Can Poetry Be Translated? Roman Jakobson on When Form Becomes Meaning

Introduction

Why is it easier to translate a scientific article than a sonnet? Most people would answer that poems are simply more difficult. Roman Jakobson offers a more interesting explanation. The decisive difference, he argues, is not between information and poetry but between two ways language functions. Sometimes words primarily direct our attention toward the world they describe. At other times, language draws attention to itself—to its sounds, rhythms, repetitions, and patterns. When this happens, form is no longer a mere vehicle for meaning; it becomes part of meaning itself.

This distinction lies at the heart of Jakobson's classic essay On Linguistic Aspects of Translation (1959). It also explains why he can make two apparently contradictory claims. On the one hand, "all cognitive experience and its classification is conveyable in any existing language" (Jakobson, 1959, p. 233). On the other, "poetry by definition is untranslatable" (p. 238). These statements are perfectly consistent once we recognize that they concern different kinds of meaning.

When Language Points Beyond Itself

Jakobson begins from a surprisingly optimistic premise. No language is incapable of expressing a thought merely because it lacks a particular word or grammatical category. Whenever direct equivalence is unavailable, languages compensate through paraphrase, borrowing, circumlocution, or newly created expressions. Translation, therefore, is not the substitution of isolated words but the interpretation of messages.

This explains his distinction between code-units and messages. Code-units are the individual elements of a language—words, grammatical forms, and other linguistic signs. Messages are the meanings communicated through those elements. Exact correspondence between code-units is uncommon, but this does not prevent successful communication because the message can be reconstructed by other means.

Jakobson's famous example concerns the English word cheese and its Russian counterpart сыр. At first glance they appear equivalent, yet they classify dairy products differently. English includes cottage cheese within the broader category of cheese, whereas Russian distinguishes сыр from творог. A Russian speaker therefore naturally asks for "сыр и творог," while the literal English rendering, "cheese and cottage cheese," sounds unnecessarily repetitive. The words fail to coincide perfectly, but the intended meaning remains entirely communicable.

For Jakobson, this is not an obstacle to translation. It demonstrates that languages differ primarily in what they require speakers to express, not in what they are capable of expressing. Cognitive meaning survives changes in vocabulary because it ultimately refers beyond language itself. The linguistic form may vary; the underlying message remains conveyable.

When Form Begins to Mean

The discussion changes direction when Jakobson turns to poetry. His point is often summarized by the slogan "poetry is untranslatable," but this obscures the argument that leads to that conclusion.

Jakobson first redefines what poetry is:

"In poetry, verbal equations become a constructive principle of the text" (Jakobson, 1959, p. 238).

The sentence deserves careful attention. In ordinary discourse, sounds, grammatical patterns, and repetitions usually serve the communication of ideas. In poetry, those same features acquire significance in their own right. Jakobson immediately expands the point:

"Syntactic and morphological categories, roots, and affixes, phonemes and their components... carry their own autonomous signification" (p. 238).

This marks a profound shift. Language no longer functions merely as a transparent medium pointing toward external reality. Its material organization becomes part of what the text communicates. Rhythm, rhyme, alliteration, syntax, repetition, and phonetic resemblance are no longer decorative additions to meaning. They help create it.

The decisive distinction, then, is not between informative writing and literature as such. It is between language that primarily refers beyond itself and language whose own formal structure becomes semantically active.

Why Poems Resist Translation

Once form becomes meaningful, translation faces a fundamentally different task. Reproducing the informational content of a text is often possible through paraphrase or explanation. Reproducing the formal relationships that generate poetic effects is another matter.

Jakobson illustrates this with the Italian aphorism Traduttore, traditore. Its force depends not only on what the words denote but also on their striking phonetic resemblance. Translating the phrase as "the translator is a traitor" communicates the basic idea while sacrificing the paronomasia that makes the original memorable.

The same difficulty appears throughout literature. Shakespeare's dying Mercutio promises that tomorrow he will be "a grave man," a line that simultaneously means "serious" and "buried." Hamlet's remark that he is "a little more than kin, and less than kind" depends on sound as much as sense. These effects cannot simply be transferred into another language because they arise from the particular verbal resources of English itself.

This is precisely what Jakobson means when he writes that "phonemic similarity is sensed as semantic relationship" (1959, p. 238). In poetry, words resemble one another not accidentally but meaningfully. Their sounds become part of the semantic architecture of the work. Alter the sounds, and one alters the poem.

Creative Transposition

“But do we not generally regard as the essential substance of a literary work what it contains in addition to information — as even a poor translator will admit — the unfathomable, the mysterious, the "poetic”, something that a translator can reproduce only if he is also a poet?” — Walter Benjamin, The Task of the Translator

Jakobson's conclusion is therefore less pessimistic than it first appears. Immediately after declaring that poetry is "untranslatable," he adds that "only creative transposition is possible" (1959, p. 238). This qualification is crucial.

The translator's task is not to reproduce the original mechanically but to create a new verbal structure capable of producing analogous effects within another language. Translation becomes an act of composition rather than duplication.

Here Jakobson unexpectedly approaches Walter Benjamin's famous claim that "translation is a form" (Übersetzung ist eine Form). Benjamin argues that a literary translation should not be understood as a secondary copy but as a new form through which the original continues its life. Jakobson arrives at a similar conclusion from a different direction. Because poetic meaning is inseparable from verbal form, translation cannot simply transfer that form intact. It must invent another one. The translator of poetry succeeds not by preserving every linguistic feature but by producing a work that functions poetically in its own language.

Conclusion

Jakobson's distinction between cognitive discourse and poetry is ultimately a distinction between two ways meaning inhabits language. In cognitive communication, words primarily point beyond themselves toward the world they describe. Their form can change without fundamentally altering what is conveyed. In poetry, however, the form of language itself becomes part of the message. Sounds, rhythms, repetitions, and verbal patterns cease to be transparent vehicles and acquire semantic force of their own.

From this perspective, Jakobson's apparently contradictory claims no longer conflict. Cognitive meaning is always translatable because it can be reformulated through other linguistic resources. Poetry resists literal translation because its meaning is embodied in its verbal form. What survives translation is therefore not an identical linguistic object but a creative reconfiguration. Translation remains possible, but in poetry it necessarily becomes an act of invention as well as interpretation.

References

Jakobson, R. (1959). On linguistic aspects of translation. In R. A. Brower (Ed.), On translation (pp. 232–239). Harvard University Press.

Nida, E. A. (1964). Toward a science of translating. Brill.

Steiner, G. (1975). After Babel: Aspects of language and translation. Oxford University Press.

Venuti, L. (2018). The translator's invisibility: A history of translation (3rd ed.). Routledge.