User:Silence/Rhythm

=Brainstorming= December 7: 30 minutes (20 presentation, 10 Q&A)

Don't dumb down. PowerPoint. Can bring handouts.

4-5 days after presentation: Group paper addressing shortcomings, 3-5 pages.

Thesis: Poetic rhythm affects cognition for physiological and not merely sociological reasons.


 * 1) What is rhythm? (defining features)
 * 2) What aspects of rhythm are universal? What aren't? (cross-cultural similarities and differences; baby-talk, lullabies)
 * 3) Why is rhythm innate? Why did it evolve--what is its survival/reproduction value?
 * 4) What is poetry?
 * 5) How does poetry relate to music, prose, discourse? What distinguishes poetry from them? Which is/are poetry based off of?
 * 6) In what ways is poetry rhythmic?
 * 7) How do these rhythms affect listeners? (somatic? mood? thoughts?)
 * 8) How do poets use and think about these effects? How does poem-writing/reciting change a poet's cognition?

Examples: Homer, Romanticism, haiku, etc.

Look at: Brain-scans of people reading poems? Close analysis of particular poems? Types of meter? Relationship of rhythm to rhyme (why does Brie hate non-rhyming iambic pentameter)?

Much of what we have to say here will be relevant to other areas, like music and language, as well. this is precisely because our argument is that poetry is based on universal and innate cognitive structures, and is a side-effect of these basic drives rather than a completely unique development.

Is rhythm essentially vocal (even in music), and other rhythms just imitate voices? Or...? What is root source, Form/exemplar which other rhythms mimic? Does voice itself just imitate body rhythms?

What is a poem?

 * A type of verbal communication, like prose, discourse.


 * Like prose, it:
 * Comes in discrete, thematically unified "packets"
 * Tells a story in sequence, or explores aspects of a thing
 * Is meaningful, linguistic, referential


 * Unlike prose, it:
 * Is divided into short "lines"
 * Makes extensive use of sound repetition
 * Regularizes some of this repetition into a sustained structure (rhyme, meter, etc.)

These 3 differences are necessary prerequisites for rhythm. (2) is the "repeat" aspect of rhythm, while (1) and (3) make it possible to employ this repetition with regularity, giving the poem structure.

Rhythm: REGULARIZED REPETITION of SOUND

Poetry is an offshoot of prose which exaggerates and formalizes two basic features of human speech:

1. chunking. because of our working memory's limitations, to follow a conversation we break discussions into smaller chunks of meaning, often separated by a small pause. Poetry increases the duration and regularity of the pause, producing 'lines.'

2. repetitive sounds. conversations quickly develop and maintain a pattern of regular stresses and beats; this pattern becomes fixed, in that any deviation from it is perceived as a change in the conversation flow as well (at least a digression, at most an outright break in the conversation). (moreover, the act of conversing often requires parallel sentence construction, a call-and-response in which new sentences, whether said by the same speaker as the previous sentence or not, echo the form and/or content of prior sentences. "Hello, my name is Bob" calls for a similar structure in response during a conversation.) Poetry exaggerates this regularity, creating parallels between (and within) lines through regular meter and rhyme.

We find both of these features basically soothing; we like small-sized chunking because it makes meaning more comprehensible and gives us time to think. We like regular repetition because it is also less taxing, being just 'more of the same'; the more consistent a poem's structure is, the more it allows us to focus on content and not waste as much energy on that structure.

However, to keep poems from just being mindless lullabies or dumbed-down prose, poets also often violate the expectations (framing effect) of readers, in small or large ways. In part because they have been 'simplified' in the above two ways, poems have a greater ability to experiment with unusual phrasing; whereas a prose paragraph would just become a jumbled mess if it deviated much from standard grammar, poetry's line divisions allow thoughts to be grouped together into smaller, easier-to-parse chunks which may therefore be more convoluted without losing the reader (cf. the ability of readers to read words as long as all the letters are present and the first and last letter are correct). This is because lines are not divided according to sentence syntax.

Comparison
5 Prose 4              Talk  Poem Semantic         3 Film  Comics Meaningfulness 2                         Music 1 Art                     Dance 1    2     3     4     5                              Rhythm

5 Music Talk 4      Poem          Film Auditory         3       Prose 2                    Comics Dance 1                          Art 1    2     3     4     5                              Visual

Poetry vs. prose

 * Poetry
 * In Xanadu did Kubla Khan
 * A stately pleasure-dome decree :
 * Where Alph, the sacred river, ran
 * Through caverns measureless to man
 * Down to a sunless sea.

Kubla Kahn did decree a stately pleasure-dome in Xanadu, where the sacred river Alph ran down to a sunless sea through measureless caverns.
 * Prose

The prose has all of the same words as the poem and imparts exactly the same information. Why is the prose simply uninteresting information while the poem is beloved by all Romantics? (Naturally, answer is rhythm + meter. Let's use this in our presentation.)

Meters

 * Anapest
 * 'twas the NIGHT | before CHRIST | mas and ALL | through the HOUSE
 * not a CREA | ture was STIR | ring not E | ven a MOUSE


 * Iamb
 * 'twas CHRIST | mas EVE | and ALL | through OUT | the HOUSE
 * no CREA | ture STIRRED | no MAN | nor TI | ny MOUSE


 * Spondee
 * IT WAS | CHRISTMAS | EVE AND | ALL THROUGH | THE HOUSE
 * NOTHING | WAS STIR | RING NOT | EVEN | A MOUSE

Heuristics

 * 1st Thesis: Poetic rhythm affects us for innate, not just cultural, reasons. (Poetry exploits heuristics, other cognitive structures.)
 * 2nd Thesis: Poetic rhythm affects us for non-semantic reasons. (Poetry exploits heuristics.)

