I’m pretty sure I intended to post on this Grant Barrett “On Language” column about “the claim that cellar door is beautiful to the ear” (which has always intrigued me) when it came out a few years ago, but it seems to have slipped through the cracks in my brain. At any rate, it’s been brought to my attention again by Geoff Nunberg’s recent Log post which starts off citing Barrett’s column and suggests that the popularity of the phrase “cellar door” may have come from the 1894 song “I Don’t Want to Play in Your Yard,” by Philip Wingate and Henry W. Petrie, with the lines “You’ll be sorry when you see me/ Sliding down my cellar door.” (You can hear a very slow performance by Peggy Lee here, and a much livelier one by Catherine Laidler-Lau here; Nunberg also mentions a song “Playmates,” but that’s a red herring — it didn’t come out until 1940.) The song and its catchy chorus were ubiquitous in the late 1890s and stayed popular for a long time; Nunberg has a parade of citations going from 1895 (“I would not let an operator that did not have a card, carry my lunch basket or slide down my cellar door”) to 1949 (“the blunt fact remained that he wouldn’t play ball in my back lot or slide down my cellar door”), and my mother, born in 1915, sang it frequently, so seeing it referred to, and hearing it on YouTube clips, gives me a tremendous rush of nostalgic pleasure. At any rate, the suggestion that it is the source of the “cellar door” motif seems plausible to me.
I recall from when I was very young being assigned the unpleasant task of going outside in the bitter cold to fetch some preserves (put up previously by my mother) from our cellar. It was a struggle to pull up and open the door. I’m sure there was a good reason for locating the cellar entrance outside.
You will have noticed the long essay in the Wikipedia: http://en.wikipedia.org/wiki/Cellar_door.
In the article I see that many people have enthused over “cellar door”, many people later researched the identity of those enthusiasts, and many people are still commenting on that research.
I would call this the dust bunny effect. Over time, a nothing turns into a something by an accretion of cognitive fluffiness: “hair, lint, dead skin, spider webs, dust, and sometimes light rubbish and debris, … held together by static electricity and felt-like entanglement”, as the WiPe puts it.
I’m not knocking it, I should add. I just feel that “dust bunny” is a more accurate description of what some people refer to as “meme”.
Although I didn’t know either this compound noun or its sound, I guessed what it means thanks to its “Latin stave”. Spanish “cillero” and French “cellier” came from “cellarium” as well.
Celadon and Amelia is the first thing that comes to my mind.
The second is how DeQuincey thought “Burke’s Works” was offensive to the ear. Probably less so if the r’s are not dropped.
The song and its catchy chorus were ubiquitous in the late 1890s
I wonder how that came about, given that this was before the days of even radio, much less You Tube.
From the NYT piece, quoting Mencken: “One is a series of words that are intrinsically musical . . .(t)he other is a series of ideas, false in themselves, that offer . . .” I suppose if Mencken chose to ignore subject-verb agreement, I have to forgive the NYT the identical sin, seen almost daily on its pages. Except that the Times isn’t consistent, and perhaps Mencken was.
As to cellar door: Inside the house a single door sufficed. The ones I recall outside a house seemed to come in pairs, as if to allow the passage of particularly large objects.
Paul, popular songs circulated pretty fast. Often you had the same people performing them all over in various vaudeville circuits, which usually intersected. You had musicians in lots of saloons and they needed material, and they were also singing whatever people requested. Another song like this was “Oh, Susanna!”, chart-topping hit for most of the 1850s in California.
This could be a long thread if we have to explain how a series can be singular even though it ends in s.
Indeed, the 1891 hit “After the Ball” was the first platinum recording, selling over five million copies — of the sheet music (a score is a “recording” of sorts, after all).
It plays a rather large role in Jake Gyllenhaal’s cult hit Donnie Darko, in which it’s called the most beautiful phrase in the English language.
I can see why “cellar door” may trigger all kinds of nostalgia, whimsey, etc, but for the life of me I can’t see why it would be “beautiful to the ear”.
