The NGI archive at La Cambre

The National Geographic Institute (NGI) is a federal public institution that produces and distributes geographic reference data for all of Belgium. Over its history, it has constituted an exceptional legacy of maps and documents. Created in 1831, the Dépôt de la Guerre et Topographie (DGT [Department of War and Topography]), was a military agency responsible for drawing up maps of Belgium. In 1878, it became the Institut Cartographique Militaire (ICM [Military Cartography Institute]) and moved into La Cambre Abbey in Ixelles, on the site where the Royal Military School had stood since 1872. For almost a century, these buildings were home to the workshops, laboratories, and offices of national cartography. The Royal Military Academy quit the site in 1908 to make way for the ICM, which became the Institut Géographique Militaire in 1947. The IGM was demilitarised in 1976, thus becoming the National Geographic Institute.

Like a cracked lithographic stone, the NGI was forced to leave La Cambre Abbey in 2020. A large part of the documents and maps relating to its history were bequeathed to the State Archives, while the collection of maps and measuring instruments went into the drawers of the Royal Library of Belgium, of the Royal Military Museum, and Ghent University. The archival work has, however, neglected a number of tracings and maps dispersed across the buildings of the La Cambre site. To the extent possible, these were gathered in an attic during a first collective operation designed to protect them from destruction. Since 2023, they have been the focus of collective artistic research (FNRS-FRArt), whose goal is not to restore the archive in line with archival principles, but to experiment with how to give it a voice, investigate its graphic codes, and redefine it uses through creating varied encounters between this cartographic archive and different audiences. The research has thus been fuelled by performative experiments, the traces of which are contributing to the rethinking of this neglected archive.

Several hours of recordings of accounts from Jean-Luc Parmentier (former cartographer at the Map Revision Department and archivist with the NGI), Valérie Gevers (architect and drafter), Virginie Pigeon (landscape designer), Sophie Boiron (graphic designer and cartographer), Julie Vanderhaeghen (geographic information systems developer), and Milady Renoir (poet-performer), as well as those of the workshop and performance participants, were captured by Frédérique Franke (musician and engraver). These recordings form an audio archive, weaving together tales of maps, gestures, spaces, and reflections on the craft of cartography. These voices, in association with the rediscovery of the collection material, give form to a collective memory that goes beyond the documents themselves. In parallel, a narrative inventory of the archive has been established, accompanied by a photographic collection of survey maps. Complementing this is the establishment of a library of shared references, bringing together cartographic documents produced by bodies other than the NGI, enabling us to broaden the scope, placing the institute’s productions within wider landscape of the territory.

As mentioned, our approach is not archival. It is situated within a practice founded on listening, collective writing, and the spatial experience of our bodies in relation to these maps and what they represent. It also engages us in thinking about classification in order to reactivate the collection, to propose a new reading of it without freezing it in a logic of passive conservation. The three visuals presented below are three readings of landscapes through a graphic prism—the representation of water—and a classification system that underpins the document’s raison d’être. Each of these readings offers a special way of “giving voice to the archive” by searching for water in the classification system and reinterpreting the cartographic grid, found in all NGI documents. This grid organises the space of representation, and opens up fragments of interviews, parts of images, and slices of history.

