Start Submission

Reading: A Heretical Defence of the Unity of Form and Content


A- A+
Alt. Display

Research Article

A Heretical Defence of the Unity of Form and Content


Daniela Glavaničová

Slovak Academy of Sciences, v.v.i., SK
X close


The received view in the debate on the form–content unity of poetry is that the possibility of paraphrase does not sit well with the unity conception. I will suggest a shift from paraphrase to translation, since the latter is substantially closer to the heart of the matter. I will heretically divert from the ‘commonplace’ view, which claims that poetry cannot be translated. However, I will argue that the possibility of translation in this sense can be reconciled, appearances notwithstanding, with the unity of form and content. A further surprising conclusion will be that, while this possibility prima facie appears to be the best argument against the unity, the contrary is the case.

How to Cite: Glavaničová, Daniela. “A Heretical Defence of the Unity of Form and Content”. Estetika: The European Journal of Aesthetics 59, no. 1 (2022): 33–47. DOI:
  Published on 15 Mar 2022
 Accepted on 03 Feb 2021            Submitted on 15 May 2020

I. The Unity of Form and Content

The unity of poetic form and poetic content has been the subject of extensive debate in the contemporary philosophy of literature. A crucial piece of motivation for the whole debate is that what is at stake is the value of poetry. Peter Lamarque rightly observes that the originator of this debate, A. C. Bradley, claims that the value of poetry lies in the unity of poetic form and poetic content: ‘It is absurd to ask whether the poem’s value lies in its form or its content; if we are to value the poem properly, as a poem, we must value it as a form-content unity.’1 Bradley argued for the unity of form and content in poetry in his inaugural lecture ‘Poetry for Poetry’s Sake’, delivered in 1901. The unity thesis has appeared in manifold formulations since then. In a nutshell, it claims that the content of a poem is (or should be) understood as inseparable, indistinguishable, and united with, if not identical to, its form.

The recent debate moves back and forth between the value of poetry and the possibility of paraphrasing it. Contemporary adherents of the unity thesis, such as Lamarque, Rafe McGregor, and Owen Hulatt (as well as Bradley himself), focus predominantly on poetry’s value.2 By comparison, the main opponent, Peter Kivy, focuses largely on the possibility of paraphrase, allowing the value of poetry to play second fiddle.3 Kivy’s main interest in the possibility of paraphrase seems to be connected to the possibility of interpretation (‘to deny that poetry can be paraphrased is ipso facto to deny that it can be interpreted’).4 In opposition to Bradley, he argues that we should not set the criterion of success of paraphrasing poetry so high that it is in principle impossible to ever meet the criterion: one should not ‘resolve’ the controversy by stipulation.5

Interest in the issue of paraphrase rests mainly on the received view that the possibility of paraphrase does not sit well with the unity thesis. Because of this, showing that it is possible to paraphrase poetry is usually seen as an argument against the unity thesis, while demonstrating its impossibility is viewed as an argument in its favour. I will argue that the possibility of paraphrase is not as relevant to the unity of form and content as the possibility of translation is. My suggestion is to reconceive the debate in terms of the possibility of translation.

Subsequently, leaving the issue of paraphrase behind and focusing instead on translation, I will propose a positive argument for the claim that we can and should translate poetry in an important sense of ‘can’ and ‘should’ (one can also proceed in a negative manner, that is, it is possible to attack arguments derived from the thesis of the untranslatability of poetry).6 The present approach thereby diverts from the ‘commonplace’ position, which suggests that translating poetry is heresy since, as Ernie Lepore puts it, ‘poetry cannot be paraphrased (or translated)’.7 But this diversion is not as radical as it may appear. I will show that the possibility of translation in this sense can be reconciled, appearances notwithstanding, with the unity of poetic form and poetic content. A further surprising conclusion will be that, while this possibility prima facie appears to be the best argument against the unity thesis, the contrary is the case.

The structure of the paper is as follows. I provide a brief overview of the recent debate in Section II. In Section III, I discuss three issues that are central to the form–content unity debate. Section IV motivates the move from the issue of paraphrase to the issue of translation. The possibility of translation is discussed in Section V. The question of this possibility generates a dilemma, since both options seem to be right (in a sense) but also wrong (in another sense). My proposal opts for a middle way: poetry is translatable, but in a sense that is reconcilable with the unity thesis (and in that sense only). The last section develops the even more radical conclusion that the possibility of translation in this sense has, as a consequence, an important advantage for the unity thesis. I also conclude that this approach gives pride of place to our precious practice of translating poetry, and in addition it offers a reasonable measure for evaluating respective translations as translations. The focus on practice will be, in spirit, close to Lamarque’s general approach in aesthetics.

