Back in the Loop: Complexity in International Relations

There is substantial agreement in the International Relations (IR) community that its object of study is ‘complex’. Notwithstanding, IR inquiry continues to operate under theoretical and methodological premises that minimize complexity effects, system dynamics, and non-linearity.  This article considers that this inconsistency has hampered systemic thinking within IR arguments and models, independently of theoretical orientation. The article argues that System Dynamics modelling (SD) provides a sophisticated and flexible meta-theoretical framework capable of tackling this problem. Drawing from this body of knowledge, the article presents a heuristic platform – a set of conceptual notions and graphic notation – to schematize and model causal mechanisms, feedback structures, and nonlinear behaviours in world politics. This platform, it is claimed, has potential to complement existing IR arguments, theories and methods and enhance the manner in which complexity effects are analyzed and communicated, while narrowing the distance between the complexity of global affairs and the conceptual and methodological tools available to deal with it.

The Holobiontic Human Genome

Our human genome defines us. It lies at the very core of our being. To understand it is to 
know, in the most intimate sense, what it means to be human. Only recently have we come to understand the human genome in sufficient depth and complexity to be able to put together an overall account of its marvellous story – only to discover that there is rather more to it than DNA alone. Many will be familiar with the celebrated announcement in 2001, by two competing scientific organisations, that the mysteries of the human genome had been fully unravelled. In fact, though it was an epochal scientific milestone, much still remained obscure. The astonishing thing was that this pioneering first glimpse into core of our being brought to light mystery after mystery. Where virtually all of the emphasis for half a century had been on protein-coding genes, only 1.5% of our DNA was found to code for this. Huge chunks of the genome comprised strange genetic entities that closely resembled the genomes of retroviruses – the same family of viruses as HIV-1, the cause of AIDS. Stranger still, fully half of our human DNA was classed as unknown. These monumental unknowns had been mistakenly dismissed as junk by generations of geneticists. Today we are beginning to unravel these mysteries and in doing so coming to understand our genome and its workings at a deeper, more enlightening level. This amounts to an important and fascinating transformation of knowledge. It also opens wide a remarkable new window of opportunity for understanding of our human evolution while making possible major new avenues of exploration of the genetic basic of human disease.

Understanding community resilience to multiple hazards in mountain areas

In the Alps, natural hazards are part of everyday life. Communities live with continuous risk and cope frequently impact events. Every year, different kinds of natural hazard events cause damages, losses and deaths. Over centuries, communities have developed means to cope. These became part of local society and culture, including aspects such as risk perception and the role of social networks, and shape the resilience of communities towards natural hazards. But how to assess, describe and measure these abilities and their influence on community resilience?
In this talk, we will define what we understand by community resilience, and present the conceptual framework for it we developed within the FP7 project emBRACE and the case study of the small alpine community Badia that in December 2012 was hit by an unusually big landslide. We would like to open our methodology for comment – and hopefully improvement – within the constraints of the project to assess, map and measure, through different inter- and trans-disciplinary methodologies, the knowledge and risk behaviour of population and risk management actors.
An important first observation from our fieldwork is that in the field of disaster risk reduction, social networks are made out of different sub networks, working at different levels. Therefore, it is important to understand multilevel networks (e.g. networks of affected population, networks of responsible authorities, local experts, and so on) and the connections and interfaces between them. As next step, we will compare the results of different data sources and different sub networks (e.g. networks from expert interviews, networks from population survey) to identify incongruence, disruptions and missing links and to answer the question how a resilient network should look like.

A design-led approach to Natural Computing

This talk explores new possibilities for making 21st century design practices as a response to a range of novel computing practices that aim to reach escape velocity from conventional digital computing paradigms and identify ways of counting, sorting and ordering that are more like Nature. A range of subjects will be discussed including a philosophical and ontological engagement with these systems, which may help us imagine and innovate differently. Additionally, design-led experiments and projects will also be presented that start to explore and suggest initial guidelines for what it means to design and engineer with living systems.

