Can I cite the same source twice?
Yes. Cite it as often as the reader needs to know where something came from. It still takes one entry in the reference list, however many times it appears. The styles split on the second citation in a row. APA tells you not to repeat it, Leeds Harvard tells you to write it out again in full, and IEEE reuses the number. Find your style, then read your paragraph and ask who your reader thinks is talking.
The part that never changes
Before the styles start disagreeing, here is what all of them do the same way. A source you use gets one entry in the reference list, bibliography or works cited, no matter how many times it turns up in your text. Cite it once and you get one entry. Cite it fourteen times and you get one entry. The back of the paper is a list of works consulted, not a tally of mentions.
What repeats is the in-text citation, the marker in your sentence that points at that entry. That is the thing you are really asking about, and it is also the thing the styles handle differently, because each of them is solving the same problem in its own notation. The problem is that your reader has to be able to tell, at every sentence, whether they are reading your idea or somebody else's.
The University of Leeds states that principle better than most manuals do. Its Harvard guide says "The key principle of referencing is that the reader should understand which information came from another source and which is your own idea, so you should provide citations as often as is necessary to make this clear." Every rule that follows is an attempt to write that sentence down as a procedure.
Nothing you do in the text changes the back of the paper. A source repeated through a whole section is still one line in the reference list, and a marker counting your sources is counting entries, not brackets. So repeating a citation cannot pad your bibliography. What it can do is pad your word count, because in-text citations count towards it at almost every university, which is the subject of do quotes count in the word count.
One answer per style
Here is what each style does on the second and subsequent citation, quoted from the manuals and from university guides that publish the rule. Find your row and stop reading the others.
| Style | The second citation of the same source | In the list at the back |
|---|---|---|
| APA 7 | A narrative citation drops the year after the first one in a paragraph, so Rahman (2024) becomes Rahman. A parenthetical citation keeps the year every time. | One entry |
| MLA 9 | The page number alone, once it is clear which source you mean. The full citation returns if another source intervenes. | One entry |
| Chicago, notes | A shortened note each time, author plus short title plus page. Ibid only for the note immediately before, and only if that note holds one citation. | One entry |
| Chicago, author-date | A full parenthetical at the first reference, then page numbers alone for later references to different pages. | One entry |
| Harvard | Depends on whose Harvard. Leeds requires the full citation written out again. Robert Gordon lets you drop it across a run of sentences. No Harvard uses ibid. | One entry |
| IEEE | The same bracketed number as the first time, forever. Never ibid, never op. cit. | One entry, numbered by first appearance |
IEEE is the simplest and worth understanding even if it is not your style, because it makes the logic visible. The number is the source's name. The fourth source you cite is [4] on page two and still [4] on page nineteen, and Murdoch University's guide states the consequence bluntly: "When citing a source for a second or subsequent time, do not use ibid or op. cit.", because "In the text, repeat the earlier reference number." There is nothing to shorten. The citation was already as short as it gets.
Chicago's notes system is the one where ibid still appears, and its conditions are narrow. The Chicago Manual's own Q&A gives them as two tests: "It must always refer to the note immediately before it; and the note it refers to must have only one citation in it." Fail either and you write a shortened citation instead, which Chicago now prefers anyway. If your note three back is the one you want, ibid is not available to you.
The paragraph rule, where two styles contradict each other
This is the case people actually hit. You are paraphrasing one source across four sentences. Do you put the citation on all four, on the first, or on the last?
APA's answer is the first sentence and then stop. The Publication Manual's line, quoted by Royal Roads University, is to "cite the source in the first sentence in which it is relevant and do not repeat the citation in subsequent sentences as long as the source remains clear and unchanged". APA has a name for doing it anyway. Centennial College's guide quotes the manual calling overcitation "distracting and unnecessary" and gives the worked example of one source repeated in consecutive sentences.
There is a second APA rule stacked on top of it, and it is the one that produces the neat-looking paragraph. Narrative citations, the ones where the author's name is part of your sentence, drop the year after the first appearance. APA's page on the author-date system puts it as an instruction: "Do not repeat the year for narrative in-text citations the second and subsequent times they appear in a single paragraph." The rule restarts at the next paragraph, and it does not apply to parenthetical citations, which carry the year every time. There is one exception on the same page. If you are citing two works by the same author from different years, keep the date on every citation so the reader can tell them apart.
MLA reaches the same destination and goes further, across paragraph breaks. Asked whether the author's name must be repeated at the start of each new paragraph, the MLA Style Center answers "If you continue to cite the same source in subsequent paragraphs and no other source intervenes, you do not need to identify the source again unless ambiguity would result." The condition is doing the work there. Another source between the two citations resets it.
Then read Leeds Harvard, which tells students the opposite in almost the same words. "Leeds Harvard does not use ibid to refer to previously cited items. If you are citing the same item twice in a row (i.e. you do not cite any other items in the text between the two citations) you must write the full citation again." The case APA calls overcitation is the case Leeds calls the rule.
