university of cambridge research horizons issue 5 · [email protected] dr louise...

R E S E A R C H

HORIZONS

University of Cambridge research magazine www.rsd.cam.ac.ukIssue 5 | Spring 2008

In this issueTHE THINKING MACHINE?plus news and views from across the University

2 | Issue 5 | Spring 2008

EDITORIAL

Foreword

A Happy NewYear to all of ourreaders andcontributors! Thisissue coincides

with a double celebration forResearch Services Division, as wemark five years of Horizon Seminarsand the twentieth in the series:‘The Thinking Machine?’ on 18March. Not only do these eventsprovide a forum for showcasinginterdisciplinary research atCambridge, but also excellentnetworking opportunities forspeakers and delegates alike.

Our front cover shows Dr MátéLengyel from the Department ofEngineering ‘in search of lostmemories’. In the Spotlight section,he describes how the intriguingprocess of memory storage andrecall is being revealed bycomputational neuroscience. Wealso hear from computer scientistswho are modelling how humansresolve ambiguities in language anduse diagrams for reasoning. Wecontemplate why it has been sohard to program a computer to playthe ancient Chinese game of Go.We consider how the brainorchestrates the ‘daily dance’ ofsleep and wakefulness, and welook back a million years to askwhen mankind first showedevidence of a thinking brain.

Cambridge has many examplesof cross-School research initiativesthat bring together the best in theirfield to address challengingquestions of the day. We will behighlighting these initiatives incoming issues, beginning with‘Great expectations’ in our Featuresection this time.

Many thanks to all of this issue’scontributors for their fascinatinginsights into their research projects:from Anglo-Irish history toecotoxicology, archaeology to law,medicine to Greek tragedy.

We look forward in 2008 tokeeping you abreast of the ground-breaking research happening acrossthe University. Please email yourcomments and suggestions forfuture coverage to me [email protected]

Dr Louise WalshEditor

ContentsResearch News 3–7Recent stories from across the University

Spotlight: The Thinking Machine? 8–17Modelling uncertainty in the game of Go 9In search of lost memories 10Can machines read? 12Can machines reason? 13Orchestrating brain activity 14The thinking hominid 16

Preview 18–19Greek tragedy: setting the stage today

Features 20–29A campaign of silent resistance 20Great expectations in pregnancy research 22Putting metabolism on the eco-map 24Finding fault 26Rebellion, repression, retribution 28

In Focus: Arts and Humanities Research Council 30–31

Inside Out: Dr Spike Bucklow 32

Research Support 33–35News from Research Services Division 33News from Cambridge Enterprise Ltd 34Forthcoming events 35

The Back PageYour way into Cambridge

Greek tragedy: settingthe stage today

Orchestrating brainactivity

Cover photograph of Dr Máté Lengyel 'in search of lost memories' by Dr Quentin Huysand Dr Máté Lengyel, with kind permission of Nature Neuroscience.Edited by Louise Walsh.Designed by Cambridge Design Studio (www.cambridgedesignstudio.org).Printed by Cambridge University Press (www.cambridge.org).©2008 University of Cambridge and Contributors as identified. All rights reserved.

DR D

ENIS BU

RDA

KO

V

CO

URTESY

OF TH

E CA

MBRID

GE G

REEK PLA

Y C

OM

MITTEE

Issue 5 | Spring 2008 | 3

RESEARCH NEWS

Using brain imaging techniques,Cambridge scientists have shown thatthose regions in the brain that understandsentences and form memories showreduced activity after sedation. Thesefindings help us to understand howspeech is decoded in the brain, and haveclinical implications for monitoringanaesthesia and patients with brain injury.

The recently published results are theculmination of a study conducted underthe lead of Dr Matt Davis at the MedicalResearch Council Cognition and BrainSciences Unit (MRC CBU) and ProfessorDavid Menon at the University’s Divisionof Anaesthesia in the Department ofMedicine.

Dr Davis, a cognitive neuroscientist,has developed a method that detectswhen the brain has comprehendedspeech, not simply heard it: ‘We look atthe brain’s response to sentencescontaining ambiguous words. Of the mostcommonly used 5000 words in spokenEnglish, over 80% change their meaningin different contexts – such as the wordshell, which can refer to a bullet or part ofa sea creature.’ Understanding thesewords triggers additional processing in thebrain as it retrieves the different meaningsand selects the one that fits the sentencecontext; this activity can be visualisedusing functional magnetic resonanceimaging (fMRI).

In this recent study, 12 anaesthetists atAddenbrooke’s Hospital volunteered toreceive varying amounts of sedation totest at what level their brain couldunderstand speech. ‘Our research showedthat although brains that are sedated tothe same level as in sleep are able to

process sound, brain processes involved inspeech comprehension are compromisedeven at quite low levels of sedation,’ saidProfessor Menon. ‘These results haveimportant implications for ourunderstanding of how and whereanaesthetic drugs work in the brain, andwhat neural processes are involved inconsciousness.’

Being able to assess comprehensionwithout spoken responses has hugeresonance for two clinical settings. ‘Asmall proportion of anaesthetised patientsreport memories of events that occurredin the operating theatre,’ said ProfessorMenon. ‘With this methodology, we couldrefine our judgement of how deepanaesthesia needs to be to preventcomprehension and memory duringoperations. The information also helps usto understand functional imaging studiesthat we have undertaken in patientsfollowing brain injury – in individuals whoappear to be in a coma or the vegetativestate.’

For patients in the vegetative state,who are awake and yet show no outwardsigns of awareness, this tool is alreadyproviding important information. ‘Beingable to show brain activity that indicatescomprehension in the patient brings hopeto carers, and might guide treatment andrehabilitation,’ said Dr Davis.

For more information, please contact Dr Matt Davis([email protected]) orProfessor David Menon([email protected]). Thisresearch was published in PNAS(2007) 104, 16032–16037.

Asking ‘are you awake?’ with brain imagingEven though we might be able to hear someone speaking,our powers for understanding what is actually being saidswitch off as we go to sleep.

DA

NIEL J M

ITCH

ELL, MRC

CBU

Magnetic resonance image of themouth, ear and brain

One year on from its launch, the RisingStars public communication course isreadying itself for training a newgroup of inspiring undergraduates,postgraduates, post-docs and earlycareer academics for the benefit of thewider community. The course is thefirst of its kind in the UK and has beenorganised by the University’s Office ofExternal Affairs and Communicationswith funding from the HigherEducation Funding Council for England(HEFCE).

The participants have been puttingtheir new skills into practice atfestivals, museums and schools, as wellas in the media. ‘The Rising Starsscheme enabled me to join acommunity of young researchers witha passion for public communication,’said Dr Nikiforos Karamanis, whowrites for Research Horizons on page 12 of this issue. ‘It gave me theopportunity to prepare and deliverseveral outreach events and made mefeel more confident in my ability topresent my expertise to the public.’

Penny Wilson, Head of CommunityAffairs, is delighted both with thesuccess of the course and with theenthusiasm shown by the participants:‘Underlying the scheme is the realneed to create dialogue between theUniversity and the wider communityand to communicate research and itscontribution to society. It is the youngacademics who make the bestambassadors for the University formany audiences, and it is they whowill become the role models for thenext generation.’

For more information on the Rising Stars programme, pleasecontact Emma Wenborn([email protected]).

‘Rising stars’

shine onWould you like to

communicate your

expertise to the public and

become an ambassador

for your subject?


RESEARCH NEWS

Unlocking thesecrets of theuniverseCosmic defects andadolescent galaxies – two research projects inCambridge are bringing us closer to understandingthe cosmos.Astronomers believe that the universe wascreated 13.7 billion years ago following acosmic explosion, the residual energy ofwhich can still be detected as cosmicmicrowave background (CMB) radiation,which fills the universe. The matter hurledoutwards from the fireball eventuallycooled enough to form galaxies across theexpanding universe.

Cambridge scientists working with theInstitute of Physics of Cantabria in Spainmay have discovered a remnant in theCMB from the Big Bang called a ‘texture’.As the universe cooled and underwenttransitions, physicists believe that vacuummisalignments or defects occurred. Ananalogy would be the defects wesometimes see in the transition of water toice. Textures are one of the most complexof cosmic defects, and when they collapsethey form ‘hot’ and ‘cold’ spots in theCMB.

The team have analysed a large coldspot in the Southern Galactic Hemisphereand published their findings recently inScience. Professor Neil Turok, of theDepartment of Applied Mathematics and

Theoretical Physics (DAMTP), and Dr MikeHobson, of the Astrophysics Group at theCavendish Laboratory, concluded that theproperties of the cold spot are consistentwith its having been formed by a texture.‘If this is the case,’ said Professor Turok, ‘itwill revolutionise our understanding ofhow the fundamental symmetriesbetween the particles and forces werebroken as the universe emerged from theBig Bang.’ Professor Turok was awardedthe prestigious Technology, Entertainment,Design (TED) prize in November inrecognition of his work in cosmology andhis efforts as an education activist. Theprize is awarded annually to threeindividuals whose work is considered tohave extraordinary potential for positiveinfluence on mankind.

About half a billion years after the BigBang, it is thought that matter started toaggregate, eventually forming smalladolescent proto-galaxies, which thenmerged to become bigger galaxies.Although astronomers have been able tostudy starlight of the progenitors of verymassive galaxies, starlight from the

building blocks of galaxies like our MilkyWay has proved elusive. In a recent study,due to be published in March in theAstrophysical Journal, an internationalteam of astronomers led by Dr MartinHaehnelt at the Institute of Astronomywere able to capture starlight from 27adolescent galaxies about 2 billion yearsafter the Big Bang. They pointed theworld’s most powerful telescopes, locatedin Chile, to the same patch of sky for theequivalent of 12 nights. ‘It is preciselybecause this was the first time the skyhad been searched with this level ofsensitivity that we succeeded where manyastronomers had failed before,’ said DrHaehnelt. The detection of these buildingblocks means that scientists can nowstudy in detail how galaxies like the MilkyWay have come together.

For more information, please contact Professor Neil Turok([email protected]), whoseresearch was published in Science(2007) 318, 1612–1614, or Dr MartinHaehnelt ([email protected]).

PROFESSO

R NEIL TU

ROK

Cosmic texture

Photovoltaic (PV) solar panels offer great promise as a source ofclean and renewable electricity generation but the high cost ofmanufacturing the silicon-based PV panels has been aprohibitive drawback to their use. A new research anddevelopment programme led by Professor Sir Richard Friend, Dr Neil Greenham and Professor Henning Sirringhaus at theUniversity of Cambridge’s Cavendish Laboratory, incollaboration with The Technology Partnership, hopes to solve

this problem. The team are using a plastic-based technology tocreate the solar cells. A prototype has already been built andthe new funding will allow scaling up to large sheets of PV filmthat can be sited on windows or roofs to capture solar energy.

By 2017, the aim is for these plastic solar cells to bedelivering 1GW of power, equivalent to carbon dioxide savingsof more than 1 million tonnes per year. ‘This is a timelyopportunity to build on technology developed in theUniversity,’ said Professor Friend. ‘We will capitalise on thelocal Cambridge strengths in taking science to manufacturing.’

The Carbon Trust is funding the initiative. Tom Delay, ChiefExecutive, explained the importance of the research: ’Webelieve this exciting new organic PV technology is our bestshot at dramatically reducing the cost of solar PV to the pointthat, in the next 10 years, it could become as cheap as thepower currently delivered to our homes.’

For more information, please contact Dr Neil Greenham ([email protected]) or the CarbonTrust (www.carbontrust.co.uk; Tel: +44 (0)20 7544 3100).

Sunny times ahead for solar powerA new initiative funded by the Carbon Trust hopes to make solar poweran affordable choice for homeowners within 10 years.

©C

AM

BRIDG

E DISPLA

Y TEC

HN

OLO

GY

Prototype solar panel


RESEARCH NEWS

As pioneers of brighter, more-permanentand less-toxic pigments from 1832 to thepresent day, Winsor & Newton has been adriving force in developing colourchemistry for art. As part of its richheritage, the Winsor & Newton archiveboasts 85 handwritten recipe books andworkshop manuscripts dating from the1830s to 1900 – a total of 17,000 pages.Now, thanks to work carried out underthe leadership of Ian McClure and Dr Mark Clarke at the Hamilton KerrInstitute, in collaboration with Dr LeslieCarlyle of Tate, a digital archive and indexof the Colourmen’s recipes and noteshave been created.

Although the use of decorativepigments dates back tens of thousands ofyears, some of the most excitingdevelopments happened in the 19thcentury, when scientific discoveries addeddramatically to the artist’s palette.Startling new pigments were developedby industrial chemists, yet traditionalmaterials were still produced by methodsoriginating in antiquity. ‘Mummy’, forinstance, required the grinding of an

Egyptian mummy before mixing withasphalt. Many details of innovations inthis period of rapid change have beenimperfectly understood until now.

The source material from the Winsor& Newton archive – from recipes forpigments and oils to watercolours andvarnishes – has been made available forresearch purposes. Additional notes onrecipes for gout and toothache, wagesand costs, travel and hotel suggestions arealso recorded. Words and passages inpublished technical literature that werepreviously baffling are becoming clear.

Each page has been carefullyphotographed and the recipes enteredinto an innovatively structured databasethat will be available in Spring 2008. Thevalue of this work, funded by the Artsand Humanities Research Council (AHRC),lies not only in historical record, but alsoin technical and authenticity studies, aswell as in evaluations of the deteriorationof paintings.

Dr Clarke explained: ‘Research intoartists’ materials and techniques is highlyinterdisciplinary and is mainly conductedby studying artists’ recipe books, scientificanalyses and historically accuratereconstructions. This archive affords afascinating insight into the art and science of the innovative Colourmen.’

For more information, please contact Dr Mark Clarke ([email protected]) at theHamilton Kerr Institute, a departmentof The Fitzwilliam Museum, or visit www-hki.fitzmuseum.cam.ac.uk/wn

Gems of colour: pigments of the‘Colourmen’Dragon’s Blood, Purple of Cassius and Scarlet Lake – allpigments concocted by the 19th-century ‘Colourmen’ ofWinsor & Newton artists’ suppliers. Now, a painstakingand methodical analysis of their hand-written recipebooks has been developed into a searchable database.

IMA

GE C

OU

RTESY O

F WIN

SOR &

NEW

TON

Winsor & Newton catalogue with painted samples

‘Sex, scandal andsermon’ inmedieval SpainA new study of the 14th-century narrativepoem Libro de BuenAmor explores how itsearthy tales of failed loveare shaped by humour.

The medieval poem Libro de BuenAmor (‘The Book of Good Love’) byJuan Ruiz tells the story of 14 failedlove affairs, combined with animalfables, sacred and profane lyrics, anddoctrinal matters. The ribald tales,colloquial language and vividcharacters (particularly the narrator,an Archpriest ‘fallen’ from gracebecause of his amorous dalliances)have led Ruiz to be dubbed the‘Spanish Chaucer’. Its 1728 stanzasare the subject of a new analysis,which is due to be published in April2008, by Dr Louise Haywood fromthe Department of Spanish andPortuguese.

