Appendix 2                  Appendix 4     Appendix 3. A Few Words on the Rhythm of Poetry Who Is This Article For? 20% of the people drink 80% of the beer The Pareto Principle

•   Rhythm check of poems; •   Meter detection; •   Built-in dictionary:     English: 44,500 words     (109,300 word forms);     Ukrainian: 136,300 words     (1,959,400 word forms);     Russian: 137,800 words     (2,967,100 word forms), •   User and online dictionaries     (English, Ukrainian, Russian); •   Rhyme markup; •   Spell check; •   Number of syllables per line; •   Service for working     with poems from the Internet.

I wrote poetry in my youth, but I only took it up in earnest in this millennium. Around 2007, I became interested in the rhythm of poetry, and about six years later, I started making my mark by studying what I found on the internet.

It turned out that poets are people too. Despite the loftiness of the material they work with, social laws (for example, the Pareto Principle) also hold sway over them, and more than once. Eighty percent of poets are perpetually stunned by the quality of their poems and don't bother with their quality. Another 16% believe their poems are brilliant (or sent from above) and therefore uneditable. And only 4% agree to work on their poems. This article is addressed to them. It's not too long, doesn't claim to be exhaustive or academic, and merely outlines the simplest techniques for creating a picture (diagram) of rhythm—perhaps some will read to the end.

The text includes paragraphs that discuss how certain things are done using a computer program. If you're not interested, you can safely skip them (these paragraphs are indented to the right). The "Rhythm in Me" program facilitates the process, but is not required.

The following applies more to poems written in one of the East Slavic languages, whose poetry is oriented toward syllabo-tonic versification. Modern poems in English are closer to tonic versification and are less regular in rhyme and rhythm.

Notations and Terms

The main thing that distinguishes poetic speech from prose is rhythm. As in many other areas of human activity, the ancient Greeks have precedence in defining rhythm. What we call stressed syllables, they called long and denoted them with "—", while our unstressed syllables were short, denoted by something similar to "". These symbols can be found in academic publications.

According to East Slavic tradition, stress marks are placed above vowels. If you do this clearly and sweepingly in a handwritten draft, you get something like a slash ("/"). For unstressed syllables, the most mnemonic (easy to remember) is a dash ("-").

In print (and when published on some internet portals), the pair of symbols "!" and "-" is more convenient.

Some computerized individuals use the symbols "1" and "0".

As you can see, there are plenty of options. Personally, for drafts, I settled on the handwritten version ("/" / "-") for the following reasons: the academic version ("—" / "") is respectable, but cumbersome and oriented toward a different linguistic environment; In the numerical version ("1" / "0"), the zeros clutter the image too much; the ("!" / "-") version is good, and I sometimes use it when working on the computer, although I don't like using "!" on drafts.

The minimal structural unit of verse in syllabo-tonic versification, which I will focus on from now on, is the foot—a combination of a stressed syllable (ictus) and unstressed syllables. The names of the feet are also borrowed from the Greeks, for example:

and many others.

Poetic lines (verses) are made up of feet. People speak of meter (meter-scheme) and rhythm (rhythmic pattern) of the poetic line. Meter is a certain ideal, rhythmic pattern is what we have in practice. For example, for the lines

meter is iambic pentameter (-! -! -! -! -!). The actual rhythmic pattern fluctuates from line to line, but the poetic meter bears the same proud name – it would be strange if critical articles stated, "William Shakespeare tried to write 'Hamlet' in iambic pentameter." Moreover, the final foot of a poetic line can be extended by adding an ending (as in the example above) or truncated at the expense of unstressed syllables at the end of the foot (trochee, dactyl, amphibrach), but even after such changes, the final foot is considered to belong to the same meter.

It is currently officially considered that East Slavic versification has five meters: trochee, iambic, dactyl, amphibrach, and anapest, consisting of feet with one stressed syllable and one or two unstressed syllables. Other two- or three-syllable feet are considered deviations from the meter, and feet of four or more syllables fall under Occam's razor ("Entities should not be multiplied beyond necessity"). Indeed, such feet can be thought of as consisting of two- or three-syllable feet.

But poetry, like the linguistics that theoretically underpins it, is a democracy. In thermodynamics, one scientist said, "There's no such thing as caloric gas!" and others thought about it, agreed, and never mentioned caloric again (except perhaps in historical digressions). Poets write poetry and don't bother thinking about whether it's an iambic or a paeon №2 meter. Accordingly, critics are free to interpret poetry according to their own preferences. And let's not even mention experimenters—I've seen a perfectly meaningful text written in a proper 7-syllable meter!