Did you read the first link? That’s pretty much what it’s about. Perhaps you can see the attractiveness better if you respell it celador or selador.
LH: Yes I did, and two things struck me. “…it is notable that he used a template according to which the story often has been told since: a person of note — brainy, foreign or both — declares the sounds of cellar door to be exceptional, to the surprise of native but less discerning English speakers.” My well-known lack of discernment…
And: “Despite more than a century of elusive commentary on this topic, the only door we can identify with certainty is the open one through which those trying to investigate the matter have haplessly fallen. If you rely also on meaning, maybe closed cellar door is the more beautiful choice. ”
I see your point that the spellings celador or Selladore may make the sound look more attractive, but I’m afraid my hopelessly literalist mind just says that are not cellar door, which I see as what it is – I’m a lost cause.
As fior relyinbg on meaning, I’m wit Dorothy Parker on the virtues and beauty of “
Sorry, hit the wrong button.
As for relying on meaning, I’m with Dorothy Parker on the beauty of “check” and “enclosed”.
I’m afraid my hopelessly literalist mind just says that are not cellar door, which I see as what it is
But then you’re not talking about sound, you’re talking about meaning.
Paul: I also agree with Dorothy Parker. She was famous for literal-minded wisecracking, so if you’re literal-minded you’re halfway to fame. “The first thing I do in the morning is brush my teeth and sharpen my tongue.”
I’m not surprised you don’t understand why “cellar door”, or “celador”, is supposed to sound beeyutifull. I don’t either, and I bet Parker was being sarcastic about it. The banality of what “cellar door” means is intended, by contrast, to suggest how elegant one must be to adore the sound of it. The whole business is an evolutionarily successful dust bunny.
Fluff is the stuff of polite discourse.
LH: But then you’re not talking about sound, you’re talking about meaning.
I find it hard to divorce the two, though I can with Selador, as it has no specific meaning for me (other than a vague notion of celadon china – though I suppose it does sound rather like a south-east Asian town, lazing in the tropical heat. But I digress …).
And I suggest that if everyone re-reads the last graf of Barratt’s piece, he’s actually taking the mickey out of the beautiful sound idea, though of course I could be misreading the subtlety of it. (Jamesal has given up trying to stir my sensitivity 🙂 ).
It said: “Despite more than a century of elusive commentary on this topic, the only door we can identify with certainty is the open one through which those trying to investigate the matter have haplessly fallen. If you rely also on meaning, maybe closed cellar door is the more beautiful choice. ”
Stu: would there was a Dorothy Parker today …
PO: “One is a series of words that are intrinsically musical . . .(t)he other is a series of ideas, false in themselves, that offer . . .” I suppose if Mencken chose to ignore subject-verb agreement
There is no problem with Mencken’s agreement:
a series of words that are: “are” agrees with “words”
and similarly
a series of ideas, false in themselves, that offer : “offer” agrees with “ideas, false in themselves”. This could be rewritten as a series of ideas that are false in themselves but that offer …
In each case, it is not the “series” that is characterized as “musical” or “offering”, but the words, ideas, etc. A series of acts as a “quantifier”, which could be replaced by a lot of, many, a few, and more, including actual numbers.
Would you say, for instance: a number of people has told me or … have told me?
Quantifiers are tricky that way and have been studied quite intensively by linguists.
Appreciation of the beauty or not of word sounds can vary considerably from one language to another, and it is often greatly influenced by meaning and connotation.
Among the words cited as beautiful in the link above are mother and love To me as a French speaker these words do not sound beautiful at all! in particular, the stressed vowel (sound) they share I perceive as rather ugly, especially if it is lengthened, as if often is in the word love at the end of a sentence, in a sentimental context. The er of mother (which varies quite a lot between regions) does not sound very good either. Change the initial consonant so that the meaning is forgotten and you can get above and bother, which lack sentimental value.