II. The Form–Content Unity Debate

In this section, I will provide a brief roadmap of the debate on the unity of form and content. Admittedly, some might be sceptical of the very debate. However, several authors suggest that the value of poetry is at stake: Bradley construes the form–content debate as a debate about where the poetic value lies and as ‘a controversy which concerns the ultimate nature of Art’; this interpretation of Bradley is suggested by Lamarque: ‘As I see it, Bradley’s central concerns are less about paraphrase, more about value’; moreover, McGregor takes the form–content debate to be central for the value of literature; as he puts it, the ‘argument for the sui generic value of literature will draw on the ancient distinction between literary form and literary content and will offer a fresh perspective on the relationship between the two, which I call literary thickness’.8 For this reason, I consider this debate worth (re)considering. This section is devoted mainly for the readers who are not aware of this debate (and can be light-heartedly skipped by those who are). A discussion between Lamarque and Kivy will be at the centre of my attention. Time and again, I will mention other interesting contributions to the debate that are relevant for the present paper.

Bradley formulated and defended the form–content unity thesis as a claim about the intrinsic value of poetry that avoids two positions that he viewed as extreme: formalism and content-reductionism. The lecture has been reprinted many times since its first publication in 1901 and has led to extensive debate. Importantly, Bradley discusses the problem of translation, claiming that a translation of true poetry is not the old meaning in a fresh dress – it is a new product.9

The more recent stage of this debate began almost a century later, when Kivy criticized Bradley’s attempt, viewing it as an extreme conclusion.10 He admitted that there is a special intimacy of form and content, but he maintained the distinction between them (I will return to this relation of ‘special intimacy’ in the subsequent section). In other words, he claimed that poetic form is separable from poetic content. As a consequence, poetry can be paraphrased. The shift from the poetic value to the possibility of paraphrase has already taken place.

More than a decade later, the criticism provoked responses from Kelly Dean Jolley and Lamarque. Jolley criticized Kivy for misunderstanding Bradley, as well as for Kivy’s Janus-faced approach, which resulted in his ultimately doing what he was trying to avoid (that is, defining poetry or defining good poetry in terms of the special intimacy mentioned above).11 Lamarque provided a complex defence of (his reformulation of) Bradley’s thesis of form–content unity against Kivy’s criticism.12

Subsequently, Kivy criticized Lamarque’s defence, switching the attention back to paraphrase. In opposition to Lamarque, Kivy argues that Lamarque’s defence of the claim that reading poetry as poetry demands form–content unity is inadequate; this is so because it would require the unbroken tradition: however, as Kivy argues, there is no such unbroken tradition.13 In fact, he suggests, there are many different ways of reading poetry as poetry (we read poetry as poetry by contrast to reading it, for instance, as a report); for Kivy, however, form–content unity is just one of numerous ways of reading poetry as poetry.14 Kivy’s main thesis in his response to Lamarque is that ‘poems that have meaning can be paraphrased; that, in other words, one can express, in other words, the meaning the poem expresses’.15 Claiming otherwise is also a denial of the possibility of interpretation, Kivy worries. Two issues need to be noted here. First, what Kivy claims is, strictly speaking, a truism for Lamarque: ‘it is misleading to say that poems as such (as opposed to parts of poems) have meaning at all (the expression “the meaning of the poem”, taken literally, never picks out something)’.16 In other words, if no poem has a unique, specifiable meaning, then anything holds trivially for any poem that has ‘meaning’. To see this, consider the claim that any poem that has meaning can be paraphrased. If there is no such poem (as Lamarque seems to suggest in the above quotation, where he says that ‘the expression “the meaning of the poem”, taken literally, never picks out something’), then any universally quantified proposition starting with this unfulfilled condition is trivially true (in the same way as any material implication with a false antecedent). Second, Bradley had already noted that the unity thesis should not be understood as an attack on the possibility of interpretation.17 While not focusing primarily on paraphrase, when adherents of the unity thesis claim that poetry cannot be paraphrased, that is not meant to suggest that we cannot deliver a ‘literal rendition’ of one of a poem’s possible linguistic meanings. What is suggested is that no such paraphrase can deliver a poetic experience equal to the experience afforded by the original poem. However, this should be acceptable for Kivy, since he claims that ‘no one who sets out to say in prose the content of what a poem says in poetic form intends as the goal of the task to provide an alternative way of experiencing the poem’.18

A few years later, Lamarque’s take on the issue was defended by McGregor, Hulatt, and Lamarque.19 These defences share the focus on poetic value, and the focus on poetry as poetry. In addition, the problem of translation is not discussed, which is an absence that is shared with Kivy’s view (a notable exception is McGregor).20 Interestingly, McGregor extends the unity thesis to the literature as such.21 Admittedly, that is a controversial suggestion. Nevertheless, it opens the debate to further generalizations: is it possible to extend the thesis to other categories of art? Is it possible to extend it to all categories of art? If the affirmative answer is correct, the thesis has an even greater importance for aesthetics than originally thought.

III. Three Issues

Unsurprisingly, there are three notions in the debate on form–content unity that are in need of clarification:

  • – the form;
  • – the content; and
  • – the relation between the two.

When it comes to form, authors face serious problems when trying to offer positive specifications. A good starting point is a negative specification of form as what remains when content is taken away.22 One can continue with a list of the formal properties of the poem – such as ‘style and versification’ or ‘rhyme, metre, rhythm, alliteration, repetition’ – which is a non-starter for both Bradley and Lamarque, however, because for them form is indivisible from content.23 Lamarque suggests that we define form as the-mode-of-realization-of-the-subject-in-the-poem, thereby mentioning ‘the subject’ in the definition of form in terms of ‘the mode of realisation’.24 We will return to this issue presently.

