Read Receipts.

Adrienne LaFrance has a post for the Atlantic that focuses on the term “read” in “read receipts” (a phrase I had been unfamiliar with; it means “the little notification that pops up for the sender of a text message once the recipient of that message has opened [and ostensibly read] the text”):

How do you pronounce the term? Do you say it in the past-tense, so it sounds like “red”? Or in the present tense, so it sounds like “reed”?

This was the subject of a brief but dizzying newsroom back-and-forth on Monday among colleagues who insisted that one or the other was definitely, absolutely, without question the right way. Our dialogue never reached the proportions of the Great Dark Chocolate Debate of last week, but we still never reached a consensus. (I asked folks on Facebook and Twitter for their opinions and received similarly passionate yet inconclusive responses.)

Team “red” had a compelling case: A read receipt is a receipt that’s generated once the text message has been read. Therefore, past tense. But there was solid logic on team “reed,” too: Just think of it like a “repair receipt,” or “pay stub,” or “mailing receipt,” none of which are in the past tense even though they indicate an activity that’s already taken place.

Apparently “the question has been floating around since at least 2010,” and having belatedly become aware of it, my curiosity has been aroused. Arguments from logic are utterly uninteresting to me, since language is not logical; the only relevant question is how the term was traditionally pronounced in the relevant speech community, before the n00bs came along and ruined everything. So if you are aware of such a community and such a tradition, by all means share your knowledge. (Thanks, Paul!)


  1. David Eddyshaw says

    I’ve always mentally said “reed”, and it would never have occurred to me to think “red” if I hadn’t seen this post. But I am fairly sure I’ve also never once actually referred to the damn things in actual speech.
    I agree that “red” is logical, once it’s been pointed out, but as Hat says, so what?

    During the years around Eternal September I lived without access to telephones, let alone the internet, so I have nothing of historical value to contribute. I saw a punched card once.

  2. At MIT in the late 1990s, it was definitely reed. I don’t know that it had ever occurred to me that another pronunciation was possible.

  3. Definitely “reed”. If someone said “red”, I’d assume that was the color of paper it was printed on.

  4. I’ve always mentally pronounced it “red” but never had occasion to test this in public. Will do so. Read/read is definitely the stupidest thing in written English.

  5. I have no idea what the traditional pronunciation is. I don’t remember using the term before, but I’d say “reed receipt.” I don’t understand what the argument is for “red”: a past participle would suggest to me that the meaning “a receipt that has been read”, which doesn’t convey the right meaning.

    This is completely off-topic, but I was taken aback by “ouroborosian,” an odd-looking adjective derived from a noun that I find vexing due to its unconventional correspondence with Greek ουροβορος (I prefer the alternate spelling uroboros, which follows the normal conventions for Latinizing ancient Greek, but regardless of the spelling, the fact that the stress is standardly placed on the penult, which was a light syllable in Greek, bothers me). “Ouroboric” or “uroboric” seems far better to me.

  6. I say reed. Although I’ve implemented such systems off-and-on since the 1970s, I claim no tradition. There wasn’t anything like that in the earliest days. ICST/CBOS 80-2 had “return receipt,” like the post office. X.400 had “receipt notification.” I suspect read came in with some desktop system in the 90s, like maybe Outlook.

  7. Eli Nelson, does “reed receipt” suggest a receipt that you imperatively must read? Why can’t the reed and the red equally well apply to the mail?

    I don’t think I’ve ever pronounced it or heard it pronounced, though.

  8. David Marjanović says

    Oh, so that’s what Lesebestätigung was translated from.

    ruined everything

    Fascinating. Turns out today is September 8468, 1993.

    Read/read is definitely the stupidest thing in written English.

    Preach it, brother.

    does “reed receipt” suggest a receipt that you imperatively must read?

    To me, that depends on the intonation and could very easily be clarified if English used more hyphens. (I’ll never get over the living room.)

  9. Michael Eochaidh says

    “Read receipt” implies that you’ve read something (generally an email).

    I’m in the reed camp.

  10. @tangent: “reed receipt” does not suggest an imperative to me because I can’t think of any cases where an imperative is used attributively. Neither of the examples LaFrance listed with the bare verb form, “repair receipt” or “pay stub,” have any imperative sense, as far as I know.

  11. I’m with Eli: “reed receipt” seems more logical to me, and is what I would intuitively guess.

    Also, how is it that read/read avoided the orthographic alternation seen in lead/led?

