People by US city directory: what to know and why

Long before anyone built a database of who lived where, cities had one in print. The city directory was a commercial book, reissued at intervals, listing residents by name with an occupation and an address, and it is still one of the most useful sources for placing a person in a city in a given year. Library research guides to local history and genealogy treat it as a core source for exactly that reason.

It is also one of the most misread. A directory looks like a census and is nothing of the kind. Knowing how the book was made tells you exactly which of its silences mean something and which mean nothing at all, in the same way that knowing how a paper was compiled changes how you read a digitized newspaper run.

What to take away

• A directory entry is evidence that a canvasser recorded a name at an address, in the season before the book went to press. That is all it is, and it is worth a lot.
• Absence from a directory is weak evidence. Whole categories of resident were never meant to appear.
• The date on the spine is the publication date. The information inside it is older, and the gap is what causes most misdating.

How the book was actually made

Directories were compiled by canvass. People walked routes, knocked on doors, and wrote down what they were told by whoever answered. The sheets went back to a publisher who set them in type, sold advertising against them, and printed the result.

Every step of that leaves a mark on the page:

• The informant was not always the subject. A neighbor, a landlady or a child could supply a name and a trade, and the spelling is the canvasser's.
• The route was not exhaustive. Coverage followed streets the publisher considered worth the cost of walking.
• Advertising shaped the business pages. A prominent listing was frequently a paid listing, so size on the page reflects a purchase rather than importance.
• Reuse was normal. Where a canvass failed, the previous edition's entry was often carried forward, which manufactures a year of residence that nobody observed.

Many directories printed a preface describing their own method and their own claimed coverage. Read it. It is the single most valuable page in the book and almost nobody opens it.

Who is missing, and why that is not a finding

The pattern of omission is the part that most often gets turned into a false conclusion.

Directories were built around households and businesses, so people attached to someone else's household tended to be listed thinly or not at all: boarders, lodgers, live-in domestic workers, adult children, and in many editions wives and other adults recorded only through the head of household. People who moved between canvass and publication vanish. People with no fixed address never enter. Coverage of poorer districts and of communities the publisher had no commercial interest in was routinely thinner than coverage of the districts where the book sold.

So "not in the directory" supports almost nothing on its own. Presence is a strong positive signal. Absence is a prompt to check the next edition, the street section, and a different source class entirely.

The sections most people never use

A directory is several indexes of the same city, and the alphabetical name list is the least powerful of them.

Section	What it is for
Alphabetical list	Name to occupation and address
Street or reverse section	Address to occupants, which lets you read a whole block
Business or classified section	Trade to the people practicing it in that city
Institutional lists	Churches, schools, societies, public offices and their officers
Advertising pages	Which firms existed, and how they described themselves

The street section is the one that changes an investigation. Working an address rather than a name shows you households in sequence, catches a person listed under a spelling you would never have guessed, and puts neighbors on the page beside them. That last part matters more than it sounds: the strongest confirmations in this kind of research usually come from the people around the subject rather than the subject.

Reading an entry without over-reading it

Occupations are abbreviated hard, in a scheme that varied between publishers, and the abbreviation list is printed in the front matter. Expand it from that book, not from another city's book.

An occupation in a directory is what was reported that season. It is not a career, and the difference between a trade practiced and a trade claimed is exactly the gap covered on the profession page. An address is where the informant said the person could be found. In a period of high mobility, that is a snapshot with a short shelf life, and two consecutive editions showing two addresses is ordinary life rather than a puzzle.

The one inference a directory supports cleanly is presence: this name, this address, this canvass. Chain several editions together and you get a residence interval with real bounds, which is far more useful than a single point and is how the record layers of a city are meant to be stacked.

Finding and citing one

Holdings are scattered. Large public libraries, state and county historical societies, university special collections and digitization projects each hold different runs, and no single catalog covers all of them. Start with the public library serving the city itself, then its state library or archive, and ask specifically which years survive: runs are almost always incomplete, and knowing the gap stops you from reading a missing year as a missing person.

Cite the city, the exact title as printed, the year on the title page, the section, and the page. Then cite where you saw it, because a scanned copy and a paper copy can differ: pages are missing from some scans, and an index built by machine from a scan is a different finding aid from the one the publisher printed. The habit of walking a citation back to the object itself is the same one described for historical records.

Why this site publishes no directory of people

The obvious page to build from all this would be a list of notable people by city. We do not build it, for a reason that comes out of the sources above rather than from squeamishness.

A directory of that kind needs an inclusion rule, and every available rule is either arbitrary or unverifiable. It also needs a place value per person, and the state data page sets out why that value is a relation rather than an attribute. What you end up publishing is a list of who has been written about in an accessible way, presented as a list of who a city produced. For living people the problem is sharper still: we do not state anyone's birthplace or residence, and a directory page consists of nothing else.

Bottom line

Treat a city directory as what it is: a commercial snapshot, taken by a stranger at a door, months before the date on the cover. Read the preface, work the street section, chain the editions, and never turn a silence in it into a statement about a person.

Common questions

Is a printed directory better than an online index of it?

They answer different questions. The index finds candidates fast and misreads handwriting and worn type. The page shows you the neighbors, the abbreviations, the preface and the section structure. Use the index to get to the page, then read the page.

Can a directory establish where someone was born?

No. It records a residence at a canvass. Birth is a separate event with a separate custodian, and nothing in a directory speaks to it.