It began with the 1961 German-Turkish recruitment agreement. Many young Turks, who had initially come to Germany as temporary guest workers, ended up staying in what would become their second home country. While many of them still feel at home – linguistically – in the Turkish language, their children and grandchildren, who often have both German and Turkish roots, have grown up with both languages and are multilingual in a very special way. Understanding them may provide information on how the brain manages the permanent usage of different languages. The Potsdam Research Institute for Multilingualism (PRIM) is examining this question.
“Early research on multilingualism used a very simple approach,” says Prof. Dr. Harald Clahsen, Director of PRIM. “It mainly considered those who grew up in their native language and learned a second language as adults. We now know that multilingualism comes in many different forms.” Descendants of the former Turkish guest workers, for example, often grew up bilingually from birth. They are considered simultaneous multilinguals. Those who acquire a second language in preschool or at school are called successive multilinguals. The “conventional” adult learners of a foreign language, on the other hand, are called late multilinguals. The researchers at PRIM are interested in all of these groups. “The point is to better understand the nature of multilingualism in an individual,” Clahsen explains. “How do people store and use knowledge from more than one language, particularly grammatical structures? Which conditions promote or impede efficient multilingualism?”
A special aspect of the PRIM researchers’ work, as Clahsen emphasizes, is their psycholinguistic experimental approach. “We are trying to establish how language is represented in the brain, which we cannot do with questionnaires. It requires experimental examinations.” The methods of choice for these tests are eye movement monitoring and electroencephalography (EEG), which investigate word recognition and understanding of grammatical functions in the range of a few milliseconds.
The researchers at PRIM are investigating the different processing pathways of language acquisition and production in a number of subprojects. One of them is the “Experimental Studies on German-Turkish Multilingualism” that, according to Clahsen, will “cover a large part of this spectrum of multilingualism” using the example of the Turkish community in Berlin. Bilingual speakers in the German-Turkish community interest the researchers for various reasons. German and Turkish belong to different language families. While German grammar uses flection to specify the role of words in a sentence, Turkish uses agglutination. Word functions are expressed by adding affixes of exclusively grammatical meaning to the word root. Speaking both languages, therefore, poses a great challenge, and these conditions lend themselves particularly well to experimental research. In addition, among the four generations of Turkish migrants in Germany are people of simultaneous, successive, and late multilingualism, enabling comparisons highly sought after but sometimes difficult to make due to many factors. One focus of the project is on so-called heritage speakers, which includes successive multilinguals. They were mainly exposed to the Turkish language at home in early childhood. Only in preschool or at school did they learn to speak and write German, while their Turkish language skills – their “heritage” – remained exclusively oral. The PRIM studies investigate these people in both languages: German and Turkish. “This is an opportunity to identify what is more important – the age at which you learn a language or how often you speak it,” Clahsen says. It has long been assumed that you have to begin learning a language at an early age to master it and that successive native speakers never really learn German. I actually think this is not true.”
PRIM focuses its research on two core linguistic domains. One group of researchers examines word-level phenomena (morphology), the other group sentence-level phenomena (syntax). This is also done in the project on German-Turkish multilingualism. As is the case at the institutional level, Clahsen heads up the morphology group and co-director and adjunct professor Dr. Claudia Felser is in charge of the syntax group.
Dr. Gunnar Jacob is one of the researchers concentrating on word recognition. Although the three-year project, which is funded by the Federal Ministry of Education and Research (BMBF), has only been running since July 2014, he already has his first results. Jacob evaluated data of from a previous study that dealt with speakers of the same community, a fact that also illustrates the interrelation of PRIM projects that help to “expand the spectrum of the bilingual,” as Clahsen says. Jacob examines the way words are recognized, particularly morphologically complex ones because they carry not only content but also grammatical meaning. The German word “gereinigt” has a word root “reinig” and the affixes “ge” and “t”, which make it a participle. “We are interested in the first few milliseconds when our brain calls upon stored information and processes it,” Jacob says. “It is important to delimit our work to the very early stage of word recognition to avoid any interference of later effects.” For this purpose he runs a so-called “priming experiment”. The test subjects see a word on a screen for only 50 milliseconds, “because we know that our brains can take in word information even within such an extremely short presentation time, Jacob explains. They then see another form of the same word, for example “geöffnet” and “öffnen”. The idea behind it is that if our brain can “chop off” the suffixes from the first presented word and “activate” the root word, then subsequently shown words will be recognized faster. Jacob carried out such a priming experiment with several multilingual speakers: conventional foreign language learners, native speakers, and “heritage speakers”. “We wanted to find out whether they are more like foreign language learners or native speakers,” the psycholinguist says. “But neither was the case: They differed from both groups.” While native speakers disassemble the word and identify its root, heritage speakers’ comprehension system concentrates much more on orthography and reconstructs a word letter for letter. This can be explained by the fact that they learned Turkish almost exclusively as a spoken language but not the written Turkish language, which is usually not taught at German schools. Heritage speakers therefore have to use additional language processing resources for written Turkish.
