In the film The Big Sleep (1946), the private eye Philip Marlowe (played by Humphrey Bogart) calls at the house of General Sternwood to discuss his two daughters. They sit in the greenhouse as the wealthy widower recounts an episode of blackmail involving his younger daughter. At one point, Marlowe interjects with an interested and knowing ‘hmm’.
‘What does that mean?’ Sternwood asks suspiciously.
Marlowe lets out a clipped chuckle and says: ‘It means, “Hmm”.’
Marlowe’s reply is impertinent and evasive, but it’s also accurate. ‘Hmm’ does mean ‘hmm’. Our language is full of interjections and verbal gestures that don’t necessarily mean anything beyond themselves. Most of our words – ‘baseball’, ‘thunder’, ‘ideology’ – seem to have a meaning outside themselves – to designate or stand for some concept. The way the word looks and sounds is only arbitrarily connected to the concept that it represents.
But the meanings of other expressions – including our hmms, hars and huhs – seem much more closely tied to the individual utterance. The meaning is inseparable from or immanent in the expression. These kinds of expressions seem to have meaning more how a particular action might have meaning.
Are these two ways of meaning – designative and immanent – simply different things? Or are they related to one another? And if so, how? These questions might seem arcane, but they lead us back to some of the most basic puzzles about the world and our place in it.
Human beings are brazen animals. We have lifted ourselves out of the world – or we think we have – and now gaze back upon it detached, like researchers examining a focus group through one-way glass. Language is what allows us to entertain this strange, but extraordinarily productive, thought. It is the ladder we use to climb out of the world.
In this way, human detachment seems to depend on the detachment of words. If words are to keep the world at arm’s length, they must also be uninvolved in what they mean – they must designate it arbitrarily. But if words fail to completely detach, that failure should tell us something about the peculiar – and humble – position we occupy ‘between gods and beasts’, as Plotinus put it.
In his Philosophical Investigations (1953), Ludwig Wittgenstein draws a distinction that mirrors the one between these two ways of meaning. ‘We speak of understanding a sentence,’ he writes, ‘in the sense in which it can be replaced by another which says the same; but also in the sense in which it cannot be replaced by any other.’ (Marlowe evidently felt his ‘hmm’ could not be replaced.)
The first kind of understanding points to a peculiar aspect of words and sentences: two of them can mean the same thing. As Wittgenstein points out, we’d never think of replacing one musical theme with another as if they amounted to the same thing. Nor would we equate two different paintings or two different screams. But with many other sentences, understanding the meaning is demonstrated by putting it in other words.
However, the meanings of the music, the painting and the scream seem to be immediately there. ‘A picture tells me itself,’ Wittgenstein writes. There is no way to replace one expression with another without changing the meaning. In these cases, there isn’t really a sense of a meaning apart from the expression itself. It would be perverse to ask someone who has just let loose a chilling scream: ‘What exactly did you mean by that?’ or ‘Could you put that another way?’
Although these two examples of ‘understanding’ might seem of completely different kinds, Wittgenstein insists that they not be divorced from one another. Together, they make up his ‘concept of understanding’. And, indeed, most of our language does seem to lie somewhere along a spectrum between simply designating its meaning and actually embodying it.
On one end of the spectrum, we can imagine, as Wittgenstein does, people who speak a language consisting only of ‘vocal gestures’ – expressions such as ‘hmm’ that communicate only themselves. On the other end lies ‘a language in whose use the “soul” of the words played no part’. Here, ‘meaning-blind’ people, Wittgenstein writes, would use words without experiencing the meanings as connected to the words at all. They would use them the way a mathematician uses an ‘x’ to designate the side of a triangle, without the word seeming to embody the meaning in any way.
‘Livre’ might mean book but it doesn’t mean it the way that ‘book’ does
But neither of these imaginary languages seems capable of anything like the range and expressive richness of actual human language. The former seems to place human language (and our world) closer to that of animals and infants; the latter, closer to that of computers, for whom it couldn’t matter less how something is said.
Still, the examples might provide some clue as to how these ways of meaning relate to each other. The language of gesture would seem to have to come before the language of signs. It’s difficult to imagine a little girl first learning to communicate her needs with arbitrary signs, and only later learning how to communicate by gesture.