The canonical work by Juan Ruizis of great importance for itscontemporary references and literaryscope. It is richly evocative ofmedieval cultural attitudes relating tolove, religion and emotion, and henceis a much-studied work amongstudents and scholars of Iberianliterature from this era. ‘It is one ofthe most challenging and thought-provokingly diverse works in worldliterature,’ said Dr Haywood, ‘and itsimpact on Spanish culture continuesto be felt today.’

Humour has long been recognisedas an important part of Ruiz’s artistrybut Dr Haywood’s study of the Librowill be the first to approach the roleof humour systematically in this iconicwork. ‘I have re-positioned Juan Ruiz’sLibro in relation to broader trends in14th-century culture, with a particularfocus on the relationship betweenhumour and scholasticism, the body,and visual culture. In essence: sex,scandal and sermon.’

For more information, pleasecontact Dr Louise Haywood([email protected]), whose bookSex, Scandal, and Sermon in theFourteenth Century: Juan Ruiz’s‘Libro de Buen Amor’ will bepublished on 1 April 2008 byPalgrave Macmillan.

IMA

GE C

OU

RTESY O

F WIN

SOR &

NEW

TON

Box for tube paints, showing the Winsor &Newton factory


RESEARCH NEWS

Vociferous debates, threatened splits,division and dispute are elements thatcan rankle, grow and ultimatelyundermine the future of any largeorganisation unless addressed in aconstructive and holistic manner.Researchers in the Psychology andReligion Research Group (PRRG) in theFaculty of Divinity are combiningpsychological research and practice tolook at conflict in the church and howthis might be transformed. Out of thishas developed a multidimensionaltraining package to enable senior churchleaders to handle conflict better.

‘There are various lines of conflict.Some echo debates betweenconservative and liberal interpretations ofScripture; these affect moral and socialissues, with varying degrees of impactupon churches in different parts of theworld. Other conflicts are more local,and may voice disagreements betweenthose who wish to maintain traditionalforms of worship and those who wish todevelop contemporary, emergingexpressions of church,’ said Dr SaraSavage, who has developed the projectwith Dr Eolene Boyd-MacMillan. Theresearch was commissioned by theFoundation for Church Leadership withfunding from the Henry Smith Charity.

The course, which has just beenlaunched, will be road tested andassessed while being used to train seniorrepresentatives from the Anglican,Baptist, Church of Scotland, Methodist,Roman Catholic and United Reformtraditions. Over the course of threetraining days, the participants will covervarious topics and exercises designed tofoster a positive attitude towards conflictas a holistic learning opportunity, whileenhancing and adding to conflicttransformation skills.

Pre- and post-assessment of theeffects of the course will inform the

research project, and the report andresources will be fed back to the keychurch denominational conferences in2008. In time, the hope is that theprogramme will be used to trainChristians at all levels to deal withdispute – whether it be a question ofchurch-wide importance, a parishconcern, or an inter-personaldisagreement – as well as by other faiths.

For more information, please contactDr Sara Savage ([email protected]) orDr Eolene Boyd-MacMillan([email protected]).

Weathering storms:transforming conflictin the churchA research initiative in theFaculty of Divinity aims totrain church leaders toconvert the negatives ofconflict into the positives oftransformation.

Researchers in the Department of Sociology led by Dr Brendan Burchell have been analysing data gathered bythe European Commission’s European Working ConditionsSurvey. The Commission’s door-to-door surveys areconducted every five years in all EU member states, and ask30,000 employed and self-employed workers about their job,their well-being and aspects of their lives outside work.

Dr Burchell’s recently reported findings show that littleprogress has been made in reducing the workplace gendergap in the EU since the early 1990s: 75% of the EUworkforce is being managed by men, with just 9% ofemployed men in full-time work being managed by women.In the UK, although women make up nearly 50% of theworkforce, fewer than 16% of senior managers are women.

‘Because women are rarely the highest earners in thehousehold,’ said Dr Burchell, ‘an economic rationale seems to

have developed for also making them responsible fordomestic duties. The result is that women tend to have alonger working day.’ Adding together time in paidemployment with time spent commuting and performingdomestic work, men in full-time employment work for anaverage of 55 hours per week; by the same method, theaverage working week for a woman is 68 hours if she is infull-time employment, or 57 hours for part-time employment.

Dr Burchell added: ‘The best way to break the cyclewould be to reduce gender inequality in employment and thehousehold together – for instance, by encouraging men to taketheir parental leave entitlements or by better enforcement ofthe working-time directive to reduce the disproportionatelylonger working hours of men.’

For more information, please contact Dr BrendanBurchell ([email protected]).

©ISTO

CK

PHO

TO.C

OM

/CH

RIS KRID

LER

The great divideDespite efforts to narrow the gender gap in the workplace in the European Union(EU), there is still ‘a great divide’, as reported in a recent study.

RESEARCH NEWS

Dr Sabine Bahn’s dynamic and growingresearch group within the newlyestablished Centre for NeuropsychiatricResearch (CCNR) in the Institute ofBiotechnology has been progressingtowards a greater understanding of themolecular processes that lead toschizophrenia and bipolar disorder.Current estimates suggest that 1 in 50people in the UK are affected, many ofwhom are incapacitated by thesedisorders.

Dr Bahn, a psychiatrist and molecularbiologist, aims to develop a means todiagnose and treat patients at an earlystage of the disease. ‘Current diagnosticmethods have remained unchanged forover a century and rely on a subjective,interview-style analysis of the patient bythe medical practitioner rather than adiagnostic test. Following diagnosis,some patients show no improvementwith current drug regimes,’ explained Dr Bahn.

The substantial support given to Dr Bahn by the SMRI is made possible

through the philanthropy of AmericansTheodore and Vada Stanley. Withprevious SMRI funding of $3.5 million,Dr Bahn’s group completed an intensivestudy of over 150 post-mortem brains,looking for differences between patientsand matched controls. They successfullyidentified ‘disease gene signatures’ and the first strong biomarkers forschizophrenia in easily testable fluids like serum and cerebrospinal fluid. The further funding of $5 million allowsDr Bahn to build on these successes withher team of neuroscientists, molecularbiologists, mathematicians, statisticiansand bioinformaticists.

Dr Bahn’s group works closely withPsynova Neurotech Ltd, the spin-outcompany she founded in 2005 withProfessor Chris Lowe with SMRI support.Psynova is developing and exploiting thenovel biomarkers discovered by Dr Bahn,as well as initiating its own drugdiscovery programmes using proprietarybiomarkers. In June 2007, Psynova wasawarded the Medical Futures Innovation

Award in Mental Health andNeurosciences Innovation.

‘With the new funding, we believethat all the ingredients are in place tosuccessfully translate high-qualityresearch into high-value clinicaladvances,’ said Dr Bahn. ‘Using the latest biotechnological advances, wehope that this seamless interface will link fundamental research at the benchdirectly with the patient bedside.’

For more information, pleasecontact Dr Sabine Bahn([email protected]). Forinvestment opportunities relating to Psynova Neurotech Ltd, pleasecontact Cambridge Enterprise Ltd (Tel: +44 (0)1223 760339).

Translating research from bench to bedsideRecent funding of over $5 million to Dr Sabine Bahn bythe Stanley Medical Research Institute (SMRI) raisesdiagnostic and therapeutic hopes for a group ofpsychiatric diseases that affects 2% of the population.

The Leverhulme Climate Symposium2008, to be held on 10–12 March inCambridge, will explore how accuratemodelling of the Earth’s present andfuture climate might be derived fromknowledge of past climate change. ‘Thereis evidence that climate change at times inthe past was very unstable. Within thespace of a few decades, the NorthAtlantic region warmed by 10°C at thesame time that smaller changes oftemperature occurred over wide areas ofthe planet,’ said symposium convenerProfessor Harry Elderfield from theDepartment of Earth Sciences. ‘It isimportant that the models for predictingfuture climate can simulate these

extremely rapidevents.’

Over 100 international experts andyoung researchers invited from both themodelling and palaeoclimate communitieswill meet to share their knowledge ontopics such as the effect on climatechange of solar activity, changes in theEarth’s atmosphere, ocean circulation, thehydrological cycle and ice sheet melting.The meeting is coordinated by theCambridge Environmental Initiative (CEI),which supports environmental research atthe University of Cambridge.

It is hoped that by bringing these tworesearch communities together, new jointresearch areas will emerge in Cambridgeand beyond. ‘Information on past climateand its variability has increased enormouslyover the past 10 years as a result ofsystematic sampling of ocean sediments byinternational drilling programmes anddrilling of the major ice sheets,’ said

Professor Elderfield. ‘The aim now is todefine the potential risks of climate changebetter, as well as to identify strategies wemight take to minimise them.’

The symposium will conclude with ameeting and exhibition at the RoyalSociety in London on 13 March to whichthe general public, policy makers, theiradvisers and the media are invited.

For further information, please go towww.leverhulmeclimatesymposium.org

If you would like the opportunity tohear more about past climate changeand how it impacts on future climatepredictions, the Leverhulme ClimateSymposium is sponsoring a talk at the Cambridge Science Festival on 14 March 2008 (seewww.cambridgescience.org).

Earth’s climate: past, present and futureThrough considering past climate change, a major international symposium inCambridge hopes to create new research opportunities for the future.


Dr Sabine Bahn


THE THINKING MACHINE?

The ability to learn continuously aboutour environment, through experienceand education, is a major advantagethat sets humans apart from present-day machines. In this session we willexplore how machines can beendowed with sophisticated learningmechanisms and ask what insightsinto biological learning are revealed bythese new developments in machinelearning.

Professor Zoubin GhahramaniDept of EngineeringDr Máté LengyelDept of EngineeringProfessor Seth GrantWellcome Trust Sanger Institute

LEARNING

We make decisions of many kinds,often subconsciously – from whereto place our hands in order to catcha thrown object, to adapting ourbehaviour to different situations, toplanning ahead. What are theelements involved in this processand can they be replicatedartificially?

Professor Daniel WolpertDept of EngineeringProfessor Christopher BishopMicrosoft Research CambridgeProfessor Nicola ClaytonDept of Experimental Psychology

DECISIONS

What we make of the world isdictated by the information we haveabout it, and human brains areexceptionally good at modelling theirown universe. Here we examine theways in which machines are able toperceive their surroundings and weexplore whether machines shape theirown environment like us.

Professor Roberto CipollaDept of EngineeringDr Mateja JamnikComputer LaboratoryProfessor Ted BriscoeComputer Laboratory

PERCEPTION

The Thinking Machine?

The Horizon Seminar 'The Thinking Machine?'

takes place on 18 March 2008 atEmmanuel College,

Cambridge, andwill showcase the latestCambridge research in

cognition and statisticalmachine learning.

Our understanding of the process ofcognition in human beings – how weperceive, think and process informationabout our environment – is highlydeveloped. Research in areas such aspsychology, neuroscience, linguistics,engineering, philosophy and computerscience has taught us a great deal aboutthe nature of learning, decision-making,perception and thought. At this event wewill explore the cross-disciplinary researchtaking place in Cambridge and attempt to answer the questions: is intelligenthuman behaviour merely the result of pre-programed activity and can cognitiveprocesses be modelled by machines?

For more information about theHorizon Seminar series, please go towww.rsd.cam.ac.uk/events/horizon oremail [email protected]

Panel discussion

REFLECTION, CONSCIOUSNESS AND THINKING



The ancient Chinese two-player strategygame Go is presenting a ‘grand challenge’to current artificial intelligence (AI)research. At first sight it appearsstraightforward: the players take it inturns to place black and white ‘stones’ ona grid with the aim of gaining control ofthe most territory. However, the gamethat emerges from these simple rules hasdefeated all attempts by AI researchers toprogram a computer to play it. Indeed,there is currently no Go program that canplay better than a beginner on the full-size board.

Tree of possible futuresIt might seem surprising that Go is sodifficult to model, given the notoriousdefeat of World Champion GarryKasparov by the chess computer DeepBlue in 1997. The trouble with Go is thatthe brute force technique that proved sosuccessful for chess simply cannot beapplied. Deep Blue worked by consideringbillions of possible future move sequences(searching about 12 moves into thefuture) and estimating the strength of theresulting game positions. This fails whenapplied to Go for two main reasons. First,the tree of possible futures is too big toexplore usefully (on average, there are220 legal moves for each turn in Gocompared with about 35 in chess).Second, it is difficult to estimate thestrength of a Go position: in chess we cansum the point values of the pieces on theboard but no such simple heuristic existsfor Go (all of a player’s stones are identicaland take on their value from theirrelationship with surrounding stones).

Predicting human playGo is an excellent test-bed for newtechniques in AI because it is welldefined by a set of simple rules yetremains sufficiently complex to challenge the state of the art. Also, a wealth of data exists about how humans play, in the form of records of historical games.

Professor David MacKay’s InferenceGroup at the Cavendish Laboratory, incollaboration with Dr Thore Graepel ofthe Applied Games Group at MicrosoftResearch, has focused on applyingprobabilistic modelling techniques to Go.By modelling the play of human experts,the aim is to predict human moves in thegame records. One successful approachhas been to represent each potential Gomove by the pattern of stonessurrounding the move location. Millionsof these patterns were automaticallyharvested from 180,000 records ofhistorical games. The decisions made by the players in all the game recordswere then used to determine the relative value of each of the patterns.This is possible because each recordedgame position contains a set of legalmoves, one of which we know wasselected by the player, so we can inferthat one move was more valuable thanall of the other moves in that position.After being trained on all the gamepositions, the resulting system was ableto predict the moves of human playersmore accurately and much more rapidlythan other published results and wascapable of playing the game surprisinglywell.

Way to GoWe are still a long way from producingprofessional-level Go play. Formidablechallenges remain, leading someresearchers to argue that a strong Go-playing computer is still decades away.However, the fact that humans learn thegame with relative ease providestantalising evidence that, one day, it willbe possible to create a strong Go programand, in so doing, perhaps increase ourunderstanding of human intelligence.

For more information, please contact the author David Stern([email protected]) at the Cavendish Laboratory, or visit theInference Group website(www.inference.phy.cam.ac.uk/is) or the Applied Games Group website(http://research.microsoft.com/mlp/apg). This work was supportedby a grant from Microsoft Researchthrough the PhD scholarship program(www.research.microsoft.com/ero/phd).

David Stern

©IS

TO

CK

PH

OT

O.C

OM

/DO

UG

CANN

ELL

Modellinguncertainty

in thegame of

GoCambridge researchers have analysed millions of patterns of potentialmoves to model theuncertainty of play in theancient game of Go.



In search of lost memories‘I raised to my lips a spoonful of the tea in which I had soaked a morsel of the cake. […]And suddenly the memory returns.’ À la Recherche du Temps Perdu, Marcel Proust.

