'clock for compliance with his demand, to enforce which he put a crew on the Harriet Lane, then the most efficient vessel afloat of the enemy's fleet, and, while waiting for an answer, ceased firing. This demand was communicated by a boat from the Harriet Lane to the
- 81. commander on the Clifton, who said that he was not the commander of the fleet, and would communicate the proposal to the flag-officer on the Westfield. Flags of truce were then flying on the enemy's vessels, as well as on shore. Commander Renshaw refused to accede to the proposition, directing the commander of the Clifton to get all the vessels, including the Corypheus and Sachem, which had recently joined, out of port as soon as possible, and that he would blow up the Westfield, and leave on the transports lying near him with his officers and crew. In attempting to execute this purpose, Commander Renshaw and ten or fifteen others perished soon after leaving the ship, in consequence of the explosion being premature. The General commanding made the following preliminary report: HEADQUARTERS, GALVESTON, TEXAS. This morning, the 1st January, at three o'clock, I attacked the enemy's fleet and garrison at this place, captured the latter and the steamer Harriet Lane, two barges, and a schooner. The rest, some four or five, escaped ignominiously under cover of a flag of truce. I have about six hundred prisoners and a large quantity of valuable stores, arms, etc. The Harriet Lane is very little injured. She was carried by boarders from two high-pressure cotton-steamers, manned by Texas cavalry and artillery. The line troops were gallantly commanded by Colonel Green, of Sibley's brigade, and the ships and artillery by Major Leon Smith, to whose indomitable energy and heroic daring the country is indebted for the successful execution of a plan which I had considered for the destruction of the enemy's fleet. Colonel Bagby, of Sibley's brigade, also commanded the volunteers from his regiment
- 82. for the naval expedition, in which every officer and every man won for himself imperishable renown. J. BANKHEAD MAGRUDER, Major General. The conduct of Commander Renshaw toward the inhabitants of Galveston had been marked by moderation and propriety, and the closing act of his life was one of manly courage and fidelity to the flag he bore. Commander Wainright and Lieutenant-commanding Lea, who fell valiantly defending their ship, were buried in the cemetery with the honors of war: thus was evinced that instinctive respect which true warriors always feel for their peers. The surviving officers were paroled. It would be a pleasing task, if space allowed, to notice the many instances of gallantry in this affair, as daring as they were novel, but want of space compels me to refer the reader to the full accounts which have been published of the cavalry charge upon a naval fleet. The capture of the enemy's fleet in Galveston Harbor, by means so novel as to excite surprise as well as grateful admiration, was followed by another victory on the coast of Texas, under circumstances so remarkable as properly to be considered marvelous. To those familiar with the events of that time and section, it is hardly necessary to say that I refer to the battle of Sabine Pass.
- 83. The strategic importance to the enemy of the possession of Sabine River caused the organization of a large expedition of land and naval forces to enter and ascend the river. If successful, it gave the enemy short lines for operation against the interior of Texas, and relieved them of the discomfiture resulting from their expulsion from Galveston Harbor. The fleet of the enemy numbered twenty-three vessels. The forces were estimated to be ten thousand men. No adequate provision had been made to resist such a force, and, under the circumstances, none might have been promptly made on which reliance could have been reasonably placed. A few miles above the entrance into the Sabine River, a small earthwork had been constructed, garrisoned at the time of the action by forty-two men and two lieutenants, with an armament of six guns. The officers and men were all Irishmen, and the company was called the Davis Guards. The captain, F. H. Odlum, was temporarily absent, so that the command devolved upon Lieutenant E. W. Dowling. Wishing to perpetuate the history of an affair, in which I believe the brave garrison did more than an equal force had ever elsewhere performed, I asked General Magruder, when I met him after the war, to write out a full account of the event; he agreed to do so, but died not long after I saw him, and before complying with my request. From the publications of the day I have obtained the main facts, as they were then printed in the Texas newspapers, and, being unwilling to summarize the reports, give them at length. Captain F. H. Odlum's Official Report. HEADQUARTERS, SABINE PASS,
