Poetry: what it is for

Dante and Virgil in Hell (1850) William Bouguereau

Poets have from ancient times enjoyed a special status, which connected them with the sacred and the sublime. They were often believed to be inspired by divinity¹ and this position as mediators between a mundane everyday existence and something higher and more noble has often given them licence to transgress social mores and to bear a certain other-worldly eccentricity as a sign of the favour of the muses. While other men are expected to conform to certain norms, poets, by their habitual contact with things beyond our realm are granted not only the right, but the expectation of oddness. This manifests differently in different societies: it can take the shape of affectation, fragility, or unwise sincerity and bluntness.

Yet it should be obvious to modern man that poetry, just like any other manifestation of culture, must be ultimately grounded on material concerns, relating to the production and reproduction of the social matrix. This reality, which ought to be self-evident, hasn’t however made much way in correcting the notion that poets bring us insights from other worlds, and that their art is, so to speak, a terra incognita where analysis is not just unable to trace maps, but also unwelcome to try. The mystique of art, and this is not exclusive to poetry, suggests that there is something to be lost in understanding what the principles which make some things beautiful or moving, and other things ugly or dull, are. Thus, a truly materialist theory of culture eludes us yet, but perhaps we may try to inquire why poetry exists, and what function it fits in social life, so that it continues existing.

Poetry as an embodied activity

The same romantic tendencies which regard poetry as ineffable, are reluctant to bear in mind that it occurs in a human context: composition happens in human minds–brains, really–, and not in gardens where muses dwell; writing and reading require of human hands and eyes; recitation could not happen without lungs, throat, tongue… nor would it have its object without ears. Poetry and its practice belong to the world of embodied experiences, and are ultimately sensual. Something which would not so easily pass us by, were we to look concretely at poems² instead of attempting to draw abstract conclusions from the intuition that poems are too great to arise from nerve and sinew.

Communication and information theory, useful disciplines as they are, can easily lead us to conceive of embodied behaviours such as speech as though they were mathematical abstractions. Hence we focus on language as a set of discrete symbols, forgetting both that oral expression–and aural perception–preceded writing, as well as all manner of paralinguistic signals that go unrecorded on paper: accent, speed, emphasis, breath–though punctuation tries–, and the whole panoply of non-verbal communication which comprises such things as face expression, gestures, posture, touch, and so on.

What can we learn from the fact poetry–like war³–is made by human beings? Given the universality of poetry, it is reasonable to believe that it arose in combination with language and music, also universal.

It was a common belief that man evolved a more capable brain in order to produce better tools to help in the constant struggle for existence against a hostile environment, but this position is hard to support given some thought. On the one hand, the human brain has very high metabolic costs–about 20% of the body’s total energy budget. On the other, assuming that inclusive fitness effects aren’t particularly large, tool creation wouldn’t confer inventors a very strong advantage in terms of reproductive fitness: once a tool is devised, copying it is considerably easier, and anyone else could derive the benefits. Hence, we must seek alternative explanations.

Another more plausible idea is that intelligence has primarily evolved in order to allow us to model our environment, but not so much in physical terms as socially. As humans became more intelligent, assessing whether to trust other (especially non-kin) humans turned out to be an important matter for survival and reproduction. In order to make such judgements, it was necessary to develop faculties like empathy, which is to say, the ability to place oneself in someone else’s mindset. Together with a model of other people, we acquired a model of ourselves, and a capacity for introspection. In fact, these two developments were probably parallel, as there is some evidence that they use similar mechanisms in the brain. From this arises the eternal social game of signalling, deceit, detection, and so the intelligence explosion results, since anyone unable to follow social cues gets excluded or betrayed.

The development of introspection and language

Most complex lifeforms have some form of world model, and some have a model of self: they know where they are located in space, and how certain actions will shift them. This is key for any group hunting strategy, though it may in some cases be hard-coded. It’s known, for instance, that ants don’t have a particular world model, but they follow chemical signals from other ants instead. So although the individual ants do not model the world, the colony, to some extent, does. In humans, modelling is much more extensive: we do not only have a very good sense of the extension of our body and the space it occupies, but we’re also modelling the moves other bodies can make, and in order to attempt to predict them, we must ascribe intentionality to them. If we can detect external signs of the mental processes of other humans we can decide if they are a threat or an aid. Some of this isn’t exclusively human, but characteristic to most social mammals. Humans, though, have a far more complex space of inner states than other animals.

Language becomes a tool to aid in coordination tasks, such as plotting against someone else, reporting bad conduct, and incidentally hunting, gathering and keeping track of resource use. We can surmise it goes through a long process, from unarticulated signals with specific meanings, towards the general and articulated recursively enumerable grammars we use today. Language allows us to do two things of paramount importance: share information about the world and share information about one another. The theory of mind we use to attempt to predict other people’s actions becomes correspondingly richer as we learn to talk about the way we feel, the things we want, and so on.

