By Gwyneth Box
English is by nature a rhythmic language: there is rhythm in all spoken English, whether poetry, prose or simple conversation.
Many of those who use language well have a natural ear for rhythmic patterns: they know how to make the language sing. If we consider the great speeches of Winston Churchill, or Martin Luther King, we can see that one of the things that raises simple public speaking to the level of oratory is the ability of the speaker to use rhythmic devices to reinforce meaning and control the emotions of the audience.
This, too, is what we as poets hope to do, whether we choose to use the traditional metrical forms or write in free verse. In order to do this better, it helps if we have an understanding of the nature of the language we are using.
Syllables and Word Stress
First we need to recognize the syllable as the basic unit of spoken speech. All multi-syllabic words in English have one primary stressed syllable when spoken in isolation. The adjectives content and happy are both two syllables, but their stress patterns are different: HAPpy is stressed on the first syllable, conTENT on the second. If we look at the noun CONtent, we see that it is distinguished from the adjective by a change in stress pattern. This change of stress according to grammatical function is quite common in English, and results in pairs of words like reCORD (verb) and REcord (noun), proDUCE (verb) and PROduce (noun).
Sometimes the stress on a word moves according to the context rather than meaning or grammatical function. Consider the word princess. Usually the stress falls on the second syllable, for example in the expression the little prinCESS, but when it is used as a title, e.g. PRINcess Alice the stress is at the beginning. How do you pronounce it in the phrase the princess was pretty? What about the pretty princess?
As well as stress patterns within individual words, we have stress at sentence level. Usually the words that are emphasized when we speak are those that contain new or important information. Consider how you would say I’m going to London in a neutral context. Now imagine that your deaf aunt keeps getting the wrong end of the stick. First she thinks your mother is making the journey, then that you went to London last week, then that you’re going to Paris. You will find that the same phrase will be pronounced differently in each case, with the emphasis on the words I, going, and London in turn.
Except when needed for this type of contrastive emphasis, grammar words (articles, prepositions, auxiliary verbs etc) are not usually stressed.
Much of the traditional analysis of poetical meter was developed for Latin verse. Latin is a syllable-timed language: although there are strong and weak syllables, each takes about the same length of time to say. Modern Latin-based languages also tend to be syllable-timed. English, on the other hand, is stress-timed: we manage to cram our unstressed syllables together, in between the stresses, in such a way that the stresses occur at more or less regular time intervals.
This means that the time taken to say a phrase in English does not depend on the number of syllables but on how many stressed syllables it contains. Consider the phrase the fat black cat. Obviously it has four syllables, but how many of them are stressed? In normal speech we would expect all the words which contain new or important information to be emphasized, which would probably mean all except the article, i.e. three stresses. Now consider the horrible cat. This phrase has five syllables, but we would only need to stress the first syllable of HORrible and the noun CAT. That is, two stressed syllables. The second phrase is longer in syllables, but shorter in time than the first.
Notice what happens to the articles in both of those phrases. They almost disappear. The vowel sound in the unstressed article is the same as that in the first syllable of the adjective conTENT. This sound is called a schwa. It’s the most common vowel sound in spoken English although it doesn’t correspond to any single written vowel, as can be seen from the following list of words in which it appears: pronounce, attempt, surprising (in the first syllable); emphasis, currently, poetry (middle syllable); picture, sister, station (final syllable).
By using the schwa sound in many unstressed syllables, we manage to cram them together in the spaces between stresses – the key to speaking a stress-timed language.
Of course, not everybody’s pronunciation is identical. Nor are words always pronounced the same in different contexts (remember princess). While we would all agree that offer has two syllables, the second of which is a schwa, offering is more problematic. Is there a schwa sound in the middle, or does the middle syllable disappear altogether? How many syllables does different have? According to the dictionary, both the two and three syllable pronunciations are acceptable.
And what about words like child and wire? How many syllables are they? Both are technically single syllable words, but many people pronounce them as two syllables. The vowel sound in child is a diphthong, whereas wire contains a triphthong. In either case it’s easy to slip a schwa in at the end. Purists may object to this “sloppy” pronunciation, but that doesn’t change the fact that it is common.
Why So Much Theory?
Much of the above is the sort of theory that you never read in a book on poetic techniques; it is far more likely to be found in a phonology textbook. However, poetry deals with the sounds of language and it is useful to understand this background.
One problem with poetry is that it tends to be written down with no guidelines to pronunciation. Gerard Manley Hopkins did attempt to use symbols to help the reader, and in older poems you may see apostrophes put in words like liv’d or heav’n to make it clear that these are pronounced as single syllables, but in general, when we pick up a book of poems, we have to work things out for ourselves. To do this, we must look at the words in the context of the poem.
Take as an example the phrase: cowboys and Indians. It’s easy to read it as six syllables DA-di-di DA-di-di, two dactylic feet. There is no doubt that Indians is a three-syllable word. But if you find it in the phrase cowboys, Indians, cops and robbers in a piece that is primarily trochaic (the trochee being a two syllable foot), it would be much more reasonable to read it as in-juns, giving four trochaic feet. This is one of the reasons it’s so important to read out loud every poem you write: by listening to the words we can often deduce which are naturally stressed and find where the rhythm falters.
Remember that the editors who consider your work are doing the same. I for one have sent poems for critique and found that my sonnets were rejected because they didn’t conform to the required 10 syllable count: either I count a word like inspired as three instead of two, or else I think oblivious can be read as three not four. There are, however, few editors who will reject an otherwise perfect sonnet simply because it includes an extra unstressed syllable. If you are unfortunate enough to find such an editor returns your poems, I suggest you keep calm, thank him for his time, and then resubmit elsewhere without changing the line in question. It’s his loss.
Unless we are being very pedantic this “natural rhythm” interpretation is probably enough. But there is one further issue which I’d like to consider briefly: that of poetry which is defined by syllable count.
There are various poetic forms in common use where accurate syllable count is essential. A haiku is defined as having 5-7-5 syllables, while a tanka has 5-7-5-7-7 syllables, and it seems unnecessary to argue with this. If your poem is 4-9-4, it may still be a wonderful poem, but why insist on calling it a haiku? If you write enough pieces using that form, you may choose to give it a name and try and encourage other poets to use it, too.
One recently invented syllabic form is the tetractys: it has 1-2-3-4-10 syllables (or 10-4-3-2-1) and is written as a centered text, resulting in a roughly triangular form. Double tetractyses also exist with either of these two patterns in combination.
Personally, I think trying to write syllabic poetry in English is inappropriate: as I’ve already explained, the language just isn’t syllabic, it’s rhythmic. My two-syllable word is your three syllables and vice versa. There is so much potential ambiguity that the exercise is close to impossible. However, if I do enter a competition that demands a strict haiku, I will avoid using words like hardening and different — the judge may pronounce them differently from me.
Gwyneth Box has been writing part-time for publication for several years: her credits include poetry and articles in small press and writing magazines in the UK as well as the Inscriptions and Inkspot websites. She has also won various awards for poetry and has had work published in a number of anthologies. She is particularly interested in the area of phonology as applied to poetry because of her training as a language teacher. She has been resident in Spain for over ten years and edits the patchword.com website for writers.