Check any line yourself

0:23What a line has to give you

Start with what a line has to give you.

A statement about a recorded instrument is checkable only if it carries the things that identify the instrument.

The office it was recorded in. The identifier that office assigned, which in New York City is a City Register File Number, and which the ACRIS master table calls CRFN. The date it was recorded, which ACRIS names recorded date time. And the parcel it was recorded against, which is a borough, a block and a lot.

Four things. And a figure with no document behind it is just an assertion. You can't test it, so there's no reason to accept it.

1:02The two dates are not the same date

Now, here's the one that trips people. The two dates are not the same date.

ACRIS carries both, and they answer different questions.

One of them, document date, is the date on the face of the instrument. The day somebody signed it. Its companion, recorded date time, is described in the dataset as the legal date the instrument was recorded. And priority runs from the second one. Weeks can separate them.

So a claim that quotes a date is incomplete until it tells you which of the two it means. A great many of them never do. Anything about who beats whom is a statement about recorded date time, and nothing else will do.

1:44The amount field means one thing

The amount field means one thing.

ACRIS names it document amount, and describes it as the principal debt or obligation.

That's the face amount at recording. It isn't a balance, and it isn't updated. No field in the dataset holds a balance. So when a page reports the amount on a mortgage, that's this number, and you read it as the figure the instrument stated on the day it was stamped.

2:13How far the data actually runs

Then there's how far the data actually runs.

Both ACRIS tables carry a field called good through date, described as the date of the latest recording or correction included in the extract.

That one field settles a question people spend a lot of time arguing about. Anything recorded after it sits outside the copy of the data that was searched. That is a fact about your extract, and the register is fine.

And New York City's own guidance puts the other edge at 1966. Documents earlier than that live at the Land Records Research Library instead. A search of a system that starts in 1966 cannot tell you about 1962.

3:02The procedure, four steps

Okay. Here's the procedure. Four steps.

Take one line and spend five minutes on it. You're not trying to check everything. You're finding out whether checking is possible at all.

Start from the parcel rather than the name, because the land records are indexed by borough, block and lot, and the legals table holds exactly those fields. Then pull the document by its file number and read the parties and the amount off the instrument itself. A summary is somebody else's reading of it. Then read its recorded date, and put it against whatever it's being compared to. And last, look for later instruments against the same parcel naming the same party.

That fourth step is the one that changes answers. Because that's where a satisfaction or an assignment will be, and a claim that ignores them is counting a document that may already be retired.

4:02What the register will not settle

Now let me tell you what the register won't settle.

Three questions look checkable and aren't, and you want to know them before you spend an afternoon.

What's owed today, because no field holds it. Whether two people are married, which decides an entireties question under EPTL 6-2.2, and is recorded nowhere in the land records. And whether a satisfaction retired a particular mortgage, which RPL 321 leaves as an inference in any county that indexes by block, because the statute doesn't require the mortgage's own record to be marked.

Those are answerable. Just not from an index. They need the instruments, and sometimes facts from outside the register entirely.

4:51What this does not cover

And here's what I'm not covering.

One city's system, named because it's the one whose field names I can quote exactly.

Every county publishes differently. Broward's official records, the Florida property appraiser files, the Department of State's filings: each has its own identifiers, its own coverage dates, and its own idea of what a date means. What transfers is the procedure. The field names don't.

I'm not a lawyer and none of this is advice. But in this one case, what I can give you is enough. The four things a line must carry before it's worth believing. And how to get behind it when it does.

So take any line off anything I've published and go behind it. Everything I've written is built to survive that, and if a line doesn't survive it, I would rather hear that from you than have you trust it.