Creator Bias: We make stuff. Use tools. Build. Design. Thus everything we encounter was similarly built (especially if complex).
 * Effect of general heuristic: When in doubt, assume agency.
 * Offshoot: The 'built for me' bias.

Biases/euristics that negatively impact understanding of poems, leading to misunderstanding or (esp. in post-modern poetry) failure to apprehend the intention of author.
 * 1. reader over-applies a heuristic which poet (accidentally or deliberately) is flouting
 * 2. reader fails to apply a heuristic which poet expects you to both know and employ


 * Affect heuristic
 * We all have different responses to particular language elements. While poets frequently utilize this to great effect, our affect heuristic operates when the poet may prefer it not to.


 * Peak-end heuristic
 * Again, frequently utilized by poets, but sometimes cases inaccurate cognitive representation of poem or message of poem. We may disregard the subtle points that convey the real message. For instance, common misinterpretation of "The Road Not Taken" as "The Road Less Traveled" by
 * "I took the one less traveled by
 * And that has made all the difference."
 * Interestingly the character in the poem makes the very same mistake. Plus rationalization


 * Availability heuristic
 * Keep us from comprehending elements of historical context/unfamiliar language/rhythms, placing more emphasis on familiar concepts/images/language/rhythms. "Wherefore art thou Romeo?" - So many high-schoolers assume "Wherefore" means "Where" and not "Why" because:
 * 'where' is more commonly heard than 'wherefore'
 * it is very common to ask "Where are you, Bob?", but not to ask "Why are you Bob?", so extraordinary contextual evidence is required to pick the stranger interpretation (which requires very close attention to and comprehension of the surrounding lines).
 * also: folk etymology, as 'wherefore' has the word 'where' in it


 * Outline:
 * Poets manipulate readers through heuristics and biases. They also employ heuristics in writing.
 * Heuristics and biases frequently interfere with communication between author and reader.
 * Hypothesize domain-specific heuristics.

=Cites=

Edwards: Sound, Sense, and Rhythm
Edwards, Mark. Sound, Sense, and Rhythm: Listening to Greek and Latin Poetry. Princeton and Oxford: Princeton University Press, 2002.

Herman Fränkel and Milman Parry, though unaware of each other's work, shared a common and largely original insight: that Homer's verse should be analyzed into larger units than dactyls and spondees, and his sentences should be analyzed into larger units than words.
 * p. 1

In 1926 Herman Fränkel published a monumental article in a German periodical demonstrating that the Homeric hexameter should be viewed not simply as a succession of six metrical feet, heavy and light syllables arranged as dactyls (DUM de de) or spondees (DUM DUM), but as a series of four (or occasionally just three) word groups whose syllables certainly add up to a hexameter, but which should be regarded as entities separated by word-end and often by a pause in sense. This means a typical hexameter might be DUM de de DUM (word break) de de DUM de (word break) de DUM de de (word break) DUM de de DUM DUM (verse-end).... Occasionally a verse occurs that illustrates this by presenting strong and obvious breaks between each of the units, as in alla soi,  o meg anaides,   ham espometh,   ophra su chaires (Iliad 1.158: "But you--you shameless thing--we followed you, to do you a favor"). It is highly significant that all the units are of different metrical shapes--a vital matter for avoiding monotony--and, of course, further variety is added both because the units in other verses will be different in shape from those given in the example, and because each group of two light syllables may be replaced by a single heavy one.
 * p. 2

In a series of articles beginning in the early 1970s and a book published in 1994, Wallace Chafe, working not from written texts but from tape-recorded spoken narrative and conversations, suggests that first in consciousness, then in speech, discourse appears in chunks, each containing only one new piece of information together with an orientation point to indicate where it fits.... These chunks Chafe calls "idea units"; and these in turn, for both physical and intellectual reasons, are verbalized in "spurts," and appear in speech as what he calls "intonation units," usually separated by pauses. Consciousness, says Chafe, both that of the speaker and that of a listener, can only focus on one idea at a time; and (again like the Functional Grammar people) he investigates carefully the extent to which information is new, or has already been shared between speaker and listener about the subject in question (and is therefore already active or semi-active in their minds), and the effects that this has upon the words used and the word-order sequence. ...
 * p. 10-12

Bakker.... pointed out that study of the grammar of spoken language was highly relevant to the orally delivered poems of Homer, and applied Chafe's insights to the Homeric text. Bakker proposed to "accept the ultimate linguistic and cognitive consequences of what it means to speak of oral poetry," and went on to "resort to modern linguistic research on orality outside the sphere of Homeric philology or even oral poetry." He noted that what I called in my 1966 article the "component units" of the sentence, which are not divided by the end of the verse, are in fact the "chunks," the "idea units" and "intonation units," of Chafe and the Functionalists. Bakker has placed a foundation of up-to-date theory beneath facts that had been observed previously but not properly explained.

Bakker shows that the intonation units of spoken discourse correspond to the metrical units of Homeric verse; the pauses or caesurae in the Homeric hexameter, identified in 1926 by Fränkel, are similar to the pauses that continually interrupt our spoken language--in his own words, "The segmentation of Homeric discourse, as evidenced by the length of the linguistic units of which it consists, can be seen as the manifestation in speech of the flow of the speaker's consciousness, each unit being the verbalization of a focus of consciousness. The length and duration of the units fits the acoustic short-term memory of the performer, or in other words, that ability to process linguistic expressions as wholes." The traditional Homeric formulas also appear in a new light, as "stylized intonation units." This enables us to see better why the pauses that mark off the units fall where they do, and to understand this not just as the poet's following of tradition and routine, but as the result of his desire to communicate orally with his audience in the most effective way possible. Bakker has, I think, made us understand for the first time what the well-worn expression "oral poetry" really signifies. ...

In contrast to the normal English prose order of "Agamemnon, lord of men, sacrificed a fat ox, five years old, to the almighty son of Kronos," Bakker's rendering of our autar ho boun hiereuse anas andron Agamemnon | piona pentaeteron hypermenei Kronioni follows the sequence of the word groups (or intonation units) of the Greek, and runs, "But he, he sacrificed a bull, ruler of men Agamemnon, fat, five years old, to the all-mighty son of Kronos." Adding a little more continuity and word-order emphasis, this might become, "But he sacrificed a bull, Lord Agamemnon did--a fat one, five years old, to the almighty son of Kronos." This is obviously conversational English, an exact parallel to "He's a nice chap, your brother is"--and much more vivid and vital than the usual prose translations: this is the way a bard would address his audience, the kind of rendering one could read aloud to students and hope to keep them awake.

Agamemnon is already in our minds (i.e., he is the Topic), since he has been speaking in the assembly, and so ho, "he," serves to refer to him and stands first. The bull is the Focus, because this is the new, significant information--the ordinary troops would have poured libations or sacrificed smaller creatures--so boun falls in the second place in Greek (and at the end of the main clause in English). The verb, "sacrificed," again is already in our minds, because that's what all the troops have been doing, so it falls in the unemphatic place--at the end of the clause in Greek, in the middle in English. Finally, the conventional name-epithet formula anas andron Agamemnon, "Lord Agamemnon," is what in Functional Grammar is called the "Tail," like "your brother" in the English equivalent. Then, now that we have the facts straight, come the details, the qualities--and hence the great importance--of the offering to the supreme god. The matter is properly dignified, epic, and religious; but the style is conversational, or oral.

Homer sometimes evokes a special effect by using more spondees or dactyls (heavy or light syllables) than usual. Well-known examples are the six bounding dactyls (- u u) describing Sisyphyus' rock rolling back upon him, autis epeita pedonde kulindeto laas anaides (Odyssey 11.598), and the six slow, dragging heavy spondees (- -) as Achilles calls on the shade of his beloved Patroclus, psuchen kikleskon Patrokleos deiloio (Iliad 23.221). Similar rhythmic variations occur in lyric meters, too, and are much less commonly observed and appreciated.
 * p. 62

Schleidt & Kien: Segmentation in Behavior
http://www.springerlink.com/content/n354122hm72n44g3/fulltext.pdf

Temporal segmentation may be related to the functioning of short-term memory. Segmentation may thus be a central feature of neuronal integration. Segment length was hitherto thought to be determined by either capacity constraints or temporal factors. Instead we show that segment length depends on the interplay between capacity and temporal factors. ...
 * p. 77

Segmentation in behavior is the ordering of movements into action units of discrete length within a narrow time range. Segmentation has been found in the actions of six mammalian species (Gerstner and Goldberg 1994), two primate species (Kien et al. 1991; Ottet al. 1994), and in human action.

Our measurements reveal that human action units are organized within a narrow and well-definable time span of only a few seconds. Though varying from 0.3 seconds up to 12 seconds or more, most of the action units fall within the range of 1-4 seconds. ... Action units with only one repetition of a set of movements, for example, two scratches, show no difference in duration from those containing no repetition of the set of movements. These action units, categorized as nonrepetitive, were significantly shorter than those containing two or more repetitions of a set of movements or repetitive action units. ... In studies of child development, Thelen (1981a) has similarly classed as repetitive behavior only those cases in which movement patterns were made three or more times; she regarded this behavior as "rhythmic stereotypies."
 * p. 80

Even children born blind and deaf (Medicus et al. 1994) structure their behavior in segments of a few seconds. In contrast to nonhandicapped children, these children show more repetitive behavior and their action units are shorter.

Thus, segmentation in the range of a few seconds appears to be a universal phenomenon in human action. We find it in various cultures all over the world, not only in conscious and intentional acts like work activities, actions of communication like greeting behavior, and ritualized behavior such as dances, but also in less conscious and less culturally influenced activities like scratching one's body or playing with a baby.
 * p. 81

If movements are to be repeated once, then chimpanzees, like humans, compress both movement repetitions to fit into one action unit (Kien et al. 1991), and so action units with one repetition are nDt longer than those without repetition. However, in baboons (Ott et al. 1994), all repetitions, particularly of goal-directed movements, result in a lengthening of the action units; i.e., action units with one repetition are longer than nonrepetitive ones. This suggests that baboons, unlike chimpanzees and humans, need to replan a repetition of a set of movements.
 * p. 91

Another approach has been to measure the durations of semantic phrases. These have been equated with a poem's lines by Turner and Pöppel (1988) because they represent the lowest level where the meaning of a string of sounds is clear to an observer, just as action units are the lowest level where a goal can be assigned to a sequence of movements. Semantic units are extremely hard to define precisely, so Turner and Pöppel turned to poetry where they could show that a line of poetry is equivalent to a semantic unit. Thus, Turner and Pöppel (1988) and Kien and Kemp (1994) have measured the time taken to speak lines of poetry which were separated by pauses (see below), and Fenk-Oczlon and Fenk (1994) have measured the number of syllables and thereby estimated the time needed to speak core sentences--or semantic phrases--in many languages (see below).
 * p. 95-100

Measurements of the semantic phrases between pauses in poetry in fifteen languages (Turner and Pöppel 1988) show that lines of traditional poetry take 2-3 seconds to speak. More detailed measurements of two unrelated languages, German and Hangul (Korean), have shown that there are small but significant differences between languages; German lines of poetry had a median duration of 1.7 seconds during fluent reading whereas Korean lines took a median of 2.2 seconds. However, the lines of both languages fell in the 1-4 seconds range of action units and the distribution of their durations is similar to those of action units in behavior (Kien and Kemp 1994).

Phrases during prose monologues have a different time structure (Kien and Kemp 1994), even though estimates based on pause rates had also suggested a mean phrase length of 2-3 seconds (Kowal et al. 1975). In Kien and Kemp's study of German, phrase length was defined as the duration of speech between pauses of ---0.5 seconds, that is, the length of pauses made between lines when reading poetry. Since these pauses separate meaningful or semantic units in poetry, they may separate comparable units in ordinary speech.

Phrase length measured in this way clearly depended on the type of speech situation (Figure 7). The longest phrases were those spoken while reading a page from a textbook. Next longest were phrases spoken by people leaving messages on a telephone answering machine. Shorter phrases were used in free speech where the subjects were asked to speak about an interesting film or their holidays, and when retelling a well-known fairy story that the subjects had just read. These significant differences, both in medians and also in the shape of the distributions, indicate that different processes are involved in the different speech situations.

The median phrase length of 2.3 and 2.1 seconds, respectively, in free speech and retelling a story means that 50% of phrases consist of only a few words, or less than 5-10 syllables (taking syllable length as 0.2-0.25 seconds; Braitenberg and Pulvermtiller 1992). A similar finding has been reached by linguistic analysis (Fenk-Oczlon and Fenk 1994). Core sentences (i.e., those with subject-verb-predicate structure) were taken as the basic semantic unit, and the number of syllables in such sentences was examined in 29 languages. Core sentences in languages where individual syllables carry little meaning (e.g., Korean, Japanese) contain a mean of 8-11 syllables, and these syllables contain few phonemes. In languages where syllables carry more meaning (e.g., German, English) core sentences contain a mean of 5-6 syllables, and each syllable contains more phonemes. Presuming, once again, 0.2-0.25 seconds for each syllable, such sentences would take a mean of 1.25 seconds (5 x 0.25 seconds) up to 2.2 seconds (11 x 0.2 seconds) to speak. ...

[T]he data suggest that time itself is less important as a determinant of phrase or segment length than language structure, content, and speech situation. Since phrases or sentences are shorter when syllables carry more meaning or when the speech situation is more complex, it seems most likely that constraints of processing or processing capacity are the major factor in speech segmentation. ...

Perception of time differs markedly depending on the interval; time estimation up to 3 seconds is rather good, after that it becomes quite bad (Pöppel 1985). ... The beats of a metronome can be heard in connection if they follow one another in less than three seconds. With longer intervals, people no longer build associations such as thinking that one beat is louder than the others or subjectively combining several beats into one group.

Epstein (1989) has shown that most of the musical themes in the music of Haydn, Mozart, Beethoven, Brahms, and Mahler are in the range of a few seconds. ... Wallschus (1991) has measured both how long the camera stays with each detail and the duration of scenes in German and American feature films. Her findings clearly show a temporal segmentation: western television directors (and video or film editors)--probably unconsciously--use time spans of mostly 2 to 3 seconds to structure film events.

Such findings led Pöppel (1978, 1985) to postulate that occurrences within this span of three seconds could be integrated by an individual into one event. It is this integration which creates our feeling of "nowness" and of the subjective present. This integration becomes more difficult or even impossible if the time span exceeds a few seconds. Human perception, according to Pöppel's hypothesis, is not continuous but in discrete quanta. Though we, phenomenologically, are mostly not aware of it, we experience the flow of time piece after piece, each moment of a few seconds one after the other.

One factor influencing segment length in behavior is whether the action unit is made up of repetitive or nonrepetitive behavior. There seems to be a general structural difference between nonrepetitive and repetitive behavior and its action units. Repetitive behavior, composed of very short sets of movements, occurs over longer time spans than nonrepetitive behavior, both in humans and in nonhuman primates. This could not only be due to the repetition as such resulting in longer time spans, it may result from the different complexity of these two kinds of behavior (Schleidt and Feldh~itter 1989). Fentress (1976) writes that stereotyped movement sequences generally seem to be autonomous and relatively independent of sensory feedback. Thelen (1981a, 1981b) argues that rhythmical stereotypes are manifestations of intrinsic central motor programs and are less responsive to sensory feedback.

Stereotypical rhythmic behavior is very characteristic for infants during certain stages of their development (e.g., Kravitz and Boehm 1971; Thelen 1981b) and precedes the full repertoire of movements. Rhythmic stereotypies become persistent in infants or social animals raised in deprived conditions. This is also the case in other abnormal conditions, such as blindness, autism, and mental retardation. We have similarly found repetitive behavior patterns to be more prominent than nonrepetitive ones in deaf-blind children.

Rhythmic repetition is also seen in adults. It is widely found in ritualized behavior (e.g., d'Aquili and Laughlin 1979). It is also scattered throughout overt everyday behavior, but it is less frequent than nonrepetitive behavior. There is no clear reason why humans recur to such rhythmic repetition in their movement patterns. Such repetition could enlarge the unambiguity of social signals, regulate levels of arousal (e.g., Hutt and Hutt 1970), or stimulate affective arousal (e.g., d'Aquili and Laughlin 1979). In sows, stereotyped behavior is associated with the release of endorphins (Cronin et al. 1986). This may also explain the abundant rhythmic stereotypies in handicapped persons. Furthermore, rhythmic repetition obviously contributes to the effectiveness of movements, for example, in work activities.

[C]orrect recalls decline after a pause of three seconds. The authors thus confirmed that the optimal working capacity of short-term memory lies within the range of a few seconds. ... As Klein (1987:374) has pointed out, the relatively brief retention span of short-term memory has "adaptive significance--it allows us to quickly shift attention from one event to the next."
 * p. 102

Tamplin: Rhythm and Rhyme
Tamplin, Ronald. Rhythm and Rhyme. Philadelphia: Open University Press, 1993.

Stress is relative, contextual, variable and part of the genius of the language, of its economy and subtlety. Poets, and not alone poets but all of us, daily, play on such possibilities for our best effects, our most persuasive understandings. Metre (ti-tum-ti-tum), and, more particularly, its less straight neighbour, 'rhythm', is not metronomic but rather the record of a voice in movement. This is part of the excitement of rhythm in poetry. It is part of the economy of its language, offering varieties of meaning in the one ordering of words.
 * p. 6

Nevertheless, there is a tendency in English for nouns and verbs to carry within their accentual make-up, a heavy stress and for words that attend them, like 'the' and 'a'/'an', 'she'/'he'/'it'/'we'/'you'/'they', to be lightly, or at least more lightly, stressed. Because of this, and the recurrence in normal English of forms like 'the hare', 'her feet', 'he jumps', there is quite a good case for saying that English is naturally an iambic language. Or, perhaps better, that English speech tends naturally towards forming iambic patterns. It is only quite a good case -- by no means an absolute one. As an alternative tendency, notice how often there also occur phrases like 'out of doors', 'in the morning', 'on the moors', 'in her mirth', 'with her feet', 'all the way', which might easily be felt to be anapaestic (x x `). What seems to be happening is that the language of the passage has a dominant iambic drive with a hint of anapaestic variation.

In the second passage, something slightly different is operating. There are a number of short is -- ' lapis ', 'give', 'villas', 'will', 'lizard's', 'quick', 'glitter'. Now, there's nothing special that we associate with short is -- they don't signal greed or anything like that. But, in the structure of the passage, I suggest they anticipate and reinforce the end of the sequence they are part of, and this image, of the glittering eyes, does, in fact, signal greed. Be careful of this, though. It is the image, the association we have with the unnaturally alert eyes, not missing a trick, that gives rise to the meaning, not any inevitable effect of the sound. Once the association is made, though, we see, as with a building, other features, in this case, the preceding short is, anticipating and echoing the most important feature. This is primarily an effect of quantity.
 * pp. 11-13

More often than not, variation in quantity is used simply to introduce variety, texture, into the writing, to pace the lines. ...

Rhyme has a very important function in poetry. It is a device which emphasizes the interrelationship of the various elements of a poem and it is a structural echo of those other relationships -- prime among them, of course, metaphor. A poem is, in a strong sense, a demonstration of the possibilities of relationship; and all its elements, however unconscious we are of them and however concealed, even subliminal, they are, conspire to that demonstration. A rhyme-scheme is one of the most important of these 'netting' effects. ... Many poets are reassured by perceptible shape and structure, however various the means they use to arrive at it. They know that, whatever else, they have 'made' something. Rhyme, its chiming recurrence, is one of the strongest shape-giving devices.

This is all very well, but I'm making it sound a bit private. How does the reader enter into that charmed circle? Easily enough, I think. Notice how the rhymes confirm a shape for the statement Wordsworth is making. The confirmation acts both on the eye, as we see the words 'doors', 'birth', 'moors', 'mirth' and so on, and also for the ear, as we speak the verse. At the same time, part of the skill of rhyming is that, if the poet wants them to, rhymes, certainly in lines of ten syllables and longer, can be made to efface themselves, not exactly hiding but not obtruding either. Wordsworth helps maintain this balance between recognition and surprise by not end-stopping the rhymes 'moor' and 'earth' but letting them run syntactically into their following lines. End-stopping -- that is, punctuating the lines so that strong syntactic pauses occur at the ends of the lines -- is, of course, one way of drawing attention to the rhymes and so to the poetic shape. Once we start to take pleasure from the existence of such shape, we begin to inhabit the mental world of the poet. Once we see that such shape is itself a metaphor for all the relationships in sound and meaning that the poem proposes, we are deep within that world. ...

I do not suggest that these effects are heavily deliberated by Wordsworth. Rather, that the mind of the poet, as it is normally working, thinks in terms of sonic and rhythmic design. In a sense all these effects are 'rhymes' -- a sounding together. Rhyme, properly so-called, is only a more emphatic type of the poet's basic mental procedure, to set the materials of a poem in relationship.

Line-length, to a large degree, governs the unit of space that thoughts characteristically occupy in a poem. It is not going too far to say that a line is, more often than not, a 'thinking-space'. As you vary the lengths of the thinking-spaces, so you regulate the way the thoughts come. In this poem they do not come regular as clockwork but broken, differentiated. Hardy is not like a tiger in a cage, marking out the day in an endless pacing of its length. Instead he stops short, goes a little further, a little further, stops halfway, goes on. Instead of rules he has constants. However he varies it, mutters and inches foward, he will arrive, the course of gloomy foreboding held up but never arrested. ... For the metronome of regular beat, Hardy has substituted the inevitability that, however the lines fall, the end must be the same. He writes not by metronome but by the remorseless music of time, which may seem to be stayed but never can be. ...
 * p. 16-17

An addition of as little as two syllables at the end of the stanza will give a poet just a little more room to move, to close the stanza's pattern and expand the dominant thought. In Wordsworth's case it was to build a line with a little pattern of four three-syllable units and a consequent sense of resolution. In Hardy's case it is to allow in each stanza a packed climactic flourish, the final words confirming the drive of the thought in the whole sequence. ...

Is one of the effects of refrain to throw another net of rhyme over not just the stanza but the whole poem? What is the effect of that?

[F]or art to be artificial is, in fact, natural. Novels, symphonies, popular songs, football, chess and so on -- most of the things people enjoy -- are characterized by structures that are artificial rather than simply 'given'. They need laws to function by, structures to keep them from disintegrating, techniques of preparation to render them 'cooked' and not 'raw'. But, second, I wish to go further and suggest that it is precisely in this pursuit of and delight in structure that they are most lifelike: that the world in its natural state, as much as the manmade world, is subject to its own firmly laid structures of chemical interaction, ecological chain, alternation and rhythm. And so when art is structured, successive, rhythmic in its pulse, it is modelled on the deep structure of the world and not simply responding to a set of manmade, arbitrary rules.
 * p. 20

Heron identifies in the work of Leach and some other twentieth-century potters a quality which he calls 'submerged rhythm'. He says of it:
 * p. 21-22


 * We feel a powerful pulse in their pots: a rhythm that seems at its most emphatic just below the glazed surface. This is also a characteristic of natural forms -- logs; boulders that have been washed by the sea; or even in the human figure, where the structural form is below the surface of the flesh -- the bone is under the muscle.

There is a new assertion here: that the natural world is also characterized by rhythmic structure. And in developing the idea, Heron again calls on Braque, who


 * has said that the painter should put himself in rhythmic or formal sympathy with nature: he should not imitate. By doing the first he gets closer to that natural reality he loves: by the second, he estranges himself from nature. ...

But it is the second element in Braque's account that I find more intriguing and suggestive in the present context: that by merely imitating the natural world we estrange ourselves from it. To express our 'rhythmic or formal sympathy' with the natural world, therefore, we must respond in our own terms, giving our shapes to celebrate the shapes we perceive as present to us but distinct from us. Our rhythms will be made by us, made by art -- that is, artificial -- but will have a life of their own, to echo the life that precedes them in the natural world. In that sense, then, they are also natural. Heron concludes his essay by speaking of 'man's will to form'. He writes, 'If I believe this sense of form is of immeasurable importance to mankind -- that may well be because, for me, the moral and the aesthetic have a single identity. Ethics are the aesthetics of behaviour.'

Now poetry renders the natural world certainly. Prime among the elements it renders is speech. But it renders speech in a heightened way, giving it an enhanced rhythm and features like rhyme and alliteration. The ideas that 'people don't talk poetry', that 'it's not poetry if it doesn't rhyme', and the folk reaction that, if by chance you stumble on a rhyme while you're talking, 'you're a poet and you don't know it' are all popular testimony to the strangeness of poetry. Poetry is a type of alternative speech, not an imitation of everyday speech. Rhyme and metrical pulse are key ways for making it different. So Braque's idea can be applied directly to the way poets work too.
 * p. 23

Local textures in music are built on repetition; those in poetry by variation. ...
 * p. 28

There are two other uses for the word 'musical' I can think of as applied to poetry. There is the highly technical one of 'singability'. Certain clenches of sound are near to unsingable. The mouth just can't get round them. They may be highly poetic. Look, for example, at Leontes's speech in The Winter's Tale:


 * I' fecks!
 * Why, that's my bawcock. What, hast smutch'd thy nose?
 * They say it is a copy out of mine. Come, captain,
 * We must be neat -- not neat but cleanly, captain.
 * And yet the steer, the heifer, and the calf
 * Are all called neat. -- Still virginalling
 * Upon his palm? -- How now, you wanton calf?
 * Art thou my calf?

This is rhythmically far too broken to be easily singable. It is far too highly charged as well.

Tsur: What is Cognitive Poetics?
Tsur, Reuven. What is Cognitive Poetics? Tel Aviv University: The Katz Research Institute for Hebrew Literature, 1983.

[P]oetic reductionism would be the doctrine that every literary "natural" kind is, or is coextensive with, a psychological natural kind. A major assumption of the present cognitive approach is that literature does have important operational principles which cannot be exhausted in terms of cognitive science.
 * p. 6

To paraphrase again Polanyi (1967:40), the principles of literature may be said to govern the boundary conditions of a cognitive system -- a set of conditions that is explicitly left undetermined by the laws of lower processes, physical, cognitive and linguistic. If one knows what is the set of boundary conditions left undetermined, and by what laws of what "lower" processes, one may get a better understanding of the principles of literature, that govern those boundary conditions. It is claimed that in this way cognitive poetics is capable of discerning and explaining significant literary phenomena which present insurmountable difficulties to other approaches. ...
 * p. 7


 * Habituation devours works, clothes, furniture, one's wife, and the fear of war. "If the whole complex lives of many people go on unconsciously, then such lives are as if they had never been". And art exists that one may recover the sensation of life; it exists to make one feel things, to make the stone stony. The purpose of art is to impart the sensation of things as they are perceived and not as they are known. The technique of art is to make objects 'unfamiliar', to make forms difficult, to increase the difficulty and length of perception because the of perception is an aesthetic end in itself and must be prolonged (Shklovsky, 1965:12).

This passage is important for our discussion in several respects. In the first place, it shows what it is that art is better suited than anything else to do; and it also suggests that it is precisely this something that prevents human life from becoming meaningless. In the second place, it presents an instance of what cognitive poetics typically does. It indicates the way in which the human cognitive system typically functions in non-literary experience, the way that "permits the greatest economy of perceptive effort". Then it suggests that in order to achieve art's end, these normal cognitive processes must be disturbed, deformed, slowed down.

One major assumption of cognitive poetics is that poetry exploits for aesthetic purposes cognitive (including linguistic) processes that were initially evolved for non-aesthetic purposes.... Such an assumption is much more parsimonious than postulating independent aesthetic and/or linguistic mechanisms. ...
 * p. 8

The reading of poetry involves the modification (or, sometimes, deformation) of cognitive processes, and their adaptation for purposes for which they were not originally "devised". In certain extreme but central cases, this modification may become "organized violence against cognitive structures and processes", to paraphrase the famous slogan of Russian Formalism. As it will be emphasized again and again in the course of the present study, quite a few (but by no means all) central poetic effects are the result of some drastic interference with, or at least delay of, the regular course of cognitive processes, and the exploitation of its effects for aesthetic purposes.

Liberman and his collaborators conceive of systems of versification as of a kind of secondary codes. "For a literate society the function of verse is primarily esthetic, but for preliterate societies, verse is a means of transmitting verbal information of cultural importance with a minimum of paraphrase. The rules of verse are, in effect, an addition to phonology which requires that recalled material not only should preserve the semantic values of the original, but should also conform to a specific, rule-determined phonetic pattern" (Liberman et al., 1972). In terms of the premises of the present discussion, we can regard paraphrase, not as a result of forgetting, but rather as an essential condition or correlate of the processes by which we normally communicate and remember. "If linguistic communications could only be stored in the form in which they were presented, we should presumably be making inefficient use of our capacity for storage and retrieval; the information must be restructued if that which is communicated to us by language is to be well remembered" (Liberman et al., 1972). Now, it is precisely this process of successive restructurings or recodings that is interfered with by versification. In a sense this increases our "capacity of retrieval" for the particular versified messages; but this is done at the price of patent interference with the cognitive economy of the system. So, we may regard versification as an instance of "organized violence" against cognitive processes.
 * p. 12

Speech sounds do have a phonetic facet and an acoustic facet. Transition between the two involves complex recoding; that is, there is no one-to-one relationship between the phonetic unit and the acoustic cues. Since, however, we are usually not aware of acoustic cues, only of phonetic units, we may simply ignore the acoustic cues in our discussions of e.g. expressive sound patterns in poetry, though we may rely on phonetic properties of the speech-sounds. Now, it has been found statistically, that such liquids and nasals as /l, m/ are in a variety of languages positively correlated with tender poems, and negatively with aggressive ones; whereas such voiceless stops as /t, k/ are positively correlated with aggressive poems, and negatively with tender ones.
 * p. 13

Now, the correlations of liquids and nasals, and of voiceless stops with tender and aggressive poems can be explained, precisely, in terms of a delay in recoding or restructuring, from acoustic cues to phonetic entities. Voiceless stops are perceived as unitary linguistic events, stripped of all precategorical sensory information. Here, the recoding process goes on with no interference. On the other hand, in relatively unencoded speech-sounds, such as liquids and nasals, the recoding process can be "disturbed", so that some of the rich precategorical auditory information becomes available to consciousness. Emotional flexibility, an openness and responsiveness to rich precategorical information, is characteristical of tender feelings, whereas the lack of it characterizes aggression. Thus, liquids and nasals may have a perceived quality of tenderness, or emotional adaptability, or sensory richness, whereas voiceless stops may have a perceived quality of rigidity that may be positively correlated with aggression, and negatively with tenderness.
 * p. 14

Liberman and his collaborators suggest that there is a speech-mode and a nonspeech-mode of processing acoustic signals. The output of the latter is experienced in consciousness as music or natural noises; in the speech-mode the output of the nonspeech mode is shut out from consciousness. I suggest that there may be a poetic mode as well, that is similar to the speech-mode in that it is focussed on linguistic categories; but, at the same time, the output of the nonspeech-mode, in the form of precategorical auditory information, becomes -- however faintly -- accessible to consciousness. ...

Normally, the conversion from surface-structure to deep structure and back is automatic and goes unnoticed. Some aesthetic effects crucially depend on a de-automatization of this conversion.

Let us illustrate these points by comparing two poetic passages, the first one from Keats's sonnet "On Seeing the Elgin Marbles", the second one from Marlowe's tragedy "Tamburlaine".
 * p. 26-27


 * 8. Such dim conceived glories of the brain
 * Bring round the heart an indescribable feud...


 * 9. Nature that framed us of four elements,
 * Warring within our breasts for regiment,
 * Doth teach us all to have our aspiring minds.

In spite of Tamburlaine's and Faustus' notorious craving for infinite things, we may expect, from what we know from sweeping generalizations, Keats's poetry to be of a more romantic, more affective mood than Marlowe's poetry. It would be interesting to see, whether and how the two passages bear out such pieces of "common knowledge".

The two passages have a considerable number of elements in common. Both refer, in a fairly direct way, to an undifferentiable feeling, in terms of a "gestalt-free" quality: a war "within our breasts" (or "feud" round the heart), and its relation to what happens in our minds (or in the brain). For Keats, as a true Romantic, this is an intense passion at a unique moment (one of his "many havens of intensity"); for Marlowe, this is rather a permanent disposition. Some readers report, that they perceive a heightened affective quality in 8., as compared to 9. One possible explanation for this may rely on the different connotations of "warring" and "feud". But far more significant seems to be the fact that whereas in Marlowe's passage it is the clearly differentiated "four elements" that are "warring within our breasts", Keats's "feud" around the heart is not only undifferentiated and gestalt-free, but is thing-free too: in referential language we expect to be told, a feud between whom or what. Moreover, the location "round the heart" is vaguer than would be "in the heart". Thus, the more passionate impact of Keats's lines has to do with the fact that they are focussed on violent actions that are stripped of things which might carry them. Furthermore, although both metaphors seem to refer to some kind of emotional turbulence, Marlowe uses rhetorical devices to heighten its conscious "linear" quality, whereas Keats uses devices to mute, or obscure, its conscious quality. Marlowe's "warring within our breasts" is endowed with the psychological atmosphere of patent purpose, generated by the purposive ingredient in the words and phrases "for regiment", "teach", "aspiring minds", as well as by the conclusive nature of "all".

Keats, on the other hand, emphasizes the undifferentiated character of the passion by the adjectives "indescribable feud", and "dim-conceived glories".

Consider the last three lines of Shelley's "Ozymandias":
 * p. 38-39


 * 15. Nothing behind remains. Round the decay
 * Of that colossal wreck, boundless and bare,
 * The lone and level sands stretch far away.

The affective impact of this physical description is derived from an unfulfilled readiness to pick up self-specifying visual information on a vast surrounding space with fewer than usual cues for orientation. The circumstances of unfulfilled readiness to pick up self-specifying information from the optic structure of the environment are amplified in the last lines of Keats's "When I Have Fears" where the scope of orientation-space is increased, whereas the possible visual cues are reduced to zero:


 * 16. then on the shore
 * Of the wide world I stand alone, and think,
 * Till love and fame to nothingness do sink.

Much romantic poetry, like those in 15 and 16, is aimed at a reintegration of the self with the environment, or at a total dissolution in it; in other words, it is an attempt to overcome, wholly or partially, the separation of self from its environment, with the help of the orientation-mechanism.

Wallenstein: The Pleasure Instinct
Wallenstein, Gene. The Pleasure Instinct: Why We Crave Adventure, Chocolate, Pheromones, and Music. New Jersey: John Wiley & Sons, Inc., 2009.

Clearly, learning plays a large role in shaping the specific musical idioms we prefer. Research throughout the past decade, however, has begun to show that certain sounds and note combinations have virtually universal effect on the emotions of listeners independent of the culture in which they were born, raised, and live. Moreover, most neurologically normal listeners, no matter where they are from, can agree on what is and is not musical, even when the sequence of tones is novel or drawn from a foreign scale. This has led some theorists to focus on the similarities between music and language development when speculating on the origins of musicality. ...
 * p. 100-101

The composer Leonard Bernstein was the first to apply Chomsky's ideas about language to music. He suggested that all the world's musical idioms conform to a universal musical grammar. This theory was advanced more formally through the work of psychologist Ray Jackendoff and musicologist Fred Lerdahl. They viewed music as being built from a hierarchy of mental structures, all superimposed on the same sequence of notes and derived from a common set of rules. The discrete notes are the building blocks of a piece and differ in how stable they feel to a listener. Notes that are unstable induce a feeling of tension, while those that are stable create a sensation of finality or being settled.

Both infants and adults can recognize a tune based on the diatonic scale as the same when it is transposed across pitch, but fail to do so when the melody comes from a nondiatonic scale. Since all primates perceive tones separated by an equal octave as having the same pitch quality, one might predict that the ability to detect transposition of diatonic melodies is also present in our hairier cousins. To date, this experiment has only been performed with rhesus monkeys and, as expected, they exhibit the same effect as human adults and infants. ...
 * p. 103-105

Experiments have shown that newborns and infants are highly sensitive to the prosodic cues of speech, which tend to convey the emotional tone of the message. Prosody is exaggerated even more so in the typical singsong style of motherese that dominates parent-infant dialogue during the first year of life. The infant trains its parents in motherese by responding positively to certain acoustic features they provide over others. Motherese and lullabies have so many acoustic properties in common -- such as simple pitch contours, broad pitch range, and syllable repetition -- that theorists have argued them to be of the same music genre.

Just as motherese shows up with the same acoustic properties in virtually every culture, so too does the lullaby. Practically everyone agrees on what is and is not a lullaby. Naive listeners can distinguish foreign lullabies from nonlullabies that stem from the same culture of origin and use the same tempo. Of course, infants make the distinction quite readily. Even neonates prefer the lullaby rendition of a song to the nonlullaby rendition performed by the same singer. Although it is tempting to attribute such preferences to experience, studies have shown that hearing infants raised by deaf parents who communicate only by sign language show comparable biases. It appears, then, that from our very first breath, we carry a set of inborn predispositions that make us seek out specific auditory stimuli. ...

In April 2003, scientists from the University of California at San Francisco discovered that newborn rats fail to develop a normal auditory cortex when reared in an environment that consists of continuous white noise. The hallmark of white noise is that it has no structured sound -- every sound wave frequency is represented equally. ...

Interestingly, the abnormalities persisted long after the experiment ended, but when the noise-reared rats were later exposed to repetitious and highly structured sounds -- such as music -- their auditory cortex rewired and they regained most of the anatomical and physiological markers that were observed in normal rats.

Music and dance production also consume vast amounts of metabolic energy used by the brain to derive creative expressions that will attract the attention of potential mates. The overwhelming majority of energetic costs associated with being a big-brained hominid are linked to that big brain. No body part or system even comes close in its energy requirements to the ongoing metabolic demands of our brain. Hence traits that are reliably linked to increasing the already high energy demands of the brain would be honest indicators of fitness in this broader sense.
 * p. 178

Miller has made some suggestions in this regard. "Dancing reveals aerobic fitness, coordination, strength, and health. Because nervousness interferes with fine motor control, including voice control, singing in key may reveal self-confidence, status, and extroversion. Rhythm may reveal the brain's capacity for sequencing complex movements reliably, and the efficiency and flexibility of the brain's 'central pattern generators.' Likewise, virtuosic performance of instrumental music may reveal motor coordination, capacity for automating complex learned behaviors, and having the time to practice."