The first reading by Marine Declève (urban designer and art historian) is a piece based on a colour-coded plan of the building occupied by the NGI at La Cambre, where the source of the Maelbeek is found. Created by the Facilities Department at a scale of 1:250, this plan uses a colour code to differentiate the various functions of the buildings associated with the NGI departments. By comparing it with other occupancy maps and the oral testimonies that have been collected, a geographic system of information has been reconstituted, making it possible to superimpose traces and to investigate the space where these maps were produced—and where they are being used to support new forms of collective writing. In the second reading, Sophie Boiron (graphic designer and cartographer) consults the Aide-mémoire à l’usage des Officiers et Agents cartographes chargés du Service du Complément [Map-making guide for cartographic officers and agents of the additions department, Phototopography Department] (1962), and investigates the lexicon. Through a montage based on the word “tourbière” (peat bog), the research shifts to the vocabulary used for cartographic legends. This approach highlights the semantic slippages between the terrain that we observe and the lexicon associated with a graphic code that denotes it. The third reading, presented by Pierre Huyghebaert (graphic designer and typographer) is inspired by the manual Instructions pour l’élaboration de la carte de la Belgique au 1:50 000 par voie de gravure et décalcomanie sur matière plastique [Instructions for the production of the map of Belgium at a scale of 1:50000 by engraving and decalcomania on plastic, Cartography Department, Institut Géographique Militaire, 1970]. This takes the form of a flow chart detailing the complete chain of operations involved in establishing the map of Belgium, from the preparatory work to the decisions to keep or destroy certain documents. This diagram follows military rigour; it shows the logic of the rationalisation of a skill that since digitisation has become anachronistic. It questions how we produce maps today—without the succession of manual steps, without the materiality of the gesture of engraving, without water-transfer printing.

These three readings trace three paths for accessing the archive and the landscape of water: the map is a spatialised entry point, the guide to a lexical entry, the flow chart of a sequential entry point. Together they constitute an attempt to give voice to the archive through cross-referencing, dialogues, and montages. The challenge is not to classify but to connect, to bring about correspondences between the archive’s textual inventory, audio resonances, and the physical wrestling with the surviving materials. The project on the archive thus becomes a space of encounters that takes shape, methodically and poetically, at the intersection of these three approaches—spatial, lexical, sequential—to restore the archive’s capacity to incite the imaginaries. The challenge is not to close the file but to reopen its folds, to invent modes of reading in which the map, the story, and the document are no longer frozen objects but become forms in perpetual motion.

Superimpositions

The page on which this text is printed was once silent. The lines of the words that have been printed on it make visible things that have crossed my mind as I surveyed the site of the La Cambre Abbey on a colour-coded map from the dispersed NGI archive. In the text, an asterisk *, indicates a reference to the map drawing.

Silences of text. Silences of maps. Silences of history.

Begin a journey through the silences of this coloured plan. Putting colour in the white spaces, making the dispersal of the archive visible in concrete sites. Accumulating traces of gestures and techniques to reconstitute the silences of history. Written, annotated to enter into the layers of the story at different scales of territory and narration. Layers of polygons, dots, and lines. Determining which text is shown at what scale to communicate a perception of these places, from page to landscape. Laid over this, the sounds*, images, and sensations connected to the presence of water*, trees*, or Brussels sand*, superimposed on the material traces of the past, present, and future, on what becomes a cartographic annotation sheet.

Cartographic annotation sheet. This document constitutes one of the elements provided to the personnel responsible for the markings to be included in the map. (Aide-mémoire, 1962, p. 6)

All these sheets of paper to be inventoried are spaces and stories to be surveyed. In this pile of papers is the colour-coded plan of the National Geographic Institute at La Cambre, a technical document, used by NGI officials. Long ago, someone printed a plan on a silent sheet of paper, on to which they then hand-coloured the choreographic organisation of Belgian cartography. On this plan, printing workshops*, photographic laboratories*, and geodesy laboratories*. No water is marked on this map; here, the Maelbeek does not flow. No sand, no trees. Put back the water*, the sand*, the trees*.

Surveying the site as we walked, we unrolled the plan in a photo lab in Block B*. The roll in which it was stored held other copies, without colour or annotations. Placed in a sophisticated storage system to enable its handling, this roll was next to other forgotten technical plans, including a site-scale layout, and numbered and captioned occupation plans of the cellar and the first and second floors. The ground-floor plan, which was most likely handled more than the others, was missing.

The original plan divides the different floors occupied by the NGI on the page. But in the version suggested here, the ground floor* has been cut out and overlaid on other documents. Each colour corresponds to a function: the offices and cloakrooms are in light blue, the paper stores in red, the print shops in dark blue, the photo labs in mauve, the computer rooms in brown, the workshops in terracotta. A legend also indicates the presence of security systems: ionic, thermal, and flame detectors, alert buttons, and alarm horns. The sounds of these devices can be heard, as can the cartographers’ poetic gestures, and the materiality of the retaining walls, recomposed in lithographic stones*.

