SATURDAY.JULY24.

2010

JOURNA THE IIALISTREET

Inc. All RightsReserved O 2010,Dow Jones& Company,

Lost in Trarclation
thn wry peoplnseethe world,;a thatlnngwqe proloundly infltrcnces New cognitiueresearchsuggests of blnme in lapaneseand Spanish dilferent sense
BY LERA BORODITSKY

Do the languages we speakshape the way we think? Do they merely in express thoughts, or do the structures languages(without our knowledgeor consent)shape the very thoughtswe wishto express? Take "Humpty Dumpty sat on a..." Even this snippetof a nursery rhyme can differ revealshow muchlanguages In English, wehaveto from oneanother. we mark the verb for tense;in this case, say"sat"ratherthan "sit."In Indonesian you neednot (in fact, you can't)change the verbto mark tense. you wouldhaveto mark In Russian, tense and also gender, changing the verb if Mrs. Dumptydid the sitting.You wouldalso have to decideif the sitting or not. If our ovoid eventwascompleted hero sat on the wall for the entire time he wasmeantto, it wouldbe a different form of the verb than if, say, he had a greatfall. In Turkish, you would have to include in the verb how you acquired this information.For example,if you saw the chubbyfellow on the wall with you'duseoneform of the your owneyes, verb, but if you had simply read or heard about it, you'd use a different form. Do English, Indonesian, Russian end up attending and Turkish speakers to, understanding,and remembering their experiencesdifferently simply because they speak different languages? These questionstouch on all the in the study of major controversies mind, with important implicationsfor politics,law and religion.Yet very little empiricalwork had beendoneon these questions until recently.The idea that language might shapethoughtwas for a untestable at best long time considered andmoreoftensimplycrazyandwrong.

Now, a flurry of new cognitivescience researchis showingthat in fact, lanprofoundly guage howwe influence does seethe world. The question of whetherlanguages backcentuthe way we think goes shape proclaimed that "to ries; Charlemagne is to havea seclanguage havea second ondsoul."But the ideawent out of favor with scientistswhen Noam Chomsky's theoriesof languagegainedpopularity proDr. Chomsky and'70s. in the 1960s grammar posed is a universal that there for all human languages-essentially, don't really differ from that languages one another in significant ways. And didn'tdiffer from one languages because another,the theory went, it made no senseto ask whether linguistic differin thinking. led to differences ences The searchfor linguisticuniversals yieldedinterestingdata on languages, of work, not a single but after decades proposed universalhas withstoodscrudeeper as linguistsprobed tiny. Instead, (7,000 or so, into the world'slanguages innuof themanalyzed). only a fraction merable unpredictable differences emerged. just because peopletalk Of course, differently doesn't necessarily mean they think differently.In the past dechave begunto ade, coglitive scientists measurenot just how peopletalk, but alsohow they think, askingwhetherour of even suchfundamenunderstanding time as space, of experience tal domains by and causalitycould be constructed language. For example, in Pormpuraaw,a remote Aboriginal communityin Ausdon't tralia, the indigenouslanguages usetermslike "left" and"right."Instead, everythingis talked about in terms of absolute cardinal directions (north, you say south,east,west),whichmeans things like, "There's an ant on your

southwest leg." To say hello in Pormpuraaw, oneasks,'Where are you going?",and an appropriateresponse might be, "A long way to the southHow aboutyou?"If you don't southwest. knowwhich way is which, you literally can'tgetpasthello. About a third of the world's languages(spoken in all kinds of physical rely on absolutedirecenvironments) As a result of this contions for space. stant linguistic training, speakersof suchlanguages are remarkablygoodat staying orientedand keepingtrack of wheretheyare,evenin unfamiliarlandfeats scapes. Theyperformnavigational scientistsonce thought were beyond This is a big differhumancapabilities. differentway of ence,a fundamentally conceptualizing space,trained by lanoltga

Differencesin how people think rely aboutspace don'tend there.People knowledge to buildmany on their spatial other more complexor abstractrepresentations including time, number, musicalpitch, kinshiprelations,morality and emotions. Soif Pormpuraawans think differently about space,do they alsothink differentlyaboutotherthings, like time? AliceGaby Tofind out,my colleague and I traveled to Australia and gave Pormpuraawans sets of pictures that showed temporal progressions (for picturesof a man at different example, ages, or a crocodilegrowing, or a bananabeing eaten).Their job was to arrange the shuffled photos on the ground to show the correct temporal in two seporder.Wetested eachperson aratesitttngs,eachtime facingin a difWhenasked to ferentcardinaldirection. do this, Englishspeakers arrangetime do it from left to right. Hebrewspeakers Hebrewis from right to left (because written from right to left).

THE PUBLISHER'S SALE OF THIS REPRINT DOES NOT CONSTITUTE OR IMPLY ANY ENDORSEMENTOR SPONSORSHIP OF ANY PRODUCT,SERVICE,COMPANY OR ORGANIZATION Nor PERMTTTED Do Nor EDIT oR aLTER REPRINTS, REPRoDucrtoNs cusroM REpRt/vrs 609) 520-4x1 PO, Box 3oo PRtNcEToN, NJ 08543-0300.