I remember one of my French teachers pointing out that a word can sound beautiful while its meaning makes it hard to notice the beauty: her examples were diarrhée and hémorrhoïdes. The same could be said of their English equivalents, or at least the first one.
You can make a hash of anything sound-related by changing the sound: there is a sensitive dependency on the total effect. Tennyson’s line “the murmuring of innumerable bees” is excellent at conveying sonically what it refers to, and this is typical of him. But change two sounds, and you get “the murdering of innumerable beeves” (cattle), whose effect on the anglophone ear is quite different and shocking.
“the murdering of innumerable beeves”
I think that’s pretty beautiful. You shoulda been a poet..
Would you say, for instance: a number of people has told me or … have told me?
Quantifiers are tricky that way and have been studied quite intensively by linguists.
Curiously, the Associated Press Stylebook has not a word to say about this; Fowler (Gowers’ revision) devotes four pages to it. It is indeed a complex subject, far more so than I had thought. Fowler gives two germane and, he says, correct examples: “The number of people present was large” and “A large number of people were present.” Also: “In Before the conclave begins in a fortnight’s time a number of details has to be settled the singular is clearly wrong; it is the details that have to be settled not the number; a number of details is a composite subject equivalent to numerous details.”
I don’t recall coming across the term ‘composite subject’ so in pursuit of knowledge Googled it. This site takes a stab at addressing the subject, and also introduces ‘attraction,’ a related term that I’d never encountered in this context. Fowler gives it a short paragraph, whose first sentence I’ll reproduce here: “A tendency less commonly operative in English (except in mere blunders) than in Latin and Greek, by which a word is changed from the correct case, number, or person, to that of an adjacent word.” Following are a few examples, and then a reference to the entry about number (where he does not use the term).
Back to my original statement: I read a print edition of the NYT daily. The paper is unquestionably sloppy in its treatment of subject-verb agreement.
Greg: The phrase is due to I.A. Richards.
On number agreement: I don’t think there is any such thing as “attraction” or “composite subject”. Such formulations are from a bye-gone day before grammarians realized that sentences have structures built up from phrases which structures have their own properties. Instead, used to be it was all about words.
The number agreement of English verbs is with the number of the subject, and the subject is a phrase, not a word. In the simplest cases, the number of the subject phrase is determined by the number of the subject’s head noun, and one can simply look at the inflection of the head noun to tell how the verb must agree. But that’s misleading. The verb is not agreeing with that head noun, a single word, rather, it’s agreeing with the entire subject phrase.
Subjects don’t have to have explicit heads. What about the simple example: “A number were present.”? The noun “number” is not somehow attracting a plurality property from something in its vicinity, or the subject achieving some sort of compositeness. Instead, the verb just doesn’t agree with “number”. Why should it? “Number” is a word, and English finite verbs agree with their subject phrases, not words.
I don’t think there is any such thing as “attraction” or “composite subject”. Such formulations are from a bye-gone day
I’m not qualified to say, but maybe these terms indicate phenomena known today by different words. Gotta be careful not to confuse the map and the territory (pace Sam Hayakawa).
AHD’s usage note on ‘number:’ As a collective noun number may take either a singular or a plural verb. It takes a singular verb when it is preceded by the definite article the: The number of skilled workers is small. It takes a plural verb when preceded by the indefinite article a: A number of the workers are unskilled.
There’s a broad parallel here with ‘data.’ AHD’s usage note: The word data is the plural of Latin datum, “something given,” but it is not always treated as a plural noun in English. The plural usage is still common, as this headline from the New York Times attests: “Data Are Elusive on the Homeless.” Sometimes scientists think of data as plural, as in These data do not support the conclusions. But more often scientists and researchers think of data as a singular mass entity like information, and most people now follow this in general usage. Sixty percent of the Usage Panel accepts the use of data with a singular verb and pronoun in the sentence Once the data is in, we can begin to analyze it. A still larger number, 77 percent, accepts the sentence We have very little data on the efficacy of such programs, where the quantifier very little, which is not used with similar plural nouns such as facts and results, implies that data here is indeed singular.