Overlay this on the cartographic annotation sheet. Record audio excerpts from the stories of Jean-Luc* [audio 1-17], Pierre and Sophie. Annotate with other sources of information.

How to start writing on these documents, whose value is unknown? A doubt, Diogenes syndrome, a disquiet creeps in. It is linked to the affective connection that we have with this archive. Like an instinctive resistance to altering the state in which these documents have presented themselves to us. Preserving an image of spatial organisation as an index to be decoded in order to understand the history of this cartographic presence. Do not impose any definitive meaning to a new story on a silent page. The space of the information system and the recordings become our draft of overlays, the resonant forms of a state of encounter between visual and sound documents, rescued from the attic.

Produce a new version of a colour-coded plan of the NGI. Agree collectively on the writing to be transferred on the cartographic annotation sheet. The gesture of saving would thus be prolonged by the collective drawing of a kind of memoir. The desire to set up a workshop for cartographic transmission in the La Cambre attic is an echo of all this. These maps are vectors of connections, interdisciplinary works that need a dedicated space to be read, where are collective interpretations of the textual overlays seeping into the silences of the maps and of history can be shown, transmitted and placed into relation with other archival fragments.

Vertical minutes

My home is in the marshes and peat bogs of Cantal and Lozère. I am immersed in them everywhere, from Ireland to Belgium, Estonia, and Finland.

For several years, we have been steeping ourselves in the NGI’s collections of blues. We move, look, classify, and marvel. The full colours, the details, the representations, the inks, the patterns, the lines, the hyphens, the grids. The first things that Jean-Luc handed to us were blue and gray overalls, but above all, the Aide-mémoire à l’usage des officiers et agents cartographes chargés du service du complément (1962). A small pink book with brass screws, which would have accompanied the entire career of each one of the all-male cartographers, pages printed on only one side, leaving the back blank for annotations over many years. And above all, with all the planimetry details to be represented on a map. Inside, marshes, swamps, peat bogs, bog ponds, and stretches of water to fill [audio 18].

The Aide-mémoire indicates the words and signs.

Peat bogs

Definition: marshy land from which peat is extracted. Representation: represent marsh vegetation according to the standard conventions, as well as bog ponds water. Text: none.

Swamps

Definition: spongy land, heavily impregnated with water almost permanently; they change the appearance of the vegetation that grows there very little (meadows, moors, thickets, etc.). Representation: represent vegetation cover according to the standard conventions; cover the marshy area with parallel rows of blue dashes, without limiting it to a particular line.

Marshes

Definition: depressions filled with mud and temporarily or permanently covered with stagnant water in which a dense vegetation of typical grasses grows: reeds, phragmites, cattails, etc. Spaces of open water might appear in the middle of this vegetation. Representation: draw a blue line around the entire marsh; also draw a blue line around spaces of open water, if there are any. The spaces of open water (dark in the photo) may be covered in parallel blue hatching (symbolising a pond); the part of the marsh invaded by reeds (very light on the photo) should be covered by rows of parallel blue dashes. No character for vegetation cover is indicated in the marsh, but rather that may be present, which will be drawn where possible. Text: Major marshes known by a particular name will have this added to the photo. This toponym is to be transferred in bistre.

Peat bog water

Definition: parts of peat bogs covered by water. Representation: the pog bonds are represented by a particular conventional sign: delimit the bog pond with a blue line; leave its interior empty. Text: none.

On maps and shelves, in stacks and folders, the symbols of swamps and peat bogs occupy a small rectangle in the legend; over the years, they transformed according to scale.

Talking about the underneath, the invisible, a double state, a misleading state, like a tunnel, a bridge. Swamp vegetation is vertical [audio 17].

Due to climate change and human occupation, peat bogs are disappearing from the landscape. Peat bogs are also disappearing from maps, with only their status as “permanent or non-permanent body of water” remaining in the legend of the NGI portal. There is no longer any internal categorisation for these two categories, only this state that is significant for our era.

