Antoinetta Angelidi takes us on a walk through the forest. She discusses how she makes her films: her constructive principles and her method, as well as the particular realisation of her films. She offers keys for understanding, leading us through theory and practice, her work and her life. She returns again and again to the same concepts and experiences, adding different levels of meaning, like a spiral leading deeper and deeper into her magical world. She presents her views on what cinema is, heterogeneity and timing and cinematic space. She speaks about her work with actresses and actors, on kinesiology and speech enunciation. She recounts her conception of film ‘schema’ – a notion that combines the senses of idea, shape, form, and structure – and the ‘schemata’ of her films. She discusses her perception of film sound, as related to timing and heterogeneity, and the work on the sound of her films. She muses about her relation to real places and shooting locations. She analyses at length the inspirations and references, and the beginnings of each of her films. The conversation ends with a meditation on life, death and art.

This exchange cannot but bear witness to the interviewer’s very long and intimate relationship with Angelidi. It cannot be avoided. Our words are interwoven with each other, as are our lives. The conversation took place in Greek, in Athens, in August 2020, in the short respite between two Covid-19 lockdowns.

On her conception of cinema

Rea Walldén (RW) – Where do you begin?

Antoinetta Angelidi (AA) – For me, cinema begins with the distortion of images by timing; there is something programmatic about it. The process of de-composing the complexity of images through the introduction of time is determinative for my conception of cinema as a means of expression. Timing is a composition of the movements inherent in the image, where movement is rather a distortion than a transportation, and each element of the image may move independently. And inversely, I started perceiving images as a raw material for their decomposition. Therefore, I could extract an element from an image and connect it with an element from another image. Moreover, since the beginning, I realised that this decomposition of images should be done according to different codes.