Someone has died: our father, mother, grandmother or grandfather. The choice 'grandmother' feels canonical, since the author dedicates the game to Elizabeth Walton Williams who I think is their grandmother. But as far as I know the choice makes no difference to the game.
The deceased was a famous painter, and as part of their will they have left us seven paintings -- though not seven specific paintings. It is up to us to choose seven pieces from among those that are left in their atelier. This is where play starts. We are given brief descriptions of paintings, often though not always accompanied by a sentence indicating its emotional mood. Then we choose whether to take it or leave it. If we take it, the painting triggers a memory of our interactions with the deceased. These memories are somewhat randomised, although they are chosen to fit the painting at least a little. Both the chosen painting and memories are saved for the player's perusal. Once seven paintings have been chosen, a non-interactive scene follows in which the player character finishes an unfinished painting, taking inspiration from the colours, subjects and moods of the chosen paintings. This final painting and the stages of its completion are shown as pictures in the game itself -- the first and only time that the game uses visuals.
The game is highly polished, both when it comes to details such as the Twine customisation and the music that plays in the background, and when it comes to what is most important, the writing. The paintings are diverse and described as well as it is possible to describe a painting in a few sentences. And the memories, which are the most important parts of the prose, are interesting, vivid, and well-written. Some feel slightly more generic -- a burst of anger after the child ruined a canvas -- but others manage to generate a real sense of individuality -- such as the one about the bakery.
Necessarily, a character portrait done in seven brief memories remains impressionistic; or, to use another metaphor from the history of painting, cubist, giving us a few snapshots of the deceased from different directions, which suggest but don't actually constitute depth. There's only so much you can do with memories that are brief, few in number, and narratively independent from each other. I would hesitate to call what results a 'character study'. We remain far too much at the surface for that.
The mechanics of the piece raise questions both mundane and philosophical. One soon finds out that the total number of paintings is not very large; the 'next' painting seems to be chosen entirely randomly, so you will start seeing paintings twice, or even see the same painting twice in a row, as if the protagonist is too confused to remember which paintings they've already looked at.
More importantly, there is something strange about first choosing whether to keep the painting, and then getting a memory. Shouldn't we choose based on what the painting reminds us of? Of course the game needs to work the way it actually does, because it would deflate the experience if we could first check out all the memories and then had to choose seven from among them. You want to have seven memories, no more, and no going back. Still, the current set-up doesn't leave the player with much agency. The main thing one can do to steer the story one way or another is choose paintings with 'positive' emotions or paintings with 'negative' emotions; this will definitely colour the memories one gets, and it may also colour -- literally -- the final painting we ourselves make.
Here the philosophical question pops up. There are paintings both negative and positive. If we choose the negative paintings, our memory portrait of the deceased will be emotionally negative; if we choose positive paintings, our memory portrait will be positive. But this is clearly and explicitly a selection effect. We can only end up with negativity by ignoring the positive, and we can only end up with positivity by ignoring the negative. Is Imprimatura trying to tell us that we can form our own relation to the past by choosing what we want to remember? That's an interesting vision, no doubt, but it would seem to require us to abandon the quest for truth and perhaps authenticity. Given the centrality of these ideas to how Imprimatura works, I would have liked it to engage with such questions more deeply.