In the poet's assistant program "Rhythm in Me", this issue is resolved by two settings: the maximum number of syllables per foot (by default, three) and the comparison method (by default, starting with disyllabic feet). For experimentation, you can increase the permissible foot size and (all other things being equal) favor longer feet. In any case, we obtain rhythmic patterns for each individual line, which add up to a rhythmic pattern for the piece as a whole (the latter can be equated to one of the meters, providing a benchmark for comparison).

Is It Possible Without a Computer Program?

Of course, you can figure out the rhythm of poetry without a computer! Since the ancient Greeks drew rhythmic pictures... We take a handwritten text and, as we hear it, place stress marks above the syllables. Or below them:

As a result, we see vertical lines at the ictus positions (also called "support syllables," "support beats," "support pillars," or "meridians"). And where the lines curve excessively, we check to see if this is a rhythm error or just the way the letters are laid out. For those who rely on their poetic ear, this check is more than sufficient.

Of course, these lines look straighter in the program, and the hint that the next step is to be corrected is noticeable,

but you need to have a computer at hand, and you can't do without additional clicks to clarify whether monosyllabic words are stressed or not... Moreover, the program will put the stress on the first syllable of the word "subject", whereas it is a verb and the stress should be on the second syllable.

However, those who tend to offer explanations other than "that's how I hear it" have to work harder. It seems simple: if a word contains vowels, at least one of them is stressed. Lord Beaverbrook is credited with the expression, "If a woman has a mouth..." But the lord, judging by the anecdotes, is a notorious vulgarian, and accents, like women, do not necessarily follow the proclaimed principle. The science of phonetics governs pronunciation rules: "in the flow of living oral speech, function words are, as a rule, unstressed." A phonetic word, or rhythmic group, forms an independent word with adjacent function words, which lose their own stress (clitics). "As a rule" is because sometimes even an independent word transfers stress to a function word.

In a phonetic word, the stressed vowel is pronounced clearly, while unstressed vowels are not (they are reduced). Reduction is one of the indicators of an unstressed vowel, although not an absolute one. In doubtful cases, it's best to check the stress in a dictionary – fortunately, the internet makes this much easier.

It's very frustrating when a poet is caught making a pronunciation error. True, some insist on so-called "poetic license" (irregular stress). In my opinion, this is simply a synonym for the author's laziness, not self-expression, and is forgiven only to classic writers. There is also word-creation or simply abstruseness, but these are chimeras of poetry. And then there are local dialects... However, everyone resolves questionable situations according to their own taste.

In the "Rhythm in Me" program, access to dictionaries (spelling, explanatory, synonyms) is simplified – through a menu called up by right-clicking on the word of interest. But online dictionaries have one drawback: they "go bad" or their interface changes... So, if any dictionary "falls off" or you suggest a better one, please email me.

Analyzing the Rhythmic Pattern

Once we've placed the stress marks, we already have a rhythmic pattern. The program does this mechanical work for us, although it sometimes needs some hints. For example, a poet knows exactly whether he's singing about "a 'present" or "to pre'sent", but the program won't always figure it out. But it will accurately count the number of syllables in 99.99% of cases.

It's good if all the stress marks are in the correct positions for the chosen meter—the poem is rhythmically perfect. But this rarely happens—somewhere, yes, there are distortions. It's necessary to determine how significant they are.

The first level, the most basic, is controlling the number of syllables in lines. Sometimes it turns out that some lines are longer than others (naturally, rhyming with different types of endings or deliberately using lines of different lengths are not considered errors). You have to "squeeze the water out." This is not a pleasant process, since you need to both maintain the meter and pay attention to the meaning. If there are many long lines, and they cannot be compressed without significantly losing the meaning, we solve the problem in a more pleasant way – as the advertising slogan says, "just add water." Adding syllables or words is always easier than removing them, and new artistic images emerge... But poems written in shorter lines, in my opinion, are more beautiful.

The second level is checking the rhythmic picture for meter. Those parts of the poem that do not conform to the meter are subject to intensive "rolling on the tongue."

The minimum rule, called the "prohibition of over-accentuation", states that the stress in a polysyllabic word and one of the ictus of the poetic line must match. If this is met, the space between the icts can be filled with anything—monosyllabic words (even if they create spondees) for iambs and trochaics, and mono- and disyllabic words for dactyls, amphibrachs, and anapests. If this requirement is met, there are no gross rhythmic irregularities.

Minor rhythmic irregularities are formed differently for each meter.

