BOOKS ET AL.
28 FEBRUARY 2014 VOL 343 SCIENCE www.sciencemag.org 970
Science is an endeavor of artful simpli- fication, of discovering general princi- ples. Fruit flies, for example, look very
different from humans, but the two species
share 75% of disease-causing genes. Thus,
we can study fruit flies to learn about our own
weaknesses under the common framework of
genetics. Similarly, the unifying principles
of allometry allow us to make sense of most
forms of life on Earth, where a cat is just a
blown-up mouse and a human
is just a blown-up cat. Multiply an animal’s size by x, and
its heart rate, lifetime, and
other features will increase in
proportion to specific powers
of x. This is the case because
the physical rules that govern
the formation of circulatory
systems hold everywhere. Turning to social
phenomena, can we find universal processes
Let us take a big leap and consider cities:
Perhaps something similar to biology’s allometric rules would allow a megacity as complicated as New York to be better understood
by looking at what goes on in a small village. Is
New York just a blown-up version of Venice?
Universal principles that govern urban shapes
and growth processes independent of particular history or geography could support a “grand unified theory of cities.” What would
it take to discover them?
The New Science of Cities
presents a herculean attempt to
bring together widely fragmented
approaches to making sense of
human social organization with
the goal of eventually establish-
ing a consolidated “science of
cities” able to answer our ques-
tions. Michael Batty bases his
argument on the interplay among
space, dynamics, and relations.
He holds that “to understand
place, we must understand flows,
and to understand flows we must
understand networks.” Batty (a
geographer at University College
London) also stresses two other
principles: an intrinsic order of scale deter-
mines a city’s form and function, and a sci-
ence of cities should not merely observe but
also predict. The book draws on the work of
urbanists, economists, mathematicians, and
physicists as well as almost five decades of his
own contributions to urban studies.
Batty’s approach rests on a physicalist philosophy: He focuses on what can be immediately observed. In a treatise that eventually
becomes technical, he simplifies cities as “sets of actions,
interactions, and transactions.”
The book develops a toolbox of
applicable to “big data” that are
being increasingly collected
about cities—from a mesmerizing potpourri of paradigms.
Its foundational part takes us on a fascinating
historical journey through these perspectives,
from urban economics and transportation to
fractal geometry, dynamical systems, and network science. All these aspects can be embedded in complexity theory, the most likely candidate to provide the consistent philosophy for
achieving the author’s ambitious goal.
The foundations for modern urban stud-
ies were laid in the 19th century, with the
first analytical, economic approaches and
the concept of agglomeration. The closer we
live together, the more opportunities of trade
we find, creating economies of scale. Today,
extensive empirical evidence supports an
urban allometry that convincingly shows cit-
ies are humanity’s socioeconomic reactors,
in which citizens produce, consume, and
interact more as the city increases in size.
These interactions can be explained mech-
anistically using Newtonian metaphors of
cities as entities that exert gravitational-like
forces on populations. Such gravity models
are used in various social sciences and have
recently experienced a boom in the study of
human mobility through mobile phone data.
Powerful metaphors populate the histories of urban studies and planning. For a
long time, the prevailing line of thinking saw
the city as a machine, a model in which a
“controller steered the city cybernetically
toward a future desired state.” Now we know
better, realizing that cities are complex systems meandering through a space of configurations, like biological organisms evolving
in a Darwinian fitness landscape. Such systems exhibit hallmark features of nonlinear
dynamics far from equilibrium. Small interventions can have massive, counterintuitive
consequences, making cities almost impossible to control. In a field where planning
is essential, this discovery is devastating.
It means that monumental top-down plans,
which dominated most of 20th-century city
planning, are a recipe for failure.
The book provides some
better options. Parsimonious
bottom-up models and agent-based simulations unveil basic
mechanisms of competition and
growth. Cutting-edge visualizations facilitate the exploration
and the interpretation of empirical and synthetic data. Through
a synthesis of the descriptive
and the normative and a combination of quantitative modeling with qualitative theories of
social exchange and collective
action, the author shows how to
model urban design to improve
the decision processes of authorities and policy-makers.
Batty stresses that the set of
approaches he tackles is a first
step and by no means exclusive:
“it would be presumptuous to
think of this effort as the only
science of cities, for the city and
its planning admits many viewpoints.” He concludes that an
The reviewer is at the SENSEable City
Laboratory, Massachusetts Institute of
Technology, Cambridge, MA 02139, USA.
The New Science
by Michael Batty
MIT Press, Cambridge, MA,
2013. 518 pp. $45, £31.95.
Urban complexity. Human flows (mapped by Eric Fischer using geotagged
tweets) illustrate the organic structure of New York City.