The New Science of Cities

New science of cities
Author(s) Michael Batty
Publisher MIT Press
Year 2013

The New Science of Cities, dripping with mathematical models and equations interwoven with an accessible narrative, provides readers with good insight into tools that can be used to model urban change. It builds on this by illustrating the possible application of these tools to contribute to a better future for cities.  Much of what Batty reveals in his book is built on well-established ideas stemming from earlier work such as Fractal Cities: A Geometry of Form and Function (with Paul Longley; Academic Press 1994) and Understanding Cities Through Cellular Automata, Agent-Based Models, and Fractals (MIT Press, 2005).  However, far from a dreary re-examination of previous research and old work, Batty provides new insights into contemporary urban analysis and, in particular, shows how mathematical modelling can be applied to provide new insight into how cities work.

The book concentrates on three interrelated themes, namely temporal dynamics (how things change), flows (how strongly locations interact with one another) and size and scale (how things change as they get bigger or smaller).  Early on, Batty challenges our conventional understanding of cities, with their structure represented as models, plans, maps and pictures and instead paints a picture of cities as more akin to complex organisms. The implication being that representing cities through the use of static media, not only loses their dynamism, but also the complex relations and interactions between activities. Batty makes a strong case regarding the shortfalls of the more conventional approaches to analysing large volumes of data and is able to use a variety of mathematical models and formulae to demonstrate interactions, flows and networks between different aspects of the city structure.

The first part of this three-part book focusses on providing an extensive background regarding interactions, connections and correlations related to cities. Batty emphasises the importance of understanding the city differently by looking at aspects such as agglomeration, decentralisation and globalisation from the perspective of understanding the complex set of interactions at play across space and time. This part of his book highlights the fact that cities cannot effectively be understood as systems.  The city systems approach implies a closed system or at least a system that freely adjusts.  Batty points to the fact that cities are not manufactured, they are grown and evolve as a result of many individual decisions being taken over an extended period of time.  Batty suggests that the city and its ever changing nature is closely aligned to the field of complexity science. His position challenges traditional highly structured top-down planning approaches and opens the reader to a more organic bottom-up, but evidence based understanding of cities.

Part two of the book moves on to the aspect of understanding cities in their current state, and deals with themes such as hierarchies, networks, urban structure, distance, fractal growth and form and urban simulation. Batty implements scaling laws in explaining city size distributions, hierarchies within and between cities and then drops down to a very detailed scale involving networks within the city. In doing so, there is a heavy focus on examining interactions, relations, connectivity and accessibility between objects, rather than physical infrastructure and hard form. The last two chapters complete this part of the book with simulations of city growth, development, flows and relationships.  The focus here is on the use of tools such as cellular automata and computer models to reflect overall patterns derived from dynamic processes occurring at finer spatial scales. This presents insights into these macro-patterns, whilst allowing for their disaggregation and a more in-depth understanding of often divergent processes at play at the micro-scale.

The final part of the book moves from understanding cities to their planning and design. The focus here is on the introduction of models that lead to better design and decision making through the involvement of all the role players in the process of reaching consensus. These chapters highlight how the use of normative perspectives can assist in opening possibilities to participants in the decision making process, regarding the potential future of the city. Aspects such as opinion pooling and weighting the importance of opinions are at the centre of this part of the book. Batty argues that for the use of models as a decision support tool, but is careful not to present them as a replacement for the decision making process. He does however present interesting perspectives on how techniques and models can be utilised to open windows on a new and interactive way of thinking about planning and design and ultimately better solutions for the challenges facing cities.

The New Science of Cities should be on the desks of every urban geographer, planner and urban researcher in South Africa.  Currently, there is a growing need for evidence based planning to support policy development and related interventions that aim to affect change in our cities and towns. In this respect, there are increasing volumes of data being collected for our cities. Indeed we have more information than ever and at a scale of detail not possible a few decades ago. However, at issue is the ability to translate this information into knowledge and then to interpret and apply this knowledge in a manner that informs intervention. The application of advanced mathematical models and geospatial and statistical analysis tools suggested in Batty’s work assists to bridge these gaps by revealing patterns, trends and possibilities that might otherwise not be available to decision makers. Whilst the models and analysis techniques reflected should not be viewed as the only way of representing and analysing data they present a clear advance in this field.  Batty perhaps sums it up best by stating that the models and design methods will attempt to provide insight into the evident complexity of designing future cities, but he notes that ‘those looking for an integrated science that is nicely packaged and available to apply immediately and without qualification to city problems will be disappointed’ (p117).


Amanda van Eeden is a GIS analyst and lecturer at the Centre for Rural and Urban Innovation and Statistical Exploration (CRUISE) at the University of Stellenbosch. She is responsible for the GIS and Spatial Statistical modules of the MPhil in Urban and Regional Science course. Amanda is a geographer with a research interest in the application of spatial analysis and spatial statistical tools and methodologies in the field of urban and regional science. 

Read older posts from this section

Leave a Reply