  12. @Lazar: I’d guess a combination of randomness, and desire to distinguish it from the adjective “red” (although from an etymological point of view, that word also might just as well be spelled “read”, so the choice of distinct spellings seems arbitrary).

  13. David Marjanović says

    Also, there’s the metal lead.

  14. Alon Lischinsky says

    I’m not really an old-timer (I first went online on September 590, 1993, and didn’t get my own email address until September 900 or so), but I have always interpreted it as /ɹiːd/.

  15. As MMcM implies, early standards / systems didn’t specify / couldn’t implement confirmation that somebody actually opened a message, not least because mail was often delivered in a file on the user’s system and nothing prevented them from just printing the mailbox on paper and taking it home to read.

    The Disposition-Notification-To header was proposed in RFC 2298 in March 1998 as a way to request a Message Disposition Notification in an interoperable manner. From the abstract:

    The purpose is to extend Internet Mail to support functionality often found in other messaging systems, such as X.400 and the proprietary “LAN-based” systems, and often referred to as “read receipts,” “acknowledgements,” or “receipt notifications.”

    So clearly the things had been implemented before that under the read receipt name. But even though I once hacked a X.29 driver into the BSD kernel I never used a X.400 compliant mail client, so I have nothing further to add.

  16. Trond Engen says


    Norw. (B) lesebekreftelse. Infinitive + noun. That’s how I parse ‘read receipt’ too.

    Read/read is definitely the stupidest thing in written English.

    I think I stated years ago that I would start writing read-red-red. And led for the metal. Didn’t happen.

    […] if English used more hyphens. (I’ll never get over the living room.)

    Or embraced its inherent germanicity and accepted compounds as compunds: readreceipt, livingroom, kitchensink, halloweenparty, thanksgivingturkey, christmastreeornaments, summerholidaycampingplans.

  17. How do you call a read receipt written on crimson papyrus?

  18. Trond Engen says


  19. And… about the chocolate thing. My problem with liking milk chocolate is that it could imply liking Hershey’s milk chocolate. Which is unacceptable.

  20. Hmm. Since the term was developed after/during Eternal September, I guess the Truth will never be revealed. I will try never to speak the phrase aloud; I can’t even tell what my native-speaker intuition is (an occupational hazard of the linguistics student).

    Read/read is definitely the stupidest thing in written English.

    Now, that’s something we can all agree on.

  21. I will try never to speak the phrase aloud

    IPA vowel chart tells me that between /i/ and /ɛ/ there is an /e/. You have to only figure out how to increase length just for a little bit. Symbol for that is /ˑ/. So you may try to say /ɹeˑd/ at see what happens…

  22. I am often sorely tempted to write “I redd a book yesterday” and “ledd poisoning”. It won’t catch on, though.

  23. Firesign Theatre has a pun somewhere about the “lead guitar,” presumably a heavy-metal version of the steel guitar.

    Regarding the living room:

    As for “ouroboros,” at least the English stress corresponds to the Greek acute in this kind of agent noun.

  24. reed receipts
    When teaching English, I once gave an exam that consisted entirely of short words in English with “ea” in them. It was a long list, and I’m sure not complete.
    I bother to participate here because I think I am in an extremely small minority of people who say “long-lyved” whenever they see “long lived”.
    I heard it, once, on the BBC, & a light shone in my darkened brain.
    Anyone else?

  25. I used to say it that way (having had a similar experience), but then I learned that that was an unwarranted correction of something that didn’t need correcting and went back to the usual pronunciation, which felt more comfortable to me anyway.

  26. I’ve always said “long-lyved,” though if it were really put into modern English it ought to be “long-lifed.”

  27. “I will try never to speak the phrase aloud.”
    Some individual words can be treated that way. I can’t find a copy through Google image search, but I’ve seen a pertinent cartoon: Two scientists in the African savannah, two small furry animals running past them, one of the men says to the other “There goes a mongoose, and there goes another one.” Substitute an aquarium for the savannah, “octopus” for “mongoose”, and it works just as well.

  28. Eli Nelson says


    but then I learned that that was an unwarranted correction of something that didn’t need correcting

    I’m curious: what makes you say “unwarranted”? As far as I know, long-/lıvd/ is indeed anomalous; so regularizing it to long-/laıvd/ seems to have some warrant.