Symbols evaporate, peat bogs disappear. Do they still unfold in our heads? How can we still imagine them if the maps are empty? How can we pay attention to them in the field if there are no symbols, no vocabulary, no language [audio 19]?

A peat bog is a particular ecosystem, principally made up of plants adapted to a waterlogged environment, the debris of which accumulates. The slow decomposition of these elements produces peat, a material containing up to 50% carbon. The formation of peat bogs in Europe dates back ± 10,000 years, that is to say, to the end of the last ice ages. ¹

You don’t rest on peat bog, you feel the ground with your feet, you feel the limits of the peat bog, its boundaries, to ensure your feet don’t suddenly end up in the water, swallowed up like a mistake [audio 20]. We walk around, outside, around the edges. Sometimes we walk above it, on wooden pontoons, to get a closer look at the peat bogs. And this summer, I swam there. I swam in a peat bog pond. Pontoons enable you to get close, and in this instance, allowed me to slip into the water. Impossible without a pontoon, without a ladder. Being in the middle of a peat bog, surrounded by soft things, on the border between Estonia and Russia. To be inside the blue line, in the centre of parallel blue lines.

Stacks

The operational sequence for establishing the 1:50000 scale map of Belgium is a stack of 22 layers of fossils, film, and zinc substrates. The diagram details the various specific supports: Astralon [audio 21 et 22], Astrafoil, Permatrace, Stabilène, Diazo paper, Peel Coat, Eidesco, auto-reversal film, offset zinc. But also the operations that happen there: drafting bases, minutes, and stéréo-minutes (drafts of future plans), collations, photocopies, primary and secondary manuscripts, overprints, photo lettering, make-ready, printing. Between these supports, movements circulate vertically and intersect to weave an overall Euclidean trajectory, running down from the metric drafting bases at the diagram’s top to the offset printing at the bottom.

Pile 5, stack. 47 -1-4.5-6-8.7. Bromide photo print, very low ISO black and white, mirrored for reproduction. If I understand well, this is an original.

Pile 8, stack 1. Metal support, white paint, perhaps used to ensure something doesn’t move. Tracing paper. You don’t want to remove the tracing paper because it is fixed to the back, and I know you can still see quite well that these look like line-drawn originals, so just black lines on a white background.

(Narrative inventory of the NGI collection at La Cambre, created during FRArt research).

It is a human endeavour repeated many, many times to form the group of 57 maps that represent the Belgian territory. Some of the markings on the corners suggest supports to be preserved as hard objects, among the flux of those that are only temporary. The objects described thanks to this laborious stacking are distributed over the seven colour layers used to “establish the map of Belgium.” Among these colours, the blue of the water network, traced by a line of increasing thickness, the width of the watercourse exceeds 15 meters, 4.5 millimetres on the map.

A small blue arrow is placed in the frame along the extension of natural watercourses represented by a line of at least 3/10 mm thick each time the source is not located on the sheet or when it is not possible to officially establish the direction of the water flow at the confluence of any tributaries. The arrow points toward the cutting mark or away from it, depending whether it is placed, by choice, at the beginning or end of the sheet.

These many and manifold operations, all respond, directly or indirectly, to the stacking of other layers, geological layers of mother rock. Three hundred million years ago, telluric thrusts raised what would become, for humans, the southern part of the country. Since then the capillary action of the rocks, evaporation, and rain have brought water to the peaks and slopes. Runoff from the Ardennes massif, shifts in the peneplain, and then tectonic uplift, carved small, wide valleys, reticulating from a few rivers into hundreds of streams and torrents that form several watersheds.

