Euroscoreboard 2.0 how to use it

"NSU 2.0": How language profilers expose threatening letter writers

The alleged author of the "NSU 2.0" threatening letter arrested on Monday was convicted based on, among other things, his writing style. Alexander M. is said to have insulted politicians, celebrities and lawyers with Nazi slogans and threatened death. Because the 53-year-old from Gesundbrunnen also spoke in chats on the right-wing populist portal PI-News and a chess portal, investigators from the Hessian State Criminal Police Office got him on the trail.

But how can you expose perpetrators based on their texts?

Experts from the Federal Criminal Police Office helped. They found a high level of correlation between the characteristics in the threatening letters and the man's statements on the Internet. How watertight the forensic linguistic report of the BKA against Alexander M. is, remains to be seen in court. The police do not want to get involved any further.

But how can you expose perpetrators based on their texts? "We work on different linguistic levels," says language profiler Patrick Rottler from the private institute for forensic text analysis in Munich. “From the text entry, to the choice of words, the peculiarities of the grammar, you go deep into the psychology of language. Does the perpetrator write 'because' or 'there'? Does he choose 'therefore', 'therefore' or 'therefore'? In this way we identify language patterns that are significant, occur systematically and are typical of the perpetrator. "

Rottler's institute works primarily for private companies who have received blackmail or threatening letters, but want silent investigations without a public prosecutor. The “linguistic fingerprint” does not have the 99.9 percent evidential value of a real fingerprint or a DNA trace. And yet it can say more about the author than a classic fingerprint, for example about regional origin, age, language skills and level of education. In forensic linguistics, the identity of the author is determined “with probability”, “with high probability”, with “very high probability” or “with probability bordering on certainty”.

Some threatening writers try to disguise themselves in order to disguise their identity, for example by writing about “we” or “a group of disgruntled employees” when only one person is behind it. But the largest part of human language is formed unconsciously, in patterns that are deeply anchored in us. “I can only adjust what I am aware of,” says Rottler. “Errors are built in, 'das' written with two Ss or commas are intentionally set incorrectly. But adjustments usually remain on the surface because language is very complex. So there is usually enough left to recognize attempts at adjustment. "

“The decisive factor is whether such errors are individual cases or whether they appear systematically,” says Rottler.

Errors in a sharp or double S and the setting of commas are not particularly meaningful because many people make them. “99 percent of all formulations, sentence constructions or argumentation structures do not allow any conclusions to be drawn about the person or personality of the perpetrator. The trick is to find the remaining one percent, ”says Patrick Rottler. That is why he looks for significant deviations from the norm: for example, for regional coloring or if phonetics is used and endings are omitted. Or when adjectives are reinforced by spelling "early" instead of "early". “The decisive factor is whether such errors are individual cases or whether they appear systematically,” says Rottler.

Sometimes forensic text analysis makes no sense. For example, if there are too few comparative texts. “The more we have, the better,” says Rottler. Or when threatening letters and comparative texts are too far apart, because a person's language changes. With Alexander M., the linguists from the BKA have no worries. In the Internet chats, as in the threatening letters, he is said to have segregated mobbing and insults in large numbers.

Also read:"NSU 2.0": The police tracked down the author of the threatening letters through his spelling mistakes >>

In search of further evidence, the Frankfurt am Main public prosecutor's office, which is conducting the proceedings, is currently evaluating the man's computer. On this he is said to have written his 115 letters to 32 people and 60 institutions. His statements on a chess portal and the connection data there had led the investigators to his IP address and finally to the Internet connection in his apartment in the Soldiner Kiez.

Forensic linguistics cannot always be used to convict perpetrators. The federal prosecutor's office experienced a setback in court in 2007. The sociologist Andrej Holm was in custody at the time. He was accused of membership in a terrorist organization, the “militant group” (mg). Arson attacks were carried out in their name, including on Bundeswehr vehicles.

The various pieces of evidence brought forward by the Federal Prosecutor included the finding that Holm had the skills to write the intellectually demanding letters of confession from the mg. According to a BKA report, he used “phrases” and “key words” in scientific papers and articles that the mg had also used in its letters of confession. One of them was the word "gentrification". However, this has been used in sociology since 1990. The arrest warrant has been overturned.