A general pattern has been noted: feet with adjacent stress (!!, !!-, -!!, !!!) are bad if the word stressed on the non-ict plays a more significant role in the sentence than the one stressed on the ict (creating competition).

For trochee (!-) and iamb (-!) meter, the spondee feet (!!) creates a roughness with competing adjacent stresses. A single pyrrhic feet (--) usually does not produce a rough rhythm. When pyrrhics are arranged identically in adjacent lines (following the pattern of one of the paeons), we speak of the art of working with pyrrhics. However, a single pyrrhic, unsupported by adjacent lines, does not enhance the verse (though it does not radically detract from it). Typically, the structure of pyrrhics in trochaic verse is similar to paeon #1 (!---) or #3 (--!-), and in iambic verse, to paeon #2 (-!--) or #4 (---!). That is, some ictus are more significant than others, forming so-called supporting meridians. The art of working with pyrrhics consists, at a minimum, of avoiding them falling within the supporting meridians, and, at a maximum, of arranging them evenly in adjacent lines.

There is no hard and fast rule regarding the specific pattern of pyrrhics' placement that is natural for a given meter. For an odd number of feet, the pattern can be determined by counting from the last meridian (it's always present and intact, as it contains rhyming endings). Beyond that, it's even/odd: the penultimate meridian is optional, the next is desirable, and so on. For an even number of feet, this approach will result in a pyrrhic in the first foot, which isn't always possible to achieve beautifully. Therefore, the first foot is made the supporting meridian, and the others are placed symmetrically relative to the center of the line. With varying numbers of feet per line, adjustments must be made.

The longer the line of a two-syllable feet, the more likely it is to crumble under its own weight, turning into "slightly rhythmic prose, raped by rhyme." And the more necessary the caesura becomes—a rhythmic pause in one of the feet, dividing the line into parts. A classical caesura is tied to a specific syllable, while a modern caesura is tied to the rhyme scheme chosen by the author for the first half-line. While the number of feet in the first half-line is fixed, a reference meridian is associated with the caesura. Sometimes there are multiple caesuras.

In trisyllabics, feet with two stressed syllables, one of which coincides with the ictus of the main meter, are usually acceptable.

For a dactyl (!--), an intercalated antibakkhia (!!-) feet is tolerated if adjacent stresses do not compete. However, the appearance of an amphimacric feet (!-!) practically does not disturb the flow of the dactyl.

Anapest (--!) tolerates a backhia (-!!) feet less well, but tolerates amphimacric (!-!) feet (it is not uncommon for poems written in anapest to have a completely amphimacric first feet). Moreover, I've encountered normally readable poems containing a mixture of dactyl, anapest, and amphimacron feet. While this approach is probably not worth relying on, versification isn't based solely on syllabo-tonicity...

The intolerance of bakhia in anapest is also questionable. For example, one 222-line poem written in anapest contains 18 bakhia feet, formed by two-syllable words whose role in sentences is less significant than those whose stressed syllables are in ictus positions. This doesn't seem to bother the author (although I personally would try to avoid such feet).

Amphibrach (-!-) is omnivorous—it accepts both the bacchia (-!!) and the antibacchia (!!-) foot, as long as the adjacent stress doesn't compete for the ictus.

The tribrach (---) – an unstressed foot – is not a very frequent (though not particularly welcome) guest in trisyllables.

Icts are equally important in trisyllables. The final meridian is always present and intact (with an equal number of feet per line). The first is more sensitive to the introduction of tribrachs, while the others are less so. Two tribrachs in a row are undesirable, as they "break" the rhythm. In the case of long lines, a caesura is also desirable.

The final foot of a poetic line deserves special mention. It deviates from the general rule and, if it doesn't fit the chosen meter, can
  1) be lengthened by unstressed syllables – for example, an iambic foot with a feminine (-!-) or dactylic (-!--) rhyme ending;
  2) be truncated to some extent; for example, a trochee (!-) foot is shortened to one (!) syllable, and a dactyl (!--) foot is shortened to one (!) or two (!-) syllables.

Sometimes poets use logaeds—verses consisting of lines with an ordered combination of feet of canonical meters. The described situations with the last foot of a poetic line are not considered logaed, although they are similar to it. If the position of the stressed syllable in the foot does not change, then the general name of the meter does not change.

Logaeds can be considered when the position of the stressed syllable within the foot changes. For example, in a verse written in anapest (--!), the final foot is either a full (-!-) or truncated (-!) amphibrach foot.

