3

Edit note: As you’ll see from the linked-to post, I’m not expecting my code here to be anything like 100% accurate. I’m after a fast and dirty heuristic that will be correct most of the time.


I’m using some formulaic rules for determining the number of syllables in an English word, then comparing those rules’ result against the result produced by some code I wrote. I’m using the “Written Method” rules that are described under Syllable Rules at the linked website. However, I’m sure the syllable rules for English are the same everywhere.

I’ve translated these rules to code and begun testing using my code and then checking with an online syllable counter to see if it returns the same result. For that I’m using this free online tool, Syllable Counter.

However, when it comes to the word every, I can’t seem to figure out why the online service is returning two syllables because when I use the rules, it’s clearly three syllables—at least to me at this point, but maybe I’m missing something.

Looking at the rules, every has two vowel-letters and also ends with the letter ‹y›, which sounds like a vowel, giving us a grand total of three syllables. There is no diphthong or triphthong in the word, nor any silent ‹e›, either.

So using those “written rules” in English, how would one come up with an answer of two syllables? I must be missing a subtraction somewhere, but this seems so simple.


Here are the three rules for counting syllable from the linked-to website. I’m using the third one.

1. The “Chin Method”

Put your hand under your chin. Say the word. How many times does your chin touch your hand? This is the number of syllables.

2. The “Clap Method”

Clapping may help you find syllables: Say the word. Clap each time you hear A, E, I, O, or U as a separate sound. The number of claps is the number of syllables.

3. The “Written Method”

  • Count the number of vowels (A, E, I, O, U) in the word.
  • Add 1 every time the letter ‘y’ makes the sound of a vowel (A, E, I, O, U).
  • Subtract 1 for each silent vowel (like the silent ‘e’ at the end of a word).
  • Subtract 1 for each diphthong or triphthong in the word.
  • Does the word end with “le” or “les?” Add 1 only if the letter before the “le” is a consonant.
  • The number you get is the number of syllables in your word.
20
  • 8
    Some people say "evry" which is two syllables. When I say "every" I think the middle syllable is barely audible, but perhaps it isn't audible at all to a listener. Few people say all three syllables cleanly and clearly. Commented Sep 24, 2022 at 21:54
  • 5
    @WeatherVane - Except when exasperated: “Ev-er-y single time...”
    – Jim
    Commented Sep 24, 2022 at 21:59
  • 11
    The "written method" syllable rules doesn't actually give the number of syllables people use when they pronounce the word (and this number can vary). Commented Sep 24, 2022 at 22:14
  • 8
    As @PeterShor points out, number of syllables does vary in real language. But in written language (using letters not sounds), there is no such thing as a "syllable" -- there is only an artificial rule about letter and hyphen placement. If you're programming it, use whatever rules you prefer -- there's no international, or even national, standard. Commented Sep 24, 2022 at 22:23
  • 9
    "However, I’m sure the syllable rules for English are the same everywhere." Were you intending to make a joke there?
    – Hot Licks
    Commented Sep 25, 2022 at 22:11

2 Answers 2

23

No fixed answers

Your problem with every is that many but hardly all words whose pronunciations end in [ɹi] or [ɹiz] have a variable number of syllables depending on how reduced the vowel sound immediately before their rhotic is, and sometimes by how reduced the vowel sound immediately after it is, too.

Every is just one of these. That’s why everybody can have both four and five syllables. It’s not a fixed number.

But even these words are not the only ones with this property. Any “written rule” that pretends to tell you how many syllables there are in actuary, adultery, alimentary, auxiliary, bakery, baptistery, battery, cemetery, coupe, extraordinary, fiery, film, laboratory, library, medicine, military, necessary, oratory, ornery, raspberry, secretary, Salisbury, territory needs to be consigned to the rubbish bin. It’s just spouting meaningless nonsense.

That’s because each of those words can be — and is — pronounced with a number of syllables that varies by region, speaker, and utterance. Many, many more such words exist. They have no single integer answer to their syllable count. It just is not possible.

Then there’s the inconvenient truth that countless word pairs like powers and hours, or higher and fire, are perfect rhymes for some people but not for others. Even the same speaker can choose how many syllables to use for the sake of poetic meter. A given word can have a variable syllable count from one line of verse to the next, even by the very same poet, when the meter demands this.

