AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ AG-7593 1 Communicating Seismic Risk: the Geoethical Challenges of a People-Centred, Participatory Approach IAIN S. STEWART JOH AN N A ICKERT Su stain able Earth In stitu te, Un iversity of Plym ou th , U n ited Kin gd om iain .stew art@p lym ou th .ac.u k joh an n a.ickert@p lym ou th .ac.u k ROBIN LACASSIN In stitu t d e Ph ysiqu e d u Globe d e Paris, Sorbon n e Paris Cité Un iversity, CN RS, Fran ce lacassin @ip gp .fr Abstract The Sendai Framework for Disaster Risk Reduction (SFDRR) encourages scientists to participate in bottom -up risk communication approaches that directly engage hazard-prone populations. Effective communication of seismic risks not only has economic impacts in terms of hazard mitigation but also provides social value in potentially empowering the marginalized populations that disproportionately live in high-risk areas. This emphasis on community-focused disaster preparedness, however, presents a novel set of communication challenges for geoscientists. Few scientists have training in or experience of translating their science for lay publics, and conveying complex risk information is especially difficult in circumstances where scientific issues are socially contested and politically charged. Recognising that disaster threats can create troublesome information battlegrounds, this paper explores the ethical and practical aspects of seismic risk communication, motivated by an early-career earth scientists’ workshop in Istanbul that voiced the concerns of young geoscientists confronted firsthand by at-risk publics. Those concerns form the basis of a wider review of the risk commu- nication issues that are likely to be encountered if community -centred participatory DRR approaches are to be adopted by earthquake science researchers. 1. IN TROD UCTION here has to be a broader and a more peo- ple-centred preventive approach to disas- ter risk. Disaster risk redu ction practices need to be multi-hazard and multi-sectoral, inclu- sive and accessible in order to be efficient and effec- tive. (...) There is a need for the public and private sec- tors and civil society organizations, as well as academ- ia and scientific and research institutions, to work more closely together and to create opportunities for collaboration.’ (Sen d ai Fram ew ork , 2015, p .7). Th e Sen d ai Fram ew ork for Disaster Risk Re- d u ction (SFDRR) offers a n ew global in str u - m en t for con fron tin g n atu ral h azard s, settin g ou t an am bitiou s h olistic strateg y th at em brac- es th e n eed for a ‘… full and meaningful partici- pation of relevant stakeholders at appropriate levels’ an d th e ‘… empowerment and inclusive, accessible and non-discriminatory participation of the most marginalised publics’ (UN ISDR, 2015). As su ch , th e Sen d ai Fram ew ork d efin es a n ew social con tract betw een th e h azard scien tist an d th e w id er p u blic (Ism ail-Zad eh et al., 2017). It is a ‘T AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ A G-7593 2 con tract th at en cou rages th e scien tific com m u - n ity to en d eavou r, alon gsid e th eir existin g tech n ical exp ertise, to ‘… support action by local communities and authorities; and support the inter- face between policy and science for decision -making’ (UN ISDR, 2015). Th is vision of citizen -orien ted research is m im icked in oth er areas of geo- societal con cern s, su ch as en ergy , clim ate ch an ge an d in frastru ctu re d evelop m en t (e.g. Pid geon et al., 2014; Kam lage an d N an z, 2017), an d is on e in creasin gly en d orsed by m ajor in - tern ation al sp on sors of scien tific research . Th e Eu rop ean Com m ission , for exam p le, ch an ged th e title of th eir ‘Scien ce an d Society’ p r o- gram m e to ‘Scien ce in Society’ an d u n d er H ori- zon 2020 d evelop ed gu id elin es arou n d ‘Re- sp on sible Research an d In n ovation ’ (RRI) w ith th e d eliberate goal of stim u latin g reflexivity an d in volvin g a ran ge of social actors - scien - tists, citizen s, p olitician s an d bu sin esses - m ore closely in scien tific en d eav ou rs th at w ere co- d esign ed an d co-p rod u ced by society (O w en et al., 2012). In h eren t in th is sh ift from th e con ven tion al ‘top -d ow n ’, ‘exp ert-led ’ ap p roach to th e em er- gen t ‘bottom -u p ’, ‘com m u n ity-led ’ ap p roach is th e ch allen ge of ‘th e last m ile’ - a term bor- row ed from th e telecom m u n ication s sector, in w h ich th e fin al con n ection betw een th e con - su m er an d th e tech n ology d eterm in es h ow ef- fective it is for th e vast m a rket of u sers. Th e ch allen ge of reach in g th e last m ile (in th is case, reach in g th ose p eop le d irectly at risk) h as b e- com e a critical n otion th at in creasin gly in form s ou r th in kin g abou t a far w id er ran ge of n atu ral risk ch allen ges (e.g. Sh ah , 2006), alth ou gh in keep in g w ith th e sh ift to a p eop le-cen tred focu s of d isaster risk red u ction d iscou rse it h as be- com e re-fram ed as ‘th e first m ile’ (e.g. Kelm an an d Glan z, 2014; Bau d oin et al., 2016). In ad d ition to in form in g civic officials an d d is- sem in atin g to p olicy m akers, com m u n icatin g th at first m ile to reach th e p eop le w h o d irectly face extrem e h azard th reats ou gh t to be a fairly u n con ten tiou s com p on en t in h azard p rep are d - n ess an d m itigation efforts. Yet, a p articip atory ap p roach m arks a m eth od ological m ove aw ay from th e p revailin g m od e of kn ow led ge tran s- fer tow ard s m ore in clu siv e tran sd iscip lin ary strategies th at in corp orate p eer -role m od els, ad op t social n etw ork -based strategies an d d i- rectly en gage w ith com m u nities in m otivatin g p rep ared n ess action s (Sch n eid ew in d et al., 2016; Sch losser an d Pfirm an , 2012; Drake et al., 2015; Ben d ito an d Barrios, 2016; Ism ail-Zad eh et al., 2017). Th is n ew , tran sd iscip lin ary scien ce h as been con troversial (Sch n eid ew in d an d Brod ow ski, 2014) an d th ere are cu rren tly n o gu id elin es for w h at con stitu tes su ccessfu l p a r- ticip ation an d w h at m easu res p rom ote bu ild - in g tru st betw een civil society an d its organ iz a- tion s an d scien ce. Desp ite th e in tern ation al p u sh for p articip atory ap p roach es, th ere is a lack of social sp aces an d in teractive form ats th at en able exch an ge an d join t learn in g b e- tw een tech n ical sp ecialists an d lay p u blics. A review of p eop le-cen tred ap p roach es for d isa s- ter risk m an agem en t d escribed : ‘… a complex landscape characterized by insuffi- cient resources at the local level, and lack of will- ingness among the public at risk to share responsi- bility for disaster risk management with author- ities. If official authorities are to implement the new people-centred approach, they must better under- stand residents’ perspectives and responsibility ex- pectations, become more competent comm unicators, and be willing to engage in long-term dialogue with communities’ (Scolobig et al., 2015, p .202). Th e ch allen ge of h ow to com m u n ica te effec- tively to at-risk com m u n ities, th erefore, lies at th e h eart of th e p eop le-cen tred ap p roach to d isaster risk red u ction . Desp ite th is, few ge o- scien tists h ave been train ed in con veyin g th eir tech n ical kn ow -h ow beyon d th e acad em ic an d p rofession al w orld , an d , for th ose th at h ave, th at train in g u su ally p rioritises p eer -to-p eer com m u n ication skills an d h ow to m an age rela- tion s w ith jou rn alists an d better access th e broad er p rin t an d broad cast m ed ia (Liverm an 2008). In con trast, ord in ary p eop le ‘on th e grou n d ’ - from local com m u n ity grou p s to civic au th orities - ten d to be less fam iliar an d m ore rem ote (i.e. h ard er -to-reach ) au d ien ces for m ost scien tists (Liverm an 2008, Stew art an d N ield 2013). In ad d ition to bein g ‘h ard er to reach ’, p u blic au d ien ces often m eet ‘scien ce’ at tim es of crisis. In em ergen cy situ ation s, scien tific u n d erstan d - in gs bu ilt u p grad u ally over m an y d ecad es are exp ected to be d elivered by ‘exp erts’ in n eat AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ AG-7593 3 m ed ia sou n d -bites an d u n am bigu ou s p u blic statem en ts (Stew art an d N ield , 2013). Distillin g com p lex tech n ical kn ow led ge in to d igest ible p op u lar risk m essages th at can be read ily co n - su m ed by lay au d ien ces is a p ersisten t ch al- len ge for th ose w orkin g at th e scien ce/ p u blic in terface. In th e im m ed iacy of an em ergin g d is- aster, p eop le ten d to beh ave in w ays an d m ake d ecision s th at are n ot an ticip ated by scien tific exp erts an d by em ergen cy m an agers. Ou tsid e of crisis situ ation s, com m u n icatin g u n - certain ty to at-risk p u blics is associated w ith several ch allen ges, su ch as id en tifyin g th e facts relevan t to recip ien ts’ d ecision s w h ile d eter- m in in g th e relevan t u n certain ties, estim atin g th eir im p act, form u latin g p ossible m essages, an d evalu atin g th eir su ccess (Fisch h off an d Davis, 2014). In ad d ition , bu ild in g com m u n ity aw aren ess of p oten tial risks can also be d iffi- cu lt if p eop le h old seriou s m iscon cep tion s abou t basic scien ce con cep ts, if th e scien tific issu es are socially con tested , an d if th e h azard th reat is p olitically ch arged (Stew art an d Lew is 2017). Th e resu lt is th at th e scien ce/ p u blic kn ow led ge in terface can qu ickly becom e m ore like th e fron tlin e of an in form ation battlefield . In su ch com bative circu m stan ces, th ose con veyin g th e risk of n atu ral h azard s to com m u n ities th rea t- en ed by th em can fin d th em selves n avigatin g a carefu l com m u n ication cou rse b etw een th e tech n ical n u an ces an d u n certain ties of extrem e n atu ral even ts on th e on e sid e an d th e n orm a- tive n u an ces an d u n certain ties of m ed ia p ra c- tice an d h u m an beh aviou r on th e oth er. Th e d ilem m as faced by geoscien tists in co m - m u n icatin g risk d u rin g seism ic crises h ave been m ost exten sively d ocu m en ted an d acu tely d issected for th e 2009 L’Aqu ila (Italy) eart h - qu ake an d su bsequ en t legal trial (Alexan d er , 2010 an d 2014; Marzocch i, 2012; Jord an , 2013; Di Cap u a an d Pep p olon i, 2014; Dolce an d Di Bu cci, 2014; Mu cciarelli, 2014; Yeo, 2014; Cocco et al., 2015; DeVasto et al., 2016). Alth ou gh in i- tially fram ed as a trial of scien tists w h o ‘failed ’ to p red ict an earth qu ake, it is m ore w id ely in - terp reted as a failu re in risk com m u n ication , as Oreskes (2016, p .254) n otes: ‘The case centered not on the matter of whether or not earthquakes can be predicted, but on political questions about the social obligations of scientists speaking in official advisory capacities, and epistem- ic questions about the appropriate manner in which risk assessments should be performed. The questions at stake were what information scientists should have offered the public, and how that information should have been communicated. They were not so much matters of scientific facts, but matters of how those facts were rendered and communicated.’ Su ch ‘fau lty’ com m u n ication s arose becau se seism ic crises are n ot sim p ly geop h ysical p h e- n om en a bu t th ey are also socio -p olitical in ci- d en ts. Th eir th reat can m otivate vested bu si- n ess in terests to op p ose th e scien ce (Gesch w in d , 1997 an d 2011) an d th eir in cid en ce can be catalysts of cu ltu ral ch an ge or triggers for p olitical u p h eaval (Clan cey , 2006). As is ev- id en t in oth er areas of scien ce, n otably clim ate ch an ge d iscou rse, con flicts of in terest can resu lt in ‘m an u factu red u n certain ty’ an d th e d eliber- ate obfu scation or m isrep resen tation of in fo r- m ation for w h ich th ere is broad scien tific co n - sen su s (Mich aels, 2005; Mich aels an d Mon for- ton , 2006; Oreskes an d Con w ay, 2007). In su ch con tested social sp aces, th e w isd om an d re- sp on sibility of geoscien tific exp erts in offerin g gu id an ce or ad vocacy h as been qu estion ed : ‘W hether scientists providing expert input into pol- icy issues should be guided by extra-scientific values (and if so, whose values, and which values) in either the conduct or communication of their science, and even whether they should go beyond their scientific competence strictly speaking and provide advice about policy options, is open to debate’ (Yeo 2014). Th is societal en tan glem en t h as p oten tially tran sform ative im p lication s for th e geoscien - tists th at stu d y th em , as ackn ow led ged by Oreskes (2016, p .261): ‘Earthquake safety has never been simply a matter of geophysics, but most earthquake scientists … have traditionally understood their job to be to study how, when, and why eart hquakes happen, and only to a lesser extent (if at all) how to communicate that knowledge to engineers and officials responsible for mitigation, or to the general public … But in the contemporary world, the inter-relationship between knowledge and safety is not easily disentangled. AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ A G-7593 4 Seismology is no longer simply a matter of geophysics, if it ever was. It involves consideration of ethics, values, and monetary and social costs. L’Aquila shows that sci- entists can no longer ignore the social factors that affect and even control how damaging a particular earthquake may be. Earthquake prediction is a social science.’ In this p ap er, w e carry forw ard Oreskes’s prov o- cation that earthqu ake science is a social science by consid ering the issu e of how to com m u nicate seism ic risk in a pu blic sphere in w hich the sci- ence is socially contested and p olitically charged . Ou r consid eration arises not from Italy bu t from Istanbu l, another trou blesom e earthqu ake hot spot, and view ed not from the perspective of sen- ior hazard specialists bu t instead from that of ea r- ly-career geoscientists. Motivated by their em er- gent concerns, w e exam ine key them es of risk com m u nication that m ight be im p ortant if neigh- bou rhood -based particip atory DRR is to be ad opted by the earthqu ake science com m u nity. 2. CASE STUDY: A SEISMIC CONFRONTATION IN ISTANBUL Istan bu l, a m ega-city of 14.5 m illion resid en ts, faces a m ajor earth qu ake th reat (H ori et al., 2017). Over th e last cen tu ry, a w estw ard - m igratin g sequ en ce of large eart h qu akes h as left on e p rom in en t segm en t of th e N orth An a- tolian Fau lt u n ru p tu red (Arm ijo et al., 1999; Stein et al., 1997; Le Pich on et al., 2003). Th at segm en t lies im m ed iately sou th of th e city, b e- n eath th e w aters of th e Marm ara Sea an d seis- m ologists exp ect th is seism ic gap to be filled by a M >7 earth qu ake in th e com in g d ecad es (Pa r- son s et al., 2000; Parson s, 2004; Boh n h off et al., 2013). Th e leth ality of th e th reat is ev id en t from a d estru ctive earth qu ake th at stru ck to th e east of Istan bu l in Au gu st 1999, killin g 17,000 p eo- p le an d m akin g ap p roxim ately h alf a m illion p eop le h om eless (Özerd em , 1999). Risk scen a r- ios for a fu tu re Marm ara Sea earth qu ake an tic- ip ate sign ifican t fatalities an d w id esp read d am age to resid en tial h ou sin g an d u rban in fr a- stru ctu re. For exam p le, a fu tu re M 7.25 earth - qu ake on th is offsh ore segm en t is exp ected to h eavily d am age or d estroy 2-4% of th e n ear 1,000,000 bu ild in gs in Istan bu l, w ith 9-15% of th e bu ild in gs receivin g m ed iu m d am age an d 20-34% of th e bu ild in gs ligh tly d am aged (Erd ik et al., 2011; Erd ik, 2013). Sh aken by th e 1999 earth qu akes, Istan bu l’s civ ic au th orities a t- tem p ted to ad d ress th e acu te seism ic vu ln er a- bility of th e city th rou gh a series of legislative Figure 1: There is a strong cultural influence on people’s perspectives on the perceived earthquake threat. Based on in- terviews with residents in earthquake-prone parts of USA , Japan and Turkey, most participants indicated a high aware- ness of seismic adjustment but there was much variation in belief about the levels of human agency and control. Turkish participants tended strongly toward fatalism, regarded earthquakes as divine providence but displayed a height ened fo- cus on vulnerabilities caused by institutional and collective failings, implying that the consequences of earthquakes were largely determined by the actions of people and society. Redrawn from Joffe et al. (2013, figure 4). AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ AG-7593 5 m easu res facilitatin g u rban ren ew al. Th e co n - troversial im p lem en tation of th e 2005 ren ew al law N o. 5366 in th e city’s h istoric d istricts a u - th orised th e cen tral govern m en t h ou sin g d e- velop m en t agen cy (TOKI) to u n d ertake regen - eration p rojects in seism ically vu ln erable gecekondu (squ atter) n eigh bou rh ood s, p rojects th at m et sign ifican t local resistan ce (Karam an , 2008; Un sal, 2015). Th at resistan ce reflects legis- lative ch an ges th at h ave been im p osed by civic au th orities. In 2012, th e u rban regen eration law N o. 6306 exten d ed regen eration beyon d th e h istoric d istricts, targetin g n eigh bou rh o od s th at are gen erally n ot th ose m ost vu ln erable to earth qu ake d estru ction bu t in stead rep resen t areas w h ere red evelop m en t is h igh ly econ om i- cally p rofitable (Gibson an d Gök şin , 2016). De- sp ite a recogn ition th at ‘seismic risk in the build- ings in