The notion of content is no less problematic. Bradley suggests that it can be either something in the poem or something outside the poem. The former is an option, but there is, according to Bradley, no such pure content (substance) that can be separated from the form it assumes in the poem.25 Lamarque defines content as the-subject-as-realized-in-the-poem.26 Again, note that this definition contains the term ‘as realized’, which seems to be nothing other than the mode of realization. I will discuss this below. First, however, let me return to the last issue.

As regards the relation between form (whatever it is) and content (whatever it is), possible candidates are as follows: identity, indissoluble fusion, unity, inseparability, special intimacy, and so forth. Bradley holds that the relation between the two is identity, unity, or indissoluble fusion. Strictly speaking, form and content are not two things for Bradley; they are just two aspects of one thing.27 McGregor is not sympathetic to this identity talk, objecting that ‘if the relation was one of strict identity for Bradley, then form and content would be identical in both the experience and analysis of a poem’.28 He aptly summarizes other options, such as being inseparable or indistinguishable, indivisible, or mutually dependent. As noted in the preceding section, an opponent of the unity thesis, Kivy, is willing to accept that there is a special intimacy between the two and that the mode of presentation tells us something crucial about the (presentation of) content: ‘we have a deep intuition that in the arts there is an especially intimate relation between form and content not exhibited in other modes of expression’; Kivy suggests that the intimate relation between form and content specific to the arts is not that of identity and that it varies across arts.29

Let us now return to Lamarque’s definitions. He proposes that the content of a poem is the-subject-as-realized-in-the-poem and that the form of a poem is the-mode-of-realization-of-the-subject-in-the-poem. The content is defined in terms of the subject, and in terms of how it was realized. However, the latter seems to be nothing other than the mode of realization. And what we find next is form being defined as the mode of realization of the subject, the very term that was employed in the definition of content. These definitions are therefore intertwined in the notion of a mode of realization and the notion of a subject.

Is this a vice if the thesis holds? No. The two are inseparable, and thus it makes little sense to define them separately. Is this a vice if the thesis does not hold? Yes. If the two can be meaningfully separated from the whole (even when poetry is read as poetry), definitions that do not presuppose the unity thesis should be preferred.

Because of this, I will move to more neutral territory and speak of the content of a poem as the subject in the poem and of form as the mode of realization of the poem. One might smell something fishy in my attempt to reframe the definitions provided by Lamarque. First, one may worry that I have shifted from presupposing the unity thesis to presupposing its contrary. However, this is not the case. The above definitions separate form from content, but they do not suggest that we should separate them when reading poetry as poetry (that is, when we are interested in the poem itself; there are other ways of reading poetry: we may read a poem to learn something about an event described in the poem, about the author, the country, and so forth; we may read a poem as a philosophical or theological work; we may read for fun; we may read to interpret or analyse, and so forth). Second, while Lamarque defines form as the mode of realization of the subject in the poem, I suggest that we speak of the mode of realization of the poem itself. It seems that this is a fair move for me to make. After all, we are talking about the form (style, versification) of a poem, not the form (again, style, versification) of the subject.30

Another reason for this move is that, if form is the mode of realization of the subject in the poem, this definition not only mentions the subject in the poem but it also presupposes its existence. Again, this is fine if the form–content identity thesis holds, but not fine if it does not. The issue is that such a definition excludes the possibility of nonsensical poems possessing some mode of realization while lacking a subject. Lewis Carroll’s ‘Jabberwocky’ or Christian Morgenstern’s ‘Das große Lalula’ are possible (albeit controversial) examples.

I take it that appearances can be misleading in cases like this, however, for one can maintain that the very form generates (or even is) the subject of such poems. But then it is virtually impossible to substitute the definiens of the definition of form (the-mode-of-realization-of-the-subject-in-poem) for the term ‘the subject’ in the definition of content (the-subject-as-realized-in-the-poem): the mode of realization of the subject in the poem as realized in the poem. The problem is that the term ‘subject’ reappears, and calls for another substitution, in which the subject will reappear again, and so forth. While saying that the subject as realized in the poem is the mode of realization of the subject in the poem seems to be a confused finding in the case of ‘Das große Lalula’, saying that the subject in the poem is (partially or wholly) the mode of realization of the poem seems to be informative.31

Note, however, that the above definitions are not meant to be reductive (in the sense of reducing the various elements of poetic content to a single element). However, they are not meant to be very informative either. Following McGregor’s starting point that ‘form is how a poet says something; content is what the poet says’, the subject in the poem can be understood as what the poem says, and the mode of realization of the poem can be understood as how the poem says something.32

To conclude, I will make use of these neutral definitions of poetic form and poetic content. In addition, I will focus largely on the modal status of the unity thesis. This will pave the way for the possibility of translation (in an important sense of the word ‘possible’) and for its interesting relation to the unity thesis.