- 84. September 9, 1863. Captain A. N. MILLS, Assistant Adjutant-General. SIR: I have the honor to report that we had an engagement with the enemy yesterday and gained a handsome victory. We captured two of their gunboats, crippled a third, and drove the rest out of the Pass. We took eighteen fine guns, a quantity of smaller arms, ammunition and stores, killed about fifty, wounded several, and took one hundred and fifty prisoners, without the loss or injury of any one on our side or serious damage to the fort. Your most obedient servant, F. H. ODLUM, Captain, commanding Sabine Pass. Commodore Leon Smith's Official Report. Captain E. P. TURNER, Assistant Adjutant-General. SIR: After telegraphing the Major-General before leaving Beaumont, I took a horse and proceeded with all haste to Sabine Pass, from which direction I could distinctly hear a heavy firing. Arriving at the Pass at 3 P.M., I found the enemy off and inside the bar, with nineteen gunboats and steamships and other ships of war, carrying, as well as I could judge, fifteen thousand men. I proceeded with Captain Odlum to the fort, and found Lieutenant Dowling and Lieutenant N. H. Smith, of the engineer corps, with forty-two men, defending the fort. Until 3 P.M. our men did not open on the enemy, as the range was too distant. The officers of the fort coolly held
- 85. their fire until the enemy had approached near enough to reach them. But, when the enemy arrived within good range, our batteries were opened, and gallantly replied to a galling and most terrific fire from the enemy. As I entered the fort, the gunboats Clifton, Arizona, Sachem, and Granite State, with several others, came boldly up to within one thousand yards, and opened their batteries, which were gallantly and effectively replied to by the Davis Guards. For one hour and thirty minutes a most terrific bombardment of grape, canister, and shell was directed against our heroic and devoted little band within the fort. The shot struck in every direction, but, thanks be to God! not one of the noble Davis Guards was hurt. Too much credit can not be awarded Lieutenant Dowling, who displayed the utmost heroism in the discharge of the duty assigned him and the defenders of the fort. God bless the Davis Guards, one and all! The honor of the country was in their hands, and nobly they sustained it. Every man stood at his post, regardless of the murderous fire that was poured upon them from every direction. The result of the battle, which lasted from 3.30 to 5 P.M., was the capturing of the Clifton and Sachem, eighteen heavy guns, one hundred and fifty prisoners, and the killing and wounding of fifty men, and driving outside the bar the enemy's fleet, comprising twenty-three vessels in all. I have the honor to be your obedient servant, LEON SMITH, Commanding Marine Department of Texas.
- 86. HEADQUARTERS DISTRICT OF TEXAS, NEW MEXICO, AND ARIZONA, HOUSTON, TEXAS, September 9, 1863. (SPECIAL ORDER.) Another glorious victory has been won by the heroism of Texans. The enemy, confident of overpowering the little garrison at Sabine Pass, boldly advanced to the work of capture. After a sharp contest he was entirely defeated, one gunboat hurrying off in a crippled condition, while two others, the Clifton and Sachem, with their armaments and crews, including the commander of the fleet, surrendered to the gallant defenders of the fort. The loss of the enemy has been heavy, while not a man on our side has been killed or wounded. Though the enemy has been repulsed in his naval attacks, his land-forces, reported as ten thousand strong, are still off the coast waiting an opportunity to land. The Major-General calls on every man able to bear arms to bring his guns or arms, no matter of what kind, and be prepared to make a sturdy resistance to the foe. Major-General J. B. MAGRUDER. EDMUND P. TURNER, Assistant Adjutant-General. The Daily Post, Houston, Texas, of August 22, 1880, has the following: A few days after the battle each man that participated in the fight
- 87. was presented with a silver medal inscribed as follows: On one side 'D. G.,' for the Davis Guards, and on the reverse Side, 'Sabine Pass, September 8, 1863.' Captain Odlum and Lieutenant R. W. Dowling have gone to that bourn whence no traveler returns, and but few members of the heroic band are in the land of the living, and those few reside in the city of Houston, and often meet together, and talk about the battle in which they participated on the memorable 8th of September, 1863. The following are the names of the company who manned the guns in Fort Grigsby, and to whom the credit is due for the glorious victory: Lieutenants R. W. Dowling and N. H. Smith; Privates Timothy McDonough, Thomas Dougherty, David Fitzgerald, Michael Monahan, John Hassett, John McKeefer, Jack W. White, Patrick McDonnell, William Gleason, Michael Carr, Thomas Hagerty, Timothy Huggins, Alexander McCabe, James Flemming, Patrick Fitzgerald, Thomas McKernon, Edward Pritchard, Charles Rheins, Timothy Hurley, John McGrath, Matthew Walshe, Patrick Sullivan, Michael Sullivan, Thomas Sullivan,
- 88. Welcome to our website – the perfect destination for book lovers and knowledge seekers. We believe that every book holds a new world, offering opportunities for learning, discovery, and personal growth. That’s why we are dedicated to bringing you a diverse collection of books, ranging from classic literature and specialized publications to self-development guides and children's books. More than just a book-buying platform, we strive to be a bridge connecting you with timeless cultural and intellectual values. With an elegant, user-friendly interface and a smart search system, you can quickly find the books that best suit your interests. Additionally, our special promotions and home delivery services help you save time and fully enjoy the joy of reading. Join us on a journey of knowledge exploration, passion nurturing, and personal growth every day! ebookbell.com