Istanbul is mostly dominated by building vulnerability, not hazard’ (Yaku t et al., 2012, p .1533), th ere is w id esp read d istru st of Ista n - bu l’s retrofittin g an d recon stru ction m easu res even am on g resid en ts of som e of th e city’s m ost at-risk qu arters (Green , 2008; Islam , 2010; Karam an , 2013; Ku yu cu , 2014). Th e roots of th is d istru st go d eep in to th e Tu r k- ish ‘earth qu ake p sych e’. A com p arison of p o p - u lation s livin g in earth qu ake -p ron e areas in Ja- p an , USA an d Tu rkey revealed th at esp ecially stron g an d varied em otion s p erm ea te Tu rkish earth qu ake p ercep tion s an d attitu d es (Joffe et al., 2013) (Fig. 1). Direct exp erien ces w ith th e 1999 earth qu akes h ave p rovoked h eigh ten ed feelin gs of w orry, fear an d an xiety, bu t in ad d i- tion th ere w ere stron g exp ression s of corru p - tion an d in com p eten ce of p olitician s, civil ser v - an ts, p lan n in g regu lators an d th e con stru ction in d u stry (Fig. 2). For m an y, it w as th is en d em ic corru p tion , greed an d selfish n ess th at w as seen to p rod u ce u rban vu ln erability, an d in tu rn created a h eigh ten ed sen se of d isso n an ce (fatal- ism ) an d w eaken ed sen se of con trol an d self- efficacy. Th u s, d esp ite a su bstan tial aw aren ess of th e earth qu ake risk, Joffe et al. (2013) rep ort th at Tu rkish resp on d en ts w ere far less likely th an th eir US or Jap an ese equ ivalen ts to ad op t seism ic ad ju stm en t m easu res, a ten d en cy also ap p aren t in oth er stu d ies (Ru stem li an d Karan ci, 1999; Özerd em , 1999; Eraybar et al., Figure 2: The emotional and moral responses of people to earthquakes vary in strength and character between three eart h- quake-prone countries – USA , Japan and Turkey. Fear and anxiety-related emotions dominate in all three countries, but Turkish participants show a greater prevalence of grief- and trauma- related emotions and display considerably more emo- tions relating to moral issues such as corruption (e.g., anger, distrust, blame). From Joffe et al. (2013, figure 3). AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ A G-7593 6 2010; Tekeli‐Yeşil et al., 2010a, 2010b, an d 2011; Oral et al., 2015; Taylan , 2015). In d eed , Joffe et al. (2013) rep ort th at th e Tu rkish resp on d en ts w h o felt th em selves m ost v u ln erable to eart h - qu akes d isp layed th e low est ad o p tion of an ti- seism ic ad ju stm en t m easu res. In su ch a socially p olarised an d p olitically sen sitive con text, d e- velop in g n eigh bou rh ood -based p articip atory strategies for seism ic risk com m u n ication w ou ld ap p ear d ifficu lt. To exp lore th is d ifficu lty, a com m u n ication train in g w orksh op for early-career geoscien - tists h eld in Istan bu l d irectly con fron ted tw elve Ph D an d p ostd octoral r esearch ers w ith th e overtly p oliticised n atu re of th e Istan bu l eart h - qu ake th reat (Ickert an d Stew art, 2016). As p art of th e w orksh op th e you n g geoscien tists visited at-risk n eigh bou rh ood s w ith in th e city to h ear from resid en ts an d com m u n ity lead ers abou t h ow th e city’s seism ic risk w as m an ifest ‘on th e grou n d ’, w itn essin g firsth an d th e effects of th e u n p op u lar u rban ren ew al tran sform ation p r o- gram m e. Follow in g th is field p rovocation , Ick- ert an d Stew art (2016) rep ort h ow th e w or k- sh op p articip an ts w ere in vited to d iscu ss th eir p erceived role as com m u n icators. Across th e grou p of you n g research ers th ere w as broad agreem en t on th e relevan ce an d im p ortan ce of th eir exp ert kn ow led ge reach in g at-risk com - m u n ities, bu t con sid erable d iscu ssion abou t th e ap p rop riate w ay an d level of en ga gin g w ith th e p u blic. Researcher 1: If you know that something will hap- pen [...] that many people could die [...] you will have to communicate that. Y ou have to com- municate that in order to prepare people. Alth ou gh th is view w as w id ely sh ared , m u ch d ebate cen tred on th e p articip an ts’ in d ivid u al u n d erstan d in gs of th e role an d resp on sibility of geocom m u n icators, an d w h at im p lication s th is h ad for th eir p rofession al life. For som e, th e d isin cen tives to com m u n icatin g m ore w id ely w ere p ractical as w ell as m oral: Researcher 8: (...) It is our responsibility. But the problem is: W e are not paid for that. W e have to maintain a career as well. A nd this is only one of the little aspects that are very relevant. W e have to do it for the sake of it. W e do a lot of things for science which are for free. A nd we al- so have a hard time to maintain a pace...and to do publications, to find the next position and so on. So it is a very difficult balance. Researcher 7: There is no real reward. Researcher 8: W ell, it depends on how you interpret reward. Som e p articip an ts fou n d it cru cial n ot to blu r th e bou n d ary betw een scien tists an d n on - scien tists an d to retain th eir role as “objectiv e exp erts”, feelin g u n easy w ith th e p rosp ect of op eratin g beyon d th e geoscien ce realm . Researcher 11: I think you should do your best to improve your analyses and get proper results and publish and explain these results to proper people. For example, the government or the ad- ministration. A nd these people should know what to do with this. Y ou can give them sug- gestions what you think is the best idea to use the results and how to protect the people, but the decision belongs to them. Researcher 4: In my humble opinion, science has something to do with knowledge. Policies, haz- ard mitigation, those are things related to judgement, to decision -making. Those are two completely different things. For som e p articip an ts, a d irect en gagem en t w ith resid en ts, p articu la rly in p oliticized con - texts, w as con sid ered as n egatively affectin g th eir role as scien tists, p oten tially riskin g a loss of rep u tation , tru st an d scien tific cred ibility d u e to actu al or p erceived ad vocacy p o sition s. Oth ers, w h ilst ackn ow led gin g th ese w orries, stressed in stead th e ‘m oral an d p rofession al d u ty’ to d irectly p rovid e th eir exp ertise to com m u n ities, esp ecially in situ ation s w h ere in - h abitan ts face an acu te risk an d op en ly requ est closer collaboration s w ith scien tists. For th em , th ere w as a ‘risk of losin g p u blic tru st’ w h en n ot reactin g to sh ortcom in gs of com m u n ica- tion , bu t at th e sam e tim e an an xiety abou t d riftin g in to th e role of ‘ad vocate’ or even ‘a c- tivist’, as th is exch an ge d em on strates: Researcher 8: A hypothetical case, let’s imagine the scientific community has a very clear view that the M armara earthquake is going to happen in five years time, and it is going to be magn itude 8. Then what is your responsibility, when people are not reached by standard geoscience comm u- AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ AG-7593 7 nication? This is how I face this problem. Then you really have to push the boundaries and tell the people that they should move away from the boundary [… ] but I am already in the activist part. Researcher 2: Y ou’re looking at the human aspect, not at the scientific aspect. A s a human being, when you see that something bad will happen very soon, then of course you will push people and try to fix the problem [...]. A s a scientist you just have to do the research, get the information and share it. Researcher 8: But I absolutely don’t feel like this – this is my scientific part and this is my human part [...] I don’t understand why geoscience should be communicated in a very specific, nar- row way, for example centred on geohazards. Then people might know something about the physics, but they don’t really do anything in their daily lives. A nd this is the challenge. Researcher 7: Y ou could make sure that you inform the public better, so that they can find a way around this corrupt system so that people are in- formed to really make decisions. Researcher 10: But this is really complicated. Th is d isagreem en t abou t roles an d resp on sibili- ties of geoscien tists in th e risk com m u n ication p rocess su ggest th at th e m u ltitu d e of factors th at in flu en ce h ow risk com m u n ication is p e r- ceived , in terp reted an d tran slated by in h abit- an ts of at-risk com m u n ities is equ ally reco g- n ised by scien tists. Desp ite th is, th ere w as co n - sen su s am on g th e grou p on th e n ecessity to m ore effectively con n ect w ith at -risk com m u n i- ties in ord er to red u ce th eir seism ic vu ln erabil- ity. In ad d ressin g th is, h ow ever, th ere w as u n - certain ty abou t w h eth er th e you n g geoscie n - tists h ad th e ap p rop riate skillset to su ccessfu lly en gage w ith lay au d ien