Anybody has idea, how much copying is allowed in a recommendation letter and whether they really run a copying test on the letters of recommendation?

  • 6
    I would expect that if the top 10% of the professor's students applied to grad school then he would say "this student is in the top 10% of my students" in all of them...
    – user541686
    Dec 25, 2014 at 1:58
  • 1
    I am not even sure a letter of recommendation can be plagiarized, merely copied. Dec 25, 2014 at 18:23
  • 1
    I edited the title from enthusiastic student's version. S/he used the phrase "copyright" which isn't applicable. No one is disputing the copyright status of letters so it's a moot point. The question is whether letter writers should write letter de novo, whether they should recycle text, or whether they should 'borrow' text (from the student or from posted letters of recommendation).
    – RoboKaren
    Dec 26, 2014 at 19:28
  • 2
    Copying WHAT and FROM WHERE? Nobody expects you to retype the name of the applicant and the exact formulation of his theorems (or whatever) or, God forbid, try to be "inventive" about the way you present them. The letter of recommendation consists of two parts: the publicly available information about the candidate (where the closer you are to how it is presented in CV, publication list, etc., the better) and your own opinion about it (where any construct starting with "copying" will be an oxymoron by definition).
    – fedja
    Dec 27, 2014 at 17:13
  • 1
    As to "running a copying test on letters of recommendation", all I can say is that the set of people who lost all touch with common sense and reality after reading our written "academic honesty policies" in our class syllabi is certainly non-empty... In short, "No". In long: "What a novel and fascinating idea!"
    – fedja
    Dec 27, 2014 at 17:26

7 Answers 7


While our university occasionally submits statements of purpose, research statements, and writing samples through plagiarism detectors (and gives the admissions committee the results), I have never seen this done to letters of recommendation.

There are just very limited ways of saying nice things about students. I use the same variation of an opening for all my letters, with only minor changes in intensity (I am very/slightly/marginally pleased to write this letter of recommendation) and close with the same sentence on all of them.

I wouldn't worry.

Addendum given the changes to the the original question: I don't know if people have seen the output of plagiarism detectors but they don't provide a simple YES/NO answer. Instead, they provide a statistical quantification of how much unquoted text might be drawn from other sources (e.g., 3-5% of the text appears duplicative). Each report is several pages long with parts of the student text highlighted and possible prior sources highlighted.

The reports are quite "noisy" with many false positives -- notably in one case that I remember, highlighting text in the bibliography because the citations matched citations in other bibliographies. [facepalm]

Running LORs through plagiarism detectors would serve no useful evaluative purpose in terms of how the admissions committee views the candidate.

  • 23
    Oh man, I would be absolutely crushed if someone gave me a letter that began with, "I am marginally pleased to write this letter..." Dec 25, 2014 at 21:38
  • 2
    That's why you always ask your professors if they will write you a strong letter of recommendation.
    – RoboKaren
    Dec 26, 2014 at 19:43
  • 2
    @Asad: That is why I decline, politely, I hope, to write letters for students when I cannot write "recommend" or "strongly recommend." I say, "You should ask professors in whose classes you did better than you did in mine."
    – Bob Brown
    Dec 26, 2014 at 20:37

There isn't much creativity involved in most recommendation letters. And frankly, that doesn't matter. The important thing to a school or employer is that the individual who signed the letter believes it accurately describes you.

Yes, it would be nice if every letter was rewritten completely de novo. It isn't very surprising when it doesn't happen that way.

Don't worry about it. The worst that will happen is that they'll contact the recommender, he'll say you're a good candidate, and that'll be the end of it.

  • 7
    -1: It's important to me that recommendation letters are written by the ones who sign them. (This is my opinion as someone who has had a lot of involvement with graduate admissions over the years, at UGA. If you are involved in graduate admissions elsewhere and make a practice of not caring about this, I would be very interested to know.) On the other hand, I don't agree with "[I]t would be nice if every letter was rewritten completely de novo". This is really not required. But taking too much of someone else's writing and claiming it as your own is never okay in an academic environment. Dec 25, 2014 at 5:59
  • 6
    Form letter endorsements for PhD programs make for poor recommendation letters, of the sort that significantly hurt the student's chances. And yes, this is a kind of academic document. May I ask again what admissions experience you have? Dec 25, 2014 at 6:09
  • 5
    I'm not sure what you're asking that my answer doesn't already address. I already said that in this case the OP very probably has nothing to worry about. However, your answer makes a broader assertion about what schools find important. I am telling you, as a graduate admissions representative, that this is not the case. When you say you disagree, what do you mean? That in your experience PhD admissions are handled differently? If people followed your advice and, for instance, wrote their own letters, it would almost certainly hurt their applications. That is a real risk. Dec 25, 2014 at 8:28
  • 5
    You wrote: "The important thing to a school or employer is that the individual who signed the letter believes it accurately describes you." The use of the word "signed" implies that it is not important to those who process admissions who wrote the letter so long as the signer believes it is an accurate description of the candidate. Since I and the vast majority of other faculty doing graduate admissions I know vehemently disagree with that, I am commenting. If that is not your intended meaning, perhaps you should edit your answer. I ask once again: how do you know what we find important? Dec 25, 2014 at 8:37
  • 5
    @Jim B: I have explained elsewhere on this site why a student writing their own recommendation letter for a PhD program would be a very poor decision. I take issues of fairness very seriously, and the position that I am taking here is identical to that of the vast majority of my colleagues at American universities. What is your experience with PhD admissions that leads you to give such divergent advice? Dec 26, 2014 at 4:59

Anybody has any idea, how much plagiarism is allowed in a recommendation letter and whether they really run a plagiarism test on the letters of recommendation?