IV. From Paraphrase to Translation

Recall Kivy’s note that ‘no one who sets out to say in prose the content of what a poem says in poetic form intends as the goal of the task to provide an alternative way of experiencing the poem’.33 The main reason why the possibility of translation (and not the possibility of paraphrase) is central to the present debate is hidden in this important observation.

This reason is the fact that there is a crucial disanalogy between the purpose of paraphrase and that of translation. The purpose of paraphrase is not to replicate or mimic the poetic experience afforded by the original poem, while a translation aims precisely at this. The goal of paraphrase may be to retell the poem in other words, to enhance the experience when (re)reading the poem, to understand the poem better, to analyse the poem, to expose a possible meaning or interpretation, to resolve whatever ambiguities the poem contains, to offer a possible clarification, and so forth. Paraphrase is not meant to be a substitute for the original poem and paraphrase does not aim to deliver an aesthetic experience equal to the original poem, though it may very well help to enhance this experience.

And, because of this, it is quite easy for the form–content unity defenders to win the argument: no paraphrase can deliver the same experience as the original poem, for that is not its purpose to begin with.

In comparison, the purpose of translation consists precisely in attempts, however hopeless they may be, to replicate or mimic the poetic experience of the original or to replicate relevant features of the poem.34 In addition, they do not aim at clarification, explanation, or disambiguation.

The usual aim of paraphrase thus radically differs from the usual aim of translation, even though, if paraphrase of a poem is understood simply as saying what the poem says in different words, translation can be understood as a specific case of paraphrase. But, even if that is the case, my suggestion is that this specific case is more relevant to the issue of unity of form and content than paraphrase as such. This is so if I am correct that the standard aim of translation is to replicate or mimic the poetic experience of the original, or to replicate relevant features of the poem.

The possibility of translating poetry thus seems to be more important in the context of the present debate. Of course, the possibility of translation is a central topic for poets themselves and for literary theory, as is well documented by Angela Leighton.35 Notably, Bradley, Jolley, Lepore, and McGregor briefly mention the possibility of translation in the very context of the present debate.36 Nonetheless, it seems that its interest for the present topic has yet to be exhausted. This is so because it has been very briefly discussed in the context of the unity debate, without any emphasis on the above disanalogy, as if the possibility of paraphrase and the possibility of translation were equal – and equally relevant – phenomena.

V. Lost in Translation?

Do not get me wrong here. As indicated above, the possibility of translating poetry is a complex and extensively debated subject. I will not even attempt to squeeze it into the present paper. However, let me note a few brief ‘truisms’ on our practice of translating poetry:

  1. There is an ‘unbroken tradition’ of translating poetry and it makes sense to evaluate translated poems as translations (as compared to evaluating them as poems).
  2. This tradition is valuable, and we do not want it to stop; quite the contrary, we want to encourage it.
  3. Something is (has to be) preserved in translations of poetry.
  4. Something is (has to be) lost in translations of poetry.
  5. Some translations of poetry are better than others (as translations): they are comparable.

To begin with, as stated in (1), there is a long, ongoing tradition or practice of translating poetry. However, Bradley suggests that

in true poetry it is, in strictness, impossible to express the meaning in any but its own words, or to change the words without changing the meaning. A translation of such poetry is not really the old meaning in a fresh dress; it is a new product.37

Similarly, but more generally, McGregor argues that a translation is a different work of literature (consequently, a translation can be a more valuable work than the original).38 Indeed, translation of poetry is a deeply creative endeavour and translators are more often than not poets in their own right. However, the ties to the original are not lost: translation is always a translation of something. If someone brilliantly translates a poem and tries to publish it as his or her own poem, not acknowledging the original, however tempted we may be to praise the translator as a translator, we cannot praise the translator as an original author. Likewise, if someone loosely translates a horrible poem into a new masterpiece, we are likely to praise that person mainly as an author.

As (2) adds, we value the tradition of translating poetry. We do not want to prohibit it. But, if the translation of poetry really were so futile, hopeless, desperate, monstrous, and outright impossible, then why bother?39 As Jolley hypothetically puts it, ‘if a sentence or a poem is uniquely effable, if only a particular sentence or poem can say what it says, then there is no possibility of paraphrase or of translation’.40 Yes, ‘in strictness’, this is all true about translating poetry. But we do bother, and we should bother. The tradition is valuable for many reasons: translation helps to preserve works, to disseminate them, to make them accessible to foreign audiences, to open new discussions and interpretations, and so on.

Putting (1) and (2) together, there is a real question about how strict we should be. We can be so strict that almost no one will have access to the poem, by, for instance, requiring that readers be familiar with all of the connotations of the words in a poem. This is similar to positing overly strict identity conditions of works (for instance, insisting that only the first, original token of a poem is the poem, or that a slight change in font destroys the fragile identity of a collection of poems) or fictional characters (for instance, insisting that a fictional character in a translated work of fiction is different from the original, or that a fictional character mentioned on one page is different from the apparently same character mentioned on the very next page). Also, something similar to what Lamarque says about the translation of literature (as a possible objection against the textualist conception of works) can be applied here: ‘Strictly speaking, no one reading only the translation has read the work.’41 Reusing his objection, we may say that ‘this might seem counterintuitive and would have the consequence that far fewer people have read’ Baudelaire’s poetry ‘than is normally supposed’.42 Indeed, this seems rather counterintuitive to me, a great admirer of Baudelaire’s poetry who has no competence in French.