ces. Som e felt con fid en t in th is asp ect, givin g p erson al exam p les of sci- en ce–p u blic in teraction s, su ch as en cou n ters w ith local resid en ts in th e cou rse of th eir field w ork, in cid en ts in w h ich th ey “h ad to get in - form ation from local p eop le”, an d w ere asked to “exp lain ” w h at “th ey are d oin g”. In th at co n - text, som e su p p ort for th e valu e of a m ore p a r- ticip atory ap p roach em erged : Researcher 6: W hy do you think that only the geo- scientists give the information? M aybe there are things that you don’t know, and that only an ordi- nary person knows. For example when you go to the field, [...] to a little village, if you are working on a recent event of that region, you go to the manager of the village, and you talk to him, for example “Have you ever had any floods in this area?” It is a com- munication situation and you learn from a person that is not a geoscientist. More gen erally, h ow ever, th e research ers felt in su fficien tly skilled in com m u n ication to reach be-yon d th e acad em ic an d p rofession al geosci- en ce com m u n ity. In teraction w ith lay au d ien c- es w as ju d ged a ‘rath er u n kn ow n territory’. Given th is p erceived skills d eficit, d ebate em erged abou t w h eth er it w as m ore effective to ‘p in p oin t th e com m u n ication talen ts’ w ith in th e geoscien ce com m u n ity or in stead to en gage in in terd iscip lin ary research collaboration s. Li- aisin g w ith social scien tists w ere p r op osed , as w as w orkin g w ith in term ed iaries or tran slators - m ed ia rep resen tatives, N GOs or even artists - to m ore effectively sh are kn ow led ge w ith p e o- p le on th e grou n d . Th is brief exch an ge cap tu res th e essen ce of th at d isagreem en t: Researcher 10: Our responsibility is to produce sci- ence and use other scientists who can talk to people, like anthropologists, sociologists or peo- ple who have studied philosophy, psychology, this kind of stuff... M y point is that we need a bridge to communicate with the people. W e can- not communicate directly. W e need a translator. Researcher 3: Or translate it ourselves. Researcher 2: It won´t be that easy for us. Th e im p lication s arisin g from th ese w orksh op d iscu ssion s are exam in ed in d etail by Ickert an d Stew art (2016), bu t h ere w e h igh ligh t th e basic d ilem m a: com m u n icatin g d irectly w ith at-risk com m u n ities is recogn ised as bein g im - p ortan t bu t th ere w as a gen eral an xiety abou t h ow easily or effectively geoscien tists can ad op t su ch a p articip atory ap p roach . In th is regard , th e early -career geoscien tists raise criti- cal qu estion s abou t h ow best to in tegrate oth er d iscip lin ary p ersp ectives, p articu larly th ose from th e social scien ces, in to th eir geo -risk ex- AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ A G-7593 8 p ertise. Stim u lated by th ese m eth od ological d ialogu es, th e follow in g section s su m m arise som e key fin d in gs th at h ave em erged from broad er social scien ce d iscou rses on risk co m - m u n ication . 3. FAULTY COMMUN ICATION S - TOWARD S A SOCIAL SCIEN CE OF CON VEYIN G SEISMIC RISK 3.1 The Risk Perception Paradox ‘The majority of people at risk from eart hquakes do little or nothing to reduce their vulnerability’ (Sol- berg et al., 2010, p .1663). Th e con ven tion al w ay of com m u n icatin g risk is th rou gh ed u cation cam p aign s th at raise p u blic aw aren ess of h azard th reats. Th is ap p roach rests on th e assu m p tion th at in d ivid u als or com m u n ities w ith h igh h azard aw aren ess are m ore likely to resp on d to w arn in gs an d u n d e r- take p rep ared n ess m easu res th an in d ivid u als or com m u n ities w ith a low er/ d eficien t h azard aw aren ess. In crease an in d ivid u al’s p ercep tion of a th reat, th e assu m p tion goes, an d you im - p rove th eir p rep ared n ess. Dem eritt an d N or b- ert (2014) d escribe th is ap p roach to risk co m - m u n ication eith er u n d er th e term “risk m e s- sage m od el”, referrin g to th e belief th at sou n d risk com m u n ication is abou t faith fu lly tran s- m ittin g risk in form ation w ith ou t d isto rtion , or u n d er th e term of a “risk in stru m en t m od el”, a com m u n ication ap p roach w ith th e goal to elicit certain cogn itive or beh aviou ral resp on ses in th e target au d ien ces th at are ad d ressed (D e- m eritt an d N orbert, 2014). H ow ever, several d ecad es of social scien ce research in d icates th at th ere can be little or n o correlation betw een th e p rovision of scien tific in form ation abou t ge o- p h ysical h azard s an d risks an d th e ad op tion of p rep ared n ess m easu res by in d ivid u als or com m u n ities (Palm an d H od gson , 1992; Palm , 1998; Sp ittal et al., 2008; Solberg et al., 2010; Wood et al., 2011). Wh ile th e h azard scien tist is steep ed in p robabilistic or d eterm in istic th in k- in g abou t th e ch an ces or im p acts of an extrem e even t, th e statistical lik elih ood of a d isaster is barely taken in to accou n t w h en ord in ary p e o- p le m ake ju d gm en ts abou t p erceived risk le v- els, an d th e p erceived m agn itu d e of a d isa ster seem s equ ally to be of little im p ortan ce . In - stead , th ere is a grow in g ap p r eciation of th e role of socio-cu ltu ral, cogn itive an d em otion al variables in risk p ercep tion an d beh aviou r. Solberg et al. (2010) p rovid e a com p reh en sive review of h ow p eop le th in k, feel an d act abou t seism ic risk ad ju stm en t, argu in g th at in d ivid u - al or collective aw are-n ess is sh ap ed by a m yr i- ad of social factors, p sych ological biases an d cu ltu ral n orm s, in clu d in g exp erien ce, op ti- m ism , d em ograp h ic ch aracteristics (gen d er, age, statu s), p ercep tion s of fate an d fatalism , in d ivid u al an d com m u n ity feelin gs of con trol, self-efficacy an d em p ow erm en t, as w ell as th e d egree of tru st in exp erts an d a u th orities. All of th ese risk m ed iators are sen sitive to cu ltu ral an d p olitical con texts, an d all n eed to be co n - sid ered if risk com m u n ication is to be tru ly e f- fective. Th e ‘risk p ercep tion p arad ox’ (Wach in ge r et al., 2013) con ten d s th at if risk p ercep tion is on ly loosely related to risk a d ju stm en t, th en sim p ly d issem in atin g in form ation on seism ic h azard an d societal vu ln erability to exp osed p op u la- tion s m ay n ot m otivate m ean in gfu l risk red u c- tion beh aviou r. Th is reap p raisal rests on a broad er rejection by com m u n ication p ractitio n - ers of th e over -relian ce on factu al in form ation in con veyin g scien tific issu es to th e p u blic (Bu rn s et al., 2003; N isbet, 2009). After all, th e ‘facts’ arou n d com p lex scien tific issu es a re of- ten con tested even by th e exp erts, an d th e sam e tech n ical p roblem can be p resen ted in very d if- feren t w ays to elicit m arked ly con trastin g r e- sp on ses. In th e febrile atm osp h ere of n atu ral em ergen cies an d crisis situ ation s it can be e x- p ected th at ‘facts will be repeatedly misapplied and twisted in direct proportion to their relevance to the political debate and decision-making’ (N isbet an d Moon ey, 2009, p .56), w ith th e resu lt th at even ‘… compelling scientific information often runs aground almost as soon as it is launched into the choppy waters of public discourse’ (Weber an d Word , 2001, p .488). Th e social p sych ology of h ow p eop le receive an d p rocess in form ation abou t risk d ecision s is com p lex an d con tested , an d th e im p lication s th is h as for scien ce com m u n ication in gen eral are d iscu ssed elsew h ere (e.g. Jam ieson et al., 2017; N ation al Acad em ies of Scien ces, En g i- AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ AG-7593 9 n eerin g, an d Med icin e, 2017). H ow ever, th e h ead lin e m essages th at h ave em erged from several d ecad es of h u m an an d beh aviou ral r e- search are n eatly su m m arised by Corm ick (2014) (Fig. 3): 1. ‘when information is complex, people make deci- sions based on their values and beliefs’; 2. ‘people seek affirmation of their attitudes or be- liefs, no matter how strange those views are’. Th is ten d en cy m ean s th at in d ivid u als w ill reject in form ation or evid en ce th at are cou n ter to th ose attitu d es an d beliefs (Ka- h an et al., 2010); th e fact th at n ew in for- m ation con sisten t w ith on e’s beliefs is m ore easily seen as reliable an d in form ative th an in form ation th at d iscred its on e’s in itial be- liefs exp lain s w h y beliefs ch an ge very slo w - ly an d are qu ite en d u rin g in th e face of co n - trary evid en ce (N isbett an d Ross, 1980). 