September 2026 – TRIFECTA

This September, Jiaqi attended the first week of the Computational Linguistics Fall School 2026 at the University of Technology Nuremberg. The Fall School is a biennial event for Master’s and PhD students from the humanities and computer science who want to deepen their knowledge of Computational Linguistics (CL) and Natural Language Processing (NLP).

The morning course, Programming for Computational Linguistics and Natural Language Processing, was taught by Selina Meyer (UTN). Walking through Python basics again was unexpectedly nostalgic, after four years in the field of DH. The afternoon course, Reading Concordances with Algorithms: a training course in key corpus linguistics methodology, was taught by Stephanie Evert and Michaela Mahlberg (FAU Erlangen-Nürnberg). This was my first formal encounter with linguistics through a course, and I was lucky to have Stephanie and Michaela as guides. They introduced the core apparatus of corpus linguistics — the concordance and the KWIC view, collocation, colligation, semantic preference and semantic prosody, and the lexical item as a unit that extends beyond the single word.

They also introduced the tools their team has been building as part of the Reading Concordances in the 21st Century (RC21) project, funded by the AHRC and the DFG. Its main output is FlexiConc, a Python library that lets researchers sort, rank, partition and cluster concordance lines using combinable algorithms — and, importantly, keeps an analysis tree recording every algorithm applied along the way.

What struck me was less any individual technique than a way of thinking. Corpus linguistics insists on focusing on the individual example and requires going back to the concordance lines and reading them. Another thing I learnt from the design of FlexiConc’s analysis tree (which aims to record every step that leads to the reading of a concordance systematically and consistently) is that a result is only interpretable alongside the procedure that produced it. For example, when I play with contextualised embeddings I extracted from a BERT model and use them to track how the usage of a word changes over time, each decision shapes what can be observed, such as which model produces the representation, which hidden layer(s) it is taken from, how occurrences are grouped, how the time axis is cut, and which metric measures the difference. None of these is neutral, and results produced under different settings are not straightforwardly comparable.

Thanks to Michael Roth and the organising team at UTN for putting the Fall School together, and to Selina, Stephanie, and Michaela for the courses. Thanks also to GSCL (German Society for Computational Linguistics and Language Technology) and DGfS (Linguistic Society of Germany), whose support made this possible. A week like this is a chance to step outside my own project and look at computational methods from another angle — and then to carry some of it back.

On 15 July 2026, Chris Groenveld, visiting researcher with TRIFECTA and final year PhD candidate at Goldsmiths, University of London, presented a paper titled Silence as Currency: Mapping Uncertainty in Dutch Maritime Archives. It was day one of a four-day conference held jointly by Performance Studies International and the Critical Island Studies Consortium, at Universitas Kristen Indonesia in Jakarta, on the theme Archipelagic Flows. Although TRIFECTA deals daily with a colonial archive that includes Dutch history in Indonesia, this marked the first time this work was presented in this regional context.

A Fitting Room for the Question

The theme Archipelagic Flows challenged delegates to think in island logics rather than continental ones. As a game designer, Chris presented a deck of cards intended to hold the silences of an archive such as this, in a room organised around resisting a single centre of knowledge. This provided an opportunity to ask a question the team in Amsterdam periodically wrestles with: who decides the ontological framework, and how are these decisions made?

Old and New Games

The talk opened with a board game printed in the Netherlands in 1875. Its route ran from Nieuwediep, a port in the north of the country, to Batavia, now Jakarta, where this conference took place. In this historical game, a player rolled dice, sailed a paper ship down a track of ports, and reached the distant city. Along the bottom edge of the board were listed the voorwaarden, or conditions of play. This game promoted a positive image of the VOC, teaching a Dutch child that the sea is theirs to cross, and the colony is their prize. Games often reflect the historical moment in which they were designed. What would be an equivalent game for this moment?

The prototype Chris presented at PSi#31 consisted of thirty cards, each holding one entity drawn from the digitised VOC archive and related open-source image banks: an image on the front, a short line of context and mostly blank space on the back. Crucially, there was no defined goal, no predefined rules. Similar to the 1875 game, it presents players with an obstacle, the partial documentation of a colonial past, and leaves off there. The difference between these games is not the transparency of rules, but who gets to decide them.

The cards

The participants were asked to pick any two cards on screen and discuss what the pairing opened and what it closed off, or to pick one card and name what was absent. A silence followed before anyone spoke. Gradually, connections began to be offered. One participant picked up on the contrast in perspectives offered by two different cards: first, a map of a killing site in Batavia from between 1740-1741 and second, an oil painting of a Dutch port. They remarked on the radically different registers of silence contained within both images, each speaking obliquely about the experience of a colonised people.

Researching, Not Demonstrating

The deck is still in development. Scheduled to follow are conversations with historians and digital humanities colleagues, and prototyping sessions with heritage staff at the Rijksmuseum, Wereldmuseum, the National Library of the Netherlands, with the hope that further insight into the silences contained within the archive can be identified and productively discussed.

Future routes

The VOC archive has its own word for loss: “Leakage” names what spills from the barrel on the long crossing, and anything above ten percent must be reported and charged. The archive counts what it loses. This deck keeps score the other way round: it pays out for naming the thing that was never counted at all, the name never recorded, the crossing never logged. The silence becomes the value held. Silence as currency is what the deck asked a room in Jakarta to hold for an hour, with no board, no rulebook, and nobody at the table trying to win.