There is no such thing as a “written” syllable: syllables are part of pronunciation, not of writing.

It is therefore fundamentally impossible to count syllables merely by looking at the letters with which a word is written. You have to know what people actually say, and exactly what people say has no single fixed answer which is invariant across time and space. Writing is immaterial and irrelevant to actual speech.

6
  • 3
    Perhaps the closest thing to written syllable rules are the algorithms used for automated hyphenation. I think one of the original TeX papers described its method.
    – Barmar
    Commented Sep 25, 2022 at 20:32
  • 2
    Then one gets into issues as to whether "jewelry" and "jewellery" are the same word. The first is generally pronounced as two syllables, while the second spelling suggests a three syllable pronunciation. If the US and UK both spell and pronounce the word differently, are they the "same" word, or are they two "different" words that have a very close origin? Should an American spell the word "jewelry" even if they pronounce it with three syllables? What about a Briton pronouncing it with two? Commented Sep 26, 2022 at 4:29
  • 1
    @Acccumulation: in that particular example, I'd argue that they are just the US and UK spellings - i.e. the same word, with two valid spellings depending on where you are - and separate from how they're pronounced, which depends on your accent (among other factors). Commented Sep 26, 2022 at 10:10
  • 3
    @Barmar Hyphenation rules are often talked about like that but they don't work according to syllabic principles. Commented Sep 26, 2022 at 16:39
  • 1
    @Acccumulation That's an interesting example... by the OP's spelling-based heuristic, "jewelry" should have three syllables, while "jewellery" should have four. As an American who uses the spelling "jewelry," I think of the word as having three syllables, although the way I pronounce it in casual speech is probably closer to two.
    – DLosc
    Commented Sep 26, 2022 at 17:44
19

Short answer:

Because of a phonological process know as compression, every can be said with either 2 or 3 syllables. This happens to coincide with the "written method" described (see full post below) which comes up with exactly that answer. There's no way round the two answer problem. Writing, of course, doesn't consist of sound and therefore doesn't have syllables, and this is part of the issue.


Full answer:

Well, all of these rules are problematic. To give an example, the chin one won't work at all if a consonant at the beginning of a syllable requires the mouth to be more open than the following vowel. This is what happens in the second syllable of the word sorry. Your chin's only going to hit your hand for the first syllable there, not the second (try it!). However, setting those many various issues aside, let's have a look at the interesting problem that the Original Poster is having with the word every.

Special features of Every:

There's a reason why the Original Poster is having a problem with every, in particular. As mentioned in the comments by @Peter Shor and @John Lawler, pronunciation varies between individuals, and also within the speech of any given single individual. So a given person might say the word every with either two or three syllables depending on many different factors, for example the number of unstressed syllables following the word, how fast they're speaking, who they're speaking to and why they're speaking in the first place. The reason this is possible is that the word every is liable to a phonetic and phonological process known as ᴄᴏᴍᴘʀᴇssɪᴏɴ.

English syllables come in two parts. There's the Onset and the Rhyme. The rhyme is obligatory, but the onset is an optional feature. The onset is usually a consonant or group of consonants at the beginning of the syllable, before the vowel. The rhyme is everything else. So in the word cat /kæt/, the onset is /k/ and the rhyme is /æt/.

Now the rhyme is further split into two parts. It has the most sonorous musical and vowellish part, the Nucleus, which is obligatory. It can also have a Coda, which involves the syllable getting both quieter and less sonorant. Codas are usually consonants or consonant clusters. So in /kæt/ the nucleus is /æ/ and the coda is /t/. The syllable itself represents a peak in sonority. If it has an onset and a coda, then these will represent the dips in sonority at either end of the syllable.

Notice that I said that nucleuses are usually vowels. This is because we can, in special circumstances, have a syllabic consonant. This is when the nucleus of a syllable is a consonant. For example, in the word criticism /krɪtɪsɪzm/, the last syllable just consists of an /m/ sound. The consonants /r, l/ and the nasals /m, n, ŋ/ also frequently occur as syllabic consonants in English.