3. ‘people most trust those whose values mirror their own’. Th ey ten d to look to oth ers arou n d th em for social clu es on h ow to act, w h ich can eith er accen tu ate or d ecrease social ac- cep tan ce of th e risk of a giv en issu e (Kah an et al., 2010). As a con sequ en ce, in d ivid u als gen erally m ake m ore risky or extrem e d eci- sion s as p art of a grou p th an as an in d ivid - u al; 4. ‘attitudes that were not formed by logic or facts, are not influenced by logical or factual arguments’; 5. ‘public concerns about contentious science are al- most never about the science - and scientific infor- mation therefore does little to influence these con- cerns.’ Th e n otion th at logical an d factu al argu m en ts m ay be su bord in ate to valu e ju d gem en ts an d in stin ctive th in kin g in d eterm in in g h ow p e op le m ake sen se of tech n ically com p lex issu es h as im p ortan t im p lication s for h ow th e geoscien ce com m u n ity con veys its scien ce to th e p u blic. After all, con ven tion ally, geoscien tists ten d to bu ild com m u n ication strategies arou n d co n - veyin g clear, sim p le exp lan ation s of th e tech - n ical d etail. Th ey d o so becau se th at is w h at geoscien tists h ave been train ed to d o, becau se it is th at tech n ical kn ow -h ow th at d efin es th eir ow n u n d erstan d in g of th e p roblem , an d b e- cau se oth er cru cial stakeh old ers - regu lators, en gin eers, p lan n ers an d law yers - d em an d it. Figure 3: Social science research highlights a few headline messages of how people make decisions about complex and con- tested environmental concerns (after Cormick, 2014). AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ A G-7593 10 Figure 4: In the context of community conflicts, Sandman (1993) argues that ‘risk’ is a product of ‘hazard x outrage’. R e- ducing risk can be achieved by lowering outrage through adjusting the levels of the primary components of community anxiety. Bu t m arsh allin g th e scien tific facts, illu stratin g th em w ith sim p le grap h ics, an d exp lain in g th em u n clu ttered by jargon , seem s u n likely to ad d ress p u blic con cern s: ‘It is not enough to assure that scientifically sound information – including evidence of what sci-entists themselves believe – is widely disseminated: cultural cognition strongly motivates individuals – of all worldviews – to recognize such information as sound in a selective pattern that reinforces their cu l- tural predispositions. To overcome this effect, com- municators must attend to the cultural meaning as well as the scientific content of information ’ (Kah an et al., 2010, p .23). 3.2 Risk = Hazard x Outrage Un d erstan d in g th e social an d cu ltu ral con stru c- tion of risk is recogn ised to be at th e h eart of com m u n ity-cen tred p articip atory ap p roach es to d isaster risk red u ction . It is a n otion rooted stron gly in th e p sych om etric m od el of risk p e r- cep tion (Fisch h off et al., 1978; Slovic, 1987 an d 1989; Slovic et al., 1991; Fisch h off, 1995), w h ich d econ volves th e com p osite p u blic view of ‘risky’ scien tific issu es. Th e ‘p sych om etric’ risk p arad igm h as been w id ely ap p lied by San d - m an (1987, 1989, an d 1993), w h o con ten d s th at m ost local en viron m en tal con troversies com - p rise tw o com p etin g fram es. Th e first is a tech - n ical fram in g of th e p roblem , in volvin g arg u - m en ts abou t th e scien tific an alysis of th e h a z- ard s th at are p erceived to th reaten a com m u n i- ty. Th e secon d relates to th e social con text w ith in w h ich th ose h azard s exist an d th e p r o- cesses by w h ich a com m u n ity’s an xieties fu els collective an ger. Accord in g to San d m an (1993), th at com m u n ity an ger is rooted in a ran ge of con cern s – in clu d in g tru st, con trol, volu n tar i- n ess, d read , an d fam iliarity – w h ich h e collec- tively term s ‘ou trage factors’ (Fig. 4). Sand m an (1993) contend s that w hen the experts and the pu blic d isagree abou t the technical as- pects (su ch as the m agnitu d e of a p articu lar threat or its p robability of occu rrence), the exp erts are AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ AG-7593 11 m ore likely to be correct. And yet, althou gh scien- tists read ily p oint ou t how the pu blic often m is- perceives the hazard , they rarely acknow led ge that they them selves p ay little attention to that com p onent of the p erceived risk that is socially constru cted . ‘The experts, when they talk about risk, focus on haz- ard and ignore outrage. They therefore tend to overes- timate the risk when the hazard is high and the outrage is low, and underestimate the risk when the hazard is low and the outrage is high - because all they are doing is looking at the hazard. The public, in precise parallel, focuses on outrage and ignores hazard. The public, therefore, overestimates the risk when the outrage is high and the hazard is low, and underestimates the risk when the outrage is low and the hazard is high’ (Sand m an, 1993, p.8). The pu blic’s concerns - the social risk - is fre- qu ently d ism issed by scientific experts as being irrational, u nfou nd ed or m anipu lated , even thou gh it is evid ent from com m u nity p rotests that the resu lting anxiety, fear and anger is argu - ably m ore tangible and m easu reable than the u n- d erlying hazard . In the context of com m u nity conflicts, Sand m an su ggests that the technical view of risk as a prod u ct of ‘hazard x vu lnerabil- ity’ is m ore u sefu lly reform u lated as being a p rod u ct of ‘hazard x ou trage’. This, in tu rn, sets the tem p late for risk com m u nication strategies. ‘Two things are true in the typical risk controversy: People overestimate the hazard and people are out- raged. To decide how to respond, we must know which is mostly cause and which is mostly effect. If people are outraged because they overestimate the hazard, the so- lution is to explain the hazard better. But if they over- estimate the hazard because they are outraged, the so- lution is to figure out why they are outraged - and change it.’ 3.3 Risk D ialogues and Honest Brokers ‘… public participation measures are probably the most effective means to create awareness of potential disasters, to enhance trust in public authorities, and to encourage citizens to take more personal respon- sibility for protection and disaster preparedness. It will be a major challenge for risk management and also an important research topic for future research to understand people’s responses to natural hazards as well as a combination of natural and technologi- cal hazards and to design the most appropriate measures for effective risk communication, stake- holder involvement, and emergency preparedness’ (Wach in ger et al., 2013). An altern ative ap p roach to con ven tion al risk m essage m od el of com m u n ication is th e “risk d ialogu e m od el”, w h ich is based on th e belief th at th e d iverse exp ertise of m ix of civil society actors m u st be system atically an ch ored if a rel- evan t solu tion to red u ce v u ln erability is to be fou n d (Dem eritt an d N orbert, 2014). En gagin g w ith th e w id er civil society in a m ore p articip a- tory w ay abou t earth qu ake th reats m ean s fra m - in g seism ic risk com m u n ication n ot sim p ly as th e con ven tion al on e-w ay tran sfer of in fo r- m ation from th e tech n ical exp ert to th e ‘en d u ser’, bu t also as a tw o -w ay exch an ge w ith stakeh old er grou p s abou t w h at con cern s th em . In sh ort, p arap h rasin g Latou r (1994), it requ ires a m in d set sh ift from con veyin g ‘m atters of fact’ to d evelop in g d ialogu es arou n d ‘m atters of con cern ’, recogn izin g th at sh ared rath er th an u n id irection al flow s of in form ation are m ore likely to p rom ote kn ow led g e an d attitu d e ch an ge (Stew art an d Lew is, 2017). Scien tific an d tech n ical in form ation are n ecessary for th is p rocess, bu t are n ot th e sole basis on w h ich d e- cision s or action s are m ad e. Wh en accessible scien tific in form ation abou t a con tested scie n - tific issu e is p resen ted in w ell-organ ized social sp aces in w h ich ord in ary p eop le can form an d exp ress th eir op in ion s, th e p u blic can actively p articip ate in scien tific d ecision m akin g: ‘It appears that people understand some things quite well, although their path to knowledge may be quite different from that of the technical experts… given an atmosphere of trust in which both experts and lay persons recognize that each group may have something to contribute to the discussion, exchange of information and deepening of perspectives may well be possible’ (Slovic, 1985, p .170). Wh eth er d irected at th e p u blic or at p olicy m akers, m ore effective com m u n ication em er g- es from p articip atory en gagem en t an d d ialogu e AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ A G-7593 12 w ith in d ivid u als an d com m u n ities (Wach in ger et al., 2012). Particip ation p rocesses allow th e p u blic to gain kn ow led ge an d p erson al agen cy w ith resp ect to risks an d p rotective m easu res, an d au th or i- ties to gain kn ow led ge from th e “lay e x- p erts”/ th e p u blic an d to collect id eas for m easu res th at are effective for th e given p op u - lation . In d ivid u als an d com m u n ities can be risk am p lifiers or in h ibitors. Th u s, as w ell as gau g- in g th e in trin sic vu ln erability of th ose livin g in h azard -p ron e settin gs, collective con versation s p rovid e th e exp ert w ith a valu able in form ation arch ive. Th ese d ialogu es can reveal a social m em ory of p ast even ts an d h igh ligh t h ow p re- ven tative m easu res m ay be bu ilt in to trad itio n - al p ractices an d vern acu lar arch itectu re. Fin a l- ly, a sh ared resp on sibility for solu tion s in su ch a d ialogu e situ ation can also p rom ote th e social im p lem en tation of th em . Yet, d esp ite its ap p aren t ben efit s, a m ore p ar- ticip atory ap p roach raises eth ical d ilem m as for a geoscien tist, n ot least of w h ich is th e p ote n tial d rift aw ay from bein g a n eu tral m ed iator of in - d ep en d en t kn ow led ge to a p articip an t th at is en gaged w ith , or p erh ap s even em bed d ed in , com m u n ity con cern s. Th e scien tist as an ‘ad v o- cate’ or even ‘activist’ for a sp ecific cou rse of action is on e th at m an y - exp erts an d n on - exp erts alike - are u n com fortable w ith . In d is- tin gu ish in g d ifferen t ‘kin d s’ of scien tist, Pielke (2007) h igh ligh ts th e im p ortan t role of th e ‘h on est broker’ - th e sp ecialist w h o can in t e- grate stakeh old er con cern s w ith available sci- en tific kn ow led ge to op en u p an d in form a ran ge of op tion s. Draw in g on th e fau lty co m - m u n ication s an d th eir ep istem ological roots d u rin g th e Tu h oku an d Fu kish im a catastrop h e, Lacassin an d Lavelle (2016) h igh ligh t th e r e- sp on sibility of com m u n icatin g u n certain ties th rou gh su ch ‘h on est brokers’. Accord in g to th e au th ors, th ey n eed n ot on ly to focu s on th e d ecision -relevan t elem en ts of th at com m u n i- cated u n certain ties, bu t also u n cover th e u n cer- tain ties th at scien tists fail or avoid to m en tion becau se of d om in an t com m u n ication p ar a- d igm s, argu in g th at: ‘… to promote real democratic and open debate and choices, [scientists] have the responsibility to com- municate an d properly explain all uncertainties and unknowns to the technical and political sphere as well as to the rest of the society’ (Lacassin an d Lavelle, 2016, p .57). As h on est brokers, effective geoscien ce co m - m u n ication becom es n ot sim p ly a case of secu r- in g p u blic accep tan ce, bu t, rath er, of secu rin g p u blic tru st. It is argu ably m ore im p ortan t to bu ild tru st th an to bu ild tech n ical u n d erstan d - in g becau se tru st is u sed by p eop le as a su rr o- gate tool - a cogn itive sh ortcu t - for red u cin g th e com p lexity of scien t ific in form ation . An d it is th at tru st th at w ill be cru cial later if w arn in gs an d oth er typ es of vital in form ation are to be taken seriou sly d u rin g an em ergen cy (Wach in ger et al., 2013). Th ose scien tific br o- kers w h o gen u in ely en gage w ith affected com m u n ities are likely to h ave a p articu larly p rivileged p lace in th e d eliberative p rocess b e- cau se, in ad d ition to th eir grasp of tech n ical com p lexity, th ey w ill be afford ed a h igh d egree of p u blic tru st. Yet th at elevated d egree of in - flu en ce, an d th e an ticip ated co-p rod u ction of kn ow led ge th at accom p an ies a com m u n ity - cen tred ap p roach , m ay m ake m an y geoscie n - tists an xiou s th at th eir m u ch valu ed in d ep en d - en ce w ill be com p rom ised . 4. CON CLUSION S Th e eth ical resp on sibility of scien tists is th e com m u n ication of balan ced factu al in for- m ation , yet th e relative p rom in en ce given to th ose facts is cru cial. To con vey a scien tific m essage in a w ay th at gain s w id e accep tan ce requ ires a sim p lified m essage strip p ed of th e u su al tech n ical caveats. Th e eth ical bu rd en th at th is p laces on th e scien ce com m u n icator as an h on est broker of kn ow led ge is obviou s. Th e clim ate scien ce com m u n icator Step h en Sch n e i- d er d u bbed it th e ‘d ou ble eth ical bin d ’, n otin g th at ‘Each of us has to decide what the right balance is between being effective and being honest. I hope that means being both’ (Sch n eid er, 2002, p .498). Th e ch allen ge for th ose w orkin g in d isaster risk com m u n ication is h ow to con vey w h at th ey kn ow h on estly an d effectively to th ose p eop le w h o can ben efit from th at kn ow led ge. In term s of th e form er, it requ ires h azard scien tists to better u n d erstan d th e social p sych ology of h ow AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ AG-7593 13 p eop le receive an d p rocess in form ation , an d in d oin g so learn h ow best to fram e th e in trica- cies, u n certain ties an d lim itation s of th eir in tr i- cate tech n ical scien ce in w ays th at are m ore easily grasp ed by lay au d ien ces. In u n iversities an d research in stitu tes it w ill be cru cial to train th e n ext gen eration of geoscien tists in th e sci- en ce an d art of com m u n ication , bein g m ore jou rn alistic an d m ed ia -savvy in th e w ay w e com m u n icate, an d m ore im agin ative in exp lo r- in g n ew com m u n ication ch an n els, su ch as w eb - based p latform s, social m ed ia. Th e n ew reality is th at in tim es of th e crisis factu al scien tific in - form ation s can be qu ickly (m in u tes to h ou rs) tran sm itted th rou gh social m ed ia (e.g. Tw itter, Facebook), th ereby h elp in g p eop le to ap p reci- ate th e geop h ysical u n d erstan d in g of an on g o- in g n atu ral even t. In th e sam e w ay, h o w ever, m isin form ation or m iscon ceived facts can equ ally p rom in en tly be in stan tly relayed , m ean in g th at p u blic tru st in h on est scien tific brokers m u st be m ain tain ed . For th is reason , alon gsid e th e obviou s ap p eal of n ew , rem ote m obile tech n ologies an d d igital com m u n ica- tion s, face-to-face en cou n ters w ill be critical for effective com m u n ity-cen tred d isaster d ia- logu es. In th ose en cou nters, geoscien tists sh ou ld exp ect to m eet th e p u blic in th eir ow n com m u n ity sp aces - in sch ools, p laces of w or- sh ip , civic h alls an d social clu bs. In ad d ition , d eliberative form ats su ch a ‘Reallabore’ (Ma r- qu ard t an d West, 2016), scien ce sh op s (Sch lierfs an d Meyer, 2013) or variou s form ats of in for- m al, d ialogu e-orien ted citizen p articip ation (Kam lage et al., 2018) are im p ortan t foru m s w h ere m ore d irect tw o -w ay com m u n ication ap p roach es betw een d ifferen t stakeh old ers can take p lace. Bu t as w ell as bein g m ore effective, geoscien - tists n eed to better ap p reciate th at th eir seism ic risk com m u n ication s are u n d ertaken in co n - tested socio-p olitical con texts an d brin g eth ical d ilem m as. Th e exp ert tech n ical kn ow led ge of th e earth qu ake geoscien tist m ay be afford ed a h eigh ten ed d egree of tru st by th ose in d ivid u als an d com m u n ities m ost at risk, bu t th at brin gs resp on sibility. Iron ically, th e m ore effectively a p oten tial seism ic th reat is com m u n ica ted by th e geoscien tist, th e m ore th at scien tific m essage w ill be n orm alized in to th e com p lex, ch aotic an d con tested d iscou rses of d aily life. Partici- p atory -based en gagem en t strategies an ticip ate tech n ical exp erts takin g accou n t of th ese local socio-cu ltu ral, em otion al an d even p olitical d im en sion s of risk in w orkin g d ir ectly w ith vu ln erable com m u n ities. In su ch circu m stan c- es, scien tific risk m essages can becom e h ijacked by or assim ilated in to social, econ om ic an d p o- litical con troversies. Th e ch allen ge for th ose w orkin g at th e con tested fron tlin e of seism ic risk com m u n ication s w ill be balan cin g th e eth i- cal bin d s th at con tin u e to con strain u s in ou r role as h on est brokers ACKN OWLEDGEMEN TS All th ree au th ors gratefu lly ackn ow led ge th e fin an cial su p p ort of th e EU Marie Cu rie In itial Train in g N etw orks p rogram m e ‘ALErT’ (An a- tolian p Lateau clim ate an d tecton ic h azard s) an d IS w as ad d ition ally su p p orted by th e N ERC-ESRC-AH RC Global Ch allen ges Re- search Fu n d p roject ‘Research for Em ergen cy Aftersh ock Forecastin g’ (REAR). We th an k D a- vid Mogk an d an an on ym ou s review er for co n - stru ctiv e com m