Status, deference, and information sharing among fishermen: How re-conceptualizing altruistic behaviour can change the way we understand natural resource use

The tragedy of the commons is ubiquitous in discussions of natural resource use and has led many to conclude that the only way to combat the depletion of ecological commodities such as fish, forests and farmland is through government intervention and privatization. However, in recent years scholars have 
begun to challenge the notion that resource users should be viewed as rational agents, locked in a doomed competition for their immediate individual benefit. In dispelling with the old model of pure self-interest a new avenue of research into the nature of other-regarding behaviour has emerged. However, as of yet there remains no systematic understanding of when and why resource users act altruistically, hindering attempts to model coupled social-ecological dynamics. Drawing on research in animal behaviour, evolutionary psychology, and natural resource governance, this paper argues that altruistic behaviour may have evolved in humans as a social learning strategy, and therefore should be heterogeneously expressed in groups based upon the status and success of members. Furthermore, these altruistic acts – which often consist of knowledge and information sharing – may allow groups of resource users to increase efficiency while minimize damaging practices compared to individuals acting independently. To test this theory, this paper introduces a new method for studying cooperation and altruism in natural resource systems through the use of multi-player video games.

Complexity analysis in social development and the social determinants of health  

Since the late 1980s, complexity theory has enjoyed a rapid adoption across many disciplines. The generality of complexity theory has led to different formulations in different areas of application. In this talk, I will focus on how complexity theory has been used in social development work and public health promotion. A brief overview of evaluation strategies used in these applications will also be given. The talk will close with a discussion of the potential benefits of mathematical and computer modeling approaches to complexity in these applications.

Summer School Presentations

The robust assembly of small symmetric nano-shells

Highly symmetric nano-shells are found in many biological systems, such as clathrin cages and viral shells. Many studies have shown that highly symmetric shells appear in nature as a result of free energy minimization of a generic interaction between their constituent subunits. In this talk, I show that the symmetric structures observed both in equilibrium simulations and in self-assembly experiments can readily grow from identical subunits under non equilibrium conditions. Our minimal model of nano-shell assembly shows that the spontaneous curvature regulates the size of the shell while the mechanical properties of the subunit determines the type of symmetry of the assembled structure. Understanding the minimum requirements for the formation of closed nano-shells is a necessary step toward engineered nano-containers, which will have far reaching impact in both material science and medicine.

Accurate Host Landscapes for Modelling the Spread of Plant Diseases

Host landscapes are used to describe the spatial patterns and locations of infectious or susceptible plants in a heterogeneous environment. Inaccuracies may exist in host landscapes because the data they are constructed from are incomplete, noisy or inconsistent. Inaccuracies typically occur due to errors in attributes, spatial positioning or temporal validity. It is unclear how much of an effect these errors have on models for the spread of plant diseases or whether some errors are more detrimental to modelling predictions than others. Errors and inaccuracies are inherent to a certain degree in all spatial data. Therefore, unless we understand how inaccuracies in the host landscape affect our models, it is difficult to have confidence in their recommendations for control. This presentation considers potential errors that may be introduced into the landscape, describes ways to detect these errors and methods to check whether the predictions of our models are correct. Although these issues are discussed from an epidemiological perspective, the techniques described in this presentation are relevant to a wide range of spatial modelling applications, including ecology, demographics and urban planning.

Public Confidence Modelling: a locally based approach to police performance management

In the UK, the ability of the police to function effectively depends on the authority that they can command, rather than the force that they can deploy. It is therefore essential that police retain their legitimacy in their role of defenders of the British Public. This can only be achieved through maintaining and improving levels of public confidence in policing.
Previous work on the determinants of public confidence has mainly used data either at a national level, or from large, metropolitan urban areas. Little research has been carried out examining public confidence within smaller urban areas. This paper presents the results of a survey evaluating the views of 1322 residents of York around a number of areas including their perceptions and attitudes towards crime, their local area, and the police. The study uses Structural Equation Modelling as the methodology to develop two models of public confidence aimed at assessing the main factors affecting public confidence in policing in York.
Contrary to previous work, it is found that only a small subset of the studied factors have a significant effect on public confidence. Instead, it is the interactions between these factors that provide evidence for a novel framework for public confidence in smaller urban areas. Due to the increasing focus on evidence based policing worldwide, several potential strategies are presented with the aim of improving both public confidence and community safety within York and other similar areas.

