it is important to note

How to spot it

The test is subtraction. Put your thumb over the opening clause, read what is left, and ask whether the sentence still stops you where it stopped you before.

"It is important to note" is the common form, and "it’s important to note" is the same phrase typed faster. The family around them is small enough to learn by ear. "It should be noted" is the formal cousin. "It is worth noting" sounds a little more casual. "It is crucial to understand" turns the volume up. Every one of them rates a claim before the claim has arrived. The rating is a promise about the sentence that follows, and the sentence that follows either keeps it or does not.

What subtraction catches is the gap between the promise and the goods. A writer who has found something genuinely odd can usually say it flat, because the oddity does the stopping by itself. A writer who has found something ordinary sometimes reaches for the wrapper instead, and the wrapper offers weight that the sentence underneath cannot pay for. Take the wrapper away and you hear which one you have. If the claim still lands, the announcement was never holding anything up. If the sentence goes slack without it, the announcement was the only thing standing there.

There is a second signal, and it costs nothing to check. Notice how often the habit shows up in the same passage. A page that rates nearly every claim it makes has quietly stopped ranking any of them, because importance is a comparison, and a writer who marks everything has kept nothing back to compare it against.

Why models overdo it

Plausibility is what picks the next words in generated prose, and this opening is about as plausible as English gets. It fits in front of almost anything. It commits to nothing. It is already grammatical before the writer has settled what the claim will be, which makes it a useful thing to say while the claim is still being assembled.

The second reason is less about arithmetic and more about diet. The prose these systems learned from is thick with judicial opinions, journal articles and technical standards, and the pause does real work there. In that writing the phrase usually sits in front of a distinction the argument genuinely turns on. What carries over into new prose is the shape of the move. What does not carry over is the judgment that decided when to make it, because that judgment was never on the page to begin with. It lived in a writer's head, as a private ruling about which of several true things mattered most here.

So the habit arrives without its governor. And unlike the claim it introduces, the rating itself can never be wrong in any way a reader could check. A sentence saying the migration finished can be tested against the logs. A sentence saying the next sentence matters can be tested against nothing at all, which is why it is so cheap to write and so easy to keep writing.

When it's fine

Two judges and a folklorist show the honest use: a pause that marks a turn nothing else on the page would mark.

Judge John E. Jones III, ruling in 2005 on whether a ninth-grade science disclaimer amounted to an official endorsement of religion, stops in the middle of his analysis to say that the reasonable, objective student the law asks about is not a real child in a real classroom but a hypothetical one, assembled by the court out of history and context. Everything that follows depends on that distinction. Nothing in the citations packed around it would have shown a reader where the hinge was. The pause is the hinge.

Justice Byron White is doing narrower work in 1991, marking where the dissenters in an earlier case stopped disagreeing with him. They had fought his reading of the law, but they never contested who got the benefit of it. Alfred Nutt, writing about Cinderella in 1893, uses the same move to point at the one detail in an old story his comparison is about to turn on. In all three the announcement is followed by something the reader could not have supplied alone, and that is the test. A pause you can justify by naming what changes after it has a job; one you cannot justify should go, and the claim can do its own standing.

Where it comes from

The gesture is older than the phrase and older than print. Medieval scribes wrote nota bene in the margin, note well, with nota as the imperative of the Latin notare, to mark. The abbreviation N.B. survived into English schoolbooks long enough that many of us met it before we met the words behind it. The modern opening does the same job with the margin folded into the sentence. It is a marginal note that lost its margin.

Linguists eventually gave the habit a shelf to sit on. George Lakoff's 1972 paper described hedges, language that adjusts how certain a claim is without changing what the claim says, and by the 1990s researchers such as Françoise Salager-Meyer were tracing that machinery through scientific writing, where a pause in front of a result is often genuine caution, guarding a finding against being read as more than it is. Ken Hyland later gathered these devices under metadiscourse, language that manages a reader's attention to the text rather than adding information to it, and separated the hedges from the attitude markers. A hedge says how sure the writer is. An attitude marker says how much the writer thinks the claim is worth. "It is important to note" is the second kind wearing the clothes of the first, which is part of why it reads as careful even when nothing careful is happening.

Editors noticed the drift long before anyone tried to measure it. Wikipedia's house style lists "it should be noted" under editorializing, on the grounds that a writer who calls a claim significant without saying who thinks so has slipped an opinion into a sentence that was supposed to stay neutral. That is a rule for one encyclopedia rather than a law of prose. But the reasoning behind it is the reasoning a careful writer applies privately anyway: if the claim is worth the pause, say why it is worth the pause, or say the claim plainly and let the reader feel the weight land.