Back to every. The most expanded pronunciation of the word every is this:

  1. /'evəri/

I've used the standard UK transcription here, as used by John Wells in the Longman Pronunciation Dictionary. You will see that the second vowel there is a schwa /ə/, a mid-central vowel which only ever occurs in unstressed syllables.

If we agree with Wells that, other things being equal, a consonant will usually be syllabified with a stressed syllable when between two vowels, then the /v/ in every belongs in the first, stressed syllable. In the pronunciation of every shown above, the second syllable has no onset and consists of just a rhyme with a schwa as its nucleus.

In English when a schwa is followed by an /l/ or /r/ or a nasal such as /m, n, ŋ/ there is usually an alternative possible pronunciation, which has no schwa, but instead uses the following consonant as a nucleus, in other words as a syllabic consonant. When transcribing the word, this can be shown by using a small diacritic under the consonant in question. With the word every this would look like this:

  1. /'evr̩i/

This pronunciation of every, like the one shown in (1) has three syllables, but this time the nucleus of the second syllable is the consonant /r/. This consonant will have a longer duration here than it would if occurring in an onset or a coda.

For most speakers, however, it is possible for a further reduction to take place and for the /r/ there to become the onset of the following syllable. In this case the /r/ will be quieter and significantly shorter in duration. This 'process' results in the loss of a syllable, of course. The resulting pronunciation is shown below:

  1. /evri/

In the second and last syllable here the /r/ is the onset and /i/ the nucleus.

This same type of variation can be seen in (4, 5) below:

  1. happening: hæpənɪŋ --> hæpn̩ɪŋ --> hæpnɪŋ
  2. pedalling: pedəlɪŋ --> pədl̩ɪŋ --> pedlɪŋ

This is one type of compression. There are others.


The Original Poster's Question

The Original Poster asks how many syllables every has according to the "Written Method." The answer here is easy: 2 or 3. If the person applying the rule arbitrarily decides that the pronunciation is /'evəri/, then they will decide that the second < e > in every has a sound and is not a silent letter. In this case every has three syllables according to the given rules. However, if they decide the pronunciation is /'evri/, then the second < e > in every is silent and they will therefore need to subtract 1 to find the number of syllables, giving them the answer 2. The reality is that writing doesn't have syllables (apart from those assigned in the mind of the reader).

I can hear some people saying "OK, but which pronunciation is the right one, the one with two syllables or the one with three?" [Sigh. Whatever happened to free speech?!]

There is no correct pronunciation. The rules of English phonology allow for both. Native speakers use both. That's all there is to it. If you're looking to beat yourself with a riding crop for using the wrong one, go ahead, feel free. Make your choice and stick with it and administer your punishment accordingly. If you feel a deep need to apply your riding crop to others, however, please do ask first.

6
  • 3
    Right on. (Though I spell it rime, because Bolinger) Commented Sep 25, 2022 at 18:31
  • 1
    Papering over the weak vowel merger by using /ə/ everywhere is good and makes things easier to explain—although personally I’m losing faith with the oft-repeated mantra that schwa can “never” be stressed. Just one thing: I know you understand it, but many North American and international readers will be even more confused by the practice of writing phonemic /e/ for phonetic [ɛ] than they will by that of writing phonemic /r/ for phonetic [ɹ].
    – tchrist
    Commented Sep 25, 2022 at 18:57
  • 2
    @tchrist Yes, but you lot know that us lot are eccentric, though. I think they'll understand. Commented Sep 25, 2022 at 23:26
  • 2
    @JohnLawler: You might want to revise this sentence in your paper, in light of well, late enlightenment: "The residue of 4 words that belong to neither of these classes comprises pump, trump, compass, and crumpet." You might reconsider one of those as a candidate for the pejorative category, in both groups: "unpleasant people" and "unpleasant sounds". ;-)
    – Drew
    Commented Sep 26, 2022 at 15:35
  • 3
    @Araucaria-Nothereanymore Phonosemantics is like that. Just repeating a list of simplex words with the same assonance or rime tends to make the semantic neutrinos swarm. When I give talks about this stuff, I usually start off just by reciting some lists and watch the audience catch on. It's interesting stuff. Commented Sep 26, 2022 at 15:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.