Primary sources are original documents or objects from the time, person, or event being studied. Secondary sources explain, judge, or place those materials in context. That is the direct answer. It sounds simple, but the difference depends on the question being asked.
A diary can be a primary source for a study of daily life. A newspaper from the same year can be a primary source for public reporting. A letter can show what one person said, believed, or wanted another person to believe. A photograph, recording, map, law, interview, or set of raw data can also serve as a primary source.
The key point is that “primary” does not mean true. An original document may contain errors, bias, missing facts, or a narrow view. It is evidence from a person, group, or time under study. It still needs careful reading.
A secondary source comes later or comes from a different role. It may explain a historical event, review a novel, assess a study, or compare several documents. Common examples include scholarly books, journal articles, textbooks, book reviews, and critical essays.
I think the word “analyze” matters most here. A secondary source does not simply repeat a primary source. It makes a claim about that source. It may ask why a letter was written, how a law worked, or what a set of data shows. It may also compare several primary sources and offer a wider view.
The same item can change category
The source type alone does not settle the matter. The research question does.
A speech about a past event may be a secondary source for a study of that event. The speaker was not present, and the speech explains what happened later. Yet the same speech can be a primary source for a study of public memory. In that case, the object of study is the speech itself and the way it presents the past.
A memoir shows this problem clearly. It may be a primary source for the author’s later account of personal experience. It is not always a direct record of events. Memory changes. The author may leave out details, reshape a story, or write for a particular audience.
A newspaper report can work in the same way. For a study of an event, it may provide a report made at the time. For a study of the newspaper’s language, audience, or political position, it becomes evidence about the newspaper.
This is why fixed lists can mislead. A diary is often primary. A scholarly article is often secondary. But a document’s role depends on what the researcher is studying. Library guides from major research institutions make this same point: the distinction is contextual, not only based on format.
Why both kinds of sources matter
Primary sources give a researcher material to examine. They show words, images, records, objects, or data tied to a person, time, or act. They can reveal details that later writers omit.
Secondary sources help readers understand those materials. They can explain terms, identify patterns, point to related records, and show how other researchers have read the evidence. They also make claims that need checking.
A strong research process often uses both. A secondary source may help locate a useful letter or document. The primary source then allows a closer reading of the original evidence. The secondary source may offer a useful interpretation, but it does not replace the document itself.
Digital collections add one more concern: the copy is not always the original object. A scan may show a page, while a catalog record gives basic facts about it. Those facts are called metadata. Metadata is information about a source, such as its creator, date, title, and format.
Metadata helps users find and understand an item. It does not prove that every detail is correct. Dates may be uncertain. Names may have several forms. A collection may include only selected items from a larger archive. A search box may find words in typed text but miss words in handwriting or images.
That limit matters. A collection can hold primary sources while presenting them through modern descriptions. The document may be primary. The title, summary, subject terms, and transcription may be later work. Those parts help access the source, but they are not the same as the source.
A practical test
When I sort a source, I begin with one question: what is being studied?
If the study concerns a person’s letter, the letter is usually primary. If the study concerns a scholar’s explanation of that letter, the explanation is secondary. If the study concerns how later scholars have discussed the letter, both the original letter and the scholarly works become evidence, but they serve different roles.
A second question concerns creation. Was the item made by a participant, witness, creator, or researcher connected to the subject? Was it made during the period being studied, or later as a memory or report? These questions help, but they do not settle every case.
A third question concerns purpose. Who made the item? Who was meant to read or see it? What did the creator hope to achieve? A government report, personal letter, and public speech may describe the same event in very different ways.
I would also separate the source from the copy. A scanned letter is a digital version of a primary source. A transcription is a typed representation. An editor’s note is secondary material. These items may appear together on one page, but they should not be treated as one voice.
The honest limit is that no short rule can classify every source. Fields use the terms in different ways. In science, a report of original experiments may count as primary research. In literary study, a novel may be primary evidence for the study of literature. In history, a later memoir may be primary evidence for memory and secondary evidence for the events it recalls.
The distinction is useful because it tells us what kind of work a source is doing. It does not tell us whether the source is accurate, fair, complete, or easy to find. Those are separate questions.
A primary source gives original evidence. A secondary source gives analysis of evidence. The best reading keeps those roles clear while allowing for context. Once the research question changes, the role of the source may change too.
That is the practical value of The Source List: one digital source worth knowing, one search tip, and one honest limitation. For this question, the limitation is central. A label can guide research, but the question and the source’s context decide what the label means.