en ts th at im p roved th e m an u - scrip t, an d to Giu sep p e Di Cap u a for su p p or t- in g th e p ap er th rou gh th e review in g p rocess. REFEREN CES Alexan d er D.E. (2010). Th e L'Aqu ila eart h - qu ake of 6 Ap ril 2009 an d Italian Gover n - m en t p olicy on d isaster resp on se. Jou rn al of N atu ral Resou rces Policy Research , 2(4), 325-342. Alexan d er D.E. (2014). Com m u n icatin g eart h - qu ake risk to th e p u blic: th e trial of th e “L’Aqu ila Sev en ”. N atu ral H azard s, 72(2), 1159-1173. Arm ijo R., Meyer B., H u bert A., an d Barka A. (1999). Westw ard p rop agation of th e N orth An atolian fau lt in to th e n orth ern Aegean : Tim in g an d kin em atics. Geology, 27(3), 267- 270. Bau d oin M.A., H en ly-Sh ep ard S., Fern an d o N ., Sitati A., an d Zom m ers Z. (2016). From top - d ow n to “com m u n ity -cen tric” ap p roach es to early w arn in g system s: exp lorin g p ath - w ays to im p rove d isaster risk red u ction th rou gh com m u n ity p articip ation . In tern a- tion al Jou rn al of Disaster Risk Scien ce, 7(2), 163-174. AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ A G-7593 14 Bend ito A. and Barrios E. (2016). Convergent agency: Encou raging transd isciplinary ap- proaches for effective clim ate change ad apta- tion and d isaster risk red u ction. International Jou rnal of Disaster Risk Science, 7(4), 430-435. Boh n h off M., Bu lu t F., Dresen G., Malin P.E., Eken T., an d Aktar M. (2013). An earth - qu ake gap sou th of Istan bu l. N atu re Com - m u n ication s, 4. Bu rn s T.W., O’Con n or D.J. an d Stocklm ayer S.M. (2003). Scien ce Com m u n ication : a con - tem p orary d efin ition . Pu blic Un d erstan d in g of Scien ce, 12, 183-202. Clancey G. (2006). Earthqu ake Nation. The Cu ltu r- al Politics of Japanese Seism icity, 1868-1930. Berkeley, CA: University of California Press. Cormick C. (2014). Community Attitudes Towards Science and Technology in Australia. CSIRO, 26p. Dem eritt D. an d N obert S. (2014). Mod els of best p ractice in flood risk com m u n ication an d m an agem en t. En viron m en tal H azard s, 13(4), 313-328. DeVasto D., Grah am S.S. an d Zam p aru tti L. (2016). Stasis an d Matters of Con cern : Th e Con viction of th e L’Aqu ila Seven . Jou rn al of Bu sin ess an d Tech n ical Com m u n ication , 30(2), 131-164. Di Cap u a G. an d Pep p olon i S. (2014). Geoeth i- cal asp ects in th e n atu ral h azard s m an ag e- m en t, p p . 59-62. In : Lollin o G., Arattan o M., Giard in o M., Oliveira R., Pep p olon i S. (Ed s.), En gin eerin g Geology for Society an d Territory - Volu m e 7, Ed u cation , Profes- sion al Eth ics an d Pu blic Recogn ition of En - gin eerin g Geology, XVII, 274 p ., Sp rin ger, ISBN : 978-3-319-09302-4. Drake J.L., Kon tar Y.Y., Eich elberger J.C., Ru p p S.T., an d Taylor K.M. (Ed s.) (2015). Com - m u n icatin g Clim ate-Ch an ge an d N atu ral H azard Risk an d Cu ltivatin g Resilien ce: Case Stu d ies for a Mu lti-d iscip lin ary Ap - p roach , Vol. 45, Sp rin ger. Dolce M. an d Di Bu cci D. (2014). Risk m an - agem en t: roles an d resp on sibilities in th e d ecision -m akin g p rocess. In : Wyss M. an d Pep p olon i S. (Ed s.), Geoeth ics: eth ical ch al- len ges an d case stu d ies in Earth Scien ces, 450 p ., Elsevier, Walth am , Massach u setts, ISBN : 9780127999357. Eraybar K., Okazaki K. an d Ilki A. (2010). An exp loratory stu d y on p ercep tion s of seism ic risk an d m itigation in tw o d istricts of Ista n - bu l. Disasters, 34(1), 71-92. Erd ik M., Şeşetyan K., Dem ircioğlu M., H an cılar U. an d Zü lfikar C. (2011). Rap id earth qu ake loss assessm en t after d am agin g earth qu akes. Soil Dyn am ics an d Earth qu ake En gin eerin g, 31(2), 247-266. Erd ik M. (2013). Earth qu ake risk in Tu rkey. Scien ce, 341, 72. Fisch h off, B., (1995). Risk p ercep tion an d com - m u n ication u n p lu gged : tw en ty years of p rocess, Risk An alysis, 15, 137–145. Fisch h off B., Slovic P., Lich ten stein S., Read S., an d Com bs B. (1978). H ow safe is safe en ou gh ? A p sych om etric stu d y of attitu d es tow ard s tech n ological risks an d ben efits. Policy Scien ces, 9, 127-152. Fisch h off B. an d Davis A.L. (2014). Com m u n i- catin g scien tific u n certain ty. Proceed in gs of th e N ation al Acad em y of Scien ces of th e Un ited States of Am erica, 111, 13664-13671. Gesch w in d C.-H . (1997). 1920s Pred iction Re- veals Som e Pitfalls of Earth qu ake Forecast- in g. Eos, Vol. 78, (38), 401-412. Gesch w in d C.-H . (2001). Californ ia Eart h - qu akes: Scien ce, Risk, an d th e Politics of H azard Mitigation . Baltim ore an d Lon d on : Joh n s H op kin s Un iversity Press. Gibson M. an d Gökşin Z.A. (2016). N eigh bou r- h ood Regen eration in Istan bu l – from Earth qu ake Mitigation to Plan n ed Dis- p lacem en t an d Gen trification . In tern ation al Plan n in g H istory Society Proceed in gs, 17(2). Green R.A. (2008). Un au th orised Develop m en t an d Seism ic H azard Vu ln erability: a stu d y of squ atters an d en gin eers in Istan bu l, Tu r- key. Disasters, 32(3), 358-376. Gu ̈n grm u ̈ş Z., Karabu lu tlu E.Y. an d Yıld ız E., (2012). Determ in in g th e kn ow led ge an d b e- h avior of th e in d ivid u als abou t earth qu ake p rep ared n ess at h om e in Tu rkey. H ealth - MED, 6 (1), 232-237. H all S.S. (2011). Scien tists on Trial: At Fau lt? N atu re, 477 (14 Sep tem ber), 264-9. H ori T., Pin ar A., N ecm ioglu O., H ori M. an d N ish izaw a A. (2017). Sp ecial issu e “Th e n ext Marm ara earth qu ake: d isaster m itig a- AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ AG-7593 15 tion , recovery, an d early w arn in g”. Earth , Plan ets an d Sp ace, 69(1), 65. Ickert J. an d Stew art I.S. (2016). From geoscien - tific “m atters of fact” to societal “m atters of con cern ”: a tran sd iscip lin ary train in g a p - p roach to com m u n icatin g earth qu ake risk in Istan bu l (Tu rkey). N atu r al H azard s an d Earth System Scien ces, 16(1), 1-1. Islam T. and Sakızlıoğlu B. (2015). The m aking of, and resistance to, state-led gentrification in Is- tanbu l, Tu rkey. Global gentrifications: Un e- ven d evelopm ent and d isp lacem ent, 245. Ism ail-Zad eh A.T., Cu tter S.L., Takeu chi K. and Paton D. (2017). Forging a p arad igm shift in d isaster science. N atu ral H azard s, 86(2), 969- 988. Jam ieson K.H ., Kah an D., an d Sch eu fele D.A. (Ed s.) (2017). Th e Oxford H an d book of th e Scien ce of Scien ce Com m u n ication , O xford Un iversity Press. Joffe H ., Rossetto T., Solberg C., and O’Connor C. (2013). Social rep resentations of earthqu akes: A stu d y of peop le livin g in three highly seis- m ic areas. Earthqu ake Spectra, 29(2), 367-397. Jord an T.H . (2013). Lesson s of L’Aqu ila for o p - eration al earth qu ake forecastin g. Seism o- logical Research Letters, 84, 1, 4-7. Kah an D.M., Jen kin s‐Sm ith H . an d Bram an D. (2011). Cu ltu ral cogn ition of scien tific co n - sen su s. Jou rn al of Risk Research , 14(2), 147- 174. Kam lage J.H . and N anz P. (2017). Crisis and p ar- ticip ation in the Eu ropean Union: energy poli- cy as a test bed for a new p olitics of citizen p articip ation. Global Society, 31(1), p p. 65-82. Kam lage J.-H ., Rich ter I., an d N an z P. (2018). An d en Gren zen d er Bü rgerbeteiligu n g: In - form elle d ialogorien tierte Bü rger- beteiligu n g im N etzau sbau d er En er- giew en d e. In : H an d bu ch En ergiew en d e u n d Partizip ation , p p . 627-642, Sp rin ger. Karam an O. (2008). Urban p u lse - (re)m akin g sp ace for globalization in Istan bu l. Urban Geograp h y, 29(6), 518-525. Karam an O. (2013). Urban ren ew al in Istan bu l: Recon figu red sp aces, robotic lives. In tern a- tion al Jou rn al of Urban an d Region al Re- search , 37(2), 715-733. Karan ci N . (2013). Facilitatin g com m u n ity p a r- ticip ation in d isaster risk m an agem en t: Risk p ercep tion an d p rep ared n ess beh aviou rs in Tu rkey. In : Joffe H ., Rossetto T. an d Ad am s J. (Ed s.), Cities at Risk, Sp rin ger N eth er- lan d s, 93-108. Kelm an I. an d Glan tz M.H . (2014). Early w ar n - in g system s d efin ed . In : Red u cin g d isaster: early w arn in g system s for clim ate ch an ge, Sp rin ger N eth erlan d s, 89-108. Ku yu cu T. (2014). Law , Prop erty an d Am bigu i- ty: Th e Uses an d Abu ses of Legal Am bigu i- ty in Rem akin g Istan bu l's In form al Settle- m en ts. In tern ation al Jou rn al of Urban an d Region al Research , 38(2), 609-627. Lacassin R. an d Lavelle S. (2016). Th e crisis of a p arad igm . A m eth od ological in terp ret ation of Toh oku an d Fu ku sh im a catastro p h e. Earth -Scien ce Review s, 155, 49-59. Latou r B. (2004). Wh y h as critiqu e ru n ou t of steam ? From m atter s of fact to m atters of con cern . Critical In qu iry, 30 (2), 225-248. Liverm an D.G. (2008). En viron m en tal geosci- en ce: com m u n ication ch allen ges. In : Liver- m an