I agree with McGregor that we may have good reasons for treating a translation as a different work of literature.43 However, I claim that we may have equally good reasons for the opposite. This is possible because, as Lamarque persuasively argued, identity conditions in the aesthetic domain can be seen as context-sensitive and interest-relative.44 If we are interested in crediting the translator, we may well opt for McGregor’s suggestion. However, if we are interested in giving the credit to the original author, or in opening the door to the author’s poetry to a foreign audience, I would advise against cutting all ties with the original.

Connecting the above with (3), which claims that something is (has to be) preserved in translation, we can already see the sense in which poetry ‘can’ be translated, contrary to the commonplace view, which treats it as ‘heresy’. Usually, much content can be preserved, along with some degree of form, and, ideally, the original unity of form and content finds its counterpart in a parallel unity of form and content of a translated poem.

But let me comment on (4) too. This ‘truism’ claims that something is (has to be) lost in translations of poetry. It is precisely the addition of this issue to (1), (2), and (3) above that makes the general question of the possibility of translation a dilemma. All of the above builds a case for the possibility of translation. But (4) crucially restricts this possibility, since something of the form–content unity is lost. Further evidence for this claim can be provided by the survival test employed by Lamarque: imagine that an original poem is irretrievably lost but many translations of it remain. Or imagine that the language of the original poem is irretrievably lost.45 Surely, we would be tempted to say that the poem as such is lost. However, one may also argue in the opposite direction. Imagine that the original poem has been translated directly into many other languages (some of which are similar to the language of the original), and also that there is enough evidence to establish that the form of the original poem was rather simple (for instance, a certain rhyme scheme, but no other artistic interplay of formal features is present), and that the work is original in terms of its (supposedly translatable) metaphors, ideas, views, and so on. It then seems that one can say that the work has survived. But then it is in principle possible to count some translations as variants of the original rather than as new poems; that is, even if something is always lost in translation, it does not follow that we have to count all translated poems as new works.

Finally, allow me to comment on (5), which has it that translations of poetry are comparable. Comparing the original to a translation, Bradley laments that ‘the charm of the original has fled’, and it indeed often has.46 Translating poetry is a balancing act: an attempt to preserve form (the mode of realization of the poem) without compromising content (the subject in the poem), and the other way around. As Lepore puts it, ‘by focusing on word meaning, a translator retains original meaning, which invariably gets twisted when beat, rhyme, meter, and the like are respected’.47 But not all translations are charmless, not all translations fail to pay attention to form, and not all translations twist meaning considerably. As our truism says, translations can be compared. There is a plenitude of comparative literature studies to motivate this claim. Let me illustrate this briefly. Consider the last stanza of Baudelaire’s ‘L’Ennemi’ and two of its English translations. The original reads as follows:48

— Ô douleur! ô douleur! Le Temps mange la vie,

Et l’obscur Ennemi qui nous ronge le coeur

Du sang que nous perdons croît et se fortifie!

A translation by Jack Collings Squire runs as follows:49

O misery! misery! Time eats our lives,

And that dark Enemy who gnaws our hearts

Grows by the blood he sucks from us, and thrives.

A translation of the very same text by Robert Lowell (surprisingly entitled ‘The Ruined Garden’ rather than ‘The Enemy’) proceeds as follows:50

Time and nature sluice away our lives.

A virus eats the heart out of our sides,

digs in and multiplies on our lost blood.

The former translation is arguably better than the latter: the linguistic meaning is captured fairly well, Baudelaire’s desperate cry is preserved (O misery! misery!), the enemy is capitalized, and the rhyme scheme is preserved. The latter does not mention an enemy at all (not even in the title) and instead refers to a virus. As such, we cannot even begin to ask who this mysterious enemy is, a question that has bothered many.51 Neither the rhyme scheme nor the cry is preserved, nature appears out of the blue, and so on.

By all this I am not suggesting that, taking any two translations, we can conclude which is better and which is worse. But there are some methods of comparison (for example, if we are evaluating translations of form as translations, then form-derivative forms – mimetic or analogical – are to be preferred over content-derivative or extraneous forms).52

All this being said, both the possibility of translation and its impossibility appear to have a grain of truth to them (but also a grain of falsity). A way out of this dilemma can be found in Kivy’s demand that we be more precise with what counts as success53 and in Lamarque’s interest-relativity of work identity. One can resolve this dilemma once one precisely specifies what counts as a successful translation of poetry. Is it a fairly poor translation – one that conveys nothing but a reasonably similar message (as a paraphrase might do)? Is it a fairly good translation – one that offers a decent balance between sacrificing form here and there and sacrificing content here and there? Is it the perfect translation – one that offers precisely the same form–content unity (which would be impossible)?

If the interests of the non-native audience are to be respected (as I think they should be) and if the tradition of translating poetry is to be acknowledged and encouraged (as I think it should be), then the possibility of translating poetry should be admitted, at least in one sense of the word ‘possible’. My hypothesis is that the poetic form and the poetic content can be fairly well preserved in translations of (at least some) poems. And these poems are fairly well translatable: you cannot have the whole cake, but you can have a reasonably sized piece.