DOWIoNES==

found, Pormpuraawans, we arrangedtime from east to west. That facingsouth,time wentleft to is, seated right. When facing north, right to left. Whenfacingeast,towardthe body,and so on. Of course,we never told any of our participantswhich direction they not only faced. The Pormpuraawans knewthat already,but they alsospontato neouslyusedthis spatialorientation their representations of time. construct time And many other ways to organize In Manexist in the world'slanguages. darin, the future can be belowand the pastabove. in South In Aymara,spoken America,the future is behind and the pastin front. In additionto spaceand time, languagesalso shapehow we understand For example, Englishlikes to causality. doing describe in termsof agents events things. English speakerstend to say things like "John broke the vase" even for accidents.Speakers of Spanishor would be more likely to say Japanese "the vasebrokeitself." Suchdifferences languages have profoundconbetween underfor howtheir speakers sequences stand events,constructnotionsof cauwhat they remember sality and agency, and how much they as eyewitnesses blameandpunishothers. In studies conductedby Caitlin of English, Fausey at Stanford, speakers of watched videos Spanish andJapanese two peoplepoppingballoons,breaking eggsandspillingdrinkseitherintentiongot Later everyone ally or accidentally. memorytest: For eachevent, a surprise who did it? Shediscan you remember differcovered a strikingcross-linguistic ence in eyewitnessmemory. Spanish did not rememand Japanese speakers eventsas ber the agentsof accidental Mind you, well as did Englishspeakers. the agentsof intenthey remembered tional events(for which their language wouldmentionthe agent)just fine. But events, whenonewouldn't for accidental normallymentionthe agentin Spanish or Japanese,they didn't encode or remember the agentaswell. In anotherstudy, Englishspeakers watchedthe video of Janet Jackson's infamous "wardrobe malfunction" (a wonderful nonagentivecoinage introby Jusduced into the Englishlanguage accompanied by oneof tin Timberlake), two written reports. The reports were identical except in the last sentence where one used the agentive phrase "ripped the costume"while the other ripped."Eventhough said "the costume watchedthe samevideo and everyone witnessedthe ripping with their own

mattered.Not only did eyes,language peoplewho read "ripped the costume" blame Justin Timberlake more, they also levied a whopping 53%more in fines. Beyondspace,time and causality, patterns havebeenshown to in language shapemany other domainsof thought. who make an extra Russianspeakers, light and dark blues between distinction in their language,are better able to visually discriminate shadesof blue. The Piraha, a tribe in the Amazonin language eschews number Brazil,whose words in favor of terms like few and many,arenot ableto keeptrack of exact quantities.And Shakespeare, it turns out, was wrong about roses: Rosesby many other names (as told to blinddo not smellas sweet. foldedsubjects) offer a window Patterns in language andpriorities. dispositions on a culture's For example,English sentencestrucand in our crimituresfocuson agents, nal-justice system, justice has been donewhenwe'vefoundthe transgressor and punishedhim or her accordingly (ratherthanfindingthe victimsandrestituting appropriately,an alternative approachto justice). So doesthe languageshape or does the culturalvalues, gothe otherway, or both? influence of course,are human Languages, toolswe invent and hone to creations, suit our needs. Simply showing that speakersof different languagesthink differently doesn'ttell us whether it's languagethat shapesthought or the the other way around.To demonstrate what'sneeded causalrole of language, lanthat directlymanipulate are studies guage in cognition. andlookfor effects in recent One of the key advances years has been the demonstration of preciselythis causallink. It turns out talk, that how people that if you change If people learn how they think. changes another language,they inadvertently also learn a new way of lookingat the world. When bilingual people switch theystart to another, from onelanguage thinkingdifferently,too.And if youtake in ability to uselanguage awaypeople's what should be a simple nonlinguistic canchange dratask,their performance makingthem look matically,sometimes no smarter than rats or infants. (For example,in recent studies, MIT studentswere showndotson a screenand askedto say how many there were. If they were allowedto count normally, they did great. If they simultaneously did a nonlinguistictask-like banging out rhythms-they still did great.But if they did a verbal task when shownthe

in thewordsspoken dots-like repeating fell apart. a newsreport-their counting In other words, they neededtheir language skillsto count.) All this new research showsus that the languages we speaknot only reflect our thoughts, but alsoshape 0r express the very thoughtswe wish to express. The structuresthat exist in our languagesprofoundlyshapehow we construct reality, and help make us as aswe are. smartandsophisticated Languageis a uniquelyhuman gift. \Mhen we study language,we are uncoveringin part whatmakes us human,getting a peekat the very natureof human nature. As we uncoverhow languages and their speakersdiffer from one another, we discover that human natures too can differ dramatically, we speak. depending on the languages The next steps are to understandthe mechanisms through which languages helpus construct the incrediblycomplex knowledgesystems we have. Understanding how knowledgeis built will allowus to createideasthat go beyond the currently thinkable.This research questions cuts right to the fundamental How do we we all ask aboutourselves. cometo be the way we are?Why do we think the way we do?An importantpart it turns out, is in the lanof the answer, guages we speak.