Grandpa died years ago, and the memories are scattering

• A list of everyone still living who knew him
• The shoebox (or attic box, or album) of photos and papers, if one survives
• A phone camera and about an hour per photo session

This playbook is for a quieter kind of urgency. Nobody is sick. There's no service to plan. But Grandpa has been gone for years, the people who knew him best are getting older, and you've noticed the stories getting shorter each Thanksgiving — details dropping off, versions blurring together. What you're sensing is real: family memory scatters on a schedule of its own, and reconstruction is genuinely time-sensitive work, just on a scale of years instead of weeks.

The good news: reconstruction is very doable, and it's some of the most satisfying archive work there is. You're not inventing anything. You're gathering what already exists — in people's heads and in a shoebox — before it finishes dispersing.

1. Inventory who remembers him

Start with a list, not with the archive. Write down everyone still living who knew him firsthand: his surviving siblings, his children, the nieces and nephews, the neighbor who's still on the same street, the men from his shop or his unit or his congregation if any can be found. Each of these people is a primary source, and each holds pieces nobody else has — his sister knows his childhood; his coworker knows a whole self the family never saw.

Then order the list by age and interview the eldest first. This is unsentimental and important: the ninety-one-year-old sister's memories are the least replaceable thing on your list. Everything else in this playbook can wait a season. She may not be able to.

2. The shoebox workflow

Almost every family has one: the shoebox, the album, the envelope of photos in the attic. Worked properly, it becomes the backbone of the archive. Here's the workflow:

1. Photograph every photo — a phone camera in good light is fine. Don't sort first; sorting is how shoebox projects stall. Capture everything, curate later.
2. Photograph the backs too. The backs are where the dead left you notes: names, dates, "Ellie and Ray, Lake Erie, '48." Half the identifications you'll ever get are written on the backs in someone's fountain pen.
3. Caption each photo with the person who knows — sit down with your mother or his sister, go through them together, and record what she says as she says it. "That's his first car, he was so proud of it, it broke down at my wedding" is a caption and a memory at once.

3. The half-remembered stories — keep every version

Every family has them: the stories everyone half-knows and nobody agrees on. Did he walk out of the interview or get thrown out? Was it a bear or a very large dog? Your instinct will be to establish the true version. Resist it — that way lies a Thanksgiving argument and, usually, a story that goes unrecorded because consensus never arrived.

Instead, record every version, from every teller, as alternate tellings kept side by side. The archive is built for exactly this: two family members remembering differently can both be kept, neither erased, each in the teller's own words. Decades from now, the disagreement itself will be part of the story — and honestly, it usually already is.

4. Decade-precision is honest history

Reconstruction runs on approximate dates, and that is fine. The archive accepts dates at exact, month, year, or decade precision — and "sometime in the 1950s" is not a placeholder to be embarrassed about; it's an honest record of what's actually known. False precision is worse than none: "June 12, 1954" invented to fill a box will mislead a great-grandchild someday, while "the early fifties" tells the truth.

5. Letters and papers that survive

If letters survive — from the war, from courtship, from the years away — photograph each page, then transcribe the text so it's readable and searchable (a photographed letter in faded pencil is safe; a transcribed one is alive). Add a note about context from whoever knows it: who he was writing to, what was happening, why this one was saved. The same goes for the other paper: the discharge papers, the recipe cards in his hand, the church program with his name in it.

6. Build the timeline last

It's tempting to start a reconstruction with the timeline — born, married, moved, died — because it feels like the scaffolding. Start with the stories and photos instead, and build the timeline last, from what surfaced. Interviews and photo backs will hand you dates and events you'd never have listed in advance (the year the house burned, the summer he drove a cab), and a timeline built afterward records the life you actually found rather than the outline you assumed.