21 years — as the Whitney-Rothschild collection’s associatecurator from 1932 to 1944, andcurator to 1953.

His early work guidedTheodosius Dobzhansky’sGenetics and the Origin ofSpecies (1937), which founded thesynthetic theory of evolution. WithJulian Huxley and George GaylordSimpson, he helped incorporatenew discoveries by naturalists andpopulation geneticists into theframework of Darwinian theory.

Harvard was important to thesecond half of his life, he wasprofessor of zoology from 1953 to1975 and also directed theuniversity’s Museum ofComparative Zoology from 1961to 1970. His interest inevolutionary biology, and thehistory and philosophy of biologygrew. His Animal Species andEvolution (1963) had new views onthe nature of species.

His later years were dominatedby continuing his argument intothe development of a philosophyof biology. Mayr was acutelyaware of the distinction betweenbiological principles and those ofthe physical sciences. As physicalscience had an increasing inputinto the workings of biology hisplangent voice grew stronger. Hisone long argument, backed by alifetime of solid practical work, issure to endure. “I’m an old-timefighter for Darwinism,” he told theHarvard University Gazette in1991. “I say: ‘Please tell me whatis wrong with Darwinism. I don’tsee anything wrong.’”

Nigel Williams

“Das eigentliche Studium derMenschheit ist der Mensch”,“Mankind’s actual study is man”;with this statement, the Germanpoet and polymath J.W. Goethe(1749–1832) captured the factmuch of human intellectualendeavour — in art, literature andphilosophy — has tried to answerthe question of what it means tobe human. Not far from Goethe’scity of Weimar, in Leipzig, theMax-Planck-Institute forEvolutionary Anthropology (MPI-EVA) provides scientists from aseemingly disparate set of fields,including psychology, linguistics,primatology, paleoanthropologyand genetics, with an idealenvironment to address humanevolution.

Leipzig is now a booming city inthe east of Germany, but theremnants of 40 years ofcommunism, seen by itsproponents as a practical answerto the human condition, are stilltangible throughout the town.There are fading signs advertisingproducts from the ‘people’sfactories’, a brass plate on a hotelannouncing the fact that KarlMarx, ‘founder of scientificsocialism’ spent a night here, andthe characteristic ready-madehousing blocks, the‘Plattenbauten’ that line the widestreets for miles.

Amidst such blocks the Max-Planck society, Germany’sleading research body, has setup a stylish and light newbuilding which has been thehome of the MPI-EVA since 2003.When officially founded in 1997,the MPI-EVA was the firstinstitute of anthropology startedby the Max-Planck society sincethe end of World War II. This

reluctance to be involved inanthropology was in large partdue to the fact that Hitler’snational socialist regime hadabused anthropology — in thosetimes called ‘Rassenkunde’ (thestudy of races) — to provide ascientific foundation for thesuperiority of a white race andthe crimes committed againsthumans all over Europe in itswake. The rehabilitation ofanthropology in Germany is likelyto have been facilitated by theclear evidence from modernmolecular genetic data againstthe idea that there are largedifferences between, and relativehomogeneity within, so-calledraces.

The tower of BabelWhat makes us human? Part ofthe answer lies in our ability to askthe question: language. Of course,other animals do have —sometimes sophisticated —communication systems, but thehuman language is unique in itscomplexity and versatility, a factthat is reflected in the sheernumber of different languages.Today something like 6500languages exist, each with acomparably high grammatical andlexical complexity. In the past,largely influenced by NoamChomsky’s enormous body ofwork, linguistics has emphasisedthe universal features of grammarand language.

A group of linguists in Leipzig,however, takes a differentapproach. Martin Haspelmath andBernard Comrie are charting thetypological differences andgrammatical complexity of theworld’s languages. The outcomeof this 5-year effort involving

Current Biology Vol 15 No 5R146

Darwin’s champion: Ernst Mayr, whodied last month, spent his researchcareer bolstering Darwinism.

Feature

Planet of the apes

What makes us humans so special? Our language, our genes, ourculture, our cognitive skills? At the Max-Planck-Institute for EvolutionaryAnthropology in Leipzig, psychologists, linguists and biologists tacklethis old question in a truly multidisciplinary way. Their results haveimplications not just for our understanding of human evolution — theyalso touch directly on many social and environmental issues. FlorianMaderspacher reports.

collaborators from all over theworld is an atlas of languagestructures, summarizing thegeographical distribution of overa hundred features of a thousandor so languages. This will, forinstance, tell the reader wherepeople put the numeral in front orbehind the noun, where the sameword is used to connect nounsand clauses, or where numbersystems based on ten asopposed to twenty are used. “Wehope that this quantitativeapproach will provide a newboost for linguistics as a whole”,says Haspelmath. His analysisreveals, for instance, thatneighbouring languages are oftenalso typologically similar, eventhough they are not related bydescent. Exchange, not only ofwords, but also grammaticalfeatures between languages maybe more common than previouslyassumed. This in turn hasconsequences for the use oflinguistic similarity to bolster dataon the common genetic ancestryof human populations.

Some populations may evenexchange their entire language, asfound by Mark Stoneking from thegenetics department of the MPI-EVA. Stoneking analysed humanpopulations in the Caucasus, aregion particularly rich in linguisticdiversity. He found that two majorpeople, the Armenians andAzerbaijanians, are more closelyrelated to their geographicneighbours than to peoplespeaking similar languages,indicative of languagereplacements.

The atlas of language structuresmay soon be mapping a lostworld, as about half of thelanguages spoken today arepredicted to vanish within the nextgeneration. This affectspredominantly many so-calledsmall languages: more than 95%of all languages are spoken byless than a million people. Apartfrom depriving linguists oflanguages for study, thisimpoverishment puts the culturalheritage and identity of thespeakers at risk.

But languages can bereanimated, “much like Hebrewwas revived as a spoken languagein Israel”, says Juliette Blevins, a

linguist at the MPI-EVA. As a sideproject, Blevins became involvedin revitalising Chochenyo, thelanguage originally spoken in theSan Francisco Bay. AfterChochenyo had stopped beingactively spoken, Blevins helpedreconstruct the language fromthousands of handwritten pagesby J. Harrington in the late 19thcentury. Now there is a smallcommunity of native Americans inthe bay area who are reviving theironce lost language. SoChochenyo and other smalllanguages may, after all, survivebeyond the bookshelves.

Language of the genesHow we came to speak in thefirst place is the questionaddressed by the group ofSvante Pääbo, who studies thegenes behind human evolution.Pääbo’s work is always good forsnappy headlines in the sciencepages of the newspapers, as forexample when two years ago hisgroup showed that a proteincalled FoxP2 has acquiredchanges that were selected for inthe human lineage. That in itselfwould have been interesting, butwhat stirred the interest inPääbo’s findings is the fact thatthis gene plays a role in speechand language: FoxP2 wasisolated from members of afamily who had specificdifficulties in articulating andunderstanding language caused

by mutations in the gene. FoxP2is a transcription factor and is, asexpected, active in the brain, aswell as a number of other tissues.

Despite its fame as a languagegene, it is not clear whichprecise function of FoxP2 isrequired for language andspeech. Even less clear is howthe human specific changes inFoxP2 will have affected itsfunction with regard to language.Wolfi Enard, who is working onFoxP2 in Pääbo’s lab, is usingmice to address this question.He has replaced the mouse’sendogenous FoxP2 gene with thehuman version of the gene; onepresumes that these mice willstill be squeaking, rather thanspeaking, but Enard hopes theystill can provide insight intowhich activities of FoxP2 havebeen selectively modified duringhuman evolution and how theyaffect our ability to speak.

Besides language, humansexhibit a number of otherbiological peculiarities — justthink of hairlessness, upright gaitor brain size, for all of which thegenetic basis is not clear. Pääbo’sgroup is trying to find more genesthat, like FoxP2, were selected forduring human evolution.Comparisons of orthologousgenes in whole genomesequences of humans,chimpanzees and mice identifygenes with coding sequencechanges specific to the human

Magazine R147

A new light: The Max-Planck-Institute for Evolutionary Anthropology in Leipzig at dawn.

lineage. Such genes are thenchecked for variation withinhuman populations: a low level ofvariation, especially in non-codingregions, is taken as a sign thatselection has been acting on thegene in question. But such datado not reveal the functionalsignificance of the gene.“Unfortunately, the evolutionarysignatures are often very subtleand we hope that mice can serveas an evolutionary model systemfor functional analyses”, saysEnard.

Despite the experimentaluncertainties, knowing even a fewcases of changes in gene functionthat apparently occurred on ourlineage is likely to considerablyadvance our understanding of thebiological factors that drovehuman evolution. And it may haveimplications for our understandingof evolution in a wider sense.Svante Pääbo gets excited whenhe talks about similar analyses hewants to apply to domesticatedanimals, comparing them withtheir wild counterparts. He hastherefore set up a collaborationwith a group of scientists ineastern Siberia, who have beenexperimentally domesticatingfoxes and rats over manygenerations (see the recent paperby Brian Hare and colleagues

‘Social cognitive evolution incaptive foxes is a correlated by-product of experimentaldomestication’, Current Biology15, 226–230). Finding a pattern ofgenes that were selected forduring domestication mightultimately even provide a freshtwist to an old idea, originally putforward by Konrad Lorenz,according to which the biologicalchanges in human evolution aresimilar to those domesticatedanimals have gone through.

The social beastWhen looking for what makeshumans special, everyone seemsto know intuitively where to look:“It’s the brains, stupid”. But at adeeper level, it is much moredifficult to actually detail thepeculiarities of the humancognitive organ and itsaccomplishments: The brains ofsome animals rival the humanbrain in size or complexity. Some,especially when trained, cancommunicate and expressthemselves. They can solveproblems or use tools. Some ofour relatives even seem to havedeveloped culture-like skills thatthey learn or are taught. So whatprecisely makes the difference?Michael Tomasello, head of thedepartment for comparative

developmental psychology at theMPI-EVA, has devised aconceptual and experimentalframework to try and answer thisquestion. The idea is that thespecial cognitive abilities ofhumans reflect differences insocial and cultural cognition.

Among all the animals on thisplanet, humans are the ones thatcompose and performsymphonies, travel into space,play football and talk about theweather. Such more-or-lesscultural activities are the result ofcollaboration, and requireindividuals sharing their intentionsand focussing their attention ontoa common goal or subject. “Thisshared intentionality is a specialversion of cooperativity”, saysTomasello. It involvesunderstanding the collaborator,sharing his perceptions andintentions, and being willing andable to engage with him.

Thus, a ‘triadic engagement’ ofthe individuals directed towards ashared goal emerges. From thiscommon ground, “hotspots ofcommunication”, as Tomaselloputs it, arise, allowing humans notonly to cooperate, but also tocommunicate about ‘topics’ or‘subjects’. This may haveultimately driven the evolution oflanguage. In evolutionary terms,the social cognitive skills, theability and willingness to shareintentions and to engage andcommunicate with the others mayhave conferred an advantage tothe group as a whole. “This issomehow reviving the concept ofgroup selection”, says Tomasello,“as cultural group selection”. Thecultural conformity within groupsand the amazing differentiation ofhuman cultures — and languages— may reflect the action of suchgroup selection.

From a less philosophical pointof view, it is clear that sharedintentionality requires certaincognitive skills, which in the firstinstance have to do withunderstanding the other’sperceptions and intentions. Tofind out what precisely these skillsare and how they develop,Tomasello’s group studies humaninfants, the only living humansstudied at the MPI-EVA. Between9 and 15 months of age, children

Current Biology Vol 15 No 5R148

In a huddle: Though chimpanzees are highly social, they may lack some particularfeatures of human social cognition.

gradually develop an ability tounderstand other people’sperceptions and intentions, andultimately they are able to sharethe other’s attention and intentiontowards a third entity — ‘triadicengagement’. The latter inparticular is predicted to accountfor the special cognitive abilitiesof humans, so these skills arecompared to those of otherexperimental subjects: apes,because they are evolutionarilyrelated to humans, and dogs,because they share, as aconsequence of theirdomestication, some aspects ofhuman social life.