Recalling events from our past is a taskwe need to accomplish several timesevery day – even if the form it takes isusually not nearly as dramatic as Marcel Proust describes in his famous novel, in which a mere tasteevokes memories long buried. Forcenturies, artists and philosophers have been intrigued by the process of memory storage and recall. Morerecently, psychologists, neuroscientistsand cognitive scientists have begun tounravel some of the principles of‘remembrance of things past’. And now, a branch of neuroscience thatworks on quantitative models of thenervous system – computationalneuroscience – is also contributing to the quest.

Dr Máté Lengyel, at theComputational and Biological LearningLab in the Department of Engineering, is borrowing ideas from machinelearning to elucidate the principles ofmemory recall. One particular challengein memory research is to unfold thesequence of events happening at thelevel of nerve cells (neurons) that leads to the retrieval of a memory. Not only is this a fundamental question inneuroscience, but it has also providedsome of the finest examples of howcollaboration between theoretical andexperimental approaches can beespecially fruitful for understanding the brain.

Attractor networks –attractive theoriesA commonly held view in modernneuroscience is that most forms ofmemories are stored in the nervoussystem because of the changing way thatconnections (or synapses) are madebetween neurons – a phenomenonknown as ‘synaptic plasticity’.

Once a memory trace has been laiddown in a set of synapses it has to berecalled by neurons interacting with eachother through these synapses. The firstcoherent picture about how this mighthappen was proposed by theorists whodeveloped a specific class of neuralnetwork models called ‘attractornetworks’. This theory has provided anelegant mathematical formulation of howa network of neurons graduallyreconstructs a complete memory tracestarting from only partial information –just as in Proust’s novel, the narratorrecalls an entire scene starting from onlythe taste of cake soaked in tea.

Despite the success of attractornetworks as a theoretical framework thatguides many scientists’ thinking about theneural bases of memory storage andrecall, there are several questions that theoriginal theory is unable to address. Is itimportant just that a neuron is active ornot, or is it the graded level of activationthat bears information about the originalmemory trace? If we take it a stage

further, do neurons only communicatethrough the average rate of the transientelectrical impulses they emit, called spikes,or is it the precise timing of theseimpulses that are important?

It is well established, for instance, thatthe graded activity level of neurons andtheir spike timings play a central role inthe functioning of the hippocampus, abrain structure in the medial temporallobes that is crucial for an intact memory.Yet, traditionally, the theory of attractornetworks assumes that memories arebinary and based on the rate of spiking,and it has proved notoriously difficult tobridge this gap between theory andbiological reality.

Bits and brainsRather than following the moretraditional path that starts from knownbiological properties of single neuronsand synapses, and proceeds by analysingthe emergent network behaviour theygive rise to, Dr Lengyel is taking adifferent approach to understanding howmemory processing is achieved byneuronal networks. This approach,pioneered by Professor David MacKay atthe Cavendish Laboratory, first studiesthe task posed by memory recall as aspecial case of ‘statistical inference’, a mathematical theory forming thefoundations of many of today’s mostpowerful machine learning applications.The core idea is that storing


DR M

ÁTÉ LEN

GY

EL

memories in the synapses that connect a set of neurons is formallyequivalent to data compression –something we are all familiar with, forexample, when storing music as an MP3 file on an audio player. Memoryrecall then becomes an act of‘decompressing’ the informationpreviously stored in these synapses.

The next step is to take the positionof an engineer who needs to construct adevice that achieves the highest possibleperformance given the constraintsprovided by biology. This allows one toattempt to predict the properties ofneurons and synapses that would beoptimal for retrieving memories. Usingthe mathematical analogy betweenmemory storage in neural networks anddata compression, it has become possibleto address the question of how neuronscould implement the optimaldecompression algorithm for recallingmemories if the memories arerepresented by the graded activities ofneurons using precise spike timings.

With recent funding from theWellcome Trust, Dr Lengyel, togetherwith Professor Peter Dayan at the Gatsby Computational NeuroscienceUnit, University College London, ispursuing this direction of research. They work closely with Dr Ole Paulsen at the University of Oxford, whose group conducts experiments to test the predictions of the model in the

neural networks of the hippocampus,and has already successfully confirmedsome of the precepts of the theory.

The aim is to understand how thechanging connections between thebrain’s neurons maximise the information that is stored, providing the brain with the ability to remember.The aim also is to show how differentrates of neural spiking might representthe level of certainty that a memorybeing recalled is correct. Both of theseissues are important if one wants todesign a device that presents an optimal solution to the task of memory recall, and allows us to askwhether nature has designed our brainsin such a way.

For more information, please contactthe author Dr Máté Lengyel([email protected]) at theComputational and Biological LearningLab in the Department of Engineering.

Dr Máté Lengyel

The same mathematical and engineering principles that can be used to understandlearning in the human brain can also be used to build artificial learning systems.This interplay between human and machine learning is the main research focus ofthe recently established Computational and Biological Learning Lab (CBL) at theDepartment of Engineering. CBL was founded in 2006 with the arrival of ProfessorDaniel Wolpert and Professor Zoubin Ghahramani in Cambridge, and has rapidlygrown to include Dr Carl Rasmussen, Dr Máté Lengyel and over 20 PhD studentsand postdoctoral researchers. CBL is investigating the computational principlesunderlying human sensorimotor control, the design of computer algorithms thatlearn, adaptive reinforcement learning controllers, statistical theories of learning,and how networks of neurons can perform computations.

For more information, please contact Professor Zoubin Ghahramani([email protected]) or visit http://learning.eng.cam.ac.uk

Minds and machines

Recall of a memory represented by spike times (white ticks) in a simulated network of neurons:recall starts from a noisy representation (left), which is then gradually ‘cleaned up’ (right)

DR G

ERGÖ

ORBÁ

N


Searching for information on the internethas become second nature. Yet, manysearches return too many documents andit takes time to examine each one andfind the information we need. Scientistsare particularly familiar with this problemwhen searching the scientific literature.How can professional curators, trained toidentify information in a scientific articleand enter it into a database, mine thetext more efficiently? A multidisciplinaryteam in Cambridge led by Professor TedBriscoe in the Natural Language andInformation Processing (NLIP) group atthe Computer Laboratory has beeninvestigating whether computerisedmethods for analysing the language ofscientific articles can facilitate databasecuration.

Resolving ambiguity inlanguageLanguage is ambiguous: in the phrase ‘Isaw Bill with the telescope’, who isholding the telescope – you or Bill? Toresolve these ambiguities, people use thesurrounding text and their knowledge ofthe world. But this is much harder for acomputer to do and has been a long-standing challenge in computer science.Natural language processing (NLP)provides a way for the computer to treateach word as a piece of a puzzle. Eachpiece is combined with other piecesusing knowledge about the structure ofEnglish in the context of the subject.Although not an exact science, NLP can



Enabling computers to understand language might helpusers to overcome online information overload.

be exploited to ameliorate tasks such ascuration by attempting to overcomeambiguity in phrasing and draw outrelationships between words.

FlySlipThe NLIP group has teamed up with theFlyBase-Cambridge curation team in theDepartment of Genetics, one of the threemembers of the international FlyBaseconsortium that provides the largestdatabase repository of genetic informationon the fruit fly. The result, FlySlip, is fundedby the Biotechnology and BiologicalSciences Research Council (BBSRC) withthe aim of developing text informationextraction tools to assist FlyBase curators.

There is a strong need to develop suchtools not only because of the enormouspublication rate in genetics but alsobecause of the particular difficultiesassociated with automatic recognition offruit fly gene names. New gene names areconstantly being introduced in theliterature and many are the same ascommon English words such as ‘not’, ‘an’,‘was’ and ‘if’. Andreas Vlachos andCaroline Gasperin, PhD students in theNLIP group, have been investigating theuse of statistical NLP techniques forrecognising gene names and fordisambiguating expressions such as ‘thisgene’.

The NLIP group has also been the firstto investigate what is the most useful wayof presenting the NLP analyses to curators.Dr Ian Lewin and Dr Nikiforos Karamanishave been working on the design andevaluation of a unique curation interface.The interface uses automatically recognisedgene names and disambiguatedexpressions as mechanisms to navigate thetext. FlyBase curator Dr Ruth Seal hasprovided them with domain-specificexpertise.

Beyond FlyBase curationThe FlySlip team has demonstrated thatthe NLP-powered interface enablescurators to interact with articles quicklyand efficiently. Worldwide, more than 80 curated databases similar to FlyBaseexist, and similar approaches are likely tobe adopted by other curation groups. Inthe future, NLP technology may also beused to support everyday tasks such asinternet searches. So, next time you feeloverwhelmed by the online ‘data deluge’,remember that a solution might be on the way.

For more information, please visit the NLIP group website(www.cl.cam.ac.uk/research/nl). The author, Dr Nikiforos Karamanis([email protected]), is a participant onthe Rising Stars scheme (see page 3).

Can machines read?

Dr Nikiforos Karamanis

©ISTO

CK

PHO

TO/PA

MELA

MO

ORE



Some of the deepest and greatest insightsin reasoning have been made usingmathematics. It’s not surprising thereforethat emulating such powerful reasoningon machines – and particularly the wayhumans use diagrams to ‘see’ anexplanation for mathematical theorems –is one of the aims of artificial intelligence.

Diagrams for reasoningDrawing pictures and using diagrams torepresent a concept is perhaps one of theoldest vehicles of human communication.In mathematics, the use of diagrams toprove theorems has a rich history: as justone example, Pythagoras’ Theorem hasyielded several diagrammatic proofs in the2500 years following his contribution tomathematics, including that of LeonardoDa Vinci’s. These diagrammatic proofs areso clear, elegant and intuitive that withlittle help even a child can understandthem.

The concept of the ‘mutilated’checkerboard is another usefuldemonstration of how intuitive humanreasoning can be used to solve problems.If we remove two diagonally oppositecorners, can the board still be coveredwith dominoes (rectangles made out oftwo squares)? The elegant solution is tocolour the checkerboard with alternativeblack and white squares, like thechessboard, and do the same with thedominoes so that a domino is made ofone white and one black square. Thesolution then immediately becomes clear:there are more white squares than blacksquares, and so the mutilatedcheckerboard cannot be covered withdominoes. This problem is very easy forpeople to understand, but no system hasyet been implemented that can solve it insuch an intuitive way.

As these reasoning techniques can beincredibly powerful, wouldn’t it be

exciting if a system could learn suchdiagrammatic operations automatically?So far, few automated systems haveattempted to benefit from their power byimitating them. One explanation might bethat we don’t yet have a deepunderstanding of informal techniques andhow humans use them in problemsolving. To advance the state of the art ofautomated reasoning systems, some ofthese informal human reasoningtechniques might have to be integratedwith the proven successful formaltechniques, such as different types oflogic.

From intuition toautomationThere are two approaches to the difficultproblem of automating reasoning. Thefirst is cognitive, which aims to devise andexperiment with models of humancognition. The second is computational –attempting to build computationalsystems that model part of humanreasoning.

Steps along the computationalapproach are being taken by Dr MatejaJamnik in the Computer Laboratory withfunding for an Advanced ResearchFellowship from the Engineering andPhysical Sciences Research Council(EPSRC). While at the University ofEdinburgh, Dr Jamnik built Diamond, aprogram that uses diagrammaticreasoning to prove mathematicalconcepts. However, there are theoremslike the mutilated checkerboard thatmight require a combination of symbolicand diagrammatic reasoning steps toprove them. In Cambridge, Dr Jamnik isnow investigating how a system couldautomatically reason about such‘heterogeneous’ proofs. This requirescombining diagrammatic reasoning inDiamond with symbolic problem solving in

an existing state-of-the-art automatedtheorem prover. The way forward is togive heterogeneous reasoning frameworksaccess to intelligent search facilities in thehope that the system will not only findnew and more intuitive solutions toknown problems, but perhaps also findnew and interesting problems.

Automated diagrammatic reasoningcould be the key to making computerreasoning systems more powerful, as wellas to providing the necessary tools tostudy and explore the nature of humanreasoning. We might then have a newmeans to investigate the amazing abilityof the human brain to solve problems.

For more information, please contactthe author Dr Mateja Jamnik([email protected]) at theComputer Laboratory.

Can machines reason?Humans often use diagrams for reasoning, but can computers do the same?

Dr Mateja Jamnik



New research inCambridge isdeciphering

neural controlsignals that

create the rightbrain state for

the rightsituation.

Orchestrating brain activity

It would be reasonable to think of thebrain as the most complex machineknown, given the diversity andsophistication of the tasks it can perform.It perceives and organises multisensoryinformation about the world, makes split-second decisions based on thisinformation, and controls sophisticatedmovements to put these decisions intoaction. It also generates seemingly infinitecombinations of thoughts and feelings tosolve abstract problems and create worksof art. What’s more, it can get better at allof this through learning and practice.

This vast range of powers is trulyimpressive, but poses a serious controlproblem. How does the brain ensure thatall of these feelings and actions happen atthe right time and don’t clash with eachother? Recent studies suggest that thismay not be as random and whimsical as itoften seems. Well-defined neuralnetworks appear to keep a close watchon overall brain activity and helpconsciousness flow smoothly from onestate to the next.

The daily dance of brainstatesWe are all familiar with two moststrikingly different states ofconsciousness: sleep and wakefulness.Most of us take wakefulness for grantedand can assume that we won’t suddenlyfall asleep in the midst of talking andlaughing. Not so for people sufferingfrom narcolepsy, a sleep disorder causingirresistible attacks of sleep and paralysisthat suddenly intrude into normalwakefulness. Here, laughter is definitelyNOT the best medicine – bizarrely, severenarcoleptic attacks can be triggeredprecisely by positive emotional states suchas laughter. For many years, the cause ofnarcolepsy was a mystery but it is nowknown to be due to a lack of brainproteins called orexins that are made by a few thousand neurons located deep inthe brain’s hypothalamus.

Orexin neurons appear to be criticalfor the smooth flow of brain states in allmammals. Loss of these cells in humans,dogs, mice and rats causes narcoleptic

Orexin neurons (red/pink) in their native network surrounded by other neural types (blue and green)

DR D

ENIS BU

RDA

KO

V

fragmentation of consciousness, withabnormally frequent and rapid transitionsbetween sleep and wakefulness. How doorexin neurons promote stablewakefulness? Even though they are few in number, orexin neurons extend theiraxon arms throughout the brain. Releaseof orexins from these axons turns on brain regions involved in cognitive arousaland attention. Somehow our brain state is prone to instability, and the orexinsystem often steps in to prevent loss of wakefulness, by orchestrating a burst of activity throughout the brain.Orexin neurons also seem to be critical for normal orchestration ofwakefulness-dependent behaviours such as reward seeking and foodconsumption.

Sweet dreamsThe evidence linking orexin neurons tobrain state control is so strong that thiscell type is especially attractive toresearchers like Dr Denis Burdakov at theDepartment of Pharmacology who wantto understand the neural basis ofbehaviour. By studying the properties ofsingle orexin neurons, it might be possible to get a glimpse of the molecular machinery underlying brainstate control.