Even once we do come to use words in an arbitrary, designative manner, they – at least, many of them – still seem to have their meanings in themselves. When I first learn that the French ‘livre’ means book, the word is associated with its meaning only in a mediated manner. I remain, at this stage, meaning-blind with respect to the word. I know what it means, but its meaning doesn’t resonate in the material aspects of the word. As I become more fluent in French, however, the word’s meaning becomes sedimented in it. ‘Livre’ begins to sound like it means what it means.
Full understanding, in Wittgenstein’s sense, seems to involve not just being able to replace ‘livre’ with ‘book’, but also in the experience of the meaning in the word. To put it another way, ‘livre’ might mean book but it doesn’t mean it the way ‘book’ does.
We can of course imagine a person (or machine) using words competently without having this experience of meaning, but is what we imagine really human language use? It’s hard to see how such a person would have access to the whole range of practices in which we use words. Subtleties in certain jokes or emotional expressions would escape them. Meaning is more sunken into words than the practice of replacing one term with another suggests.
The idea that words themselves might harbour meaning used to be more intellectually respectable. Deciphering the relationship between what words mean and how they sound, which seems absurd with all but a small subset of our vocabulary, used to be of great interest. In Plato’s Cratylus, the title character indulges in the speculation, common at the time, that certain words are correct: that they name the things they refer to accurately. Etymology can therefore provide insight. ‘Anyone who knows a thing’s name also knows the thing,’ Cratylus says.
Plato’s Socrates prefers to gain insight into things by grasping the ‘forms’ behind them, instead of through the contingent, and often mistaken, names given to them. The production of names or words – ‘onomatopoieo’ in Ancient Greek – tells us only how an individual name-giver saw things, Socrates tells Cratylus. There’s no way to adjudicate the ‘civil war among names’ and decide which get at the truth.
Today, we use the concept of onomatopoeia in a more restricted way. It is applied only to words for sounds – ‘boom’, ‘gasp’, ‘splash’ – that bear a mimetic relationship to a sound in nature. The connection might be more indirect and tenuous in other cases, as in apparently onomatopoetic words for motions such as ‘slither’ or ‘wobble’ that seem through a kind of synaesthesia to imitate a sound that might accompany the motion.
But Socrates and Cratylus were also talking about what we now call sound symbolism, a much wider range of connections between sounds and what they mean. These include things such as the association in English and related languages between the ‘gl-’ sound and light, as in ‘glisten’, ‘glint’, ‘glimmer’ and ‘glow’.
Does this sound have an onomatopoetic connection to light? Or is it just an arbitrary connection that has come to ‘feel’ nonarbitrary to native speakers? The question is difficult to even ponder. Asking how ‘gl-’ relates to light is a little like enquiring after the connection between sad music and sadness. We can point to feelings and sensations that suggest they belong together, but we struggle to come up with an objective arbiter of the connection outside our own experience. This, as I’ll come to, is because the articulation itself produces or constitutes the connection.
Studies have established that the connections between things and sounds are nonarbitrary in many more of these cases than it would seem at first glance. There seem to be universal or near-universal synaesthetic connections between particular shapes and sounds. But, in a certain sense, the objectivity of the connection is beside the point. These connections still won’t underwrite the kinds of hopes that Cratylus had for etymology. At most, they indicate certain affinities between aspects of things – shape, size, motion – and particular sounds that the human vocal apparatus can produce. But even if the connection between ‘glow’ and glowing were not based on any verifiable affinity, the word’s meaning is still accompanied by its sound. ‘Glow’ would still glow with glowing.
Can we describe the depth of a deep thought without drawing in some way on the concept of depth?
What is noteworthy here is the human capacity not only to recognise but to produce, transform and extend similarities as a way of communicating meaning. In a short and opaque essay from 1933, Walter Benjamin refers to this capacity as the ‘mimetic faculty’ and suggests it is the foundation of human language. Language is an archive of what he calls ‘nonsensuous similarities’ – similarities produced by human practices and human language that don’t exist independently of them. Nonsensuous similarity, Benjamin writes, establishes ‘the ties between what is said and what is meant’. He suggests that fully developed human language emerges out of more basic imitative practices both in the life of the child and in the evolution of language itself.
The suggestion that all language is onomatopoeic becomes here not a thesis about any independent relationship obtaining between words and the world, as it was for Cratylus, but one about human creativity and understanding: our ability to produce and see correspondences or, as Benjamin talks about elsewhere, to translate into words the meaning communicated to us through experience.