The actions performed on the final foot can alternate across lines, which, combined with the varying number of feet in the lines of poetry, makes the rhythm and rhyme pattern of the verse quite capricious. For example, the logaed of anapest and amphibrach mentioned in the last variant is quite tolerant of situations where the last foot "straightens" to anapest (--!), or where the amphibrach encompasses not only the last foot but also the penultimate one. This looks unsightly on the diagram and clearly has no connection to classical versification, but the ear doesn't particularly object, and critics either ignore these liberties, determining the poetic meter by the skeletal mass of the feet, or declare it a logaed (static or dynamic).

After checking the rhythm, we begin correcting questionable passages. Sometimes we can get away with a little effort – using synonyms; sometimes we have to rework adjacent lines as well. Pay attention to unstressed feet and instances of possible stress competition. For example, if an iambic (-!) contains a spondee (!!) or pyrrhic (--) foot, we should analyze the meaning of the words that form it (or simply "test it out"): is there a trochee (!-) hiding there? And if so, it's no longer a roughness, but a rhythmic breakdown... The prospect of such a check seems depressing, but with time, skill develops, and we no longer waste time on harmless spondees and pyrrhics.

Homographs – polysyllabic words that are spelled the same but have different stresses – pose a particular headache. We also have to contend with the phenomenon of "authorial deafness." You've written a poem, and you like it. But in one line, you used a word with the wrong stress. Of course, gross semantic errors ("`content" instead of "con`tent") are unlikely—they are weeded out at the writing stage. But pronunciation difficulties and common speech errors are not so easy to catch. Another problem is monosyllabic words perceived as unstressed (clitics).

"Stuck in one's own head" (author's deafness) and simple laziness sometimes push us to the simplest solutions. What, a foreign foot in the line, stress competition? But what emotions, imagery! And if you read quickly, you won't notice anything... The recipe works: pop singers veil the flaws of the text in precisely this way. And people sing songs with these flaws with pleasure... Whether to refine the poems or let them do as is is up to the author.

In creating the "Rhythm in Me" program, I tried to simplify the work of constructing rhythmic patterns as much as possible. Unfortunately, manual work is essential for those using it. This is due to both monosyllabic words, whose meaning changes depending on their role in the sentence, and polysyllabic words with stress variations (like "`content / con`tent"). Although the program takes care of 99% of the mechanical work, 1% remains for humans. Trust, but verify. For example, through online dictionaries, links to which are built into the program.

When the Computer Fails

In any case, one must deal with cases of adjacent stress, conflicting word meanings, and homographs. Even in matters such as syllable counting, there is a certain amount of uncertainty. Some dialects are characterized by cases of "swallowing" vowels, which reduces the number of syllables. Dictionaries won't help here—only a poetic ear. And, preferably, with verification by reading aloud.

The "Rhythm in Me" program program adopts a formal approach, whereby a word has as many syllables as vowel sounds in transcription. Exceptions include words consisting of several consonants ("psst"), single-letter abbreviations, and initials with a period. For the especially lazy, Appendix 4 describes how to delegate the work of reading text to a robot.

About Caesura

Rhythm in poetry manifests itself on many levels: horizontal rhythm in the line, vertical rhythm in the rhyme scheme, and macro rhythm in the stanzas. But sometimes poets (particularly to save on the second level) combine the first and third: they write several lines of poetry (usually two, but sometimes three or even four) into a single physical line. Poets are within their rights: they save lines, and readers have less stringent expectations for so-called internal rhymes than for those clearly visible at the ends of lines.

Critics, upon seeing a caesura (classical or modern), rant along the lines of "a samurai without a sword is like a samurai with a sword, only without a sword," and, out of deep respect for the author's rights, hesitate to say, "Let's divide the verse into poetic lines!" Besides respect, laziness also plays a role: just think of how much rewriting is required... And it's easy to screw up: some authors write and write with a caesura, and then insert a long word in its place...

The "Rhythm in Me" program offers two tools for working with caesura. In the gentle mode, you point to a syllable, the program draws a blue line across the entire verse, and you can see whether there's a caesura (you might have to correct some manually) or if it's just an illusion. Or you can take a more ruthless approach: after identifying a caesura, tell the program: "Cut!"

So, here's a tip: if you've constructed a rhythmic scheme for a poem (either using a program or manually) and find that the rhythm is clearly visible at the beginning of the lines, but a mess at the end, don't be too skeptical. It's possible that the poem contains a caesura (or even several). Divide the poem into lines by caesura—and your lips will stretch into a smile! Although I wouldn't do this manually.

 
 
© Serhii Zihulia (Tom d`Cat)
2017-2026

Appendix 2. Keyboard and Mouse

Appendix 4. Reading Poetry Using Computer Voices