Getting started

Starting your family tree: what to gather first

Before you record a single name, gather what only your living relatives can tell you. A practical order of work for beginning a Muslim family tree.

Most family trees fail in the same way. Someone starts with an ancestor they are proud of — a scholar, a migration, a title — and works downward toward themselves. It feels like progress, and it produces a document that cannot be checked by anyone, including the person who made it.

Work in the other direction. Start with yourself, move outward to what living people can confirm, and only then reach for the generations nobody alive remembers. The early work is unglamorous and it is the part that determines whether anything you build later is worth keeping.

Talk to the oldest people first

This is the only genuinely urgent task in genealogy. Documents wait; people do not. An eighty-year-old relative holds names, nicknames, village names, marriage sequences and the reasons behind family ruptures that exist in no archive anywhere.

When you sit with them, record the conversation if they agree to it, and ask about specifics rather than generalities. “Tell me about our family” produces a speech. “What was your mother’s father called, and where was he buried?” produces data.

Questions that reliably yield more than they seem to:

• What was this person’s full name — including names they were called only at home, and any name they took after Hajj?
• Who else lived in the house? Siblings who died young are the most commonly lost people in any tree.
• Where is this person buried, and who arranged it?
• Which relatives did we stop speaking to, and roughly when? Estrangements explain gaps that otherwise look like missing records.
• Who else in the family is “the one who knows all this”? There is almost always such a person, and they are often not the eldest.

Separate what you know from what you were told

Keep three categories apart from the very first entry, because merging them later is almost impossible:

1. Documented — you have seen a record that states it.
2. Attested — a named living person told you, and you noted who.
3. Traditional — “the family has always said”, with no identifiable source.

The third category is not worthless. Oral tradition preserves real information, sometimes across centuries, and dismissing it would discard most of what any family knows about itself. But it has to be labelled, because the failure mode of family history is a traditional claim quietly graduating into a documented one after enough retellings.

Gather documents in order of fragility

Paper in a relative’s cupboard is at more risk than anything in a state archive. Photograph it first, sort it later. A phone photo of a document you did not have time to read is worth more than a careful plan to return next year.

Records that tend to exist

• Nikah namaنکاح نامہ — marriage contracts, often naming the bride’s and groom’s fathers, and the witnesses, which quietly gives you three or four people per document.
• Land and revenue records. In much of South Asia these are the deepest continuous record of ordinary families, frequently reaching further back than anything else available.
• Migration paperwork — passports, refugee registration, identity cards, ship manifests. Partition, the Gulf labour migrations and later movements to Europe and North America all generated documents that name parents.
• Graves. A headstone is a primary source, and photographing family graves is one of the highest-value afternoons in this whole undertaking.
• Endowment records. A waqfوقف deed names its founder and often the line intended to benefit, which is precisely a genealogical claim.

Get the names right

Muslim naming conventions carry more information than a Western given-name-plus-surname model can hold, and flattening them loses the part that is genealogically useful.

• Nasabنسب — the patronymic chain, linked by bin/ibn (son of) or bint (daughter of). This is the tree written as a name.
• Nisbaنسبة — an adjective of belonging: a place, a tribe, a school. al-Bukhari means from Bukhara.
• Laqabلقب — an honorific or descriptive epithet.
• Kunyaكنية — Abu or Umm plus a child’s name.

Record the full form as it was actually written or spoken, then note the modern simplified version separately. A relative who appears as “Muhammad Yusuf” on a passport may appear in a village record as a five-link nasab chain, and recognising them as the same person later depends on your having kept both.

Dates need their calendar attached

A year written without its calendar is a year you will misread. Hijri and Gregorian dates appear side by side across these records, and the gap between them is large enough that an unlabelled conversion can misplace a person by decades. Note the calendar with every date, and keep the original as written even after you convert it.

Then start entering it

Once you have living memory captured and documents photographed, building the tree itself is mechanical. On Shajrah you would start a private tree, enter yourself, and add people outward — attaching each document to the person it evidences as you go, so the reasoning stays visible rather than living in your head.

What makes a record trustworthy later is not how confidently it was entered but what is attached to it. That is the subject of the next guide: what counts as evidence in nasab.

Nothing you enter is public by default. Trees start private, and living relatives are never published even after you make one public — see how living relatives are protected.