In terms of interest-relativity and context-sensitivity, the present argument is that there are such interests and such contexts where a translated poem counts as an acceptable variant of original, rather than a different poem. For instance, consider a literature class in a Slovak high school where students – who are rarely fluent in French – are required to read and interpret a poem by Baudelaire. Such interests and contexts where reading a translation of a poem counts as reading the poem itself are very common in practice. Owing to interest-relativity and context-sensitivity, this is sufficient for showing that it is possible to translate poetry in an interesting sense of the word ‘possible’.

Further support for the claim is that claiming otherwise undermines the very practice of translating poetry. That is, if translating poetry is impossible, it is not clear why we should continue our futile efforts. In other areas of human endeavour, we usually stop doing things that we learn to be principally impossible to realize.

Admittedly, our successful and valuable practice of translation is not denied in the discussion on the unity thesis. In other words, it is not denied that the translation of poetry is valuable. Rather, the claim is that a translated poem is not the same poem as the original. As a consequence, I cannot read Baudelaire’s work unless I learn French. According to the claim, ties with the original are severed in translation and the translated poem is a new work of art. However, it seems that, if such a claim is accepted, our existing successful practice of translation is traded for the ‘perfect’ notion of translation, which is so strict that it makes translation flat out impossible. My suggestion is to adopt a more realistic notion of translation.

VI. A Heretical Defence

To conclude, let me note that, while I (heretically) claim that poetry is – in a sense – translatable, it is translatable in a sense that is reconcilable with the unity thesis (and in that sense only). In this way, one can maintain the core of the unity thesis while giving the practice of translating poetry the credit it deserves.

An even more radical conclusion is to come – namely, that the possibility of translation in this sense has an important benefit for the unity thesis as a consequence. While it seems to be impossible to perfectly (or ‘in strictness’) preserve the unity of poetic form and poetic content in translation, this impossible goal seems to be what translators of poetry as poetry are up to.

This can be seen as further evidence for the unity of form and content: the identity of the poem (‘in strictness’) is given by the unity of form and content. The translator aims (per impossibile) at delivering the same poem in a different language. However, even though this goal is impossible (and so is perfect translation), it is possible to follow this goal, and it is possible to deliver imperfect but still very good translations.

And the pursuit of precisely this goal is what has delivered the best translations we have. I am speaking of translations that do not merely provide a literal word-for-word translation but which give due attention to form. Any translation of Apollinaire’s rain-shaped poem that presents it in shapeless plain text simply ignores its beautiful form and is clearly not an instance of the best of our available translations evaluated as translations. Equally inadequate are translations of verse into prose. The same goes for any creative twist of the original content that shows off the translator’s poetic proficiency. However good a translated poem may be on its own, it is deficient as translation if the original content is ignored in a substantial way.

What I am suggesting is that good translations of poetry (at the very least) do and should count as translations of poetry. Moreover, since such translations do actually exist, it is possible that they exist (since actuality is usually understood as implying the possibility).

What is more, this approach offers a reasonable measure for evaluating translations as translations. Imagine a graph of an inverse square function, advancing towards the asymptote from both sides but never intersecting it. My hypothesis is that both curves (‘translating poetic content’ and ‘translating poetic form’) are asymptotic. This is so because, even if one largely ignored the form (focusing on content), some connotations of some words would differ; even if one largely ignored the content (focusing on form), some words would look or sound differently; and some formal poetic properties would not be captured due to the differences between the two languages. Granted, it is, ‘in strictness’, impossible to translate poetry: the only possible intersection would involve mere repetition. But it is possible to translate poetry in another, less strict but very important sense: it is possible to move towards the natural equilibrium of the ideal – towards the original unity.


1Peter Lamarque, ‘The Elusiveness of Poetic Meaning’, Ratio 22 (2009): 407. On the value of poetry and, more generally, on the value of literature, see also Rafe McGregor, The Value of Literature (Lanham, MD: Rowman & Littlefield, 2016). 

2Lamarque, ‘Elusiveness of Poetic Meaning’; ‘Poetry and Abstract Thought’, Midwest Studies in Philosophy 33 (2009): 37–52; Rafe McGregor, ‘Poetic Thickness’, British Journal of Aesthetics 54 (2014): 49–64; Owen Hulatt, ‘The Problem of Modernism and Critical Refusal: Bradley and Lamarque on Form/Content Unity’, Journal of Aesthetics and Art Criticism 74 (2016): 47–59; Andrew Cecil Bradley, ‘Poetry for Poetry’s Sake’, in Oxford Lectures on Poetry (London: Macmillan, 1923), 3–27. 

3Peter Kivy, Philosophies of Arts: An Essay in Differences (Cambridge: Cambridge University Press, 1997); Peter Kivy, ‘Paraphrasing Poetry (for Profit and Pleasure)’, Journal of Aesthetics and Art Criticism 69 (2011): 367–77. 

4Kivy, ‘Paraphrasing Poetry’, 368. 