Dogs are, at least in the eyes oftheir keepers, obviouslyintelligent. One striking example isthe border collie Rico, Germany’s,if not the world’s, smartest dog.Rico had his 15 minutes of famewhen he proved on a German TVshow that he can recognisesomething like 200 objects byname. Not only that. JulianeKaminski from Tomasello’s groupshowed that Rico can learn wordsfor new objects, that he cancorrelate unknown objects withunknown words and that he cancorrelate the objects not only withwords, but also with smallerreproductions.

But dogs are also remarkable ata less catchy level. When it comesto interpreting social cues, such asan experimenter pointing to orlooking at a hidden food source,they perform better thanchimpanzees. These socialcognitive skills are most likely to bea consequence of domestication,because dogs also performspecifically better at such tasksthan wolves. Domestication for thepurpose of taking part in humancollaborative activities, such ashunting or herding, has thusresulted in dogs adopting some ofthe social skills of humans.Intriguingly, domestication mayhave this effect in general, not onlyon the herder, but also on theherded, as Kaminski recentlyshowed that goats also are able tofollow gaze cues.

Planet of the apesOne way of asking what it is liketo be human is to ask what it islike to be not quite human. And in

this case, the ones to ask are ourclosest living relatives, the greatapes. Research on apes istherefore a main focus of the MPI-EVA. Its associated Primatecentre at the Leipzig zoo providesa unique opportunity to study allgreat ape species in parallel. In avast open space, the largest of itskind in Europe, chimps, gorillas,orang-utans and bonobos live insmall colonies in a more opulentand naturalistic environment thantheir conspecifics in thedepressing tiled apehouseselsewhere. This has yielded theLeipzig zoo an enormous increasein popularity. Visitors can evenwatch the experiments, which incontrast to previous set-ups arevoluntary for the apes and do notinvolve food deprivation.

Josep Call, who heads theprimate research at the zoo, looksrather unconventional for apsychologist in his green overalland rubber boots, and yet hisresearch on apes affects the waywe think about humanpsychology. “Most psychologistsare centred on humans”, saysCall, “but the mood is changing.We need to broaden our scopeand look at outgroups to betterunderstand humans”.

Call’s research has already ledto a reconsideration of thecognitive skills we attribute toapes. It has been known for a longtime that chimpanzees can makeinferences about the physicalworld around them, and evenunderstand concepts such as

cause and effect. But it wasassumed that they have little or noconcept of what others see or ofwhat goes on inside other minds.

As a striking example, Call hasbeen able to show that, in thepresence of another, dominantindividual, chimps prefer to grab apiece of food, which the dominantcannot see or of which thedominant doesn’t know. Thissuggests that the lower rankingape must have some concept ofthe dominant’s perception and ofhow perception will influence hisactions. Moreover, the chimpsseem also able to distinguish if anexperimenter trying to hand themfood acts intentionally oraccidentally.

But if they do show this sort ofinsight into the other’s mind, whydo apes then not engage in thesame sort of shared intentionalityas humans do? Despite theirremarkable social cognitive skillsbecoming more and moreappreciated, apes still seem tolack the level of sophisticationeven young humans display at‘mind reading’. And, moreimportantly, apes show littlemotivation to really becomeinvolved in sharing other’sperceptions and intentions. In thisregard, they are somewhat similarto some autistic children, whoalso seem to primarily lack themotivation to engage with others,despite the fact that they can tosome extent understand theiractions and perceptions.Whatever the distinctive social

Magazine R149

Gorillas at rest: The primate centre at the Leipzig zoo allows study of all four great apespecies in parallel. Here, Bebe and Ruby are taking a break.

cognitive features of humans mayturn out to be, it seems clear thatthe research on non-human socialcognition will provide fruitful inputto our understanding what makesus so special.