  29. David Marjanović says

    Hershey’s has almost managed to produce disgusting chocolate. I hadn’t known this was physically possible.

    Interesting about long-lived; I used to think it was straight from the verb live… but then live/(a)live is pretty much another read/read situation.

  30. I’m curious: what makes you say “unwarranted”?

    The fact that /ˈˌlɔŋˈlɪvd/ is the only pronunciation in widespread use (the /laɪvd/ form was given as an alternative in the First Ed. OED but has vanished in the Third); “anomalous” is essentially meaningless in a language as refractory as English, and the main reason people use such a minority variant is to emphasize their own superiority to the hoi polloi (cf. the ostentatious use by such folks of “hoi polloi” without the article, when it’s got the article even in the earliest OED citations, from back when people not only knew Greek but wrote the phrase in Greek: 1668 Dryden Of Dramatick Poesie 65 “If by the people you understand the multitude, the οἱ πολλοὶ”; 1791 in C. Wordsworth Scholae Academicae 323 “Poor Quiz Carver is one of the οἱ πολλοί”; c1821–2 Byron in Lett. I. 633 “[We] put on masques, and went on the stage with the οἱ πολλοι”). As far as I’m concerned, when it comes to linguistic matters, vox populi vox dei.

  31. the main reason people use such a minority variant is to emphasize their own superiority to the hoi polloi

    I hope it’s obvious that I’m not talking about present company but about high-flown varieties of peever.

  32. David Eddyshaw says


  33. @David Marjanović: It’s become a truism that Europeans find Hershey’s milk chocolate to taste of vomit (due to its use of butyric acid, shared with parmesan cheese) – while Americans, though not usually in love with the taste, are usually at least tolerant of it. Worldly though I may like to be, I fall in the latter camp: I’ve never been able to detect the vomitous quality that I’ve heard so many non-Americans describe.

  34. Does that mean parmesan cheese also tastes of vomit? And what do they think of Hershey’s in Parma?

  35. It’s more the smell that used to remind me of vomit — took me straight back to my kindergarten days, little kids’ milky vomit — but either I’ve lost those olfactory cells or it was more a feature of factory packed, finely grated parmesan like we used to get. I was grating some today off a whole piece and didn’t even think about it.

    We can get Reese’s Peanut Butter Cups here, is that the same sort of chocolate as Hershey’s?

  36. Reading back further, I see long-lyfed as a spelling that somehow unambiguously indicates one of the possible pronunciations of long-lived. But I have no idea which of them it is, while the native speakers clearly do.

    How does that even work?

  37. I have had parmigiano reggiano that tasted unquestionably like vomit.

  38. Eli Nelson says

    I’m not sure why “y” as a vowel tends to be associated by native English speakers with /aɪ/ rather than /ɪ/. It might be because in non-Greek-derived words, y-as-vowel tends to only be used word-finally (e.g. “try”) or before word-final “e” (e.g. “dye”), both situations where /ɪ/ is not possible for most English speakers. Unstressed final “y” does represent another vowel, the “HAPPY” vowel that can be /ɪ/ and /iː/ depending on the speaker, but it somehow seems obvious that this value shouldn’t be used in “lyfed.”

    Or maybe it’s more that with “i”, there are many examples that can easily be brought to mind where it represents /ɪ/ before a single consonant letter and a silent e, such as “live,” “give,” “accusative,” “infinite,” whereas with “y”, there are no examples like this since there are so few words in general spelled with “y” as a vowel.

  39. David Marjanović says

    I’ve never been able to detect the vomitous quality

    Oh, me neither. From what I remember (my last encounter was 3 years ago and not memorable), it’s just bad chocolate, much worse than I thought was possible.

    But then, I don’t like cheese – any cheese, indeed any fermented milk products whatsoever. Butter is great, though (bad butter appears to be extinct, last sighted in Hungary some 20 years ago).

  40. So long-lyffed will make native speakers’ brains explode?

  41. “bad butter appears to be extinct”

    I rɛd, probably in this survey of English, that the comination “rancid butter” is standard, and thought to myself that I had no idea what “rancid butter” smelt or tasted like, and that a different example might be needed in future editions.

  42. Leave some butter out of the refrigerator for a few days. You can still eat it if you don’t mind the smell, and you can cook with it, which undoes the randicification. It’s the result of the oxidation of fats into short-chain fatty acids: refrigeration (or even freezing) slows it but does not stop it. Hard cheeses are highly rancidified.