Dr Burdakov’s group has shown thatorexin neurons act as intrinsic generatorsof wakefulness-promoting electricalsignals, and that the frequency of theseelectrical impulses is fine-tuned by theenvironment. Strikingly, orexin neuronsappear to be specialised in detectingchanges in key physiological variablessuch as sugar levels. This research hasshown that glucose opens special pores in the membranes of orexin neuronsknown as potassium ion channels, therebyblocking their electrical signals.

This mechanism is striking in itssensitivity – the activity of orexin neuronscan accurately track even the tiny changes in sugar levels that occur in thebrain between breakfast and lunch.Orexin neurons become activated whenenergy levels fall, stimulating wakefulnessand activity to ensure an animal seeks outfood when hungry. In turn, their block byglucose is perhaps the strongest candidatemechanism for after-meal sleepiness.

Glucose sensing by orexin neuronsrepresents an unusual physiologicalpathway through which sugar may affectsleep and appetite, and could provide anattractive target for pharmacologicalintervention in body weight disorders. The importance of research in this newarea was recently recognised by theBiotechnology and Biological SciencesResearch Council (BBSRC), who awardedDr Burdakov a three-year research grantto unravel molecular steps in theregulation of orexin neurons by sugar.

Awake and hungry How do orexins stimulate appetite? Dr Burdakov’s research has shown thatorexins inhibit the activity of a keypopulation of neurons that suppressappetite – the melanocortin neurons of the arcuate area of the hypothalamus.

The activity of every neuron in the brain is set by a push–pull balance between synaptic excitation and inhibition.In the appetite-suppressing melanocortinneurons, which like many brain cells areinnervated by orexin-containing nervefibres, orexins tip this balance in favour of inhibition, thereby silencing themelanocortin cells. This mechanism mayhelp ensure that alertness, attention andappetite are appropriately synchronised:when energy levels fall, cognitive arousaland appetite will be stimulated together,increasing the likelihood of successful food seeking. Conversely, when food iseaten, then appetite, arousal and activitymay be turned off together to createoptimal conditions for converting the food into fat stores. A potential downsideis that the activity of orexin neurons during wakefulness may also drive hunger, which would fit in with the well-known association between insomnia and obesity.

Neurons on acidConsidering the widespread connections oforexin neurons, it is hardly surprising thatmore and more behavioural roles for thesecells are being discovered. One of the latestis their role in breathing – orexin neuronsprovide an excitatory stimulus to keybrainstem areas involved in ventilation. Thisseems especially important when breathingneeds to be increased, for example to getrid of excess carbon dioxide in the body. Inmice with a deficient orexin system, thesevital breathing responses are reduced by asmuch as 50%.

How do orexin neurons ‘know’ whento stimulate breathing? Dr Burdakov’s teamfound that rising levels of acid, such asthose that occur during sleep apnoea (a pause in breathing), release an electrical‘brake’ in orexin neurons. This allows theneurons to fire faster. Similar to the effectsof glucose, this brake consists of potassiumpores in the orexin cell membrane – whenopen, they stop neurons from firing. Risingacidity shuts them down, making orexinneurons fire very fast; falling acidity opensthe pores, silencing the firing. In the body,rises and falls in acidity are controlled bybreathing. Acid is constantly made by thebody’s metabolism, but is normallyprevented from building up because webreathe out ‘acidic’ carbon dioxide. Duringsleep apnoea, the breathing stops, leadingto a dangerous build up of acid in thebrain. Acceleration of orexin firing by aciddissipates this threat by causing awakeningand increased breathing.

Beyond appetite and arousalIn just a few recent years, orexin neuronshave emerged as a powerful, multitasking control system of the brain, important for coordinating vitaladaptive behaviours such as wakefulness,appetite, reward and breathing. The aim now is to unravel how orexin cells control multiple interacting circuits byacting as polymodal sensors of theirenvironment and how these circuits ‘talk back’ to orexin neurons in different behavioural states.

For more information, please contactthe author Dr Denis Burdakov([email protected]) at the Departmentof Pharmacology.


Dr Denis Burdakov




Thethinkinghominid

The discovery insouthern India ofa well-preserved

quarry datingfrom a millionyears ago is

helpingresearchers toanswer: how

intelligent wereour ancestors?

We are a species capable of remarkabletechnological and cultural achievementsthanks to our big brains and our extendedcooperative social networks. Of course, ithas not always been this way. Our firstupright, ape-like ancestors lived between7 and 6 million years ago, and perhaps ahalf dozen human-like species withincreasing cognitive abilities span theevolutionary period to the present-dayHomo sapiens.

Palaeoanthropologists like Dr MichaelPetraglia in the Department of BiologicalAnthropology are interested indocumenting and explaining ourevolutionary history, and particularly inexamining the cognitive and behaviouralchanges that have occurred along theway. One fundamental question hasalways been: how did advanced levels ofhuman cognition evolve in our earlyancestors? In searching for clues, Dr Petraglia is using the artefactual record left behind by early hominids.

Acheulean toolsOne particular period, known as theAcheulean, has fascinatedpalaeoanthropologists and archaeologistsalike because it is the longest enduringtime of technology and manufacture, andtherefore of comparatively rich artefactualrecord, in our evolutionary history.Beginning in Africa at about 1.6 millionyears ago, and lasting until about 250,000years ago, the Acheulean is characterisedby the manufacture of pear- and oval-shaped stone handaxes and cleavers.Often found in association with animalcarcasses, these tools were probably usedas efficient cutting devices for detachingand slicing meat for food.

During the prolonged period of thisdominant technology, Acheulean toolusers migrated from Africa and colonisedEurasia. Examples of tools have beenfound from Arabia to Asia, and in Europefrom the Mediterranean to central Britain.

Acheulean tools dating from a million yearsago found at the Isampur Quarry, India

DR M

ICH

AEL PETRA

GLIA



Discovering debrisThe long-lasting and supposedlyunchanging technological tradition ofthe Acheulean has been thought of asdemonstrating a period of biological andcultural stasis. But did it also denote amentally and behaviourally conservativeAcheulean mind? Since the mid-1990s,Dr Petraglia has been looking at newways of evaluating the evolution ofcognition in Acheulean hominids byclosely analysing the evidence of theirmanufacturing and landscape practicesused to create these tools.

An opportunity to do this presenteditself when he and Professor K Paddayyaof Deccan College (Pune, India)discovered a well-preserved Acheuleansite in the semi-arid Hunsgi Valley ofsouthern India. The Isampur Quarry wasuncovered when the local IrrigationDepartment cleared much of the metre-deep thick black silt; extensiveexcavations since that time have exposedan extremely well-preserved Acheuleanquarry, the first to be discovered on theIndian subcontinent.

Crucially, the site was found to belittered not just with tools but with tool-making debris: here was a rare chance toexamine tool-making procedures andlandscape behaviours of a million yearsago.

The quarriers’ tale Our ancestors came to the quarry toprise up the limestone bedrock to maketheir handaxes and cleavers. Research isshowing that the tools they made werenot the end result of chance andserendipity but were the outcome ofintention, thought, planning andmemory.

The quarriers understood themechanical properties of their materialsand they approached the natural rockslabs with specific goals in mind:handaxes were crafted from tabularslabs; cleavers from flakes struck fromlarge boulder-sized slabs. As theyworked, the remnants of theirendeavours littered the occupation area,some 60 square metres in excavations, as chipped stone cores, flakes andchunks. They planned and anticipatedevents – making cleavers from largeflakes needed complex procedures toachieve the desired result; hammerstonesof different sizes and raw materials had to be brought into the quarry from sources more than a kilometreaway. All of this suggests the hominidswere capable of deliberate calculation,adjusting techniques to circumstances,remembering activities and solvingproblems. Intriguingly, what it alsosuggests is that socially learned andtransmitted behaviours were passed from individual to individual: in short,

some form of communication wasoccurring.

Interestingly, though, these hominidspractised behaviours that are unfamiliar –once tools were used, they were quicklyabandoned. Stone tools found at sitesaway from the quarry showed limitedevidence of resharpening. It seems thatearly humans made tools, carried themacross the landscape for an intendedgoal, and then discarded them. Thisdiscard behaviour has few analogiesamong modern foragers, who typicallyretain and repair their toolkits over thelong-term.

Thinking hominidsAcheulean hominids lived in a range ofecological settings with an assortment ofdifferent resources available to them –limestone, flint, quartzite, and so on. A comparison of Indian tool-makingprocedures and landscape behaviourswith other sites around the globe showsthat groups adapted to novelcircumstances in different ways and yetstill employed the same technology. Suchevidence implies a learned behaviourrevolving around a set of conventions orrules in tool-making. Some of theabilities that we see in the artefactualrecord are also found in the tool-makingand tool-using capabilities of someprimates and certain bird species.However, there are also relativedifferences in tool-using behaviours,suggesting that early hominids had anextended ability for planning beyond theimmediate future, relying on abstractionand memory for satisfying future needsover space and time. While this may bethe case, the hominids were not asinnovative as we might expect,producing similar tool forms overhundreds of thousands of years, perhapslacking the planning depth to anticipateconditions in the far-off future.

Comparative research should allowus to interpret how the mind evolved inhominids and how their behaviourdiverged from other intelligent animals.Further detailed archaeological andpalaeoanthropological studies of theAcheulean record will continue to shedlight on cognition in these earlyhominids, whose brains were three-quarters the size of our own.

For more information, please contact the author Dr MichaelPetraglia ([email protected]) at the Leverhulme Centre for Human Evolutionary Studies(www.human-evol.cam.ac.uk).

Excavating the Isampur Quarry, India

Dr Michael Petraglia

DR M

ICH

AEL PETRA

GLIA


PREVIEW

Every three years since 1882, Universityof Cambridge students have broughtancient Greek tragedies to life againthrough their performances in theCambridge Greek Play, a showcase oftheatrical and academic expertise that isspoken entirely in the original language.

The first play – Sophocles’ Ajax –was, as the publicity of 1882 boasted,the first full performance of a Greektragedy in ancient Greek in the modernworld, and the show rousedextraordinary interest. It was reviewed inall the national newspapers, and specialtrains had to be put on from London tobring the fashionistas up to Cambridgeto see the event of the season. Englandwas still in the grip of an intense‘philhellenic’ love of all things Greek;classics took up 80% of the curriculumat the best schools and universities; theneo-classical paintings of LawrenceAlma-Tadema and Frederic Leightondrew crowds of thousands; Greek lovewas the ‘dirty secret’ of the fin-de-siècledecadents. For Victorian England, theCambridge Greek Play represented a rare

chance to see an art form that featuredvividly in the cultural imagination.

Archaeological accuracy reallymattered to the Victorian audience – theplay had to embody the best scholarship,the most recent research. In 1882, thiswas ensured by the involvement of theworld-famous Greek scholar Sir RichardClaverhouse Jebb, Regius Professor ofGreek. This connection with researchcontinues today, with a thrivingacademic interest that both feeds intoand benefits from the performances.What can the surviving plays tell us ofancient Athenian society? How can weknow how to pronounce a long-deadlanguage? How can the ancient worldinform our understanding of the modernworld? What is at stake when Greektragedy is staged in the theatre today,and how are its most difficult problemsto be faced? It is this final question thathas been of particular interest to me –how audiences might see ancient Greektheatre accurately realised on stageagain, 2500 years after it was born inAthens.

Resurging interest That first astounding show in 1882heralded one of the most surprisingdevelopments in modern westerntheatre. Since the turn of the 20thcentury, ancient Greek plays havebecome part of the repertoire of allmodern theatres and, since the 1970s,there has been the most remarkableexplosion of performances of Greektragedy across the world – not just inEurope and the USA, but also in Japanand Africa and Russia. In London, Parisand New York, almost no year goes bywithout a revival of one of these classics.In 2001 alone, there were 17productions of Aeschylus’ great trilogythe Oresteia in the USA, which is morethan there were in the whole world inthe first 65 years of the 19th century. InLondon, three separate productions ofSophocles’ Electra were staged over afew months. When theatre director PeterSellars wanted to stage his anguish atthe Gulf War in the early 1990s, heturned to Aeschylus’ Persians – inCalifornia, Edinburgh and Austria. There

Greek tragedy: setting thestage todayWith thecurtains justclosed on the40th CambridgeGreek Playsince the 1880s,Greek classicistSimon Goldhillreflects on howthis creativegenre stillspeaks to amodernaudience.


PREVIEW

is no sign of this growth slowing, oncampus or in the professional theatre.Greek tragedy seems once again tospeak urgently and authoritatively to amodern audience.

A voice in modern timesWhy does Greek tragedy speak to ustoday? As with the 5th century BC, ourage is an era of great confidence in theprogress of science and knowledge:Greek tragedy ruthlessly exposes thepretensions in human claims to controland certainty. As with the 5th centuryBC, our age is obsessed with the tensionbetween the brutal realities of war andthe rhetoric of politicians: Greek tragedyanatomises this tension with painfulinsight. Moreover, Greek tragedy isobsessed with conflict between thegenders, between public and privateduty, between self-control and a sense ofhelplessness in the face of the world’sviolence: all this too finds a powerfulecho with modern audiences.

CO

URTESY

OF TH

E CA

MBRID

GE G

REEK PLA

Y C

OM

MITTEE

Staging Greek tragediesIt is far from clear how these greatmasterpieces of theatre should betranslated from the page into thetheatre. When the genre first flourishedbetween 500 and 300 BC, theconvention was for actors to wearspecially crafted masks. All the actorswere male, with a limit on how manycould appear on stage at one time, andthe chorus had to be composed ofAthenian citizens.

How can the old conventions of thechorus work without looking like aHollywood musical? Can masks evokeanything but bad clowns for today’stheatre? Is Greek tragedy destined to becrushed by its own formality, and end upas no more than men in black yellingportentous clichés at each other?

The book How to Stage GreekTragedy Today stems first from myresearch into ancient theatre and thehistory of theatre performance: I havebeen engaged for many years withexploring the political and social impactof theatre in ancient Athens, as well aswith how these old plays became soimportant in the cultural life of Europe,especially around the turn of the 20th century. But my concerns in thisbook also come from a more direct setof experiences. I have been deeplymoved by some great performances inthe theatre; I have also been annoyed,bored, outraged by others. I wanted toexplore why so many productions failed,and why the truly great productionswere great.

I also had the hands-on experience ofproducing the Cambridge Greek Playover 12 years, with two outstandingdirectors – Dr Jane Montgomery, whowas the Leventis Visiting Fellow in GreekDrama, and Annie Castledine, from theComplicite Theatre Company and whohas also directed at the National Theatreand the Royal Shakespeare Company.Seeing how professional theatre is madeat the ‘down-and-dirty’ level is notsomething most academics are privilegedto do, and anyone who writes abouttheatre can learn a lot from such anexperience. But the immediate stimulusto write my book was when I was askedto provide some suggestions for VanessaRedgrave to read about tragedy – shewas rehearsing a production of Hecubaat the time. I found to my chagrin (andto the detriment of my dignity as aCambridge professor) that there wasnothing I could really recommend to anintelligent modern actor or director tohelp them when daunted by the task ofperforming Greek tragedy. So I sat downand wrote what I hope will answer thatneed.