The reason for this excursus on the development of language is to take ourselves outside of the information theorist’s mindset of language being an exchange of symbols that happens between disembodied minds, and instead to focus on language as an activity of particularly intelligent social mammals. When the logical positivists tried to get a hold on language, they attempted to transform all its terms into formal logic, and held that ultimately all meaningful statements should be possible to rewrite by referring exclusively to sense data. This is a representational view of language, which holds that language is a vehicle which, through some isomorphism, embeds information about the state of the world. There is little doubt that language can function that way, but there is no doubt that it doesn’t have to. While I can represent information about the world by saying “there is an object with a soft texture I am touching with my left hand”, there are plenty of statements which most people would–justly–regard as meaningful that cannot be transformed into strings of sensual perceptions joined together by logical connectives. In spite of the efforts of the positivists to rewrite statements about states of mind into statements about sense perception, when I speak about what other people believe and feel, and particularly about their deceit, I’m inevitably relying on unobservables: as much as instrumentalists and operationalists try to wish it away, when I say Laura lied about her age I am making statements about things going on inside Laura’s mind, to which I have nor can have sensual access.

Another viewpoint is that language can be used as a depository of information, in which case it is representational, but it can also directly embed an extralinguistic reality through evocation.⁴ This position holds that language can work either as a mirror, encapsulating through isomorphism the meaning it is to communicate, or as a window, presenting the extralinguistic experience directly. This position is less unsatisfactory, and it is getting at something important. However, it in my view overstates the power of language to convey states of mind directly.

The limits of language: introspection as an incomplete model

The fundamental problem with the evocative view of language is that we cannot truly give each other a genuine insight into the way we feel. First off, introspection is itself, as an evolved mechanism to help us with empathy, notoriously incomplete. While it serves us well most of the time, we only have a model of the way we think, feel and behave, and not access to how we truly do so. This is why we can only to a point predict our own actions, and why self-deception isn’t just possible but common. Only, deception implies intent, and this is more of a case of being unable to know ourselves in fullness.

Because of this reason, stating that we could somehow communicate things of our inner self through evocation appears overly optimistic. Since we do not know ourselves very well, much less others, and since human language is not particularly precise at describing mental states–hence why describing a smell can never substitute for the experience–claiming that we can evoke the same state in another mind through language seems to require an appeal to sympathetic magic.

Furthermore, language is itself an interface to very complex modules in our minds. I’m reminded of a discussion regarding von Neumann machines. Someone asked, sceptically, if there was any instance in which humans had been able to create something which reproduced itself unaided in the physical world, to which someone else replied how mothers do this all the time. Of course, mothers give birth to self-replicating systems, but if you ask them how it all works, they won’t have any better insight than anyone else. Human reproduction is an interface: we push a button, and out comes a new person, but the process is outside our conscious control and understanding. Likewise with language: that we can form sentences and engage in conversation far from entails that we understand, in any sense of the word, the effects those utterances will have on other people. History is too full of misunderstandings for me to have to belabour this point.

Poetry as deixis

So if language isn’t purely representational, through isomorphism, nor strictly evocative, where does poetry fit? I would argue that there are at least 2 modes of using language: a representational mode describes not only sense perceptions, but also beliefs, assertions, and other first-class entities we have linguistic categories for; and a deictic mode. I’m using deixis in a slightly non-standard way, so I should explain quite what I mean. A word like “this” doesn’t have a representational content: it is a deictic word. This means that it is pointing or referring to something. It cannot be understood outside the context of a conversation.

What I mean by deixis follows as a slight extension of this notion: language, in particular poetic diction, needs not be limited to representing something else. It can serve as a pointer towards a state of mind. Just like music can be purely abstract, yet it has an impact on our thoughts and feelings, poetry is a path constructed with language, which points the way a mind can follow towards a given place in the space of thought. Because language isn’t unlimited, such a path isn’t always possible to follow. Different people react differently to the linguistic and paralinguistic effects in a poem, and may get to slightly different places, or even become lost entirely. But that is the core of poetry: to step outside representation to guide us into a particular manner of thought.

Given that I have insisted on poetry serving the needs of social reproduction, it may appear I have now abandoned this core principle. On the contrary, some thought will easily elucidate the social advantages of embedding directions to get to a given state of mind in a linguistic vehicle. Social synergy can be greatly aided by a uniformity in thought and feeling. Tasks that are arduous, dangerous or repugnant can be given a mystique by poetry, which can associate them to positive feelings. Action which may not be successfully induced through argument may be sometimes induced through emotion. Hence, poetry serves as a means for society to construct the maps of its myths: to build cohesion, to instil valuations of things, and to overcome scepticism, selfishness or inertia.

Why I’m unconvinced of poetry as a mere mnemotechnic aid

It is true that, generally, poetry is a lot easier to remember than prose. This is partly because it has a lower entropy than everyday language: it transmits less information per symbol. It’s indubitable that such a characteristic of poetry was a key to the formation of certain oral traditions which preserved an immense amount of material composed in verse without the aid of writing. However, much of the world’s poetry does not use the mnemotechnic devices at issue, such as rhyme. Latin and Greek poetry didn’t make use of rhyme, for instance. The degree of formal rigour–and hence memorability–of poetry can vary significantly across time and space.

While some of the poetry of the world may exist as an aid to preserve important stories, this seems far more related to music and song than poetry sensu stricto. Many rhetorical devices to be found exclusively or primarily in poetry have little utility in enhancing our memory: allegory, tropes, images, hardly seem to help us recite text.

Hence, while I don’t deny poetry and memory often had a tight bond, I find this hypothesis insufficient to explain the prevalence for less memorable poetic forms. For those, deixis appears to be the most plausible answer.

It's only fair to share...

Print

email

Facebook

Twitter

Linkedin

Pinterest

Reddit

Tumblr

StumbleUpon

VK