A literature note that is worth keepingরাখার মতো পড়ার নোট

What a highlight is for

A highlight marks the sentence, not the meaning. It is fast on purpose: you see a claim, a number, a phrase you might quote, and you mark it without stopping to think, because stopping is exactly what a first read is meant to skip. A highlight remembers where a thing was said. It does not remember why you cared, and six months later that is precisely what you will have forgotten.

A paper you finish with forty highlights and no note is a paper you will have to read again. The highlights say the sentence existed; they do not say what to do with it. That decision, what a sentence means for your own question, is the note, and it is worth making while the paper is still open, because a second reading costs the same hour the first one did.

Seven things a literature note needs

1. The claim. What the paper says it is arguing or testing, in one sentence, in the authors' own terms.
2. The method. What kind of study this is: a randomised trial, a household survey, a regression on published accounts, a set of interviews. One line is enough.
3. The sample. Who or what was studied, how many, where, and when. A finding about six villages in one district travels differently from one about a national sample.
4. The finding. What was actually found, in plain words, kept separate from what the authors say it means.
5. The effect size. The number that says how much, with its unit attached. "Significant" is not a size.
6. The limitation. What the authors admit, or what you noticed reading it, that narrows how far the finding can travel.
7. How it bears on my question. One sentence joining the finding to the thing you are actually trying to answer.

Leave one out and the note quietly turns into something else. No claim, and you have a fact with no argument attached to it. No sample, and a finding from six farms reads the same as one from six thousand. No limitation, and the note is more certain than the paper it came from.

Source notes and permanent notes

The seven-field note above is a source note: it is filed under the paper and makes full sense only next to it. That is right for most of what you read. A note saying that a shock is not always independent of the treatment in an event study is worth keeping attached to the paper that raised the point.

Some notes are worth more than that. An idea you find yourself reaching for across three different chapters, a distinction you keep having to explain again, is worth rewriting as a permanent note: one idea, one sentence that states it without leaning on the paper's name, filed under the question or claim it answers rather than under the source it came from. A permanent note can then sit beside notes drawn from other papers that bear on the same question, which is the point of writing one: a source note is filed by where you found it, a permanent note by what it is for.

Why your own words outlive a quotation

A quotation is exact, and it is frozen. It says what the author said, in the order the author said it, and returning to it in a year tells you only that. It does not tell you why you kept it, what you thought it meant, or which of your own sentences it was meant to support. Writing the finding in your own words forces the decision a quotation defers: you have to decide, now, while the paper's argument is still in your head, what the sentence actually means for your question. That decision is the thing a note preserves and a quotation, on its own, does not.

Keep the quotation where the exact wording matters: a legal test, a definition, a phrase you expect to cite directly. Keep it inside the note, though, beside your own sentence about what it means, never in place of that sentence.

A highlighted sentence with a page number and no sentence of your own beside it is not a note. It is a promise to reread the paper, and the promise is rarely kept.

Worked example: a highlight becomes a note

Where it goes wrong

• Highlighting a whole paragraph and calling it read. A note is what turns a highlight into something you can use later; the highlight alone is not that.
• Copying the abstract's sentence as the finding. The abstract is the authors' claim; the finding is what the tables actually show, and the two do not always agree.
• Leaving out the effect size, so "significant" is the only number on the page, and significance says nothing about how large an effect is.
• Skipping the limitation because the paper is useful, which leaves the note more certain than the paper it is drawn from.
• Writing a permanent note that still opens with the paper's name, so it can never be read without going back to find out which paper that was.

• All seven fields are filled, in my own words, not copied from the abstract.
• The effect size carries its unit.
• A limitation is named, the authors' or my own.
• One sentence says how the finding bears on my question.
• A quotation, if kept, sits inside the note beside my own sentence, not in place of it.
• An idea I keep returning to has been rewritten as a permanent note, filed under the question it answers.

The reading room is where the highlight happens, anchored to the sentence itself; the notebook is where the note gets written afterwards, in your own words, and filed under the question it answers rather than under the paper it came from.