Getting answers to unasked questions using modelling to unpack ideas and assumptions about peasant farming and fishing in Africa

Games play presents implicit answers to questions that the game 'asks' of the player. Using Netlogo and modelling4all we are developing approaches which can be customised in situ and which facilitate data collection (data on livelihoods and the costs of production) and on how people conceive of their environment in a context of increased climate variability. Replay offers the prospect of improved feedback loops

Perception-for-action, Action-for-perception

In this talk I will review the evidence that when we simply look at objects, such as a coffee cup, the actions we would typically produce, such as grasping it, are automatically activated in the brain.  Similarly, when we observe another person’s actions we covertly simulate their actions.  For example, when we observe a person kick a soccer ball, motor areas in our brain that control our own feet become activated and actions with the same body part (foot) are made more quickly.  It has been proposed that such action simulation processes aid our understanding of objects and people. For example, preparing to produce an action can facilitate perception of certain objects, pain in a body part can impair the perception of another person’s action, and motor simulation of another person’s actions can aid empathy and prediction. Hence states of our motor system aid understanding of our visual world.

 A composite Bayesian hierarchical model of compositional data with zeros

We present an approach to modelling compositional data with large concentrations of zeros and several levels of variation, applied to a database of elemental compositions of forensic glass of various use types. The procedure consists of (i) partitioning the dataset in subsets characterized by the same pattern of presence/absence ofb chemical elements and (ii) fitting a Bayesian hierarchical model to the transformed compositions in each data subset, using MCMC methods.
We derive expressions for (a) the posterior predictive probability that a newly observed fragment of glass is of a certain use type and (b) the evidential value of recovered glass fragments relating to two competing propositions about their source. The model is assessed using cross-validation and it performs well in both the classification and the evidence evaluation tasks.

Dancing bees improve colony foraging success as long-term benefits outweigh short-term costs

Waggle dancing bees provide nestmates with spatial information about high quality resources. Surprisingly, attempts to quantify the benefits of this encoded spatial information have failed to find positive effects on colony foraging success under many ecological circumstances. Experimental designs have often involved measuring the foraging success of colonies that were repeatedly switched between oriented dances versus disoriented dances (i.e. communicating vectors versus not communicating vectors). However, if recruited bees continue to visit profitable food sources for more than one day, this procedure would lead to confounded results because of the long-term effects of successful recruitment events. Using agent-based simulations, we found that spatial information was beneficial in almost all ecological situations. Contrary to common belief, the benefits of recruitment increased with environmental stability because benefits can accumulate over time to outweigh the short-term costs of recruitment. Furthermore, we found that in simulations mimicking previous experiments, the benefits of communication were considerably underestimated (low food density) or not detected at all (medium and high densities). Our results suggest that the benefits of waggle dance communication are currently underestimated and that different experimental designs, which account for potential long-term benefits, are needed to measure empirically how spatial information affects colony foraging success.

Social and Economic Models and Equilibrium 

In 1776, Adam Smith envisioned the presence of an invisible hand playing an active role in economic markets. These ideologies have supported the concept that all economic models eventually reach an equilibrium, supporting the traditional research of using differential equations encouraging the birth of rational people making rational decisions at the correct time. However, the 2008 credit crunch has highlighted the flaws in these theories.
Agent-based modelling allows us to go beyond this assumption of a utopian society by incorporating learning, location, relationships and many more to construct more realistic models removing some of the assumptions. However it presents various challenges in itself when working with disciplines of social scientists and economists with issues of model validation and verification. This talk will discuss the results and models of socio-economic models with methods on how software engineering principles can provide an answer to these problems.

