The Memory of Treblinka Foundation is dedicated to commemorating those who were murdered in the Treblinka extermination camp. Its mission is to restore the names and identities of the victims and ensure they are never forgotten. At the heart of this work is the Book of Names – a digital database of victims that records biographical information, deportation routes, and family connections. The project was launched by computer scientist Paweł Sawicki and his wife Ewa Teleżyńska-Sawicka. In over 15 years, they have collected more than 130,000 names of Holocaust victims from over 500 sources with help from academics, archives, and dedicated individuals. This comprehensive database forms the basis for a planned memorial wall on the site where the Treblinka camp used to stand.
During a visit to Bad Arolsen in May 2025, Paweł Sawicki spoke about his personal motivation, the methodological challenges involved in reconstructing the victims’ life stories – and about how the alphabetical-phonetic system used by the Arolsen Archives is helping to further improve the quality and the accuracy of the Book of Names.
How did your research begin?
It all started by chance. Many years after my father’s death, I learned that he was a Holocaust survivor. He’d never spoken to us children about it. Nearly everyone in his family was murdered in Bełżec. But this personal story is not what triggered our work in Treblinka. In 2010, my wife and I had the feeling that something vital had been lost in Poland – the memory of the Jewish population. We began reading Jewish names aloud at the places where these people had lived and died. In 2011, we started doing the same thing in Treblinka every last Saturday of the month. This was at the invitation of Father Wojciech Lemański, a Catholic priest who regularly said prayers at the memorial site. He welcomed the idea, and since then, reading out names has become an integral part of the monthly commemorations held there.
We soon realized that if we didn’t want to read the same names out every month, we would need to write them down. That’s how the Book of Names was born.

What exactly is the Book of Names?
The Book of Names is more than just a database – it’s really a place of remembrance. Right from the start, our focus wasn’t just on individual names, we were also interested in family connections. People were deported and murdered together. We want them to stay together in remembrance too.
We’ve collected more than 130,000 names so far, from over 500 different sources – including eye-witness accounts, transport lists, archival material, court records, and oral histories. But this number only represents a fraction of the 900,000 people or thereabouts who were murdered in Treblinka.
What are the biggest challenges you face in your work?
There are three main challenges. The first is identifying duplicates. Many of the sources we use – such as Yad Vashem’s Pages of Testimony – contain a huge amount of personal data. The same person can appear more than once with different spellings or incomplete information. All these different sources and transcriptions can lead to multiple entries for one and the same person. We examine each case individually and try to bring them together using a combination of comparison algorithms and manual research.
The second challenge is establishing a person’s fate. Many relatives didn’t know exactly where their relatives had been deported to. Sometimes the names Treblinka, Auschwitz, and Theresienstadt were used interchangeably. We try to check every entry, but many documents have been lost or never existed in the first place.
The names themselves are the third challenge. They were often misspelled, mis-transcribed or mistranslated – sometimes they were changed beyond recognition. We compare different versions, reconstruct the most likely form of the name, and consult experts if needed. This is time-consuming, but necessary.

What led you to contact the Arolsen Archives?
The team at the Arolsen Archives had developed an alphabetic-phonetic system to help identify individuals despite different spellings, languages, or transcription errors. This was exactly what we needed.
We’re currently cross-checking our database using this system. This process generates “match scores” – for example, a 100 percent score indicates that the name, date of birth, and place of birth are an exact match. Lower values indicate deviations. This enables us to identify possible duplicates and cases where someone may have been wrongly listed as having been murdered in Treblinka.
The alphabetical-phonetic system is an incredibly valuable tool for us – especially in view of the memorial wall which will include the names. We want our work to be as accurate as possible, and collaborating with the Arolsen Archives helps us to get closer to this goal.
How do you deal with errors in the database?
We accept that mistakes can happen – and we correct them. That’s the only way to carry out this type of research. Sometimes families write to us saying, “You’ve listed our relatives as murdered – but they survived.” Moments like this are always a time to pause and reflect, and we adjust our data accordingly. Some people say we shouldn’t include names on the memorial wall because errors are inevitable. But that’s exactly what the perpetrators wanted. They wanted to erase people completely, leaving no trace. If we don’t name them, we’ll lose them a second time. Of course we want to avoid mistakes. But recording the victims’ names is the most important thing of all. If we wait for perfection, we will never get round to remembering them. The memorial wall won’t be a monument of certainty, it will be a memorial of presence. Memory is always imperfect, and that imperfection is what makes it so precious.
What does this work mean to you personally?
When I enter names in our list, I see the people in my mind’s eye. A family with three children: an 18-year-old daughter, a 17-year-old son, and a 14-year-old child. Or a baby that’s just 2 weeks old. Or a grandmother aged 102. These aren’t just numbers – these are lives that were violently cut short. But if you get too emotional, you can’t do this work. And if you feel nothing at all, you shouldn’t be doing it. It’s a delicate balance, and fortunately we seem to be able to achieve it.