I examine the six most pressingquestions any company faces with the

task of staging a Greek tragedy: thetheatre space, the chorus, the actor'srole, the relationship between tragedyand politics, the translation, and therepresentation of the gods and heroes. I look at what we can learn from theancient world about these issues, howthe most successful modern productionshave dealt with them, and how acompany can negotiate a way throughsome of the most difficult problemsthese texts provide.

My hope is that actors and directorsembarking on the journey of staging aGreek play might have some guidance. I hope too that, for the reader wishing toknow more about these truly remarkableplays and their extraordinary re-emergenceon modern stages, this might inspirethem to consider what makes Greektragedy so exciting and so relevant agenre today.

For more information, please contact the author Professor SimonGoldhill ([email protected]) at theFaculty of Classics. Please go towww.the-medea.co.uk for moreinformation on the recent play,Euripides' Medea. How to StageGreek Tragedy Today is published byUniversity of Chicago Press.

Olga Tribulato (right) as Teiresias and Marta Zlaticas Oedipus in Sophocles’ Oedipus the King, 2004

Professor Simon Goldhill

PERRY H

ASTIN

GS

'Simon Goldhill's new book isenthralling. A 'can't put down'and a 'foreverre-read'. Hisdetailedanalyses of somany pastproductionsare rare andexciting. Hisunfolding ofthe Greektexts and the manydifferent translations is bothinstructive and exhilarating.' Vanessa Redgrave CBE, actress


FEATURE

Often considered to be wartime kitsch,frequently ending up in junk shops or in rubbish skips, the artefacts made bypeople of the Channel Islands duringWorld War II (WWII) have only recentlybeen recognised as being worthy ofacademic enquiry. Like the ‘trench art’created in the trenches of WWI, theseartefacts were made from whateversupplies were at hand – coins, tin cans, spent shells, even barbed wire – in a time of military conflict and itslegacy of deprivation and defiance.Thanks to funding from the BritishAcademy, the artefacts held in Islandmuseums and private collections arebeing catalogued and photographed by Dr Gilly Carr from the Department of Archaeology. They tell a story of the quiet yet determined resistance to occupation, as well as the patrioticunderpinning of a society under threat.

Occupied IslandsThe Channel Islands, the only part ofBritain that was occupied during WWII,endured a period of history that left suchan impact on the collective psyche of itspeoples that it forms an integral part ofChannel Island identity and heritagetoday. Island newspapers still featurearticles about the occupation; manyIslanders still talk about it daily.

The Islands were consideredindefensible by the British forces but were deemed by Hitler to be of strategicvalue – their occupation would effectivelystick a thorn in Britain’s side and woundmorale. Seeing the smoke rising from the coast of nearby France and hearingthe approaching shell-fire, around a thirdof the population of the Islands evacuated to the UK in the days beforethe Germans arrived. By the end of June1940, the Germans had landed in theIslands following a bombardment of

A campaign of silent resistanceA fascinating

study ofwartime

artefacts isuncovering a

story ofsymbolic

resistance andcreative

necessity inthe ChannelIslands 60years ago.

PHO

TOG

RAPH

S BY D

R GILLY

CA

RR; ARTEFA

CTS C

OU

RTESY O

F CH

RISTINE BA

ILEY (TO

P LEFT) AN

D PETER A

ND

PAU

L BALSH

AW

, VALLETTE M

ILITARY

MU

SEUM

,ST PETER PO

RT, GU

ERNSEY

(OTH

ERS)


FEATURE

St Helier and St Peter Port, the capitals ofJersey and Guernsey. The occupation wasgruelling. In September 1942 and February1943, a total of 2200 Islanders weredeported to internment camps on thecontinent and those who remained weretrapped.

In the last year of the occupation,supply lines to the Islands were cut offafter the Allied invasion of France, and theIslanders and occupying army alike beganto starve. In the last six months of the war,during the bitter winter of 1944, gas,electricity, wood-fuel and soap ran out.Had the Red Cross ship the S.S. Vega notarrived in December 1944 bearing parcelsof food, the population might have starvedto death.

‘V’ for victoryAlthough the occupied Islanders wereunable to mount a proper resistance, theartefactual record provides compellingevidence for a campaign of ‘silentresistance’ using the ‘V for victory’ sign,which has previously been undocumentedin such detail. Following a BBC broadcastat the time, the people of occupied Europewere encouraged to use this as a symbolof resistance, and a range of artefacts thatwere made in 1941 and 1942 depict Vs.The most intriguing of these are badgesmade from coins, where the maker filedaround the King’s profile and scored theletter V underneath before attaching asafety pin to the reverse. These were wornunderneath jacket lapels and were flashedto trusted friends in the street.

The V-sign campaign was also used asa form of resistance by the English-bornIslanders from Guernsey, Jersey and Sarkwho were deported to civilian interneecamps in Germany and Austria. MontyManning shaved his beard into a V-shape;Byll Balcombe engraved a mug made froma Red Cross tin with a V so that he coulddrink a toast to victory; one woman wovea V into the plaited string sole of her shoe;and Nellie May Faulder embroidered atablecloth with Vs around the edge, with adedication to George V in the centre –even though George VI was on thethrone!

Internees were largely kept alive byRed Cross parcels, the contents of whichthey recycled – including cardboard, string,tins, cellophane packing material andwooden parcel crates – to make a range ofartefacts to distract themselves from thecircumstances of their internment. Passingthe time was a key objective, as shown bymany of the items such as chess andcribbage boards and sewing baskets. Otheritems, including women’s fashionaccessories such as shoes, handbags, hatsand belts, and items made from Red Crosstins, such as plates, trays, a communionchalice and a football trophy, are amongsurviving artefacts from the camps.

A symbolic warOne particularly fascinating element of theunfolding story has been the evidence of a‘symbolic war’ waged silently between theGermans and the locals over the use ofGuernsey’s crest. The Crest – three lionspassant guardant – was stamped or carvedby the Germans on nearly every item theymade as a symbol of their control of theIslands. However, the Islanders reclaimedthis symbol of their identity, making ringsand badges out of Island coins thatdisplayed the crest. They quietly andunobtrusively wore these on their bodyand clothes, fighting back against theappropriation of their islands and symbols.

Rewriting historyThis research has overturned some long-held beliefs about the occupying army. Forexample, it is widely believed that Germansoldiers, like the Islanders, had also run outof gas, electricity and wood-fuel monthsbefore the end of the war. However, manyGerman-made artefacts were carved out ofsolid lumps of wood as late as 1945. Onesoldier had carved an electric lamp standwith the date 7th May 1945 – two daysbefore liberation – which raises thequestion of whether the soldiers sufferedas much as the locals in terms of a lack ofsupplies.

Resonance of troubled timesThe physical artefacts speak volumes of apeople living under occupation. Notknowing when the war would end, orhow they would survive until that time,some of the artefacts show how Islandersused their creative talents to make lifemore bearable. An ‘Occupation Monopoly’board was produced, to be played duringthe hours of curfew, with instructions suchas ‘exceeded gas ration, go back 2’,‘receive Red Cross letter, go forward 3’ and‘identity card lost, go back 4’. Findingssuch as this offer a snapshot of life underthe occupation and illustrate the proud,defiant and patriotic spirit for whichChannel Islanders have long been known.

For more information, please contact the author Dr Gilly Carr([email protected]) at the Departmentof Archaeology. Although Dr Carr wasbrought up in the UK, her family isfrom Guernsey and were variouslyevacuated, occupied and deportedduring WWII. She is currently writing a book and preparing two museumexhibitions on the subject of artefacts of the occupation.

Dr Gilly Carr

Examples of the wartime artefacts held inmuseums and private collections in the ChannelIslands. Above: Monty Manning shaved hisbeard into a 'V for victory'; drawn by fellowinternee Eric Sirett. Top left: 'V for victory' mugengraved by Byll Balcombe in the internmentcamp of Biberach. Top right: Barbed wire broochmade in Biberach. Bottom left: 'V for victory'badge made from a coin by Alf Williams.Bottom right: George V coin on the side of acigarette lighter.

CO

URTESY

OF PETER SIRETT


Complications in pregnancy representa persistent and major problem inpublic health. The first three monthsafter conception are known to be themost critical, with as many as 20% ofpregnancies lost during this time. Forpregnancies that develop beyond 24weeks, between 0.5 and 1% result indeath of the baby, either in the wombor in the first four weeks of life.Premature birth can incur majorcomplications associated with delivery,immediate care of the infant,childhood diseases, and educationaland social problems in later life. Notonly is there an emotional cost tofamilies, but an economic assessmentin the USA reported that thecumulative subsequent healthcare andsocial costs associated with one year’sworth of pre-term deliveries was $26 billion. Understanding andintervening to prevent these events isclearly crucial.

Although some advances havebeen made, the dismaying fact is thatthe rates of stillbirth have generallyremained static over the past 20–30years. This partly reflects an incompleteunderstanding of the biological eventsthat lead to these complications of

pregnancy. Determining what thesemechanisms might be is essential fordevising new strategies of intervention, andapplying in-depth scientific studies tohuman pregnancy is now seen as vital.

Two multidisciplinary initiatives inCambridge have recently embarked onimproving our understanding of pregnancyand its outcomes: a large antenatalscreening of women at the Rosie MaternityHospital in Cambridge and the recentendowment of a Centre for TrophoblastResearch within the School of BiologicalSciences. Both initiatives build on thewealth of expertise in the biology ofpregnancy that exists across Cambridge.

Screening for adverseoutcomesA four-year research project that aims tomonitor 5000 pregnant womencommenced in 2007 under the leadershipof Professor Gordon Smith in theUniversity’s Department of Obstetrics andGynaecology. A multidisciplinary team oftranslational researchers in both theSchool of Clinical Medicine and the Schoolof Biological Sciences are participating inthe project, which is funded under theWomen’s Health theme of the UK

Department of Health’s CambridgeComprehensive Biomedical ResearchCentre.

Women enrolling in the study arescanned and give blood samples at 12, 20, 28 and 36 weeks of gestation,allowing detailed characterisation of thebaby’s growth and development. Thanksto an industrial collaboration with GE Healthcare, the long-term loan of twostate-of-the-art scanners will enable real-time three-dimensional scanning of thebabies in utero. At birth, samples ofplacenta and cord blood will be obtainedand stored.

The study is prospective; for thosewomen whose pregnancies sadly havecomplications or adverse outcomes (suchas pre-eclampsia, spontaneous pre-termbirth, stillbirth or low-birth-weight babies),the stored samples will be retrieved andcompared with controls. These samplesthen become the focus of extensive clinicaland biological analyses to try to establishthe cause. Studies will analyse thedevelopment and function of the placentaand the effect of oxidative stress; theexpression or silencing of genes in relationto whether they came from the mother orthe father (known as genomic imprinting);the maternal–foetal immune interaction;

Great expectations inpregnancy research

Most pregnanciesdevelop normally but

when complications arisethey can have

devastating effects. Tworecent initiatives inCambridge hope to

deliver a newunderstanding of eventsduring this critical period

of human life.Human placental villi showing evidence of

oxidative stress

FEATURE


FEATURE

PROFESSO

R GRA

HA

M BU

RTON

Three-dimensional scan of a baby in utero

Participating researchers

Antenatal screening initiative (Principal Investigator: Prof Gordon Smith)Dr Steve Charnock-Jones and Dr Miguel Constância (Dept of Obstetrics andGynaecology); Prof Graham Burton, Prof Abby Fowden, Dr Dino Giussani and Dr Anne Ferguson-Smith (Dept of Physiology, Development and Neuroscience); Dr Ashley Moffett (Dept of Pathology); Prof David Dunger (Dept of Paediatrics);Dr Ian White (MRC Biostatistics Unit); Dr Ken Ong (MRC Epidemiology Unit).

The project is within the Women's Health theme of the CambridgeComprehensive Biomedical Research Centre – a partnership between CambridgeUniversity Hospitals NHS Foundation Trust and the University of Cambridge, andcreated by the National Institute for Health Research (NIHR). These themes focuson translating advances in basic medical research from the laboratory to thehospital clinic.

Centre for Trophoblast Research (Director: Prof Graham Burton)Participating researchers will be announced in 2008. The Centre will facilitateresearch by providing flexible and responsive funding for seminars, workshopsand visiting scholars, as well as laboratory space in the Department of Physiology,Development and Neuroscience. The Centre also aims to encourage the nextgeneration through graduate studentships and postdoctoral fellowships.

GE H

EALTH

CA

RE

and the genes that are expressed in theplacenta. The MRC Epidemiology Unitwill conduct follow-up studies of thegrowth and development of the babieswho have been carefully monitoredduring the pregnancy.

The hope is that this detailedcharacterisation of foetal development,on such a large scale, will lead tomechanistic studies on the causes ofcommon clinical problems in pregnancy.As well as providing refined riskassessment, novel treatments might beidentified that could improve theoutcome of pregnancies in womendeemed to be at higher risk.

Centre for TrophoblastResearchThe recent endowment of the Centre for Trophoblast Research, due to belaunched on 9 July 2008, is a highlyinnovative initiative aimed at promotingresearch into trophoblast biology bothwithin Cambridge and on the widernational and international stages. Thetrophoblast is the cell type that forms theinterface between the foetus and itsmother, supplying nutrients to supportthe growth of the foetus. It isfundamental to successful pregnancy andmust interact intimately with thematernal cells lining the uterus, leadingto the formation of the placenta.

In humans, this interaction isparticularly invasive and, during the firstfew weeks of pregnancy, the foetusbecomes completely embedded withinthe wall of the uterus. This form ofplacentation, seen only among the greatapes, poses unique immunological andhaemodynamic challenges. The invadingtrophoblast cells, which are geneticallyrelated to, but distinct from, those of themother, must negotiate passage with her

immune system to allow them to reach their target – the specialisedblood vessels in the wall of the uterus.As a result of this invasion, the vesselsundergo major structural changes that ensure the placenta has a plentiful and continuous supply ofblood in later pregnancy.

There is now abundant evidencethat the major complications ofpregnancy are associated with deficient trophoblast invasion, resulting in aberrant maternal bloodflow to the placenta. Researchperformed in Cambridge hasdemonstrated that, paradoxically, too much flow in early pregnancyresults in miscarriage, whereas too little in later pregnancy is associatedwith low birth weight and pre-eclampsia. These new insights haveradically changed our understanding of human pregnancy and have helpedto explain why miscarriage and pre-eclampsia are virtually unique tohumans. Studying trophoblast biologyis therefore not only of basic scientific interest but is also key tounderstanding the root causes of these pregnancy disorders.