Greg: You are right about the subject in a sentence being a phrase, not a word (although a phrase may consist of a single word, depending on the type of word).
“A number were present.”
A sentence like this is meaningless in isolation: a number of what? It presupposes that the entities understood in this incomplete sentence have already been identified, either in the linguistic context (what words were used earlier) or the concrete context. The plural form of the verb also shows that a number is a short form of the quantifier phrase a number of X’s.
With a singular verb, a number was present would imply that number referred to a single identifiable entity, such as “the number five”, which had a grammatical and semantic role in the sentence.
The similarity with data is only apparent, since data is not a quantifier but a plain noun. From the examples, it looks like data can be used either as a plural noun (the plural of datum0, or as a mass noun (like water or sand and many others, which agree with a singular verb when used with generic meaning).
As a collective noun number may take either a singular or a plural verb.
I see that I haven’t persuaded you that verbs don’t agree with nouns. Well, want to try your hand at: Some like it hot. The poor have always been with us. Jack and Jill are climbing up. Green are better than red.
Oh, and try this (adapted from an example cited by James McCawley): Why can you have either singular or plural agreement of the verb in the sentence: His wife and faithful companion of 40 years is/are giving him a birthday party.
“As a collective noun number may take either a singular or a plural verb.” I see that I haven’t persuaded you that verbs don’t agree with nouns.
Don’t shoot the messenger. The target of your persuasion attempt should be the AHD.
“A number were present.” A sentence like this is meaningless in isolation.
Fowler was being concise. Given the context I don’t think clarity is lost.
PO: It is true that the sentence as an example is clear, but if found in an ordinary text (oral or written) without a context which would indicate what a number was a number of it is not wholly understandable. This is what I mean by in isolation.
Greg: His wife and faithful companion of 40 years is/are giving him a birthday party.
I would use:
His wife and faithful companion of 40 years is giving him a birthday party. (Only one person is meant).
His wife and his faithful companion of 40 years are giving him a birthday party. (Two different people are meant).
On the other hand there can be instances where context allows the lack of a possessive or other determiner. What do native speakers think of:
She is a wife and mother
She is a wife and mother of three ???
*She is Bob’s wife and mother (impossible)
Bob came with his wife and mother (understood as ‘his mother’ ??)
Bob came with his wife and children (“wife and children” are a unit)
Bob came with his wife and his children (suggests that the children are not the wife’s)
I would interpret His wife and faithful companion of 40 years are giving him a birthday party as a speech error meaning that his wife and his companion are two different people and the speaker omitted the second his.
marie-lucie: Now I’m intrigued. Can you recommend a paper on the subject (as it applies to English)?
With respect to your examples (starting with “She is a wife and mother”): These don’t address the issue of subject-verb agreement. But they do bring up the use of possessive pronouns, and they made me realize that different languages must deal with this in different ways. I don’t think French, for example, would allow “Bob came with his wife and mother” without insertion of an additional possessive pronoun before mother. (To my endless shame, I have never been able to remember whether the French possessive pronoun takes its form from the owner’s gender or from that of the possession. Pour ce péché je vais pourrir en enfer pour plusieurs éternités.!)
Verb agreement in English is predominantly semantic, not grammatical, so a rule making reference to the inflection or categorization of nouns occurring somewhere in the subject phrase is not going to work. It’s the interpretation of the subject phrase that matters, not the form of nouns that occur to the left of the verb.
Of course She is his wife and mother works fine if “she” is Jocasta.
Romance possessive determiners agree in gender and number with the possessed; that is, with the noun phrase they govern, just like any other determiners. (In some languages, like Italian and Catalan, they aren’t actually determiners: the Italian for mon livre is il mio libro.) In Germanic languages, possessive determiners agree in gender with third person possessors only; however, in German all possessive determiners are inflected in accordance with the gender and number of the possessed using the strong adjective pattern.