5Kivy, Philosophies of Arts, 104–6. 

6Daniela Glavaničová and Miloš Kosterec, ‘The Fine-Grainedness of Poetry: A New Argument against the Received View’, Analysis 81 (2021): 224–31. 

7Ernie Lepore, ‘The Heresy of Paraphrase: When the Medium Really Is the Message’, Midwest Studies in Philosophy 33 (2009): 177. 

8Bradley, ‘Poetry for Poetry’s Sake’, 9 and 8–13; Lamarque, ‘Elusiveness of Poetic Meaning’, 403; McGregor, Value of Literature, viii. 

9Bradley, ‘Poetry for Poetry’s Sake’, 17. 

10Kivy, Philosophies of Arts, 116. 

11Kelly Dean Jolley, ‘(Kivy on) The Form-Content Identity Thesis’, British Journal of Aesthetics 48 (2008): 203–4. 

12Lamarque, ‘Elusiveness of Poetic Meaning’; ‘Poetry and Abstract Thought’. 

13Kivy, ‘Paraphrasing Poetry’, 368–71. 

14Ibid., 371–72. 

15Ibid., 367. 

16Lamarque, ‘Elusiveness of Poetic Meaning’, 403. 

17Bradley, ‘Poetry for Poetry’s Sake’, 33. 

18Kivy, Philosophies of Arts, 105. 

19McGregor, ‘Poetic Thickness’; Hulatt, ‘Problem of Modernism’; Peter Lamarque, ‘Semantic Finegrainedness and Poetic Value’, in The Philosophy of Poetry, ed. John Gibson (Oxford: Oxford University Press, 2015), 18–36; ‘Philosophy and the Lyric’, Journal of Literary Theory 11 (2017): 63–73. 

20McGregor, Value of Literature, 80. 

21Ibid., ix. 

22Angela Leighton, ‘About About: On Poetry and Paraphrase’, Midwest Studies in Philosophy 33 (2009): 168. 

23Bradley, ‘Poetry for Poetry’s Sake’, 18; Lamarque, ‘Semantic Finegrainedness’, 27–28. 

24Lamarque, ‘Elusiveness of Poetic Meaning’, 407. 

25Bradley, ‘Poetry for Poetry’s Sake’, 16. 

26Lamarque, ‘Elusiveness of Poetic Meaning’, 407. 

27Bradley, ‘Poetry for Poetry’s Sake’, 14–15. 

28McGregor, ‘Poetic Thickness’, 50. 

29Kivy, Philosophies of Arts, 116; ‘Paraphrasing Poetry’, 376. 

30Bradley, ‘Poetry for Poetry’s Sake’, 18. 

31An interesting example of a Slovak poem, ‘Most’ (‘Bridge’) by Peter Macsovszky was suggested to me by Michal Šedík. The subject in this poem seems to be the very unity of form and content. However, a discussion of this example is for a later occasion. 

32McGregor, Value of Literature, 48. 

33Kivy, Philosophies of Arts, 105. 

34Anna Christina Ribeiro, ‘Toward a Philosophy of Poetry’, Midwest Studies in Philosophy 33 (2009): 61–77; ‘Relevance Theory and Poetic Effects’, Philosophy and Literature 37 (2013): 102–17. 

35Leighton, ‘About About’, 167–76. 

36Bradley, ‘Poetry for Poetry’s Sake’, 19–20; Jolley, ‘(Kivy on) The Form-Content Identity Thesis’, 199; Lepore, ‘Heresy of Paraphrase’, 177–97; McGregor, Value of Literature, 80–82. 

37Bradley, ‘Poetry for Poetry’s Sake’, 19. 

38McGregor, Value of Literature, 80. 

39Leighton, ‘About About’; Carol Jacobs, ‘The Monstrosity of Translation’, MLN 90 (1975): 755–66; Lepore, ‘Heresy of Paraphrase’. 

40Jolley, ‘(Kivy on) The Form-Content Identity Thesis’, 199. 

41Peter Lamarque. The Philosophy of Literature (Oxford: Blackwell, 2009), 73. 


43McGregor, Value of Literature, 80–82. 

44See Peter Lamarque, Work and Object: Explorations in the Metaphysics of Art (Oxford: Oxford University Press, 2010), where this idea is researched extensively throughout the book. 

45Ibid., 68–69. 

46Bradley, ‘Poetry for Poetry’s Sake’, 20. 

47Lepore, ‘Heresy of Paraphrase’, 185. This claim needs to be put in the context of Lepore’s claim that poetry is hyperintensional. Hyperintensional theories model meaning in a fine-grained manner. Poetry (like quotation) requires the finest-grained approach. For a hyperintensional theory of this kind, see Pavel Tichý, The Foundations of Frege’s Logic (Berlin: De Gruyter, 1988). We have explored the connections between hyperintensionality, unity of form and content, and untranslatability of poetry in Glavaničová and Kosterec, ‘Fine-Grainedness of Poetry’. 

48Charles Baudelaire, Les fleurs du mal et autres poemes (Paris: Garnier-Flammarion, 1964), 44. 