Darwin’s last challenge

Beyond its theoretical success, the development of molecular biology has brought about the possibility of extraordinary progress in the historical study of classification and distribution of different species and different human populations, introducing a new level of evidence (molecular genetic markers) apt, among other things, to quantitative and computational treatment. Even in the cognitive sciences, purely theoretical progress in a certain discipline, such as linguistics, may have analogous historical impact, equally contributing to Renfrew’s so-called ‘New Synthesis’. Thus, I will argue that the parallelism between the study of linguistic history and genetic evolution goes beyond methodological similarities, and may begin to encompass substantive results. Darwin (1859) had hypothesised that the ultimate genealogical tree of languages should correspond to the biological one of human populations: however, testing this challenging hypothesis has so far been practically impossible. Over the past 30 years, geneticists developed ever more sophisticated techniques to compare the biological layout of even distant populations, but this is not enough; so far linguists have relied on comparing words from different languages, to reconstruct their ancestry: this method, though sometimes extraordinarily successful, goes back in time for few millennia, too little to relate distant languages on a continental or global basis. However, fortunately, languages are not mere word lists, but also sets of abstract syntactic rules to combine words into correct sentences. Exploiting dramatic progress in the theory of universal grammar, I have developed for the past 10 years a radically new method to compare languages and quantify their distances in order to reconstruct their histories from syntactic differences/similarities. We are currently applying it successfully to areas as vast as Europe and parts of Asia. On these grounds , I am now cooperating with geneticists, as the PI of an ERC grant, to finally compare genetically and linguistically distant populations, and we discovered that some results of the two disciplines may for the first time be correlated and complement each other for reconstructing ever deeper histories of human culture and biology.

Atoms, Patterns and Polyhedra: A Journey through Complexity in Art and Science

The presence of symmetry in nature has fascinated both scientists and artists for centuries. Geometry has been used across the disciplines by artists, engineers, biochemists, physicists and mathematicians, as a tool to understand, explain and order phenomena in the world around us. In the late 19th century biologist and philosopher Ernst Haeckel made detailed studies of microscopic life forms exhibiting unusual symmetric characteristics, illustrating over 4,000 species of Radiolaria. In 1940, inspired by these forms, French structural innovator Robert le Ricolaris proposed a geodesic shell structure based on the triangulated networks of radiolaria. Buckminster Fuller’s independent innovation of the geodesic dome, dating from 1948, also displays a similar structure to many radiolaria. Fuller’s ideas on geodesic structures stimulated significant scientific developments, with the 1985 discovery of a super-stable all-carbon C60 molecule, appropriately named Buckminsterfullerene. Fuller’s ideas also had an impact in the field of virology when scientists again drew ideas from the structure of his geodesic domes in the quest to understand the structure of virion protein shells. Although science and the creative disciplines of art and design are usually considered as polar opposites, both rely on a process of observation, experimentation and synthesis. An elegant example of the connections between science, art and design is shown through the parallels between crystallographic theory and pattern analysis and construction, both of which are underpinned by geometry. This paper discusses the cross-fertilisation of ideas between crystallography and design, from the early-twentieth century perspective on pattern analysis pioneered by the Russian crystallographer Federov to recent work exploring the translation of pattern from two to three dimensions.

How does the distribution of Leishmania in its mammalian host affect transmission dynamics to the sandfly vector (and is there value in modeling this)?

Leishmania parasites have varied distributions within their mammalian host (skin, different internal organs, bone marrow) and are transmitted between humans by female sandflies that take blood meals.  My lab has a long-standing interest in the pathology of leishmaniasis and how immune responses operate in different tissue sites.  We have developed a therapeutic vaccine that has just finished Phase I clinical trial, which might have value in reducing transmission efficiency, but we know very little about what governs "efficiency".  We have established a sandfly transmission model in York to test the vaccine experimentally and to ask fundamental questions about how different types of immune response (e.g that might sequester parasites together or that result in greater dispersal in tissues) impact on transmission.   I will be applying for a WT SIA at the end of February.  Is there a fruitful modeling dimension to this work (e.g. as a PhD or RA project)?