H

Mandala Q&A — Goro Murayama × Harm van den Dorpel

Questions from Goro Murayama (村山 悟郎) in response to van den Dorpel's solo exhibition Cloud Writings at Takuro Someya Contemporary Art, Tokyo, 2026. Replies by Harm van den Dorpel.


Q1 — On specific mandala references

Goro Murayama: Do you use any specific mandalas as references? If so, please tell me.

Harm van den Dorpel: I have primarily looked at Tibetan sand mandalas and Japanese Ryōkai Mandala mandalas. My first attempt to construct them algorithmically was in my work Anicca, of which there are two — see https://harm.work/nft/anicca. I went deeper into their construction later in the NFT series Quantizer, https://harm.work/nft/quantizer; the traits reflect this, and are even named — see for example https://harm.work/nft/quantizer/trait/root/tibetan and https://harm.work/nft/quantizer/trait/root/thangka.

It was in reading Carl Jung that I was made aware of their urgency, for example in this quote:

"I saw that everything, all paths I had been following, all steps I had taken, were leading back to a single point — namely, to the mid-point. It became increasingly plain to me that the mandala is the centre. It is the exponent of all paths. It is the path to the centre, to individuation."

The algorithm from Quantizer I extended to transform surfaces to lines by hatching, into something I could mechanically draw with a plotter. This can be seen in the works Anobium, Nethermind, and Nethermind Quilt.

Anicca 1
Anicca 1 (2025) — first algorithmic attempt at constructing a mandala; Pali term for "the absence of permanence and continuity"
Anicca 2
Anicca 2 (2025) — companion token; the planned series is an "exercise in patience and reflection on impermanence"
Quantizer #208 — Tibetan trait
Quantizer #208 (2025) — a token whose root trait is named tibetan; mandala construction explicit in the algorithm's vocabulary
Quantizer #066 — Thangka trait
Quantizer #066 (2025) — a token whose root trait is named thangka, after the Tibetan painted scrolls that frame mandala imagery

Q2 — On the algorithm

Goro Murayama: Could you explain the details of the algorithms used in your mandala-inspired works in a bit more detail?

Harm van den Dorpel: Mandalas are often recursive structures. Shapes become containers for multiple, smaller shapes, in careful symmetrical subdivision. They communicate deep spiritual truths using highly structured and organised visual means. That in itself already sounds like generative art. I often use Markov Chains: the distribution of a certain step or operation happening is stochastically determined by the parent the new shape is in.

Nethermind
Nethermind (2025) — ink on watercolour paper, 98×118cm; the mandala algorithm extended from surface to line, drawn by plotter. Photo: Shu Nakagawa

Q3 — On reconciling local algorithm with macro cosmology

Goro Murayama: I believe many of your works utilize mechanisms that generate patterns through localized algorithms. On the other hand, mandalas can be described as images that apply a macro-scale cosmic view to the human world. If you have any thoughts on how you reconcile these multiple scales, please share them.

Harm van den Dorpel: That is a beautiful observation. I myself generally do not assign a meaning to the outputs of algorithms. Sometimes I title them after creation, based on how they appear to me, but it happens largely beyond me. I provoke the work, structure the process, curate outputs, and highlight particular features that emerged. The hermetic observation "as above, so below" is at work in that I sometimes experience awe by staring through a simple construed surface, into the depths of complexity.

Anobium
Anobium (2025) — Sakura fineliner on Hahnemühle watercolour paper, 70×90cm; the mandala/Quantizer algorithm hatched into plotter line. Photo: Shu Nakagawa

Notable claims


Connections to existing wiki pages