Harvard is not even settled with itself, which is what you would expect from a style with no central authority. Robert Gordon University's Harvard guide opens with "If citing a source throughout your work, give a full citation each time.", then immediately makes room for the paragraph case: "If discussing a source's content for several sentences in a row, you do not have to repeat citations." The one thing every Harvard guide agrees on is the Latin. Birmingham City University's answer is that ibid "is not used in the Harvard system, where works appear only once in the alphabetical list of references".
Two things follow. Your department's Harvard guide is the authority for your paper, not a general Harvard page and not this one. And where the guides disagree, both of them are trying to keep the reader oriented, so the tie-breaker is whether a reader who skipped your first sentence could still tell where the claim came from.
A different page of the same source
The other common case is one book carrying several points, each from a different page. All the styles that use page numbers handle this the same way, which is that the source is introduced once and the later citations are page numbers alone.
In MLA, Western Washington University's library lays out the three shapes. Concurrent references collapse into one citation at the end, "i.e., (Smith 23, 32)". Separated references keep the name once, so "If you break your citations up within the paragraph, you can cite the first, (Smith 23) and then later add (32) after the second citation." Or you name the book in your prose and let bare page numbers follow.
Chicago's author-date system does the same thing. The University of Queensland's guide gives both halves: "When the same page (or page range) in the same source is cited more than once in one paragraph, the parenthetical citation can be placed after the last reference or at the end of the paragraph", and "When referring to different pages in the same source, however, include a full parenthetical citation at the first reference; subsequent citations need only include page numbers."
IEEE puts the location inside the bracket instead, so the same source at different depths is [3, pp. 5-10] and then [3, p. 42]. The number never changes. In APA, the page number is required with every direct quotation, so a paragraph quoting one source three times carries three page numbers even though the author and year may have dropped away.
When the second citation is padding
So when is repeating a citation actually wrong? Three cases, and none of them is about frequency.
The source has not changed and the reader knows it. This is APA's overcitation, and it reads as nervousness. Four sentences of one argument with the same bracket at the end of each tells the reader nothing after the first one, and it interrupts the prose three times to do it. Put the citation where the source enters the paragraph, name the author once in your own sentence so the thread is visible, and let the paragraph run.
The claim did not need a source. Centennial College's guide puts the test as whether something is well enough known to stand on its own, noting that "some facts and ideas are so well-known that they are considered to be common knowledge and do not need to be cited", while arguments, foundational texts and anything contested always get one. A citation on "many organisations expanded remote work after 2020" is decoration. A citation on the size of the emissions saving is load-bearing.
One source is doing the work of a bibliography. This is the real version of the padding worry, and repeating the citation is the symptom rather than the disease. If the same name is at the end of every second sentence for three pages, the problem is that you have written a summary of one paper, and no citation pattern fixes that. What fixes it is a second and third source that agree, disagree or measure something else, which is the argument in how many sources a research paper needs.
Notice what is not on the list. Citing one source in your introduction, again in your methods and again in your discussion is not padding. It is a source that earned its place, and a marker reading a reference list can see the difference between a source you returned to and a source you touched once.
Same author, same year, different work
One question gets mistaken for this one often enough to be worth separating. If you cite two different works by the same author published in the same year, the author-date styles cannot tell them apart, so you give them letters.
The University of Wolverhampton states the rule and the ordering: "The first source with the same author and date combination that you mention in your assignment will get an 'a' added after the year, the second will get a 'b' added, the third a 'c' and so on." You get Miller (2006a) and Miller (2006b) in the text, and two entries at the back carrying the same letters. APA and Harvard both work this way.
The trap is applying it to your own question. Wolverhampton is explicit that the letters are for distinct publications, not for repeat appearances, so if you are citing one document several times, even from different chapters, there is no a and no b. It stays Miller (2006) throughout. The letters only exist to separate two works that would otherwise look identical.
If the two works are both yours, that is a third question again, with its own bracket and its own honesty problem, and how to cite your own previous paper covers it.
Read the paragraph as a reader, not as a rule
Every rule on this page is downstream of one test, and you can run it in thirty seconds. Read your paragraph as somebody who arrived at it cold, then ask, sentence by sentence, whether you can tell who is talking. If the answer is yes all the way through, you have cited enough. If there is a sentence where a reader could reasonably think the claim is yours, that sentence needs a citation, and it does not matter that you cited the same source two sentences earlier.
That is also why the styles disagree without either being wrong. Leeds asks for the full citation because a full citation can never be ambiguous. APA lets you drop it because a well-built paragraph already says whose argument it is. Both are chasing the same reader.
Where this quietly goes wrong is not the pattern of the brackets. It is a paragraph that carries one citation, four sentences and three claims, only two of which the source actually supports. The citation at the top makes the whole paragraph look sourced, and neither you nor your marker can see which sentence the source is really standing behind. That is the failure in I cited a paper I never read, and it survives every formatting rule here.
CiteOwl is built the other way round. Each claim in your draft holds its own source and the quote from that source sits beside the sentence, so a source used five times is five visible pieces of evidence rather than one bracket doing five jobs, and the reference list is assembled from what is actually cited. If you just want the entry itself, the citation generator is free and turns a DOI or a title into a finished reference in any of eight styles. Either way, what you want at the end is the same thing, a draft whose citations resolve and whose edits you reviewed before they went in.