Possessive pronouns are really fused determiners; that is, they incorporate their own head, like the English fused relative particle what in What you want is not important, meaning That which you want is not important. Consequently, possessive pronouns follow the same rules as possessive determiners. English is fairly unusual in distinguishing between determiners my, your, her, their and pronouns mine, yours, hers, theirs. This is mostly a relic that has become grammaticalized, but my/mine, thy/thine once worked like a/an, with the nasal lost before a consonant in the next word.
PO, I don’t know what to recommend, but Greg Lee alluded to something by James MacCawley. Perhaps he can suggest a source.
wife and companion
I see that I got off the subject-verb agreement topic into that of coordinated nouns or noun-phrases, which (depending on their place in the sentence) have a bearing on subject-verb agreement depending on how the phrase is to be understood.
Bob’s wife and mother were present (= … and his mother)
Bob’s wife and companion *was/were present (both)
In the second sentence, I agree that companion has to be a different person from wife, because there is no reason to add “companion” to reinforce “wife”. But in his wife and companion of 40 years … I would be inclined to think that this refers to the same person, a partner a 40-year marriage (although his wife of 40 years is more common). It seems to me that “companion” as a near-synonym of “wife” reinforces the connotation of mutual understanding, love and cooperation which are not always present in a marriage.
The meaning of wife and mother, father and mother, wife and children, etc is unambiguous because the members of the coordinated phrase are semantically incompatible, but in wife and companion the word companion could refer to a wife or to another person.
In French the possessives are not pronouns but determiners, so they work like articles, agreeing with the following noun:
ex. Paul a un livre : voici son livre.
Paul a une voiture : voici sa voiture.
In English you need to know the gender of the possessor, hence the difficulty of the agreement with epicene names:
Dana has a car : here is Dana’s/his/her (??) car.
James MacCawley appears to have published prolifically. One of his books is titled ‘Thirty Million Theories of Grammar,’ which would certainly keep the milliners here busy for a few days. ‘On identifying the remains of deceased clauses’ is another, as is ‘The nonexistence of syntactic categories.’ A festschrift was published in his honor, and another work labeled as authored by him is curiously titled ‘To the memory of James David McCawley : (30 March 1938 – 10 April 1999).’
Aha! One for the Hatter-in-Chief:
[Неэлементарные задачи в элементарном изложении.] Challenging mathematical problems with elementary solutions … Translated by James McCawley, Jr. Revised and edited by Basil Gordon.
by A M Yaglom, and YAGLOM (Isaak Moiseevich); Basil GORDON; James David MACCAWLEY; Isaak Moiseevich YAGLOM
Caveat lector: a) Though labeled as books, some of the items at the linked site seem to be articles. b) It’s hard to tell whether James McCawley and James MacCawley are different people or merely the result of sloppy copyediting.
Over to you, Greg Lee!
Reference: James D. McCawley, “The Syntactic Phenomena of English”, second edition, Chap. 11 on Anaphora, section f. “Choice among Pronouns; Morphological Indeterminacy”, page 372, has the example I was thinking of, which M. says is adapted from Quirk et al. 1972:
Her longtime companion and the future editor of her papers was happy to give his approval to her request.
Earlier in the section, page 371, McCawley says:
Part of the grammar of a language consists in a set of conventions for the resolution of conflicts between morphological number and semantic number (likewise between morphological gender and semantic gender, etc.) whenever a linguistic form needs to be put in a particular number, gender, or other form. (Besides choice of pronouns, this will also cover agreement of verbs with their subjects, of adjectives with the nouns they modify, etc. …)
Then M. mentions examples of the sort that we have discussed. And others.
Thanks for the reference, Greg Lee, though it may be a work too advanced for me. The information you provided led me to a review of the book in Springer’s Journal of Logic, Language and Information (03-2001, Volume 10, Issue 2, pp 263-266), partly accessible at no charge, and a Wiki entry about its author that says “in 1965 he received his doctorate for a dissertation under Noam Chomsky on The accentual system of modern standard Japanese.”
Back to my original remark about subject-verb agreement in the NYT: Does the following sentence pass grammatical muster?