Raising hopes for futurepregnanciesThe aim of these multidisciplinaryinitiatives across Cambridge is to arriveat a better understanding of thebiology of normal and complicatedhuman pregnancy. Only by doing socan scientists hope to develop newdiagnostic tests to identify women atincreased risk of complications and,potentially, new interventions thatmight prevent the life-long effects ofthese complications on mothers andtheir children.

Professor Gordon Smith (left)

and Professor Graham Burton

For more information, please contactthe authors Professor Gordon Smithat the Department of Obstetrics and Gynaecology ([email protected])or Professor Graham Burton at the Department of Physiology,Development and Neuroscience([email protected]). Please go towww.trophoblast.cam.ac.uk for moreinformation about the Centre forTrophoblast Research.


To what extent do living organismsabsorb pollutants in our environment?Are particular ‘chemical cocktails’ morerisky than others? Do currentecotoxicological risk assessmenttechniques adequately protect theenvironment? These are the sorts ofquestions that interest Drs Oliver Jonesand Julian Griffin in the Department ofBiochemistry, who are working as part ofa Europe-wide integrated researchproject to develop better tools toevaluate the chemical risks we face ineveryday life.

EcotoxicologyIt is generally acknowledged that manyorganisms in the environment areexposed to a large variety of pollutantsduring their lifetime; a fact borne out byadvances in analytical technology. Forexample, many people will have heard ofthe effects on fish populations caused byendocrine-disrupting compounds insewage, whereby some male fish livingdownstream of sewage treatment plantswere found to have developed femalecharacteristics, leading to a reduction intheir ability to reproduce. In recent years,a plethora of other anthropogeniccontaminants such as pharmaceuticals,personal care products, pesticides andflame retardants, and the potential forthese man-made products to work theirway into the food chain, have alsobegun to be of concern to environmentalchemists.

However, the majority of thesepollutants are present at extremely lowconcentrations and so it is difficult toascertain whether or not they have anoverall effect on ecosystem health,especially if outward effects are minimal.An added complication is the fact thatthe interaction between the environmentand organism health is extremelycomplex, with chemical, biological,physical and geographical stressors eachcontributing to toxicological effects overtime.

It’s important therefore to developmethods for assessing the cumulativerisks for a range of species that are beingexposed to mixtures of pollutants at non-lethal levels. In this way, steps can betaken both to improve safety in theenvironment and to safeguard ecologicalhealth.

MetabolomicsOne technique that shows a great dealof promise in the area of ecotoxicology ismetabolomics. This rapidly emergingdiscipline measures the thousands ofnaturally occurring small molecules(metabolites) such as sugars, organicacids, amino acids and lipids that are theproducts of cellular metabolism. Anorganism’s ‘metabolome’ is its fullcomplement of metabolites, in the sameway that its genome is its completegenetic content.

Why study metabolic changes? Well,these changes often happen much earlier

Puttingmetabolism

on the eco-map

One of the latest‘omic’ technologiesto emerge –metabolomics – isbeing used to createa snapshot of howenvironmentalchemicals affectliving organisms.

FEATURE


in an organism than either tissueaccumulation of pollutants or inducedhistopathological changes. The techniquecan be used to give a biochemicalsnapshot of a cell, tissue or indeed whole organism at a moment in time.When an organism is stressed ordiseased, its metabolic pathways areperturbed. Advanced computer-assistedpattern recognition techniques can then be used to assess the differences in metabolic profiles between samplegroups. Metabolomics therefore offers aparticularly sensitive method to monitorchanges in a biological system and isproving to be an outstanding tool forstudying ecotoxicology.

NoMiracleThe environmental research in Dr Griffin’sgroup in the Department of Biochemistryis part of a European Union (EU) researchproject involving 38 laboratories spreadacross 16 countries and is known asNoMiracle (for ‘Novel Methods forIntegrated Risk Assessment ofCumulative Stressors in Europe’). The project seeks to improve ecologicaland environmental risk assessment in the EU, and to help scientists gauge the impact of chemicals on theenvironment and human health.

The Cambridge team are developinganalytical techniques based on high-throughput analysis of metabolites from organisms at different positions inthe food chain, such as earthworms,nematodes, slime moulds, marine mussels and water fleas. Being able tostudy such a broad set of experimentalspecies has been possible because oflong-term collaborations with the Centre for Ecology and Hydrology (part of the UK Natural EnvironmentResearch Council), King’s CollegeLondon, the University of PiemonteOrientale in Italy and the University ofAntwerp in Belgium, all developed aspart of the NoMiracle project.

Using state-of-the-art nuclearmagnetic resonance spectroscopy andgas chromatography mass spectrometry,long-term studies are being run toestablish a basal metabolic profile foreach of these species, as well as howthese profiles change in response to toxic insult. By looking at the differentpatterns of metabolic profiles betweenorganisms, a comprehensive description is being built up of how each of them responds to stress andtoxicity. One important finding has been that biochemical effects are often observed at lower chemicalconcentrations than were previously thought to cause any effect whenassessed using traditional toxicologytesting techniques.

Assessing the risksWhy is there a need for improved riskassessment in ecotoxicology? In currenttoxicity tests, an organism is typicallyexposed to a single chemical in a strictlycontrolled laboratory setting, over arelatively short period of time (typicallydays or weeks). Yet, in the environment,organisms will clearly be exposed tomany different pollutants possiblythroughout their entire life. An accuraterisk assessment must take into accountcumulative effects rather than just directeffects and single factors. Organisms arealso often likely to be stressed by otherfactors not present in a laboratorysetting. For instance, work within theNoMiracle project has demonstrated thatorganisms can be affected by pollutantsat much lower levels than thosepredicted from traditional toxicity tests ifthey are also stressed by other factorssuch as co-exposure to pollutants,temperature extremes or food restriction.

The work in the Cambridge sectionof the NoMiracle project is moving intoits third and final year. The research isshowing that the accurate assessment ofchemical mixtures is more complex thancurrent testing regimes allow for and theaim now is to use these results todevelop a new framework for assessingthe effects of complex mixtures ofpollutants. The ultimate goal of theNoMiracle partners is to changeecotoxicology policy in the whole of theEU, so that long-term, multi-stressorexposure testing is considered asstandard. This will offer greatimprovements in understanding andmitigating the effects of cumulativepollution exposure on the health of our ecosystem.

For more information, please contact the authors Dr Oliver Jones([email protected]) or Dr Julian Griffin([email protected]) at theDepartment of Biochemistry. Further details of the NoMiracleproject can be found online(http://nomiracle.jrc.it/).

The tricarboxylic acid (TCA) cycle, a metabolicpathway involved in energy generation, is oftenperturbed following exposure to pollution

Dr Oliver Jones (left)and Dr Julian Griffin

MISS K

ARO

LINA

LAD

A A

ND

DR O

LIVER JO

NES

FEATURE

CH

RIS GREEN


FEATURE

Examining how western legal systemshave developed, and understanding thefactors that have shaped the dynamics oflegal change, has been at the heart of athree-year study that reached itsconclusion in December 2007. TheEuropean Legal Development project,funded by the Arts and HumanitiesResearch Council (AHRC), was led byProfessors John Bell and David Ibbetsonat the Faculty of Law. It brought togethersome 70 academics, including eminentscholars and early career researchers,from 10 different jurisdictions acrossEurope.

Although concerned with legaldevelopment and involving the researchinput of many lawyers, the project alsobenefited from the insights of historians,philosophers and scholars in otherdisciplines. It stands as a model for hownational scholars can be supported anddeveloped in an international context inthe humanities and social sciences.Regular meetings and websiteinteractions have built a network ofresearchers that will continue long afterthe project itself.

Illustrations of legaldevelopmentOf course, the topic ‘How do westernlegal systems develop?’ is very broad.Does the law change principally in stepwith the state of the economy and ofsociety at large, or does it respond to itsown internal dynamics of change? Evenwhere there are strong similarities

between the social and economic factorsthat fuel changes in the law, how dolegal systems in different Europeancountries respond? To make these areasof enquiry more manageable, they weretackled through the lens of a particularbranch of law that illustrates legalchange over a specific period ofEuropean history: the liability for harmcaused to others by fault in the years1850–2000.

In England, the core principles fordeciding when one individual has tocompensate another for the harm theyhave caused date back to the 14thcentury; in continental Europe, they dateback to the 3rd century BC. This researchhas sought to chart how these oldprinciples have needed to change and toseek explanations for what happened.

The law relating to fault began toalter around 1850. Enormoustechnological and economic changestook place as industrialisation andurbanisation occurred in differentcountries, influencing many aspects ofsociety, including the law. With the rapidrise in the use of steam boilers to powerfactories, boats and trains, accidentsbecame more frequent. Machinery wasboth more complex and less predictablethan before, causing injuries toemployees and passengers. Although in1850 there were many similarities inapproaches to liability for fault across thelegal systems of western Europe,significant divergence began to occur inthe years that followed.

Constructing the caseTo understand the forces behind legalchange, we have to look beyond theexperience of one country and investigatemany. To this end, six case studies wereinvestigated by academics fromjurisdictions across Europe: England,Scotland, The Netherlands, Austria,Germany, Sweden, France, Italy and Spain.

The six case studies, each convened bya different project member, focused onthe problems that the law has faced inthe 150-year period under study: liabilityin relation to technological change;liability between neighbours; liability fortraffic accidents on rail and road; liabilityfor products; liability for medicalnegligence; and legal doctrine, or thewritings of legal academics trying to setout the principles of this developing law.

Factors for changeIn a second stage, the project examinedthe factors that actually shaped legaldevelopment in the fields studied. Certainlegal institutions, such as law reformbodies, have been important in fosteringchange in the law. But these bodies havedepended on key individuals who havepromoted change, typically by persistenceover many years. Governments oftenpromote legislation in response to recentprominent crises or disasters, and at suchmoments proposals that are alreadyformulated are often seized upon.

To what extent does the law reflectdevelopments in social and political ideas?Sometimes there is a clear connection. For

KEITH

HEPPELL

Finding F ult a

A multicentre project ledby the Faculty of Law has

reached its conclusion,having studied over a

century’s worth ofEuropean legal changes

relating to liability.


example, the French social theorist ÉmileDurkheim influenced key French legalwriters of the late 19th century to arguethat the law should provide compensationout of social solidarity with those whowere injured, rather than focusing simplyon whether a responsible individual was atfault. But, in other areas, it is the opinionsof specialist technical experts that shapethe law. For example, the law governingasbestos was strongly influenced byrelatively small numbers of people ininspectorates, rather than by a broadmovement of opinion or ideas.

Changes in the economy clearly havean impact on law – new problems arisethat the law has to solve. But it is less clearwhether the pattern of solutions directlyresponds to economic interests. For

example, the easy availability of insuranceis frequently invoked in arguments forlegislation or in court. But it does notdictate a solution. Although Germany andFrance introduced insurance-basedcompensation for road accidents, Englanddid not, even though the economicconditions and the availability of insurancewere similar. And so, similar economies donot necessarily adopt similar legalsolutions.

This study has illustrated that privatelaw can operate with a kind of relativeautonomy from contemporary social andpolitical ideas or economic interests.History and legal ideas can be powerfuldeterminants of how far private lawcontributes to the solution ofcontemporary problems.

Professor John Bell

Members of the European LegalDevelopment project met in December 2007

to discuss findings of the three-year study

Three main trends in the law relating to liability have become apparent from this study:• Victims of accidents have gradually found it easier to obtain compensation, either

because the burden of proof has shifted towards the person causing the injury orbecause liability no longer depended on proof of fault.

• Simpler and less expensive compensation systems have gradually been createdoutside private law (the law of relations between individuals). For example,although the victims of boiler and railway accidents tended to be employees, they rarely gained compensation through private law but instead through state-created insurance-based workmen’s compensation systems. In Germany, Swedenand France, such schemes have also replaced private law for most road accidents,and Sweden and France have now adopted similar schemes for medical injuries.

• Although private law has played a minimal role in incentivising accidentprevention, other forms of regulation could have an impact. For instance, stateregulation on the siting of boilers, or of buildings or crops alongside railway lines, as well as regulation related to determining who can practise as a doctor, hasplayed a very important role in reducing the incidence of harm.

Coming to a conclusion

For more information, please contactthe author Professor John Bell([email protected]) at the Faculty ofLaw or visit http://eld.law.cam.ac.ukand http://www.findingfault.co.uk

KEITH

HEPPELL

FEATURE


FEATURE

Re bellion, rep

John Morrill exploresone of the most

extraordinary and leastunderstood aspects of

Anglo-Irish history –the rebellion of 1641.

The true course of events of the Irishrebellion of 1641 has never been fullyknown. Initiated by disaffected IrishCatholics rebelling against Protestantsettlers, the rebellion quickly escalated inviolence, resulting in widespread killing.But was the rebellion intended to be abloodless coup that spiralled out ofcontrol, or were the thousands ofProtestants deliberately driven out andmassacred? What’s clear is that the yearsthat followed were a time of savagerevenge for the events of 1641 – OliverCromwell arrived with 30,000 Englishtroops to conquer Ireland in the name ofthe English Republic and to exact ‘a justjudgement of God upon those barbarouswretches, who have imbrued their handsin so much innocent blood’ – and thegroundwork was laid for Ireland’sCatholic–Protestant divide.

A curious aspect of the rebellion isthat although it is the least understoodof all the great massacres of Europeanhistory, it is amongst the best recorded.Historical narratives in the form ofeyewitness accounts of those who livedthrough the rebellion are still in existencein the library of Trinity College Dublin,where they have remained largely

unstudied. This is chiefly because there istoo much of a record of what happenedand it has taken until now, withimprovements in technology and thepolitical climate, to conspire finally tomake it possible for the secrets of the‘1641 depositions’ to be unlocked. Ateam of scholars in Cambridge, Dublinand Aberdeen are poised to do just this.Professor John Morrill from Cambridge’sFaculty of History is chairing the three-year project, working alongside ProfessorJane Ohlmeyer and Dr Micheál Ó Siochrú(Trinity College Dublin), and ProfessorTom Bartlett (University of Aberdeen).

Roots of an uprisingThe 1641 rebellion had roots stretchingback to the mid-16th century, when theIrish provinces were heavily colonised byEnglish settlers. Throughout the reign ofQueen Elizabeth I, the Englishgovernment, fearful that continentalCatholic kings would use Ireland as aspringboard for invading England toexploit the dynastic weaknesses (Elizabethwas, in Catholic eyes, a heretic bastardtyrant, unmarried and the last of her line),sought to impose strong Protestantcontrol of Ireland. This led to a dreadfulcycle: Catholic rebellion, repression of theuprising, replacement of Irish landownersby English as part of a ‘Plantation’ policy,then more rebellion, more repression andfurther Plantation.

In and after 1610, the largest of thePlantation policies, in which not only theIrish landowners but also the tenantfarmers and urban elites were displaced,affected large parts of Ulster in the farnorth of Ireland. Previous Catholic ownersand occupiers were driven into exile,where thousands either becamemercenary soldiers (‘Wild Geese’) in thearmies of the Habsburg kings or fell intodestitution.

