The Wikipedia on Roman Jakobson says

The true hallmark of poetry is according to Jakobson "the projection of the principle of equivalence from the axis of selection to the axis of combination".

I think I know that the "syntagm" is an ordered series of units of the poem (be that a syllable, sound, word, etc.), that syntagms make the "axis of combination", and that the "axis of selection" is what we say and how we say it, in so far as usually, outside poetry, we can select a different word.

"syntagmatic axis" is along the bottom of a sentence diagram, while "paradigmatic axis" is along the side. The sentence "a boy cried" can be read left-to-right, while columns provide options which change the meaning but fulfill the same grammatical function, such as substituting "boy" for "man" or changing the exact ending verb.

Image from "Paradigms and Syntagms" by Daniel Chandler

My question is quite specific, and I don't want to read another non-specific explanation of Jakobson on this.

What does the equivalence to the axis of combination mean? Does it mean that in e.g. "a rose is a rose is a rose" the word "rose" is an equivalent "token", and means the same thing each time it appears? Or is the word "rose" an equivalent "type", so that it is an occurrence of something that means the same thing? Or are both characterisations wrong?

2 Answers 2


For those unfamiliar with Roman Jakobson, the correct quote is (with italics from the source)

The poetic function projects the principle of equivalence from the axis of selection into the axis of combination.

The quote comes from Jakobson's influential closing statement "Linguistics and poetics" at the conference on "Style in Language" held at Indiana University in 1958. It was revised and published in the conference proceedings Style in Language, edited by Thomas Sebeok, in 1960. The noun "hallmark" does not occur in Jakobson's text.

At the beginning of his talk, Jakobson identifies six basic functions of verbal communication, including the referential, the metalingual and the poetic function. The poetic function is not limited to poetry; it is also used, for example, in "advertising jingles" and versified scientific treatises. But only in poetry does it have a "coercing, determining role" (Jakobson).

To turn to the question more specifically, there is no such thing as an "equivalence to the axis of combination". The type of equivalence that Jakobson sees at work on the paradigmatic axis is based on sound, metre and rhythm. He describes "equivalence" as follows:

In poetry one syllable is equalized with any other syllable of the same sequence; word stress is assumed to equal word stress, an unstress equals unstress; prosodic long is matched with long, and short with short; word boundary equals word boundary, no boundary equals no boundary; syntactic pause equals syntactic pause, no pause equals no pause. Syllables are converted into units of measure, and so are morae or stresses.

So Jakobson does not mean that the semantically based selection that works on the paradigmatic axis (roughly an equivalence in meaning) now also functions on the syntagmatic axis. The equivalence on the syntagmatic axis is based on sound; examples of equivalence in sound include rhyme and metrical patterns. Caesar's famous quote "Veni, vidi, vici" is an example: three disyllabic words (one type of equivalence based on syllable counts), each stressed on the first syllable and beginning with the same consonant (alliteration) and ending with the same vowel (a rhyme).

With this in mind, the word "rose" in "a rose is a rose is a rose" is an example of sound-based equivalent, but of an easy sort, because the word is simply repeated. "A rose is a rose, I s'pose" would also achieve equivalence on the paradigmatic axis. The verb "s'pose" is not part of the same word class as "rose", nor is it semantically related, so it is not even on the same paradigmatic axis as "rose". In that sense, it is not of an "equivalent type" as "rose". I don't think C. S. Peirce's distinction between types and tokens, if that was an intentional reference by the question, is helpful here.

  • I can't follow your reasoning at all. it may be me being dim, but I don't see it
    – user5641
    Jun 19, 2022 at 22:14
  • unless I've misread you, you need to show both that equivalence is only phonemic and that the axis of combination does not consist of phonemes. I would suggest both claims are false! how eliding the 'in' of 'into' makes the statement false is also completely beyond me..
    – user5641
    Jun 19, 2022 at 22:18
  • Rhyme can be described as equivalence of phonemes on the paradigmatic axis. Why anyone would need to show that the paradigmatic axis does not consist of phonemes is beyond me, since there is no speech without phonemes. (There are larger units than phonemes on the paradigmatic axis, but these can be broken down into smaller units, including down to the level of phonemes.)
    – Tsundoku
    Jun 20, 2022 at 10:23
  • why is there no such thing as an "equivalence to the axis of combination"? I cannot follow what you are saying, sorry. what is equivalent in poetry?
    – user5641
    Jun 21, 2022 at 11:48
  • There are two axes, the syntagmatic one ("combination") and the paradigmatic one ("selection"). These are not equivalent to each other, as the image you copied from Daniel Chandler's page shows. There is equivalence on each axis, not to them.
    – Tsundoku
    Jun 21, 2022 at 13:11

equivalence means that at least in one characteristic the two elements are similar, and in at least one characteristic they are opposite. Therefore, identity is not equivalence, because it lacks the opposition requirement. If you have several occurrences of "rose", they could, however, form an equivalence if in the respective context they contain an opposition, like in, for example, "the thorny rose, the tender rose".

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.