Featured article: Karin Bijsterveld on “Slicing Sounds”

Karin Bijsterveld is a Professor of Science, Technology & Modern Culture at Maastricht University article. Her article, “Slicing Sound: Speaker Identification and Sonic Skills at the Stasi, 1966-1989,” appears in the June 2021 issue of Isis.

Our manuscript assistant, Mississippi State University graduate student Samuel Green, spoke with Dr. Bijsterveld about her research on the sonic techniques for voice identification developed by the East German Ministry for State Security.

GREEN: What first drew you to the topic of Stasi sound analysis?

Karin Bijsterveld

During a Maastricht University reception, an acquaintance introduced me to a forensic specialist who worked on speaker identification. When this specialist, Ton Broeders, heard about my background in the history of science and sound studies, he mentioned that Alexandr Solzhenitsyn had quite realistically written about the early years of speaker identification in the Soviet Union in his novel The First Circle (1968). At home, I checked out the novel, and realized the details had much in common with what I had read about the history of speaker identification in the United States in Mara Mills’ work. I was immediately intrigued, as speaker identification involved both visual (spectrographic) and audio analysis of recorded voices, but I do not read Russian. I considered it likely, however, that the Stasi would have worked on this as well, given their extensive practices of eavesdropping and wiretapping. I therefore contacted the former Stasi archives (BstU) by writing several letters. I did not get any response. One and a half year later, however, our department secretary gave me a call saying that an impressive package had arrived from Germany. I said I did not expect anything but could not resist checking what it was. Well, the package included over 1000 of copies of documents from the Stasi, all about the history of speaker identification as of 1966! It had taken so much time because the archivists had to screen all documents for names of those monitored, which they had to delete before they could share the documents with me.       

Were there any interesting finds that turned up in your research?

I have done my research in several rounds of archival visits, and each round came with its own surprising findings. In 2016, I came across preparatory documents for courses that introduced Stasi employees to voice identification. In those courses, they learned to distinguish between relevant characteristics of recorded voices, such as tempo, pitch, speaking style or dialect. Some of those documents were preparatory meetings between course leaders and Stasi employees about their everyday experiences with the auditory analysis of recorded voices. These were very detailed, and displayed such a frankness about the difficulties the employees experienced with voice identification that I could almost “see” the Stasi employees struggling with how to listen systematically. This material filled the gap between policy documents describing the “ideal” of voice identification and the numbers that showed that voice identification had not brought the Stasi what it had hoped to achieve. Another important moment, in 2019, resulted from help by Anita Krätzner-Ebert, a former BStU archivist. She was in the audience during one of my Berlin lectures about the Stasi, and helped me to get access to the personal Stasi files of Roland Jahn, a former GDR opposition member. These showed the actual voice analysis reports in his case by a key voice identification expert who worked at the Humboldt University, informing me both about the University-Stasi relations and about the style and content of these reports.   

Stasi surveillance is at the heart of this article. How does the current discussion of government surveillance resonate with what you found in your research?

The validity of speaker identification techniques, especially that of the voice spectrogram or voice print, soon became highly contested in the international scholarly world. Nonetheless, the Stasi kept trying to make the audio and visual analysis of speaker identification as “objective” as possible.  Constantly refining and mechanizing the system, however, also implied that earlier classifications of voice characteristics became less useful, as the system was not consistent anymore. This is a more generic issue that also plays out in government surveillance. The unique access to the Stasi archive enables us to trace such processes though. 

As historians, we often feel as though we’ve had to leave part of our stories untold. What sort of question about your work do you wish people would ask?

“How is it to work on such a recent and still sensitive history?” Each time I am presenting on this topic, audience members approach me to share stories about surveillance or other GDR experiences they or their family members have had. One of my Berlin lectures was in the building of the Berlin-Brandenburgische Akademie der Wissenschaften in the former Eastern part of the city center. When I entered that building, one of the doorkeepers said he knew what I was going to talk about, and that he would like to show me something. He then literally took me by the hand and pointed to particular spots on the walls and ceilings. “There was a microphone hidden there,” he said, “there, and there.” He looked at me as if he was still sharing a dangerous secret. It made the terror of the Stasi surveillance palpable again.