But of course, apes are not onlyinteresting as a standard by whichto measure humans, they arecomplex interesting creatures intheir own right — creatures aboutwhose lives we still knowsurprisingly little. Although somefamous ape populations, such asthe chimpanzees of Gombe andthe mountain gorillas in Virunga,have been intensely studied,much less is known about other,even nearby populations. Otherspecies, such as bonobos, are sopoorly understood that not eventheir precise geographicdistribution is known. Theprimatology department of theMPI-EVA is therefore geneticallyanalysing ape populations.

When the genetic data, mainlyderived from faeces samples, arecombined with behavioural andecological observations, asometimes surprising picture ofape life-style emerges. Forinstance, when Linda Vigilant whodoes most of the molecularanalyses in the primatology group,found that groups of lowlandgorillas, in contrast to mountaingorillas, are headed by singlemales that are often related toeach other. Ultimately, comparingape populations will provide afuller picture of the range of socialsystems in apes and how they are

influenced by ecology. With thisknowledge at hand, one can tryand make educated guesses abouthow our own ancestors may havelived, but more importantly, theresearch can be beneficial for theapes themselves, as it may help tosteer conservation efforts moreefficiently.

Christophe Boesch, head of theprimatology department, has beenstudying chimpanzees in the Taïrain forest of the Ivory coast formore than 25 years. Boesch’s fieldwork has helped to appreciatethat chimpanzee behaviour,especially with regard to tool use,has culture-like features.Regardless of whether or notthese behaviours match allacademic criteria applied to‘culture’, their complexity isbewildering. Like human culture,these behaviours show featuressuch as tradition and — notably— innovation. “The chimpanzeesinvent a new behaviour everyyear”, says Boesch “but thedissemination is difficult anddepends on the social status ofthe inventor”. Once again,chimpanzees seem all too human.

But, like the other great apes,chimps are under threat. This ismainly due to deforestation —where there were lush greenforests when Boesch first arrivedin Taï now farmland stretches —and poaching, especially in thenotoriously conflict ridden centralAfrica. Though eating apes asbush meat is not necessarily asign of scarcity of food, indeed

chimpanzees are in some partsseen as a delicacy: “much like wehave a goose at Christmas,poaching goes up around newyear’s” says Linda Vigilant. Onemajor line of Boesch’s efforts istherefore to increase theawareness and understanding forchimps. Counterintuitive as it mayseem, “the locals don’t naturallyhave a good opinion of thechimpanzees”, says Boesch. Soamong other things, Boesch’swild chimpanzee foundation(www.wildchimps.org) hasrecruited a theatre companywhich travel the villages aroundthe Taï reserve and playfullydepict what it is like to be achimpanzee. The people who sawthe play seem indeed to realisethat rather than easy meat,chimpanzees are our ‘cousins inthe forest’. Little steps like thesemay help ensure that this won’tone day be the planet of just onekind of ape.

Mankind’s study is man, and sothe MPI-EVA was recentlycompleted by addition of apaleoanthropology group, headedby Jean-Jacques Hublin. Butresearch at the institute showsthat a very promising way oflooking at humans is to studyother animals and comparing theircharacteristics to ours. At thesame time researchers learn moreabout what makes these othercreatures so special and how wecan help them to survive. Thiscomparison involves genetics,ecology and psychology, allworking within an evolutionaryframework. The multitude ofscientific approaches reflects thecomplexity of the question andthe answer lies somewhere in theas yet hazy territory between thedisciplines. It must be an evolvedproperty embedded somewhere inour genes that makes us willingand able to engage in culturalactivities with others and to shareand exchange intersubjectively,by means of our sophisticatedlanguage. The answer will for awhile be difficult to find, perhapsalso because in this case the onewho studies and what is studiedare the same thing — human.

Florian Maderspacher is CurrentBiology’s reviews editor.

Current Biology Vol 15 No 5R150

‘Man of the forest’: our most distant relatives among the great apes, orang-utans.