For 30 years, the strong authoritariangovernment, softened by a blind-eye toprivate Catholic worship, kept thedispossessed of Ulster and elsewhere in check. But in 1641, England wasparalysed by the disputes that were to lead, a year later, to civil war. King Charles I’s puritan opponents hadplans to introduce much more effectivereligious persecution of the Catholic Irishand to make Ireland increasingly part ofan enlarged English state. This provoked,

Propaganda image of the type circulated incontemporary Europe depicting atrocitiesallegedly committed in the rebellion

REPROD

UC

ED BY

PERMISSIO

N O

F THE BO

ARD

OF TRIN

ITY C

OLLEG

E DU

BLIN

Deposition taken from a witness to the 1641 Irishrebellion; Trinity College Dublin, MS 833 f.8

REPROD

UC

ED BY

PERMISSIO

N O

F THE BO

ARD

OF TRIN

ITY C

OLLEG

E DU

BLIN


FEATURE

ression, retributionfrom late October 1641, a series of pre-emptive strikes by members of theCatholic nobility and, in the ensuingchaos, a series of what (unless thisresearch project tells otherwise) appear to be spontaneous revenge attacks onProtestant settlers that quickly got out of control.

An imperfect accountAlthough we have no idea how manypeople were killed during the events of1641, the most prudent estimates arethat 4000 died through acts of violenceand that 6000 more died of theconsequences of being driven out nakedinto the winter cold, while many morefled from their homes and made their wayeventually back to England. So much isclear. But the precise chronology andgeography of the rebellion have remainedhazy at best.

The English government had to dosomething to protect the EnglishProtestant settlers, but their own countrywas in chaos. They could not raise taxesto fund the army. So they borrowedmoney from 2000 venture capitalists (the‘Adventurers’) against the promise thatthey would receive two million acres ofIrish land once Ireland was conquered. To establish which land was to beconfiscated, all (mainly Protestant)witnesses to the rebellion werequestioned by government-appointedcommissioners and their accountsrecorded as ‘depositions’ that could beused in court.

Today, 3400 depositions are inexistence, providing the fullest and mostdramatic evidence we have for any eventof this kind before the 20th century. Theyadd to up 19,000 pages of testimony incrabbed 17th-century hands. TrinityCollege Library acquired the documents in1741 and for centuries there they haveremained, far too extensive for any onescholar to explore them all and in toopoor a condition for widespread access.Even with a team of researchers, it willtake a total of more than eight personyears to transcribe the accounts.

A new kind of historyThe spirit of co-operation between theUK and Irish governments following theGood Friday agreement has made it

possible to fund a project of this size –the most ambitious British-Irishcollaboration in the humanities everundertaken. Separate but linked fundingstreams in the UK and Ireland haveraised more than 1 million euros fromthe Arts and Humanities ResearchCouncil (AHRC) in the UK, the IrishResearch Council for the Humanities andSocial Sciences (IRCHSS) and TrinityCollege Dublin.

Once the depositions are capturedand online, they will constitute adatabase that can be arranged and re-arranged in any way a scholar wouldlike: by date, by map reference, even byact of violence. Many of the depositionsgive detailed inventories of goods takenand destroyed, affording unique insightsinto the material culture of a colonialsociety. Members of the general publicmight even use depositions to tracefamily trees. There are endlesspossibilities for further study, bothlooking backwards to the pattern ofexploitation that provoked the explosionof Catholic violence, and forwards to theway in which these massacres resulted inthe confiscation of 40% of the land ofIreland and its transfer from Catholicsborn in Ireland to Protestants born inEngland. These are events thattransformed Irish history and thereforeBritish and world history. Thiscollaborative project represents a newkind of history: one where the mediumand the message can change how weunderstand ourselves in time.

For more information, please contactthe author Professor John Morrill([email protected]) at the Facultyof History.

Professor John Morrill


IN FOCUS

Established in April 2005 from the Artsand Humanities Research Board (AHRB),the AHRC has an annual budget ofaround £90 million to fund research andpostgraduate study, as well as museumsand galleries associated with highereducation establishments. In 2006–2007,the value of awarded grants to theUniversity of Cambridge was £4 million.

The AHRC recognises not only theimportance of sustaining the arts andhumanities research base, but also ofensuring that the knowledge andunderstanding it generates is widelydisseminated. The two Cambridgeprojects highlighted here – the ShahnamaProject and Accessing Virtual Egypt –address this strategic priority in differentways. Indeed, the Council has

established itself as a leading authorityon research-based knowledge transfer(KT), with several new initiatives(including the KT Fellowship) launchedspecifically for the AHRC researchcommunity. Because the AHRC’sdefinition of KT is broad and flexible inimplementation, its impact has extendedto key societal and economic challenges.

Several new strategic initiatives havebeen planned for 2007–2008. In 2007,the AHRC launched a joint £5mprogramme on Religion and Society withthe Economic and Social ResearchCouncil (ESRC) and a £5.5 millionprogramme entitled Beyond Text:Performances, Sounds, Images, Objects.In 2008, work will begin with theEngineering and Physical Sciences

Research Council (EPSRC) on a Scienceand Heritage programme.

Alongside research grant activities,strategic programmes and KT, the AHRCis keen to highlight another priority: thesupport of postgraduate and early careerresearchers to protect the long-termsustainability and health of the UK’s artsand humanities research. A recent surveyshowed that 74% of AHRC awardholders go on to pursue academic andresearch careers, with the majority ofothers pursuing careers in creative andcultural sectors, non-profit organisationsand public services.

For more information on the AHRC,please go to www.ahrc.ac.uk

The Arts and HumanitiesResearch Council (AHRC)supports research within awide subject domain, fromtraditional subjects such ashistory, modern languagesand English literature, throughto the creative andperforming arts.

The project Accessing VirtualEgypt is breaking new groundin knowledge transfer betweenmuseums and prisons, withempowering results.

A pioneering programme is enablingprisoners to unlock the secrets of the pastas part of an initiative to improve theirprospects for the future. For the past fiveyears, Dr Sally-Ann Ashton, Curator of theEgyptian Collection at The FitzwilliamMuseum, has taken her knowledge ofancient Egypt into prisons, working closelywith prisoners and prison educationdepartments. The study of ancient Egypt isbeing used as a means of inspiringlearning through literacy, numeracy andart at several prisons including HMPEdmunds Hill. These new skills have alsosupported many prisoners of North Africandescent in exploring their own culturalheritage.

‘Many of the prisoners were servinglong sentences and asked whether therewas a way I could bring the museum tothem more directly,’ said Dr Ashton. ‘Theidea they came up with was a VirtualGallery – something they could access oncomputers in the prison and then carry onusing when they are released.’

As the idea grew, Dr Ashton realisedshe needed to spend more time in prison.

Winning an AHRC Knowledge Transfer(KT) Fellowship in 2007, one of the first 12to be awarded, has allowed Dr Ashton todo just this: ‘I needed to find out at first hand what the important issues are foreducating people serving different lengthsof sentences and I needed to work withprison education departments to see hownew resources could complement buildingbasic reading and writing skills.’

The KT Fellowship scheme is designed to support academics whowould like to embark on a programme of knowledge transfer that will ‘make asignificant difference beyond the world of academia.’ As Dr Ashton explained, it was important that any new resourcesshould have as wide a benefit as possible:’I wanted to do something that wassustainable and could reach potentially all of the prison population, as well as the wider public.’

Among other resources, one outcomeof the KT Fellowship will be the VirtualGallery, which was funded by the HeritageLottery Fund (HLF) and is due to belaunched later in 2008. The look and feelof this facility – a virtual walk through thereal museum, allowing the user to stop atany of the cabinets and examine theartefacts – has been driven entirely by theprisoners. ‘They wanted a resource thatlooked both like a computer game andthe outside world. The result is the nearest

thing we thought you could get to beingin the gallery yourself.’

Dr Ashton views the exchange ofknowledge as very much two-way: ‘Theyare a very inspiring audience. Theirquestions often make me go back and re-think, and I’m looking at my ownsubject quite differently as a consequence.’By evaluating the impact of her workduring the Fellowship, Dr Ashton hopes toarrive at a clearer idea of the best waymuseums can contribute to societythrough prison education.

For more information, please contact Dr Sally-Ann Ashton([email protected]).

Egyptology in prisons

BY PERM

ISSION

OF TH

E SYN

DIC

S OF C

AM

BRIDG

E UN

IVERSITY

LIBRARY

Dr Sally-Ann Ashton


Almost 1000 years ago, the Persian poetFirdausi created an epic poem of suchunparalleled sweep and power that, afterhis death in 1020, it continued to live onas a seminal expression of Iranian art,literature and history. The Shahnama (Bookof Kings) is the longest poem ever writtenby a single author and narrates the historyof Iran from the first King until the Arabinvasions in the early 7th century AD.

The story goes that the sum of moneyFirdausi was paid by the Sultan for hiswork of 35 years was so pitiful that hegave it to an attendant at the baths andleft the country. Little did he know that,for the next 800 years, his epic tale wouldbe fêted by successive Persian rulers andaristocracy, whose scribes would fashionprecious copies and illustrate them usingthe finest materials – lapis lazuli, gold,ultramarine. Many of these manuscriptssurvived, became scattered throughoutthe world and have now been brought

Bringing together the Book of Kings

Dr Charles Melville

together in an online environment by theShahnama Project at the University ofCambridge.

Dr Charles Melville, from the Facultyof Asian and Middle Eastern Studies, wasawarded a five-year grant from the AHRBin 1999 to photograph, catalogue andproduce an electronic corpus of thethousands of paintings in the versions ofthe Shahnama still existing worldwide. A second stage commenced in 2006, withthe award of a three-year grant from theAHRC’s Resource Enhancement Scheme to Dr Melville and John Norman of theCambridge Centre for Applied Research in Educational Technologies (CARET) todevelop the Shahnama Project website.

Thanks to the technological expertiseand digital know-how provided by CARET,international researchers are able to usethis uniquely interactive online resource togather together a global record of extantShahnama manuscripts. Visitors arealready building their own workspaces forteaching and research, and will be able touse the site to engage in researchdialogue with other users. With theaddition of a sophisticated behind-the-scenes approvals process, the website willultimately even allow visitors to correctthe database directly. ‘Although it hasbeen vitally important to collate andpreserve these images digitally, the resultis much more than a catalogue,’ said JohnNorman, Director of CARET. ‘It is a newtype of research tool.’

One of the many fascinating aspectsof the research is tracing the ‘transmissionhistory’ of the texts and their illustrations.The oldest surviving copy of the poemdates from 200 years after Firdausi’sdeath. ‘Thereafter, seeing sections of themanuscripts side by side and comparingthem over time, one can explore thecontext in which they were made andwhy they were commissioned,’ saidDr Melville.

By the end of the project, it is hopedthat the corpus will include 10,000images, representing about 70% of thesurviving manuscripts. ‘My main aim wasto stimulate and promote research inPersian history and culture. With thisfunding, we have had a fantasticopportunity to encapsulate a key elementof this and to make it accessible for all touse,’ said Dr Melville.

For more information, please contactDr Charles Melville([email protected]) or John Norman([email protected]) or visit theShahnama Project website(www.oriental.cam.ac.uk/shah).

IN FOCUS

Full-scale battle between the armies of Iranand Turan, from a version of the Shahnamacreated circa 1580 held at the UniversityLibrary, Cambridge; Add. 269, f.178v

The ShahnamaProject is buildinga powerful onlineresource that willstimulate research

and interest inPersian cultural

history.


From organic chemist, through maker ofrubber monsters, to conservator of fineart, Spike Bucklow describes his careerpath as ‘something of a drunkard’swalk’. But you could say that it hasalways been leading to the same end:the application of science for thebenefit of art.

Coming across an article in New Scientist 15 years ago thatdescribed the contribution science can make to art proved tobe the catalyst for a career change that Spike Bucklow hadunknowingly been waiting for. Until then, his career had takenhim from the organic chemist’s bench, where he synthesisedcockroach sex pheromones, to the film industry, where hedesigned and built prosthetic puppets, to artificial intelligenceand technology transfer at Cambridge Consultants.

The New Scientist article led him to the Hamilton KerrInstitute – a place where science and art come togetherthrough the conservation of easel paintings and the analysis ofartists’ materials and methods. In the 15 years since joining theInstitute, a department of The Fitzwilliam Museum and linkedto the University’s Faculty of Architecture and History of Art, hisresearch has allowed him to combine his background as anorganic chemist with his passion for how paintings have beencreated and how they can be preserved. Now a renownedexpert on craquelure – the cracks that naturally form in oilpaintings over time – he has also developed a solid groundingin the conservation of medieval art through working on thetwo oldest altarpieces in the UK: the Thornham Parva Retableand the Westminster Retable. His methods require an intriguingmix of state-of-the-art scanning electron microscopy, 3Dimaging and analytical chemistry, together with anunderstanding of fine paintings, the creation of pigments and Aristotelian physics.

What would others be surprised to learn about you?Working for Spitting Image and the film industry in the 1980s,my role was to take a combination of prosthetics andanimatronics and create rubber monsters that flexed andmoved – Jabba the Hut in Return of the Jedi, and MargaretThatcher and Ronald Reagan in Spitting Image were all mymodels. People might also be surprised to know that my namereally is Spike.

Who or what inspires you?People who know what they’re doing and the evidence theyleave. When I was a kid I went to Wookey Hole in Somersetand I saw someone making paper – he just scooped up thisslurry and suddenly he’d made a sheet of paper. It was such anelegant, relaxed gesture and resulted in this perfect product.It’s the same with art: when I look at fine paintings, I seesomeone who knew what they were doing.

Have you ever had a Eureka moment?About two decades before the film Pirates of the Caribbeanappeared, I had an idea of how to connect computer-generated imagery with live action imagery and I envisaged itall in terms of a computer-generated parrot sitting on anactor’s shoulder. I was half-way through the patenting processwhen I discovered the conservation of paintings and I made thedecision to become a conservator. So you could say thatdiscovering the Hamilton Kerr Institute was my second Eurekamoment.

What’s the best piece of advice you’ve ever been given?When I joined the Institute I was a trained chemist who knewno art history. I went along to an art history lecture as part ofre-training and the lecturer said to me: ‘Just because it’swritten in a book doesn’t mean it’s true. Always go back to thesource.’ Maybe I was a late developer with this, but it openedmy eyes about how to think critically.

If you could wake up tomorrow with a new skill, whatwould it be?A bareback horseman and world-class dressage champion.

What motivates you to go to work each day?The opportunity to get really close to first-class works of art –the evidence of people who knew what they were doing – andto discover things about them that other people might nothave noticed. I’m interested in the physical aspects of thepainting today but also in the how and the why of its history,which is why I’m interested in alchemy and Aristotelian physics.I look at the painting from the point of view of 21st-centuryscience but I also think about the person who painted ithundreds of years ago.