49Jack Collings Squire, Poems and Baudelaire Flowers (London: New Age, 1909), 41. 

50Jeffrey Yang, ed., Birds, Beasts, and Seas: Nature Poems from New Directions (New York: New Directions, 2011), 38. 

51Alan S. Rosenthal, ‘Baudelaire’s Mysterious “Enemy”’, Nineteenth-Century French Studies 4 (1976): 286–94. 

52James S. Holmes, ‘Forms of Verse Translation and the Translation of Verse Form’, Babel 15 (1969): 195–201. 

53Kivy, Philosophies of Arts, 104–6. 


I am grateful to Carolyn Benson, Miloš Kosterec, Peter Lamarque, Matteo Pascucci, Michal Šedík, Martin Vacek, and Marián Zouhar. I am also indebted to the reviewers of Estetika for their helpful comments. This paper was supported by VEGA grant No. 2/0117/19 Logic, Epistemology and Metaphysics of Fiction and by the University of Oxford project ‘New Horizons for Science and Religion in Central and Eastern Europe’ funded by the John Templeton Foundation. The opinions expressed in the publication are those of the author and do not necessarily reflect the view of the John Templeton Foundation.

Competing Interests

The author has no competing interests to declare.


  1. Baudelaire, Charles. Les fleurs du mal et autres poemes. Paris: Garnier-Flammarion, 1964. 

  2. Bradley, Andrew Cecil. ‘Poetry for Poetry’s Sake.’ In Oxford Lectures on Poetry, 3–27. London: Macmillan, 1923. 

  3. Glavaničová, Daniela, and Miloš Kosterec. ‘The Fine-Grainedness of Poetry: A New Argument against the Received View.’ Analysis 81 (2021): 224–31. DOI: 

  4. Holmes, James S. ‘Forms of Verse Translation and the Translation of Verse Form.’ Babel 15 (1969): 195–201. DOI: 

  5. Hulatt, Owen. ‘The Problem of Modernism and Critical Refusal: Bradley and Lamarque on Form/Content Unity.’ Journal of Aesthetics and Art Criticism 74 (2016): 47–59. DOI: 

  6. Jacobs, Carol. ‘The Monstrosity of Translation.’ MLN 90 (1975): 755–66. DOI: 

  7. Jolley, Kelly Dean. ‘(Kivy on) The Form-Content Identity Thesis.’ British Journal of Aesthetics 48 (2008): 193–204. DOI: 

  8. Kivy, Peter. ‘Paraphrasing Poetry (for Profit and Pleasure).’ Journal of Aesthetics and Art Criticism 69 (2011): 367–77. DOI: 

  9. Kivy, Peter. Philosophies of Arts: An Essay in Differences. Cambridge: Cambridge University Press, 1997. DOI: 

  10. Lamarque, Peter. ‘The Elusiveness of Poetic Meaning.’ Ratio 22 (2009): 398–420. DOI: 

  11. Lamarque, Peter. ‘Philosophy and the Lyric.’ Journal of Literary Theory 11 (2017): 63–73. DOI: 

  12. Lamarque, Peter. The Philosophy of Literature. Oxford: Blackwell, 2009. 

  13. Lamarque, Peter. ‘Poetry and Abstract Thought.’ Midwest Studies in Philosophy 33 (2009): 37–52. DOI: 

  14. Lamarque, Peter. ‘Semantic Finegrainedness and Poetic Value.’ In The Philosophy of Poetry, edited by John Gibson, 18–36. Oxford: Oxford University Press, 2015. DOI: 

  15. Lamarque, Peter. Work and Object: Explorations in the Metaphysics of Art. Oxford: Oxford University Press, 2010. 

  16. Leighton, Angela. ‘About About: On Poetry and Paraphrase.’ Midwest Studies in Philosophy 33 (2009): 167–76. DOI: 

  17. Lepore, Ernie. ‘The Heresy of Paraphrase: When the Medium Really Is the Message.’ Midwest Studies in Philosophy 33 (2009): 177–97. DOI: 

  18. McGregor, Rafe. ‘Poetic Thickness.’ British Journal of Aesthetics 54 (2014): 49–64. DOI: 

  19. McGregor, Rafe. The Value of Literature Lanham, MD: Rowman & Littlefield, 2016. 

  20. Ribeiro, Anna Christina. ‘Relevance Theory and Poetic Effects.’ Philosophy and Literature 37 (2013): 102–17. DOI: 

  21. Ribeiro, Anna Christina. ‘Toward a Philosophy of Poetry.’ Midwest Studies in Philosophy 33 (2009): 61–77. DOI: 

  22. Rosenthal, Alan S. ‘Baudelaire’s Mysterious “Enemy”.’ Nineteenth-Century French Studies 4(1976): 286–94. 

  23. Squire, Jack Collings. Poems and Baudelaire Flowers. London: The New Age Press, 1909. 

  24. Tichý, Pavel. The Foundations of Frege’s Logic. Berlin: De Gruyter, 1988. 

  25. Yang, Jeffrey, ed. Birds, Beasts, and Seas: Nature Poems from New Directions. New York: New Directions, 2011. 

comments powered by Disqus