I have read literally thousands of academic recommendation letters, and I have never run one through plagiarism detection software. I don't understand why you did that after the letter was submitted: what were you hoping to gain?

You also sound slightly naive about how "plagiarism detection software" works. Such software does not directly detect plagiarism; it only makes a more or less compelling case for it according to the degree of similarity. The letter is a highly structured, ritualized writing style. Having 10% of a letter be the same as some other letter is not itself problematic, and it certainly need not imply "10% plagiarized".

A recommendation letter is not an academic paper, and most people who write many (graduate school) recommendation letters keep a basic template and fill in information accordingly. Therefore a lot of recommendation letters are going to be a lot more than 10% alike without there being anything problematic. This is another reason why putting a graduate recommendation letter through a plagiarism detector is not a very plausible thing to do.

In summary: I think it is overwhelmingly unlikely that anyone except you will know what percentage of this letter the plagiarism software reported as copied. I doubt you'll have a problem.

On the other hand, I am not quite willing to say that including an entire paragraph in a letter that one found on the internet is a good practice. It depends a lot on what the paragraph is. If the language in that paragraph is boiler-plate or standard, then I certainly don't care. (In particular, I hope you would have told us if the copied paragraph was the very last one, e.g. something like "In summary, Ms. X is a very strong candidate and deserves the highest consideration. Please do not hesitate to contact me if you require further information.") However, if the paragraph contains distinctive, poetic or especially clever language and your writer lifted it from the internet because he liked someone else's clever language better than his own: well, it has nothing to do with you, but I don't think is a good practice, and it doesn't fill me with confidence about his writing skills or professional ethics.


Almost every prof I know uses a template for recommendation letters, and adds a paragraph or two specific to the student. I'm not surprised a about the reuse of a paragraph and similar wording.

Remember, the purpose of a letter of recommendation is to make the case that you are a good student, and will be a dedicated, ethical researcher, not to introduce new research or new ideas. A letter of recommendation would likely not pass several requirements of publish, professional research as it does not present a novel idea, nor does it have external references.

A recommendation letter is a vote of confidence in the individual. Its difficult to see how similar content between letters would hurt either the recommend-er or the recommend-ee, so long as the recommendation accurately reflects the opinion of the recommend-er.

  • 3
    One person who could be harmed by a plagiarized recommendation letter is the person whose distinctive language and/or ideas the plagiarist stole. Dec 25, 2014 at 6:06
  • @PeteL.Clark is right - and the reputation of the plagiarist could also be severely harmed, I would think.
    – A E
    Dec 26, 2014 at 19:36
  • Come to think of it, it could also do harm to the person being recommended, depending on the amount of content that's copied.
    – A E
    Dec 26, 2014 at 20:07
  • While all the comments would be true about plagiarism, most recommendation letters (in my experience) don't meet other requirements of professional writing as they do not usually have any novel idea or new research presented. Its the opinion of a current researcher usually without references, which would be unpublishable anyway. Dec 27, 2014 at 16:41
  • The concept of plagiarism doesn't just apply to academic papers. The practice of copying others' distinctive language or ideas without attribution is viewed as unethical by (e.g.) academics in whatever context. As I (and others) have said, whether the duplicated paragraph is problematic depends on what it is. We have already said that this hunting expedition is poorly motivated, and the paragraph would have to be something truly artful and distinctive for it to be even worth shaking a finger at. But, for instance, lifting a whole paragraph from someone else's short story would be bad. Dec 27, 2014 at 18:15

This seems like a strange question to me, because if I were a new-ish prof asked to write a letter of recommendation, the very first thing I would do would be to type "letter of recommendation template" into Google (12.9 million hits), and use the first one that seemed to fit. Since I think I am not much lazier than the average new prof (and in STEM, have the advantage of being a native English speaker), I'd expect the majority of letters to be created the same way.


I am surprised that nobody has mentioned the possibility that the supposed "plagiarist" recommender used his/her own text, which was also coincidentally found online.

While a cut-and-pasted paragraph of 50 words strongly suggests copying of some kind, it is not necessarily the case that the 50 words were plagiarism (i.e. usage of another person's text without attribution).

It could very likely be the case that the recommender posted up one of his/her old recommendation letters online, and it got picked up as a template text by the anti-plagiarism software, which then detected the similarity. I would not at all find it surprising that my recommender(s) were reusing parts of their old recommendation letters, as they clearly have much better things to do with their time than write a completely new recommendation letter for each application.

In fact, it could even be the case that your recommender did not even post his/her letter online. Assuming your recommender has also forwarded their recommendation letter to his/her past students, it is not a stretch to consider that the student could have posted that letter online.

With the limited information, I find it rather uncalled for to immediately jump to the conclusion that your recommender has committed plagiarism.


Well, the title isn't quite asking the same question as the question body.

The direct answer to "Is there any copyright on letters of recommendation?" is that it depends on the country you're in (because copyright law differs from one country to another), but if you're in the UK or the USA then copyright automatically attaches to the creation of an original written work such as a letter. No need to register the copyright or declare it. Letters of recommendation are not exempt by virtue of their subject matter.

copyright in the United States automatically attaches upon the creation of an original work of authorship

WP: Copyright law of the United States

So the general answer is: Yes, letters (whatever the topic) do fall under copyright law.

"how much copying is allowed in a recommendation letter"?

Copying from oneself is fine (so long as one still owns the copyright in the work being copied). Copying from someone else is not, unless they've given you permission to do so.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .