Volume VII - 2002-03

Dual Coding Theory and Reading Poetic Text
     by Mark Sadoski

     Mark Sadoski is a Professor and Distinguished Research Fellow in the College of Education at Texas A&M University, College Station. His primary research interest is in the way language and mental imagery combine in literacy. His 2001book, Imagery and Verbal Processes: A Dual Coding Theory of Reading and Writing, co-authored with Allan Paivio and published by Lawrence Erlbaum Associates, extends Dual Coding Theory to literacy in detail and reviews years of research in this area.

EDITOR 'S NOTE: Serendipity —or at least truly surprising coincidence —seems as predictable as rain. One can 't say for certain when or where it will come, but one can predict with certainty that it will occur again. Over the years, this Journal has had more than its share of serendipity. And it continues. For example, in the following article, Sadoski discusses the mental processing of language and that of imagery. He says that they constitute two great mental codes: the verbal and the non-verbal. He proceeds to explain fundamental concepts that are of immediate relevance to language acquisition and to the language classroom. Enter serendipity. In another article in this issue of the Journal , Ana Robles points out the difficulty students have when the word in the target language calls up, not an object or concept, but a word in the first language, which then calls up its own object or concept. Often, she says, this results in the failure to communicate. There is no one-to-one correspondence available, but rather a new concept that must be learned. In this note, I simply wish to direct your attention to these articles and to ask you to read each with the other in mind.

The way to read, then, seems to be: use imagination and a dictionary. --Louis Simpson (1986, p. 14)

     Intentionally or unintentionally, that lilting line from an introductory volume of poetry metaphorically captures much of the essence of Dual Coding Theory (DCT), a scientific theory of cognition that has evolved from laboratory psychology to a practical theory of reading and writing. DCT deals with the way cognition occurs —both as language and as mental imagery —and it can be applied to the understanding and appreciation of all text including poetry. The theory has particular application for second language cognition. This article will briefly (a)overview DCT in nontechnical terms, and (b)apply it to examples of poetry.

Overview of DCT

     The substance of DCT is covered in three volumes. The general theory was presented by Paivio (1971), further developed by Paivio (1986), and specifically applied to literacy by Sadoski and Paivio (2001). Detailed explanations of all the principles of the theory, with supporting evidence, are presented there for the interested reader. The most basic principle of DCT is that all cognition involves an intricate interplay between two great mental codes, the verbal code and the nonverbal code. In the case of multilinguals, separate but connected verbal subsystems exist for each known language.

The Verbal Code

     The verbal code is a mental code specialized for dealing with verbal language. Verbal language provides a remarkable means for the encoding, communication, and decoding of messages that symbolize our experiences and ideas, both abstract and concrete. The specific units and arrangements of verbal language that appeal to our different senses are well known. In the auditory and articulatory senses, the units and arrangements are phonemes, word pronunciations, stress intonations and rhythms, and so on. In the visual sense (tactile in the case of Braille), the units and arrangements are letters, written spellings, punctuation marks, lines of verse, and so on. General language concepts such as morphology, grammar, and usage apply to both sense modalities.

     One can readily appreciate the intricacy of verbal cognition alone because these units and arrangements are not automatically linked in cognition. One can have an advanced command of speech but still be illiterate. In the case of English and many other languages, written language maps onto spoken language imperfectly (e. g. , silent letters). Morphology can affect pronunciation and spelling (e. g. , the pronunciation but not the s pelling of read depends on its tense, but example and its close derivative, exemplary , differ in both spelling and pronunciation). Moreover, different languages use different sets of graphemes and phonemes and have different degrees of grapheme-phoneme correspondence as well as other features such as written directionality (i. e., up, down left, right). Yet they can be readily translated to each other.

     A key variable in verbal language is concreteness. Concrete language readily evokes mental imagery (e. g. , steaming jungle)whereas abstract language does not (e. g. , climatic variation). Some words hover in between —equator is a physical place but also an abstract concept —there isn 't literally a line around the earth there. Decades of laboratory studies have determined that concrete language in words, phrases, sentences, and texts is comprehended and remembered better than abstract language. DCT explains why: When language can be encoded verbally but also as nonverbal mental imagery, the potential for comprehension and memory increases by a factor of two (i. e. , dual coding). Furthermore, different languages appear to draw on a common code of nonverbal images as part of their meaning base. For example, the English phrase steaming jungle and its German translation dämpfendschungel would likely refer to the same images in a bilingual reader 's non-verbal code.

The Nonverbal Code

     The nonverbal code is specialized for dealing with nonlinguistic knowledge of the world. It is commonly referred to as the imagery code because the generation of mental images is a chief function. The nonverbal code provides a remarkable means for us to retain, manipulate, and transform the world around us mentally, in imagination. Even more than language, the imagery code is represented in multiple senses. We can imagine sights, sounds, smells, tastes, and touch sensations, although visual imagery is most apparent to most people. Sometimes imagery is multimodal and approaches actual experience, if vicariously. Imagine slogging through an equatorial jungle –the sticky, humid, heat; the slippery mud underfoot; the dense, green foliage; the calls of the jungle birds; the vivid color and exotic scent of a delicate jungle orchid. The units and arrangements in this code are more fluid than in the verbal code. Images tend to occur in overlapping nested sets and are not constrained by the stricter sequences found in phonology, grammar, or verse. We can mentally switch perspective or surreally –“dissolve ” one image into another without regard to logical convention. . Imagery is the stuff of dreams, fantasies, and nightmares.

     The extraordinary effects of imagination are impressive, but imagery is basic to more mundane cognitive functions as well. How many windows does your home have? Most people report taking a mental tour of the premises, counting windows as they go. Similarly, we could organize a shopping trip through a memorized alphabetical listing of items, but we more typically organize such a trip from the locations of the items in a mental tour of the supermarket. We don 't feel we really remember someone until we can match a name with a face. In reading, we often do not fully comprehend unless we can generate a mental model of the text, a situational instance of the content of the text. This what is meant by “making sense ” of the text –literally, making it a quasi-sensory event. Hence, a teacher 's quick explanation of the textbook phrase equatorial climatic variation might be “mostly like a jungle. ”This quickly “makes sense. ”

Connections Between Codes

     The connections between the two codes are many, but they vary with both language and experience. Abstract language evokes little mental imagery, concrete language more. Hence, abstract language is understood primarily as an intraverbal puzzle, through mental paraphrasing, translation, and other verbal elaboration. Dictionaries or textbooks in any language might define the equator in the abstract as a latitude in a plane perpendicular to the axis of the earth and equidistant from the poles with climatic maxima determined by the equinoxes. They might also define the equator in more concrete terms, as in describing the river basin geography and rainforest climate of the Amazon or much of central Africa. Concrete language therefore enjoys the advantage of verbal elaboration, but it also evokes images formed from our experience. Our fund of experience limits the images that we might form in response to a text, so that someone who has seen an equatorial jungle firsthand can imagine its sensory character better than someone who has only experienced the frozen Arctic, and vice-versa. Films, videos, and still pictures serve as approximations for most of us. But we are not prisoners of our experience. Imagination is the act of taking apart and putting together our images in new and perhaps novel arrangements. We have all been hot and sweaty, felt mud underfoot, seen green plants and flowers. Hence, we can vicariously imagine steamy jungles even if we have only experienced them in National Geographic specials. Likewise, we can imagine what is only fictional or fantastic, such as the huge, man-eating orchids of the equatorial jungles of the planet Xanus-3!

     In this way, meaning emerges from the intricate interplay of activity within and between the verbal and nonverbal codes. The more elaborate, organized, and connected our complementary systems of language and imagistic world knowledge, the more potential for meaning. This interplay also has implications for memory and learning, emotional response, motivation, inspiration, and so on.

      The broad outline of the theory as it applies here should now be evident. Cognition in reading text is in the evocation of language by other language, and mental imagery by language. In reading literary and poetic text, mental imagery will virtually always be evoked.

DCT and Appreciating Poetic Text

     The idea that poetry is largely concerned with communicating thoughts and feelings through verbal rhythms and images is widely accepted and dates back centuries. Literary historians speculate that poetry itself evolved as an oral mnemonic system for preserving and transmitting the stories, values, and traditions of early societies through rhythmic recitation before the advent of writing. The work of Parry (1971)on the Homeric epics suggested, for example, that picturesque Homeric epithets such as ”“wine-dark sea ” or “white-armed goddess ” were probably improvised to meet the demands of the dactylic hexameter of Greek heroic poetry. That is, the phrasings fit the verbal intonations that served as a memory vehicle for the oral recitation of the poem, and they evoked a vivid story image as well. Together, they formed a verbal-nonverbal poetic unity. Later poetry, developed in the context of writing, was less constrained by oral mnemonic traditions and developed forms such as free verse. Poets also began to use the written, visual structure of the poem itself as a tool (e. g. , variations in line length and spacing).

     However, orally-based verse conventions persist in poetry for cognitive as well as aesthetic reasons. The rhythm provided by repeated stress patterns, pauses, and rhymes serves a “chunking ” and organizing function that is effective in memory. It keeps phrasings in working memory as the poem is being read, and it appeals to long-term memory as well. As Parini (1999)noted of Robert Frost 's lyrical “Stopping by Woods on a Snowy Evening, ” one can hardly not memorize it!!

     While always acknowledged as important in poetry, imagery seems to have grown in stature as a poetic tool during the last century. Literary figures in the early twentieth century took a profound interest in the poetic image. The Imagist poets including Ezra Pound and Amy Lowell concentrated on evoking imagery in their readers through vivid descriptions, metaphors, and the juxtaposition of disparate objects. Imagism as a movement was short-lived, but it is credited with beginning modern poetry. Virtually every poet since, including Robert Frost, has been influenced by the imagists and employed vivid images that are juxtaposed without specifying their relationships (Abrams, 1999). T. S. Eliot (1960)maintained that the only way to express emotion in poetry was to find an “objective correlative, ” a set of concrete objects or events that evoked an emotion in the reader without actually stating that emotion (i. e. , show us don 't tell us). C. Day Lewis (1948)maintained that a poetic image is a picture made from words and that a poem is itself an image composed of a multiplicity of images. His statement is highly consonant with the DCT principle stated earlier that images tend to occur in overlapping nested sets.

     Images serve both cognitive and aesthetic purposes in poetry and literature. Images are symbolic – they stand for often unstated ideas that may embody the theme of a poem. The deserted, broken statue of the forgotten king in Shelly 's “Ozymandias ” is a symbol of the impermanence of human power and vanity. . As with that poem, even if the words are forgotten, the images remain in memory as thematically meaningful mental pegs. Images are central to metaphors, and metaphors on both local and global scales are central to the meaning of poetry. In short, the separate but unified contributions of the verbal and nonverbal codes as postulated in DCT may be what poetry is all about. The great classic poets – Homer, , Dante, Shakespeare, and Milton, among others –– are all known for both their eloquent use of language and their vivid imagery. . Any standard textbook on the subject is replete with explanation and examples (e. g. , Simpson, 1986). Such examples will not be further repeated here; let us turn to a more complex issue.

Language and Imagery as Collaborators and Contenders

     The issue of appreciation may be more subtle that a simple combining of verbal eloquence and vivid imagery. Stylistic uses of language can finesse a balance between language and imagery, so that the reader 's attention is not completely on either but on both in a mental model where they imply each other. For example, consider two similar metaphors. The first is from Yeats 's (1976, p. 111)“Sailing to Byzantium ”:

An aged man is but a paltry thing,
A tattered coat upon a stick, unless
Soul clap its hands and sing, and louder sing
For every tatter in its mortal dress …

This metaphor becomes a conceit in Proust 's (1981, p. 1106-1107) Remembrance of Things Past :

…when he rose to his feet and tried to stand firmly upon them, swayed backwards and forwards upon legs as tottery as of some old archbishop with nothing solid about him but his metal crucifix …and had advanced with difficulty trembling like a leaf, upon the almost unmanageable summit of his eighty-three years, as though men spend their lives perched upon living stilts which never cease to grow until sometimes they become taller than church steeples, making it in the end both difficult and perilous for them to walk and raising them to an eminence from which they suddenly fall.

     The images of dancing scarecrows, archbishops propped up by their metal crosses, and men trembling on stilts taller than church spires depend for their effect on the fact that they are not to be taken literally. Converted to a literal image, these metaphors would seem absurd. The proper meaning resides in the understatement or suggestion between the language and the images, not in a literal translation. Perhaps T. S. Eliot (quoted in Valentine, 1968, p. 367)had something like this in mind when he remarked that in reading Milton 's Paradise Lost “our sense of sight must be blurred, so that our hearing may become more acute. ”That is, we must not take the imagery too literally at risk of losing the meaning and force of the words. If the appreciation of poetry resides in the willing suspension of disbelief, it resides in the willing suspension of literal belief as well.

     A poet may take advantage of this fact and actually play language and imagery against each other to achieve a counterpoint effect. Consider Masefield 's (1951, pp. 20-21)often-misunderstood poem “Sea-Fever ”:

I must go down to the seas again, to the lonely sea and the sky,
And all I ask is a tall ship and a star to steer her by,
And the wheel 's kick and the wind 's song and the white sail 's shaking
And a grey mist on the sea 's face and a grey dawn breaking.
I must go down to the seas again, for the call of the running tide
Is a wild call and a clear call that cannot be denied;
And all I ask is a windy day with the white clouds flying,
And the flung spray and the blown spume, and the sea gulls crying.
I must go down to the seas again to the vagrant gypsy life,
To the gull 's way and the whale 's way where the wind 's like a whetted knife;
And all I ask is a merry yarn and a laughing fellow rover,
And a quiet sleep and a sweet dream when the long trick 's over.

     The images evoked are the sights, sounds, and feel of adventurous seafaring: the sail heaving as the clouds fly by, the foam and spray of the waves, the cry of the gulls, the kick of the wheel, the whistle and bite of the wind. The narrator yearns for “the vagrant gypsy life ” and a “merry yarn from a laughing fellow rover. ”The images suggest the wild, free life of the wanderer. Even if we have never directly experienced these things, we can imagine them.

     But the poem is curiously conflicted in this regard. Each stanza begins briskly but then slows and becomes sluggish in its meter. The first two lines of each stanza have a rollicking, sing-song meter that suggests the rhythm of the waves or the pitching of a ship. Their masculine rhyme is hard and punchy. (Technically, these line pairs suggest the heroic couplets of the eighteenth century or the dactylic hexameter of Greek heroic poetry). But the meter changes in the second set of lines. Consonants are pushed against consonants, accented syllables against accented syllables. This has the effect of slowing the rhythm, holding it back. The rhyme in these pairs is feminine, ending on softer unaccented syllables. The difference between the first and second pairs of lines in each stanza is marked.

     Why? It would have been easy to continue the rocking rhythm instead of slowing it down (e. g. , “the kick of the wheel and the song of the wind ” instead of “the wheel 's kick and the wind 's song ”). The slowing effect is therefore intentional, and it seems inconsistent with the adventurous mental images. Perhaps the poet was suggesting something by this conflict, implying that there are forces within us, as there are forces within the poem, that hold us back from the “vagrant gypsy life. ”We hold ourselves back, and dream. Among the many things the narrator asks for is “a quiet sleep and a sweet dream when the long trick's over ”—the narrator doesn 't seem to be sleeping too well. (A trick is a turn of duty at the helm; figuratively, life, and perhaps an irony; the poem 's a trick. The title may also have an ironic meaning. )

     In short, this poem isn 't about going to sea, it 's about not going to sea. It 's about the forces within us that hold us back from living out our dreams. But the effect is nowhere achieved literally. It is all implied in the disharmony between the visual imagery and the auditory meter of the lines, a play of the nonverbal and verbal codes off each other for rhetorical and poetic effect.


     DCT is a general theory of cognition that has been applied to text and can explain a great deal of the comprehension and appreciation of literary text including poetry. Language and imagery working together, or even in intentional disharmony, produce the effects we experience in evoking a text including rhetorical effects (for additional explanation and several examples from nonfiction, see Sadoski, 1992). Certainly more analysis of the examples cited here could be undertaken and other valid literary interpretations are possible. But all would deal in fundamental ways with the imagery evoked by the text and the author 's
skillful use of language. To rephrase the quotation that leads this article, the way to read, then, seems to be: to creatively use our mental imagery and our command of language.


Abrams, M. H. (1999). A Glossary of Literary Terms (7th ed. ). Fort Worth, TX: Harcourt Brace College Publishers.

Eliot. T. S. (1966). “Hamlet and His Problems ” in Selected essays . London: Farber.

Lewis, C. D. (1948). The Poetic Image . London: Jonathan Cape.

Masefield, J. (1951). Poems . New York: Macmillan.

Paivio, A. (1971). Imagery and Verbal Processes . New York: Holt, Rinehart, and Winston. (Reprinted 1979, Hillsdale, NJ:
     Lawrence Erlbaum Associates)

Paivio, A. (1986). Mental representations: A Dual Coding Approach . New York: Oxford University Press.

Parini, J. (1999). Robert Frost: A Life . New York: Henry Holt and Company.

Parry, M. (1971). The Making of Homeric Verse: The Collected Papers of Millman Parry (A. Parry, Ed. ). New York: Oxford       University Press.

Proust, M. (1981). “Time regained. ” In Remembrance of Things Past (vol. III). (C. K. S. Moncrieff, T. Kilmartin, &A.
     Mayor, Trans. ). New York: Random House.

Sadoski, M. (1992). “Imagination, Cognition, and Persona. ” Rhetoric Review , 10, 266-278.

Sadoski, M., &Paivio, A. (2001). Imagery and Text: A Dual Coding Theory of Reading and Writing . Mahwah, NJ: Lawrence      Erlbaum Associates.

Simpson, L. (1986). An Introduction to Poetry (3 rd ed. ). New York: St. Martin 's Press.

Valentine, C. W. (1968). The Experimental Psychology of Beauty. London: Camelot Press.

Yeats, W. B. (1976). The Poems of W. B. Yeats (W. Y. Tindall, Ed. ). Norwalk, CT: Easton Press.

back to content page