D. (ed .), Com m u n icatin g En viron m en - tal Geoscien ce, Geological Society, Lon d on , Sp ecial Pu blication s, 305(1), 197-209. Le Pich on X., Ch am ot‐Rooke N ., Ran gin C. an d Sen gör A.M.C. (2003). Th e N orth An atolian fau lt in th e Sea of Marm ara. Jou rn al of Geo- p h ysical Research : Solid Earth , 108(B4). Marqu ard t E. and West C. (2016). Co-Prod u ktion von Wissen in d er Stad t. Reallabor “Urban O f- fice–N achhaltige Stad tentw icklu ng in d er Wissensgesellschaft” an d er Universität H ei- d elberg. Technikfolgenabschätzu n g–Theorie Und Praxis, 25(3), 26-31. Marzocch i W. (2012). Pu ttin g scien ce on trial. Ph ysics World , 25(12), 17. Mich aels D. (2006). Man u factu red u n certain ty. An n als of th e N ew York Acad em y of Sci- en ces, 1076(1), p p .149-162. Mich aels D. an d Mon forton C. (2005). Man u fac- tu rin g u n certain ty : con tested scien ce an d th e p rotection of th e p u blic’s h ealth an d en - viron m en t. Am erican Jou rn al of Pu blic H ealth , 95(S1), p p .S39-S48. Mu cciarelli M. (2014). Som e com m en ts on th e first d egree sen ten ce of th e “L’Aqu ila trial”. In : Wyss M. an d Pep p olon i S. (Ed s.), Geo- eth ics: eth ical ch allen ges an d case stu d ies in Earth Scien ces, 450 p ., Elsevier, Walth am , Massach u setts, ISBN : 9780127999357. AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ A G-7593 16 N ation al Acad em ies of Scien ces, En gin eerin g, an d Med icin e (2017). Com m u n icatin g Sci- en ce Effectively: A Research Agen d a. Wash - in gton , DC: Th e N ation al Acad em ies Press, d oi: 10.17226/ 23674. N isbet M.C. (2009). Com m u n icatin g clim ate ch an ge: Wh y fram es m atter for p u blic en - gagem en t. En viron m en t: Scien ce an d Policy for Su stain able Develop m en t, 51, 12-23. N isbet M.C. an d Moon ey C. (2009). Fram in g scien ce. Scien ce, 316. N isbett R.E. an d Ross L. (1980). H u m an in fer- en ce: Strategies an d sh ortcom in gs of social ju d gm en t. Pren tice-H all, En glew ood Cliffs, N J. Oral M., Yen el A., Oral E., Ayd in N . an d Tu n - cay T. (2015). Earth qu ake exp erien ce an d p rep ared n ess in Tu rkey. Disaster Preven - tion an d Man agem en t, 24(1), 21-37. Oreskes N . an d Con w ay E.M. (2011). Mer- ch an ts of d ou bt: H ow a h an d fu l of scien tists obscu red th e tru th on issu es from tobacco sm oke to global w arm in g. Bloom sbu ry Pu b- lish in g USA. Oreskes N . (2015). H ow Earth Scien ce H as Be- com e a Social Scien ce. H istorical Social Re- search 40, 2, 246-270. Ow en R., Macn agh ten P. an d Stilgoe J. (2012). Resp on sible research an d in n ovation : From scien ce in society to scien ce for society, w ith society. Scien ce an d Pu blic Policy, 39(6), p p .751-760. Özerd em A. (1999). Tiles, tap s an d earth qu ake - p roofin g: lesson s for d isaster m an agem en t in Tu rkey. En viron m en t an d Urban ization , 11(2), 177-180. Palm R. (1998). Urban earth qu ake h azard s: th e im p acts of cu ltu re on p erceived risk an d r e- sp on se in th e USA an d Jap an . Ap p lied Ge- ograp h y, 18(1), p p .35-46. Palm R. an d H od gson M.E. (1992). After a Cali- forn ia Earth qu ake: Attitu d e an d Beh avior Ch an ge. Th e Un iversity of Ch icago Press: Ch icago. Parson s T., Tod a S., Stein R.S., Barka A. an d Dieterich J.H . (2000). H eigh ten ed od d s of large earth qu akes n ear Istan bu l: an in tera c- tion -based p robability calcu lation . Scien ce, 288 (5466), 661-665. Pid geon N ., Dem ski C., Bu tler C., Parkh ill K. an d Sp en ce A. (2014). Creatin g a n ation al citizen en gagem en t p rocess for en ergy p oli- cy. Proceed in gs of th e N ation al Acad em y of Scien ces, 111 (Su p p lem en t 4), p p .13606- 13613. Pielke Jr R.A. (2007). Th e h on est broker: m ak- in g sen se of scien ce in p olicy an d p olitics. Cam brid ge Un iversity Press. Rossetto T., Joffe H . an d Solberg C. (2011). A d ifferen t view on h u m an vu ln erability to earth qu akes: lesson s from risk p ercep tion stu d ies. In : Sp en ce R., So E. an d Scaw th orn C. (Ed s.), H u m an Casu alties in Eart h qu ake, Sp rin ger N eth erlan d s, 291-304. Ru stem li A. an d Karan ci N .A. (1999). Corre- lates of earth qu ake cogn ition s an d p rep a r- ed n ess beh aviou r in victim ised p op u lation . Th e Jou rn al of Social Psych ology, 139 (1), 91-101. San d m an , P.M. (1987). Risk com m u n ication : facin g p u blic ou trage, EPA J., 13, 21-22. San d m an , P. M. (1989). H azard versu s ou trage in th e p u blic p ercep tion of risk, in : Covello, V.T. (Ed itor). Effective Risk Com m u n ica- tion , N ew York: Pu tn am Press, 45-49. San d m an P.M. (1993). Resp on d in g to com m u - n ity ou trage: strategies for effective risk com m u n ication . Am erican In d u strial H y- gien e Association , Fairfax, Va. Sch lierf K. an d Meyer M. (2013). Situ atin g kn ow led ge in term ed iation : In sigh ts from scien ce sh op s an d kn ow led ge broker s. Sci- en ce an d Pu blic Policy, 40(4), 430–441. Sch losser P. an d Pfirm an S. (2012). Earth sci- en ce for su stain ability. N atu re Geoscien ce, 5 (9), 587-588. Sch n eid er S.H . (2002). Keep in g ou t of th e box. Am erican Scien tist, 90 (N ov -Dec), 496-498. Scolobig A., Prior T., Sch röter D., Jörin J. an d Patt A. (2015). Tow ard s p eop le-cen tred ap - p roach es for effective d isaster risk m a n - agem en t: Balan cin g rh etoric w ith reality. In - tern ation al Jou rn al of Disaster Risk Red u c- tion , 12, p p .202-212. Sh ah H .C. (2006). Th e last m ile: earth qu ake risk m itigation assistan ce in d evelop in g cou n - tries. Ph ilosop h ical tran saction s - Series A, Math em atical, p h ysical, an d en gin eerin g scien ces, 364(1845), 2183-2189. Slovic P. (1985). In form in g an d Ed u catin g th e Pu blic abou t Risk. Disaster Research Re- p ort, Oregon , 80-85. AN N ALS OF GEOPH YSICS, 60, FAST TRACK 7, 2017; DOI: 10.4401/ AG-7593 17 Slovic P. (1987). Percep tion of risk. Scien ce, 236 (4799), 280-285. Slovic P. (ed .) (2000). Th e Percep tion of Risk. Earth scan , Lon d on . Slovic P., Flyn n J.H . an d Laym an M. (1991). Perceived risk, tru st, an d th e p olitics of n u - clear w aste. Scien ce, 254 (5038), 1603-1607. Solberg C., Rossetto T. an d Joffe H . (2010). Th e social p sych ology of seism ic h azard a d - ju stm en t: re-evalu atin g th e in tern ation al lit- eratu re. N at. H azard s Earth Syst. Sci., 10, 1663-1677. Stein R.S., Barka A.A. an d Dieterich J.H . (1997). Progressive failu re on th e N orth An atolian fau lt sin ce 1939 by earth qu ake stress trig- gerin g. Geop h ysical Jou rn al In tern ation al, 128(3), 594-604. Stew art I.S. an d N ield T. (2013). Earth stories: con text an d n arrative in th e com m u n ication of p op u lar geoscien ce. Proceed in gs of th e Geologists’ Association , 124(4), 699-712. Stew art I.S. an d Lew is D. (2017). Com m u n i- catin g con tested geoscien ce to th e p u blic: Movin g from ‘m atters of fact’ to ‘m atters of con cern ’. Earth -Scien ce Review s, Vol. 174, p p .122-133. Tekeli‐Yeşil S., Ded eoğlu N ., Brau n‐Fahrlaend er C. and Tanner M. (2010a). Factors m otivat- in g in d ivid u als to take p recau tion ary a ction for an exp ected earth qu ake in Istan bu l. Risk An alysis, 30(8), 1181-1195. Tekeli‐Yeşil S., Ded eoǧlu N ., Tan n er M., Brau n ‐Fah rlaen d er C. an d Obrist B. (2010b). In d ivid u al p re-p ared n ess an d m itigation a c- tion s for a p red icted earth qu ake in Istan bu l. Disasters, 34(4), 910-930. Tekeli-Yeşil S., Ded eoğlu N ., Brau n-Fahrlaend er C. and Tanner M. (2011). Earth qu ake aw ar e- n ess an d p ercep tion of risk am on g th e resi- d en ts of Istan bu l. N atu ral H azard s, 59(1), 427-446. UN ISDR (2015). Th e Sen d ai Fram ew ork for Disaster Risk Red u ction 2015-2030, Un ited N ation s In tern ation al Strat egy for Disaster Red u ction , h ttp :/ / w w w .p reven tion w eb.n et / files/ 43291_sen d aifram ew orkford rren .p d f (accessed 31 Jan u ary 2018). Un sal B.O. (2015). State-led Urban Regen er a- tion in Istan bu l: Pow er stru ggles betw een in terest grou p s an d p oor com m u n ities. H ou sin g Stu d ies, 30(8), 1299-1316. Wach in ger G., Ren n O., Begg C. an d Ku h licke C. (2013). Th e Risk Percep tion Parad ox - Im p lication s for Govern an ce an d Com m u - n ication of N atu ral H azard s. Risk An alysis, 33, N o. 6. Witte K. an d Allen M. (2000). A m eta-an alysis of fear ap p eals: Im p lication s for effective p u blic h ealth cam p aign s. H ealth Ed u cation & Beh avior, 27(5), 591-615. Wood M.M., Mileti D.S., Kan o M., Kelley M.M., Regan R. an d Bou rqu e L.B. (2011). Com - m u n icatin g Action able Risk for Terror ism an d Oth er H azard s. Risk An alysis, 32, 601- 615. Yeo M. (2014). Fau lt lin es at th e in terface of sci- en ce an d p olicy: In terp retative resp on ses to th e trial of scien tists in L’Aqu ila. Earth - Scien ce Review s, 139, 406-419.