Gloria-Mona Knospe is also looking at the linkage of two languages in heritage speakers. The PhD student is working on a subproject dealing with sentence-level processing. She examines how these speakers process the grammatical function of personal and reflexive pronouns in a sentence. This is done with an eye tracker – a device that measures the eye position and eye movement of a test subject – and with ‘visual world’ eye tracking. The subjects hear sentences while looking at pictures of people to which the pronouns refer. “The eye tracker records where a subject looks while listening to a sentence, Knospe explains. “We can measure how long it takes until they look at the correct person. In principle, the device opens a door to the mind, a window to the brain.”
The PhD student then compares the results of the German-Turkish bilingual speakers with those of German native speakers and Russian-German bilinguals, migrants who belong to the group of late multilinguals. Over the past months Knospe has been able to get 56 German-Turkish bilinguals to participate. One of the project’s challenges is finding suitable test subjects. Clahsen, Felser, and their teams have been aware of this from the outset. It was clear from the beginning that you have to work with the community in which they live: in Berlin. That is why the researchers decided to rent a room in the Berlin district of Wedding for the experiments to keep distances short and the psychological barrier low for participants. Once involved, many of them actual enjoy taking part in the study, Knospe says. “The Turkish group has been very open-minded because it was an interesting experience for them but also because their Turkish linguistic identity is rarely the focus of attention, and they find it great that it is being investigated.”
The project involves not only German researchers. It is a collaborative project, a so-called scientific and technical exchange with Prof. Dr. Bilal Kırkıcı and his research group at the Middle East Technical University in Turkey. Knospe and Jacob went to Turkey in October 2014 and took the eye tracker with them to test Turkish students as a native-language reference group. They also trained the Turkish researchers to use the device. The Turkish partners will visit Potsdam to support the researchers on designing the tests and questionnaires for the Berlin German-Turkish speakers. “We would never run a project on Turkish without a native-language project partner,” Clahsen says. “Perhaps they can also better reach the cool kids,” he jokes.
It is a goal that demands social skills but mainly hard work and using the right instruments. Understanding the realities of multilingualism requires being near the subject. The complexity of the investigated phenomena necessitates a wide range of monitoring instruments that also have to be usable. Jacob only needs a laptop for the priming experiments. Knospe, however, needs a lab for her examinations due to the eye tracker. Clahsen considers the mix of examination methods a prerequisite for the overall project’s success. “When you want to examine such a community, you cannot just show up with a brain scanner. You need a combination of practical and informative field experiments as well as laboratory ones. “
Despite their sophisticated equipment, the researchers are well aware of the fact that a lot of work still lies ahead. Knospe has finished the first series of experiments – on German. The second one – on Turkish – is being carried out and evaluated. Jacob is also working on another experiment – a priming experiment on Turkish morphology. It will also include the recording of language processing but this time using electroencephalography (EEG). Although the researchers use a detailed and differentiated approach to the forms of multilingualism, they never lose sight of the overall assignment, Clahsen stresses. The community is like a window that offers insight into the brain’s ability to manage the acquisition and use of several languages.”
Prof. Dr. Harald Clahsen studied philology and sociology and earned his doctoral degree on the subject of language acquisition in 1981. Since 2011 he has been Alexander-von-Humboldt Professor at the University of Potsdam and Director of PRIM.
Potsdam Research Institute for Multilingualism
Dr. Gunnar Jacob studied psychology at the University of Münster and earned his doctoral degree on the role of the native language in foreign-language syntactic processing. Since 2011 he has been a Postdoctoral Research Fellow at PRIM.
Gloria-Mona Knospe studied psychology at the Philipps University of Marburg. Since 2012 she has been a PhD student at PRIM.
Experimental Studies on German-Turkish Multilingualism
Participants: at the University of Potsdam – research group of Prof. Dr. Harald Clahsen and PD Dr. Claudia Felser / at the Middle East Technical University – research group of Prof. Dr. Bilal Kırkıcı
Funded by: Federal Ministry of Education and Research, The Scientific and Technological Research Council of Turkey (TÜBİTAK)