There is a particular kind of silence that falls over a studio after a work leaves it. The piece goes to a group show, then into a collector’s hallway, then — often — into storage, a divorce, an estate sale. What remains in the studio is not the work but whatever the artist happened to write down about it. That residue, more than the object, is what determines whether the work has a life ten years later.
Most artists learn this the hard way. A curator asks for the dimensions of a piece sold in 2019 and the answer lives in a WhatsApp thread with a gallerist who no longer replies. A grant application requires an installation image and the only surviving photograph was taken on a phone, at an angle, with a stranger’s elbow in the frame. A museum considers a work for a survey and cannot establish who owns it. None of these are catastrophes individually. Collectively, they are the reason careers with equal talent diverge sharply after the first decade.
The archive is not nostalgia; it is infrastructure
It helps to separate two things that are usually confused. The first is memory — the sentimental impulse to keep sketchbooks and exhibition postcards. The second is documentation: a structured, retrievable record of what you made, when, from what, at what size, in what condition, and where it went. Sentiment is optional. Documentation is the load-bearing structure of a professional practice, and the moment you need it is always a moment when you have three days to respond and no time to reconstruct anything.
Consider what the record actually does. It makes work legible to institutions, which cannot exhibit what they cannot describe in a checklist. It makes work sellable, because provenance is the difference between an object and an asset. It makes work insurable and, when damaged, restorable — conservators work from the artist’s material notes, and in their absence they guess. And it makes work interpretable, because the meaning of a piece is frequently carried by conditions that vanish when the piece is de-installed: the height of the hang, the direction of the light, the sequence in which viewers encountered it.
Sol LeWitt understood this to an extreme. His wall drawings exist as certificates and diagrams — the instruction is the work, and the documentation is the object. Felix Gonzalez-Torres’s candy spills and stacks likewise depend on certificates specifying ideal weight, replenishment, and permissible installation. Both artists made the archive constitutive rather than supplementary, which is why their work can still be installed correctly decades after their deaths. Most artists are not making conceptual work of that kind, but the lesson transfers: whatever you do not write down, someone else will eventually invent.
The reverse case is instructive too. Eva Hesse’s latex and fiberglass pieces have deteriorated in ways that force conservators into continuous interpretive decisions about what she would have accepted — a debate made harder by the brevity of her working life and the incompleteness of her material notes. Ana Mendieta’s Siluetas survive almost entirely as photographs and Super 8 film she made herself; the ephemeral earth-body actions are gone, and her documentation is the practice as we now have it. In both cases the archive did not merely record the work. It set the terms on which the work could continue to exist.
What a working archive actually contains
Artists who do this well tend to converge on roughly the same structure, arrived at independently. It is worth stating plainly, because it is less elaborate than people fear.
A single object record per work. One row in a spreadsheet, one entry in a database, or one text file per piece. It contains: title, year, medium stated the way you want it printed, dimensions in both metric and imperial, edition information if relevant, and an inventory number you assign yourself. That last item matters more than it appears. A consistent identifier — 2024-018, say — lets you connect a photograph, an invoice, a shipping crate, and a loan agreement to the same object across a decade of scattered systems.
Photography at three levels. A clean documentation shot on neutral ground, shot straight on with accurate colour. Detail shots of surface, edge, and any signature or verso information. And installation views showing the work in relation to architecture and other pieces. The first sells the work, the second protects it, the third is what publications and curators actually want and what artists most often lack. Hire a photographer once a year and shoot everything at once if budget is the constraint; it is cheaper than commissioning individual shoots later and far cheaper than not having images.
A materials and process note. Two or three sentences written while the work is still on the table: what paint, what ground, what varnish, what adhesive, what supplier, what you did that was unusual. This is the document conservators pray for. It also serves you personally — the number of artists who cannot reproduce a surface they achieved four years ago is not small.
A provenance and movement log. Where the work went, to whom, on what date, for how much, through which gallery, under which agreement. Include loans and returns, not only sales. If a piece has been reframed, restretched, or repaired, that goes here as well.
Context documents. Exhibition invitations, wall texts, press mentions, jury feedback, studio-visit notes, correspondence about the work. Keep them attached to the object record rather than in a general folder, or they become unfindable at exactly the moment they would be useful.
A practical protocol
The failure mode is almost never ignorance of the above. It is that documentation is treated as an occasional project rather than a step in production — and occasional projects lose to deadlines every time. What works is attaching the record to a moment that already exists in your workflow.
The most robust habit I have seen artists keep is a completion ritual: nothing is considered finished until it has been photographed, measured, numbered, and entered. Twenty minutes at the end of a piece, on the same day, before the work is wrapped. Photograph against the same wall in the same light every time, so that four years of documentation is visually consistent and can be published together without looking assembled from strangers.
Second, a pre-shipping checkpoint. Before anything leaves the studio, the record is verified and a condition photograph is taken. This single habit resolves the majority of damage disputes, and it costs the time of one phone photo.
Third, a quarterly hour. Once every three months, sit down and reconcile: which works moved, which sold, which images are missing, which entries are incomplete. An hour every three months is four hours a year, which is roughly one-fiftieth of the time it takes to reconstruct a decade of records under pressure.
On tools: the sophistication of the software matters much less than people hope. A well-maintained spreadsheet plus a disciplined folder structure — INVENTORY / 2024 / 2024-018 / [images, notes, documents] — outperforms an expensive collection-management platform used inconsistently. Dedicated systems (Artwork Archive, Artlogic, Airtable configured by hand) earn their cost at the point where you have hundreds of works, multiple galleries, and edition tracking. Below that threshold, they are mostly a way of feeling organised. Whatever you choose, keep two copies in two physical locations plus one in the cloud, and export your data to a plain, non-proprietary format once a year. Platforms fail. CSV does not.
The archive as a form of authorship
There is a quieter argument for all of this, and it is the one that tends to persuade artists who find inventory management spiritually unpleasant.
Documentation is the last place where you retain narrative control. The account of your practice will eventually be written by other people — curators, critics, estate executors, algorithms — and they will write it from whatever material exists. If your archive is thin, they will work from the fragments that happen to survive: the pieces that sold, the shows that were reviewed, the images that circulated. Those fragments are not neutral. They over-represent whatever was commercially legible at the time and erase whatever was difficult, unfinished, or ahead of its market.
Artists who keep serious archives can intervene in that process. They can show the abandoned direction that turned out to be the origin of the mature work. They can date the idea. They can prove the sequence. On Kawara’s date paintings, with their newspaper clippings and journals, are an extreme instance of an artist authoring his own record so completely that no external account can displace it. Tehching Hsieh’s year-long performances survive as meticulously witnessed and notarised documentation because he understood that without it the work would be, in the most literal sense, unbelievable.
You do not have to build a monument. But it is worth recognising what you are actually deciding when you postpone the record. You are not choosing between organisation and freedom. You are choosing who gets to describe your work when you are not in the room — and, eventually, when you are not there at all.
The archive is the part of the practice that outlives the studio. It deserves the same seriousness as the surface of the painting.
