|This site maintained by: Aomar Boum. Site last updated on October, 2001.||
Journal of Political Ecology:
Case Studies in History and Society
VOLUME 6 (1999)
Human Demography and Disease, by Susan Scott and Christopher Duncan (1998) New York: Cambridge University Press. 354 pp.
Reviewed by Robert D Hoppa, Department of Anthropology, University of Manitoba, Winnipeg, Manitoba, Canada
Demography and Disease represents what I believe to be a much welcome
and much needed addition to the broad field of historical demography.
As a reference work or teaching manual for quantitative methods for analysing
historical demographic data, the book is excellent. Drawing heavily from
their own research on epidemic diseases and time series analysis of parish
records, the authors present a series of case studies to illustrate the
application, methods, and interpretation of various statistical approaches
to understanding the relationships between mortality, disease and environment
(e.g. weather, economic conditions etc) over long periods of time. While
their case studies are very specific, for the most part they are easily
transferable to other kinds of analyses that individuals might be interested
book is not for the total novice, however. In places it becomes extremely
technical and requires a solid grasp of the material within the chapters.
All the same, its technical depth is what makes the volume most useable
as both a reference work for individual researchers and a training manual
for courses in quantitative history, historical demography or related
topics like population ecology/historical geography or epidemiology.
two chapters of the book essentially provide an introduction to the general
subject areas and the sources of analysis. Chapter 1 provides an extremely
brief and generalized overview of the fundamental relationship between
mortality and morbidity in populations, and acts as a backdrop for demonstrating
the very broad readership that this book is geared to. Chapter 2 is also
an introductory chapter, but here presenting some comments on the principal
source of data most readers will be using (parish records) and the principal
methodological approach explored throughout the book ö time series
analysis. This chapter sets the overall tone for the volume, and in my
opinion is the litmus test for readers. If you shy away from the book
after reading this brief chapter, this book probably isn't for you. However,
I suspect that for many readers like myself, reading Chapter 2 will result
in a response along the lines of "Hey, that's what I'd like to
be able to do with my data. Let's see how this works."
next eleven chapters provide illustrations of various aspects of time
series analysis for historical demography. While they appear at first
glance somewhat modular in structure, this is in fact a strength of the
book. Each chapter contains numerous cross-references to either previous
or forthcoming chapters, and because each deals with a specific analytical
approach, many people will find that they are able to jump to individual
chapters that are more directly relevant to their own research questions
and the type of analysis they would like to undertake.
Of key importance in this book is the idea of short wavelength versus long wavelength oscillations in historical demographic data. Understanding and interpreting these differences form the basis of Chapters 3 through 8. Chapter 3 provides an introduction to the anatomy of time series analysis using a familiar and relatively standard data type.
way, it is an extremely useful chapter for those not familiar with time-series
analysis. Specifically, this chapter introduces the concept of intrinsic
versus extrinsic mortality cycles. While much of the later chapters increase
in complexity, Chapter 3 provides a solid introduction that acts as a
foundation on which subsequent chapters build. Chapters 4 and 5 explore
one of the real advantages of mathematical epidemiology for historical
demography ö the ability to model population processes and then compare
model prediction or outcomes to observable outcomes from the data themselves.
Chapters 6 through 8 provide a series of illustrations for understanding
the impact of short-wave or exogenous mortality cycles (grains prices,
wool prices) and their interactions with the general demographic features
of a population (fertility rates, infant mortality etc) and long-wave
or exogenous cycles associated with epidemic diseases. One of the nice
features of this book is again illustrated in this section whereby subsequent
chapters continue to build on evidence presented in previous case studies,
while at the same time remaining independent enough for use as individual
9 through 13 present a series of cases studies focused primarily on smallpox,
measles and Whooping Cough in London, to illustrate various approaches
to modelling disease processes in historical populations. These are probably
the most rigorous chapters for the non-specialist as they provide a series
of in-depth illustrations of the power of mathematical epidemiology for
addressing historical demographic questions. Key issues addressed include
inter-epidemic interval, changes in age-specific susceptibility, population
structure, the effects of co-dependent variables like temperature or nutrition
and even interactions among multiple diseases. The latter is particular
relevant as too often analyses model the impact of specific diseases independent
of one another. While this adds to computational ease it is an undesirable
approach since we know that causes of death do not operate independently
of one another.
14 presents a large-scale illustration of the analytical techniques presented
in the previous 5 chapters. This forms as a lead-in to the final chapter
that argues succinctly for the power of time-series methods for meta-analyses
in historical demography. Here the authors reiterate that one of the strengths
of the techniques presented in this volume is the ability to differentiate
endogenous mortality cycles related to broader interactions of demographic
structure versus exogenous mortality cycles that can be detected independently,
but likely have interactions with other changes socio-economic factors.
Overall, the book is highly recommended to anyone interested in learning more quantitative approaches to disease-mortality-environment interactions in historical demographic data. Useable as both a reference source and a teaching manual, it will serve as an invaluable research tool for historical demographers, quantitative historians, historical geographers, historical epidemiologists, and related specialists.