  43. And after all these years I learn that rancor is the noun based on rancere ‘to stink’ (as rancidus is the adjective).

  44. Apparently not linked to Spanish “rencilla”, a quarrel, a squabble.

    If you boil animal fat in water and let it stand for several days, it becomes rancid, and the stink is basically identical to the stink of rotting flesh (subjectively worse for me than just the smell of rancid butter, I think because of the water). I don’t particularly recommend doing this intentionally!

  45. David Marjanović says

    rancor – rancid should be added to the classic table that I just learned of a few minutes ago.

  46. Instant classic 🙂 Geoff Pullum posted it on the Log yesterday, but comments there are closed as per usual.

    I’m trying to think if each column has sets that are only attested in that specific column. There are several examples for column 4 in the table already, risible should work for column 5.

    Correction: GKP is in the UK, so he posted it on Sunday.

  47. David Marjanović says

    Geoff Pullum posted it on the Log yesterday, but comments there are closed as per usual.

    He’s so scared in advance! Does all of the stuff he’s afraid of really happen on the other LLoggers’ threads, and they moderate them 24/7? If so, they’re really good at it…

  48. I think it’s because of a tradition, or an old charter or something. GKP has had one post on LL with open comments that I can recall.

    (I think there was a post once explaining why — not suffering fools gladly is clearly a part of it. But his posts on Lingua Franca do take comments, and he does answer).

  49. Alon Lischinsky says


    GKP has had one post on LL with open comments that I can recall.

    not really. He routinely used to leave comments open until early 2011. I think the post that marked the change was The One Where The Passive Is Explained.

  50. That’s five and a half years. I don’t remember anything but true trivia for that long, in fact I can be pleasantly surprised by finding posts on this blog that I wrote about a year ago.

  51. Me too!

  52. “Correction: GKP is in the UK, so he posted it on Sunday.” Is it now always November 2011? I’ve only just found out it was always September 1993.

  53. J.W. Brewer says

    Prof. Pullum used to routinely have quite lively comments threads on his LL posts. His change to a comments-generally-closed approach seems to have closely coincided in time with the terminal cancer diagnosis of his wife. He has never afaik linked these two developments, but the latter development is the sort of thing that can have lots of spillover effects in terms of what sorts of other sometimes-rewarding-yet-other-times-irksome things in your life you do/don’t still have the willingness to deal with.

  54. Indeed.

  55. I also had always assumed that his partner’s cancer and subsequent death had caused him to reevauate things, although he never actually said so explicitly. He did post about his general ambivalence toward continued posting here, shortly before he stopped allowing comments on most of his posts. Evidently, not generally having to deal with the comments was the best compromise he found he could deal with.

    He does occasionally still leave comments open, although sometimes he goes back and closes them after a few days. As I recall, the first time he left them open after his change in policy, it was by accident, and he showed up in the comment thread himself, to thank the commenters for providing such constructive responses.

  56. November 2011 Oops. But David linked to Gretchen McCulloch and she posted it last night. So when I saw November 6 on the LL post I jumped to a conclusion.

  57. Team reed here. I don’t think it’s a matter of logic, but more of grammaticality – where else is a past verb form used in a noun phrase like this? To say it like this forces me, at least, to add implicit quotation marks on “read” because the referent has to be meta.

  58. But what is “a noun phrase like this”? Can you think of any other examples of verb + noun in this rather odd conjunction? I’m not sure there are good parallels.

  59. > Can you think of any other examples of verb + noun in this rather odd conjunction?

    Well, “read” /ri:d/ is not always a verb; for example, consider the idiom “to get a read on [someone]”.

    And in computing circles — which would likely be where “read receipts” originated — it’s not uncommon to use “read” /ri:d/ and “write” as nouns, especially in noun + noun compounds; for example, a “database read” is when you read something from a database, a “read miss” is when the CPU cache doesn’t have the memory location that you wanted to read from, and so on. (And these are definitely /ri:d/ — I’ve heard many people say them — and they have direct analogues using “write” rather than “written”.) The person who invented the term “read receipt” may not even have realized that this use of “read” is a bit jargony.

  60. That makes sense, thanks. OK, I guess I’ll mentally say /ri:d/.

  61. /ri:d/ for sure.

Speak Your Mind