Each species of occupants of these reliefs, the flora on the few decimetres of soil landscape and the fauna on and under the surface, including the water, are present on a small scale. Thus every crumb of earth passed through the intestine of an earthworm at least once every fifty years. For two millennia, a species has been erecting buildings, building transport routes, accumulating waste, and carving canals. The human inhabitants of the political territory of Belgium entrusted a very small number (in 1970, about one person in 50,000) with the task of establishing a sequence of operations to produce images from a bird’s eye perspective of the geological, biological, and technical transformation of the surface, at a scale that follows the same proportion. Building this stack requires great care, and a very small number of this very small number, probably just a few people, drew a map of the operations required to produce the map.

Despite their efforts, despite their dogged resolve to produce a standard free of ambiguities, unanticipated situations will surface. The variety of situations to be described means making corrections on the dense, ordered, printed diagrams, with ballpoint pens and correction fluid. Materials marked to be destroyed should, on the contrary, be kept. Despite the efforts of the printers in Building A to match the colours to each other, the line of the river partly extends over that of the bridge. The course of the water, indifferent to borders and thus to whichever organisation describes it, forces collaborations between institutions from neighbouring countries so as to not interrupt its representation and its flow.

Reading and writing landscape-classifications

As this research period draws to a close, it is not so much a question of closing a file rather than of opening up its continuations. Our exploration of the archival collection of the National Geographic Institute at La Cambre has revealed the extent to which cartography, far from being a simple measuring tool, can become a means of memory and relating. The documents found—maps, plans, manuals, flow charts, instruments—form a network of material traces, which in dialogue with oral accounts and cartographic documents, sketch another territory: that of transmission. This work, which began as a rescue operation, transformed into a collective writing experience founded on listening and determination for reactivation.

One of the major challenges for the future is in the preservation and remobilisation of this archive. The NGI’s archival collection at La Cambre is, as a whole, fragile, dispersed, and often lacking context. The issue is not just to conserve, but to design a space where consultation is itself a form of activation. Several possibilities emerge from the artistic devices experimented with during the research period: the creation of a conservation and consultation space at La Cambre, the constitution of a decentralised, hybrid collection associating physical media in various private homes with digital reproductions and sound archives accessible online, putting performative devices in place, where the public could experience the materiality of the corpus.

Additionally, considering the preservation of the archive, a key-word index has been proposed to enable circulation between various groups. Built from interviews, annotations, legends, and technical documents, this index is not designed to order the archives according to a hierarchical logic, but to open cross-disciplinary readings based on the approaches explored above: a spatial entry, connecting the space of the page with the accounts of uses of the landscape; a lexical entry, based on the special vocabulary of cartographic legends (such as the word “peat bog,” which was the starting point for the montage mentioned above); and a sequential entry, which follows the production chain of the maps.

This index functions like an open grid: it enables navigation from one document to another, from a word to a voice, from an image to a place. It makes the archive porous, alive, traversed by unexpected correspondences. It is a type of research tool, but also a space of invention, where reading is a creative act.

The interest in conserving this archive rests in its potential activation. It is not a question of fetishising the objects found, but of recognising what they may make possible: a rereading of the history of cartography in Belgium from its graphic clues, a sensitive understanding of the work of the former NGI personnel, and reflection on the transformation of sites of knowledge production.

The sound creations realised as part of the project tested how visual representation can be translated into words and sounds. A way of inhabiting the map. These experiments explored a new spatiality of the archive. Working with sound material—the voice of Jean-Luc Parmentier, the sounds of the workshops, the echoes of the building—the project shifted the map toward an immersive dimension, where listening becomes a way of surveying. The map fixes, sound connects. The map delimits, the voice crosses.

This intersecting of map and sound sketches a form of augmented writing of the archive. It engages body, space, and time in the same experience. By confronting the graphic documents with the dimension of sound, the project interrogates the possibility of making the mark audible, the precision of the line resonate, the voice of cartographic legends vibrate.

Research into the NGI collection at La Cambre has been conceived as a process in movement, at once artistic and memorialising. It is not seeking to bring closure, but to invent forms of continuity. The continuation of the project will consist of deepening these connections, experimenting with new forms of restitution—installations, publications, listening systems­—where each fragment of the archive will find its place, not as a relic, but as material for thinking our relation to the earth and to cartographic representations of living things.