18 studies / 54 · Warsaw — Wola

The Hankiewicz houses: setting buildings full of people on fire

On Saturday, 5 August 1944, the Hankiewicz houses on Wolska Street became one of the places where Wola residents were murdered en masse. In a witness account, the buildings were surrounded, hit with grenades and set on fire without any prior order to leave. The Warsaw Rising Museum cites an estimate of about two thousand people killed in this group of houses. [1] [2]

Editorial account based on the sources listed below. Sources checked: 5 September 2026.

Victims and toll
About 2,000 people — an estimate for the group of houses, not a named list.
Perpetrators
German soldiers; the cited sources do not determine the exact subunit that carried out the arson.
Documented methods
setting inhabited buildings on fire · throwing grenades · surrounding houses and limiting the possibility of escape

Basis for the facts and their qualifications: [1] [2]

People in the surrounded houses

The houses at 105, 107 and 109 Wolska Street held both permanent residents and people seeking shelter from the fighting. An account published in English as record no. 59 describes the sudden encirclement of the buildings in the afternoon. The witness did not remember an order to leave the houses. The account indicates that the attack was directed at buildings in which people remained. [1]

Arson and grenades

According to the same account, soldiers threw grenades and used a substance to start a fire. We do not assign it a specific chemical composition: the witness described the appearance of the substance, not a laboratory examination. The combination of encirclement, arson and grenades shows how a place of shelter was turned into a trap. The account says people died in the flames or in explosions. [1]

The toll and the wider setting

The Warsaw Rising Museum places the Hankiewicz houses within its account of the Wola massacre of 5 August and cites about 2,000 victims. This estimate concerns the group of buildings. We do not add it to the overall number of people murdered in Wola, because that would count the same victims twice. [2]

Limits of the testimony

We do not have a complete named identification of those killed here or a reconstruction of the fate of every flat. The witness's strong statement that there were no survivors should be read as a description of what the witness knew, not as the result of a later, complete investigation. We also keep separate the events in these houses from other crimes discussed by the museum in the same text. [1] [2]

Limits of the findings

The number of victims is an estimate. Record no. 59 does not provide the witness's name or the date of examination in the available transcription. We do not transfer to this address every method of killing described generally for Wola. [1] [2]

Sources and notes

Footnote markers identify the basis for specific findings. Bibliographic records remain in their original language; the notes indicate how each source is used.

The bibliographic record and source-use note are retained in Polish to preserve the original scholarly citation.

  1. świadek niepodpisany w internetowej transkrypcji. Warsaw Uprising Witnesses: German Atrocities IV — Record No. 59. WarsawUprising.org — publikacja dokumentu historycznego.

    zeznanie świadka · Record No. 59 · Dostęp: 2026-09-05.

    Use of this source: Otoczenie domów, granaty, podpalenie bez nakazu opuszczenia.

  2. redakcja MPW. Powstańcze dni: 5 sierpnia 1944 r. Muzeum Powstania Warszawskiego.

    opracowanie instytucjonalne · Akapit o domach Hankiewicza i kolonii Wawelberga · Dostęp: 2026-09-05.

    Use of this source: Umiejscowienie zbrodni w rzezi Woli oraz szacunek około 2000 ofiar.

How we assess sources

The evidence base consists primarily of Polish scholarly publications, investigative findings, archives, museums and documented witness accounts. Reliability is assessed through the document’s provenance, scholarly method and the ability to check the finding. Statements by perpetrators and family accounts of perpetrators require especially critical comparison with other material. Two pages from the same institution may repeat one finding; they are not automatically treated as independent confirmation.

Full methodology