“Many conservatives live inside an intellectual bubble of think tanks and captive media that is ultimately financed by a handful of megadonors.” (Paul Krugman, Wealth over work,” NYT, Page A21, New York edition, March 23, 2014)
PO: “Many conservatives live inside an intellectual bubble of think tanks and captive media that is ultimately financed by a handful of megadonors.”
It sounds all right to me, the intellectual bubble … is financed …. On the other hand, one could be literal and say that it is not the bubble that is financed but the think tanks and captive media, so … that are financed would be OK too. But the words ultimately financed must apply to the bubble as the end product of the direct financing of think tanks and media by megadonors.
If indeed they really do. Perhaps the donors fund foundations, which fund think tanks and media.
OK, there can be more than one intermediate stage between the donors and the bubble, but is ultimately financed shows that the bubble is understood as the end result.
From 1947 to 1960 Groucho Marx hosted a quiz show on American television called “You Bet Your Life.” Each week, before the quiz started, he would chat with the two contestants for a few minutes.
One time, probably during the second half of the 1950s, one of the contestants was a woman named [spelling?] Selladore [spelling?] Wintasia (the contestants’ names were never shown on the screen).
Naturally, he asked her about her given name and she responded that her father thought cellar door was the most beautiful sequence of words in the English language. The pronunciation I recall is /ˈsɛlədɔr wɪnˈteɪʒə/ and she seemed to be in her forties at the time. .
If that was in fact her given name and that is how she acquired it, her father would presumably be the first on record to hold that opinion about the words.
I’ve now looked for the program on Youtube but only one with Celladoor Crockett turned up (https://www.youtube.com/watch?v=koZnDyMd-UE), who is not the person I remember. Nor is the name Wintasia mentioned.
@David L. Gold:
Despite the fact that the Youtube title for the Apr 23, 1959 episode of You Bet Your Life spells “Celladoor” as you show it, and the pronunciations certainly leave the “r” hard to hear, according to the IMDB, it should be “Cellardoor”.
Also, at 1:18 in the show, Groucho asks about it, and she explains her name:
[Groucho talks to the other contestant for a bit]
Then he gets back to CC at 5:07:
@David L. Gold:
Well, dang. Maybe it was the IMDB that got it wrong (and thus my transcription is wrong as well). I thought I heard the “r” at least some of the times her name was said, though.
Celladoor Tulie Poinsetta Artensia Houston Crockett, 1928 ~ 2020
Note the “Artensia”. Could that have turned into “Wintasia” in your memory? Oddly, she never mentions her additional names that I could hear in the show (granted, I skimmed the show rather than watching), but perhaps that occurred on the $64,000 Question?
The photo of the gravestone was just added to findagrave a few weeks ago (14 May):
https://www.findagrave.com/memorial/226951585/celladoor-tulie_poinsetta-crockett
David L. Gold, May 27, 2021 at 9:25 pm: “If that was in fact her given name and that is how she acquired it, her father would presumably be the first on record to hold that opinion about the words.”
(second half of 1950s) minus (seemed to be in her forties) = 1910 at the earliest. But Grant Barrett’s column already cited “the most beautiful combination of English sounds was cellar-door” in a 1903 novel, so even if the story’s true, it wouldn’t be the first on record — though it *would* beat H.L. Mencken in 1920.
As for the Celladoor Crockett who was born in 1928 and named by her sister who was going to college, I wonder if the professor might have been David Allan Robertson of the University of Chicago, who’s on record in 1921 writing about the “assonance and cadence” of “Cellar door, oleomargarine; oleomargarine, cellar door.”
How remarkable that Celladoor Crockett lived until 2020, and what a shame that no one made that connection to Geoff Nunberg — I can imagine him interviewing her and posting about it on Language Log.
such a lasting family commitment to fantastic (and abundant) names: tulie! poinsetta! dawna joy! bonnie blue!
anthony had better have some good middle names, since he seems to be the last of the line.