What will the future look like in 2050?My job as a conservation scientist is to ensure that thepaintings we work on will look good in 2050. In fact, ourtimescales are 50–100 years from now. We have theoreticalmodels about how things will behave – how pigments age,how varnish yellows, and how the dynamic nature of thepainting’s physical structure responds to light, humidity and theenvironment. But, although these are the timescales I aim forprofessionally, I know that all of the physical predictions that Imake pre-suppose certain social and political structures, andmy feeling is that these structures are a bit on the shaky side.

What’s your favourite research tool?The University Library: it’s amazing what it contains. I’ve beenlooking for information on 17th-century alchemy in northNorfolk and they’ve got it!

Dr Bucklow’s research uses state-of-the-art technology to study medievalworks of art

CH

RIS TITMU

S

INSIDE OUT

Dr Spike Bucklow

RESEARCH SUPPORT

As we celebrate five years of showcasingCambridge research through the HorizonSeminar series, we take a look back at itsdevelopment and its value to the Universitytoday.‘The Thinking Machine?’ on 18 March will be the twentiethSeminar in the series of Horizon Seminars that stretches backalmost five years to the day. The events, organised by ResearchServices Division (RSD) with support from Cambridge EnterpriseLtd, have continued to provide participants with unique insightinto the cutting-edge of interdisciplinary research at Cambridge.Each year, Horizon Seminars address four themes, attractingparticipants from academia, industry, research sponsors,government and the media.

Helping us to know ourselvesMany of the Seminar themes are chosen to reflect research areasof strategic importance within the University to an externalaudience, particularly from industry. ‘Nanoscale Science and NewMaterials’, the inaugural Horizon Seminar, coincided with thelaunch in 2003 of Cambridge’s interdisciplinary NanoscienceCentre; ‘Towards a Sustainable Earth’ was timed with the launchin 2004 of the Cambridge Environmental Initiative (CEI); and‘Neuroscience and Society’ in 2006 introduced the developingCambridge Neuroscience initiative. The majority of delegatesattending Seminars in the first few years were from industry, andregular coverage was given to R&D, knowledge transfer,technology and innovation. However, an unexpected butwelcome realisation has developed over the years: ‘We werefinding more and more that academics would say howinteresting it was to find out what their colleagues were doing.Something that we hadn’t anticipated was that Horizon eventswere also helping the University to know itself,’ explains TamsinPert, a member of RSD’s Partnership Group, the team who puttogether the programme for Horizon events.

Although the importance of promoting the University’sresearch to an external audience is just as relevant today, andindeed nearly 300 external companies have attended HorizonSeminars so far, the number of academics attending has alsorisen dramatically. This has made Horizon the perfect opportunityto network with peers and to facilitate knowledge exchangeinside and outside the University.

Choosing themesThemes are chosen not just to reflect the University’s strategicinitiatives but also its emergent research strengths (‘PersonalisedMedicine’ in 2005), funding calls by the research councils(‘Ageing’ in 2007), key issues for the government (‘Foodomics’and ‘Energy in Cities’ in 2007), or topics of global significance(‘China’ in 2008).

But the choice of theme is a comparatively small steptowards the end result, explains Gordana Najdanovic, Head ofthe Partnership Group: ‘Any preconceptions of the programmechange as we start talking to people about their research, andthe Seminar can take quite a different direction from the originalidea as a result – it’s the academic input that then shapes theevent.’ Speakers from industry or the public sector add anotherperspective on the chosen theme, and often help to highlightthe potential for research collaboration and applications.

Diversity is keyBringing together researchers from diverse areas but with acommon interest has proved to be one of the most stimulatingaspects of the Horizon series. And it’s not just about science andtechnology; the arts, humanities and social sciences have playeda vital part in the Horizon mix too. In 2007 alone, biotechnologistsrubbed shoulders with computer scientists at ‘A Sensory World’;archaeologists with specialists in obesity and diet at ‘Foodomics’;architects with chemical engineers at ‘Energy in Cities’; andhistorical demographers with cognitive scientists at ‘Ageing’.Through such serendipitous encounters, contacts are made thatsow the seeds of cross-disciplinary research and applications. Forthe external audience, here is a valuable chance to discover boththe depth and the breadth of the University’s research portfoliowithin a chosen theme.

Coming up on the HorizonSeries 6 commences on 5–6 June with ‘China’ and continueswith ‘Bioengineering’, ‘Materials’ and ‘Reproductive Health’(details of which will be announced in subsequent issues). Wewelcome comments, suggestions and ideas about the Seminars,and we look forward to many more opportunities forshowcasing the excellence and diversity of research across theUniversity.

For more information about the Horizon Seminar series,please go to www.rsd.cam.ac.uk/events/horizon or email [email protected]


Expanding Horizons

‘... a very stimulating and utterly absorbing day... I had no idea that so many people are active with

great ideas across such a range of fields.’Delegate, ‘Green Frontiers’, June 2006

NEWS FROM RESEARCH SERVICES DIVISION

I found the day very useful and highly interesting.Above all, I had an opportunity to meet a lot of

great people... and get to know the fascinating andchallenging culture of Cambridge.Speaker, ‘A Sensory World’, March 2007

‘The meeting was an excellent opportunity tohear how much is going on in Cambridge, from

understanding how we use energy to thedevelopment of new technologies that can

reduce emissions, and – importantly –recognising the commercial and behavioural

drivers behind the uptake of new ideas.’Delegate, ‘Energy in Cities’, October 2007


Too often researchers discover that prospects for patenting andcommercial exploitation of their inventions are compromisedbefore they have even begun, either because they haveprematurely disclosed the information or because ‘prior art’ existsthat invalidates their application. What if researchers couldinstead build into their research process a review of the patentlandscape at an early stage, rather than wait until issues areforced by the desire to publish? The IP position and commercialpotential could then be used to inform strategic decisions aboutthe direction of their research.

Cambridge Enterprise is working to do just this withProfessors Chris Abell and Wilhelm Huck in the Department ofChemistry. With funding from the Engineering and PhysicalSciences Research Council (EPSRC), the Cambridge MicrodropletsProject will be used as a test-bed for proactive IP analysis. A fullanalysis of the relevant patent space will be conducted, togetherwith an exploration of the best criteria for identifying andprotecting IP during the course of the project; business modelsfor commercialisation that are synergistic with the research willthen be built.

The Microdroplets Project is ideal to model this approach toIP because of its numerous and diverse IP and commercialopportunities. Microdroplets – small water droplets generated inmicrofluidic systems – have the potential to act as individual

reaction chambers in which discrete chemical or biologicaltransformations can be conducted. ‘The power of this discoveryplatform is that it offers the prospect of a completely newapproach to experimental science by allowing quantitativeanalytical experiments to be carried out in a high-throughputway,’ explained Professor Abell.

Microdroplet research is developing rapidly and has stronginternational competition from the USA, Europe and Asia.‘Understanding the patent landscape in any emerging field caninform the strategy for patenting and partnering forcommercialisation,’ explained Teri Willey, Chief Executive ofCambridge Enterprise. ‘In this case, the intention is to optimisethe value of the research results as the programme evolves. Itrepresents a strategy by which dissemination of the results canbe maximised in parallel with development of a longer termcommercialisation strategy. In traditional approaches, thesethings have sometimes been mutually exclusive.’

This model builds on processes generally adopted inacademic research and is one that may well become an exemplaras more IP-sensitive translational research is undertaken by theUniversity.

For more information, please contact CambridgeEnterprise Ltd (email: [email protected]; Tel: +44 (0)1223 760339; www.enterprise.cam.ac.uk).

RESEARCH SUPPORT

NEWS FROM CAMBRIDGE ENTERPRISE LTD

Proactive IP analysis: helpingcommercialisation not to(micro)drop out of sightAn innovative new project spearheaded byCambridge Enterprise Ltd and researchersin the Department of Chemistry is taking aproactive approach to intellectual property(IP) and commercialisation.

PROFESSO

R CH

RIS ABELL

Should my invention be patented?Deciding whether to file a patentapplication depends on the invention andthe opportunity.Only some research results are patentable or worth patenting.However, assessing the patentability and commercialopportunity for your invention is worth considering because agranted patent, like any other form of property or businessasset, can be licensed (leased) or assigned (sold) and doing somay improve the impact your research has on the world.

What is patentable? For a patent application to succeed, youridea must be: novel – it must not have been made public inany way (‘prior art’) before the patent application is filed;inventive – the invention must involve an inventive step and benon-obvious to someone with a good knowledge of the area;and industrially applicable.

How can I tell? Reviewing scientific publications, patentapplications and granted patents can help to determinewhether your invention fulfils the first two requirements forpatentability. Published patent applications and granted patentsmay be searched using websites such as the UK IntellectualProperty Office (www.ipo.gov.uk); please visitwww.enterprise.cam.ac.uk for a list of other websites.

What isn’t patentable? Scientific or mathematical discoveries, theories or methods; literary, dramatic or artistic works; ways of performing mental acts, playing games or doing business; presentation of information and some computer programs; animal or plant varieties; and methods of medically treating human or animal bodies.

What about software? Software may be patentable if it can be shown to involve some form of technical effect. Even if it can’t be patented, software is protected by copyright after its creation.

Can I patent and publish? Yes, you can do both if yourpatent application is filed prior to publication.

We can help! In addition to the above, patenting decisionsalso need to take account of the University’s current intellectual property policy (downloadable fromwww.enterprise.cam.ac.uk). If you are an employee of theUniversity, Cambridge Enterprise Ltd is happy to assess thepatentability and commercial opportunities for yourinvention.

For more information, please contact CambridgeEnterprise Ltd (email: [email protected]; Tel: +44 (0)1223 760339; www.enterprise.cam.ac.uk).


RESEARCH SUPPORT

Horizon Seminars are organised by Research Services Division. For more information, please go towww.rsd.cam.ac.uk/events/horizon or email [email protected]

FORTHCOMING EVENTS: SAVE THE DATES!

10–20 March 2008Cambridge Science Festival‘The World of Science’The 15th Cambridge Science Festival will feature science from allpoints of the globe and beyond in 2008. There will be a series ofevents celebrating Chinese science and technology past, presentand future, from kite-building and gunpowder demonstrations, todiscussions of sustainable construction. To coincide withInternational Polar Year, there will also be a series of eventscelebrating polar science, including the opportunity for families togo ‘Into the Freezer’ at the Scott Polar Research Institute. Speakers appearing at the Festival include Professor Sir David King,Dr Nick Baylis and Professor Barbara Sahakian. Over 100 freeevents are on offer. The full programme is available atwww.cambridgescience.org or please email Fred Lewsey([email protected]) to request a printed programme.

18 March 2008Horizon Seminar‘The Thinking Machine?’Our understanding of the process of cognition in human beings –how we perceive, think and process information about ourenvironment – is highly developed. This Horizon Seminar willexplore cross-disciplinary research on machine learning andcognition, which is relevant to disciplines such as psychology,neuroscience, linguistics, philosophy and computer science. In theoverlap between these disciplines, what will we learn about theextent to which man is a machine or whether a sentient machinecan ever be developed? This Horizon Seminar will be held atEmmanuel College, Cambridge.

5–6 June 2008 Horizon Seminar‘China’As the global spotlight falls firmly on China in 2008, we showcase the diversity and richness of China-related research and expertise at the University of Cambridge. This Horizonconference will highlight themes of culture, religion, education,literature and language, alongside business, technology, economyand the environment. In bringing together academic and businessleaders to share their knowledge and experience, this event will be aunique opportunity to discover China from a variety of perspectives,and to meet individuals and organisations who are keen to deepentheir understanding of this complex and fascinating country. Theconference will take place at Buckingham House, New Hall,Cambridge. Supported by Cambridge Network(www.cambridgenetwork.co.uk).

22 October–2 November 2008 Arts, Humanities and Social Sciences Festival‘Cambridge Festival of Ideas’The new Cambridge Festival of Ideas celebrates arts, humanities andsocial sciences at the University and many partner organisations. Over80 free events will be on offer to visitors of all ages. Activities includeeverything from Stone Age cooking to a day of Viking culture forfamilies. Talks by high-profile speakers include Baroness Onora O’Neillon press freedom and Dr Paul Binski on medieval art. The fullprogramme will be available from Summer 2008 or please emailJoanna McPhee ([email protected]) to be keptinformed about the Festival and to join the mailing list.

NIG

EL LUC

KH

URST

THE BACK PAGE

All materials in Research Horizons, including but not limitedto text, data, designs, logos, illustrations, still images, areprotected by copyright, design rights, database rights,trademarks and other intellectual property and proprietaryrights.

The materials in Research Horizons are either ownedby the University of Cambridge or have been licensed bythe owners to the University of Cambridge for inclusion inthe publication.

Contributors to this publication have asserted theirmoral right under the Copyright, Designs and Patents Act1988 to be identified as authors of their respective works.

Except for the purposes of private study and non-commercial research or ‘fair dealing for the purposes ofcriticism or review’ as permitted under the Copyright,Designs and Patents Act 1988, this publication or any partthereof may not be reproduced, stored or transmitted inany form or by any means without the prior permission inwriting of the University of Cambridge and the authors ofcontributions as identified.

Requests for further reproduction or reposting orreproduction for commercial purposes should be addressedto the Editor, Research Horizons, Research Services Division,University of Cambridge, 16 Mill Lane, Cambridge, CB21SB, UK; email [email protected]

Research Horizons has made all reasonable efforts toensure that the reproduction of all content in thispublication is done with the full consent of copyrightowners. Research Horizons would be grateful to hear fromintellectual property rights owners who are not properlyidentified in the publication so that Research Horizons maymake the necessary corrections.

©2008 University of Cambridge and Contributors asidentified. All rights reserved.

We are looking for article ideas for Issue 6 of ResearchHorizons and welcome suggestions from all areas ofresearch across the University. Please send ideas to theEditor at [email protected]

General enquiriesTel: +44 (0)1223 333543Fax: +44 (0)1223 332988Email: [email protected]

Edna Murphy, DirectorTel: +44 (0)1223 766964Email: [email protected]

Jo Ryan, Events and Marketing Tel: +44 (0)1223 765404Email: [email protected]

Louise Walsh, Communications and Research HorizonsTel: +44 (0)1223 765443Email: [email protected]

Research Services Division16 Mill Lane, Cambridge, CB2 1SB, UKwww.rsd.cam.ac.uk

Your way into CambridgeResearch Services Division (RSD) helps academics to identify, secure andmanage research funding from external organisations.We identify funding opportunities through our relationships with regional, nationaland international sponsors and then support academics through every step of theawards process, from applying for a research grant and checking applications arecorrect, through negotiating contracts to protect the interests of academics and theUniversity, to supporting departments in managing funding throughout the life of aresearch project.

RSD also encourages collaboration between the University and industry, and fosterslong-term research partnerships between sponsors and academics for mutual benefit.

university of cambridge research horizons issue 5 · [email protected] dr louise...

Documents