This mimetic faculty is not just active in the ‘name-giving’ that establishes connections between language and objects, but in the ways in which established language is extended. The philosopher Charles Taylor writes in The Language Animal (2016) about ‘the figuring dimension’ of language: the way we use language from one domain to articulate another. Physical language of surface and depth, for example, permeates our emotional and intellectual language – ‘deep thoughts’, ‘shallow people’. Words of physical motion and action permeate more complex and abstract operations. We ‘grasp’ ideas, ‘get out of’ social obligations, ‘bury’ emotions.
These produced similarities between the physical and social or intellectual word fit into a similar space as the ‘gl-’ of ‘glow’. They are certainly not arbitrary, and yet they can’t really be justified by any criteria outside the figuration itself.
In many cases, they are much more than metaphors, since they are indispensable for our very conception of the matters they describe. Can we describe the depth of a deep thought without drawing in some way on the concept of depth? Language, here, as Taylor puts it, constitutes meanings: ‘The phenomenon swims into our ken along with its attribution.’ These cases suggest that, not only is meaning sunken into words, it is simply unavailable without their articulation.
This kind of articulation is more familiar in arts like painting and music. Words such as ‘deep’ and ‘glow’ can be thought of as analogous to particular notes that figure a particular kind of experience in a particular way, and so bring it into greater relief. Wittgenstein writes that ‘understanding a sentence in language is much more akin to understanding a theme in music than one may think’.
Moreover, as they become conventional, both linguistic and musical phrases open up new avenues for variation and combination that enable ever more fine-grained articulation and even the expression of entirely new phenomena and feelings. As Herman Melville wrote in his novel Pierre (1852): ‘The trillionth part has not yet been said; and all that has been said, but multiplies the avenues to what remains to be said.’
There is, of course, something very different about understanding a sentence and understanding a theme in music. We can replace the words ‘glow’ and ‘deep’ in most contexts in which they appear without much fuss – with, say, ‘shine’ and ‘profound’. We can even imagine using utterly different words in their place, stipulating, for example, that ‘rutmol’ will replace ‘deep’ in the dictionary. After a period of consistent use, ‘rutmol’ might even be as expressive of depth as ‘deep’ is now. We might begin to ponder ‘rutmol thoughts’ and be shaken by ‘rutmol wellsprings of emotion’.
Now, imagine a film scene portraying a cheerful family gathering around a sun-soaked table. Instead of the customary bright melody, the soundtrack is random musical notes. No matter how many times we watch it and try to make the score express ‘cheerfulness’, it would never feel right. The music couldn’t be heard as cheerful if it wasn’t. It might convey, instead, that not everything is as it seems with this family.
The meaning of ‘deep’ is detachable from ‘deep’ in a way that the meaning of the melody is not. Words can stand for things in a way that music can’t. This is what drove Socrates in Cratylus to stop wondering how words such as ‘deep’ related to deep things, and ask instead what deep things could tell us about the idea of depth. Once we get the messy words out of the way – which might barely have anything to do with the things anyway – then we can contemplate the things abstractly.
If I ask my dog to get the leash when we’re already on our walk, it means nothing
This ability to put the expressions for things to the side to focus on the things themselves seems integral to the unique relationship we have to the world. While animals (and small children) might be able to respond to signs as stimuli, and even use them in a rudimentary way to advance particular ends, they don’t seem to have the objects the way we do. They are too close to things, unable to step back and abstract from their concrete appearance. This makes them, as Heidegger puts it, ‘poor in world’.
When I tell my dog to get his leash, what he ‘understands’ remains an element of the immediate environment in which the expression appears, indicating a path to a particular end – in this case, a walk. For the dog, words have meaning the same way that the sound of my car pulling into the driveway has meaning. The word ‘leash’ might have only an arbitrary connection to the object for me, but for my dog it is inseparable from the situation. If I ask him to get the leash in the wrong context – when we’re already on our walk, for example – it means nothing.
But what is it exactly for ‘leash’ to symbolise a leash? This might seem self-evident: human beings come along, find the world and its furniture sitting there, and simply start tagging it arbitrarily with signs. But we forget the role the words played in lifting us into this perspective on things. Would we be able to conceive of the leash the way we do if we couldn’t call it anything else? In other words, what role does the arbitrariness of the word – the fact that we can replace it – play in the constitution of the things as independent objects?
And, moreover, how exactly do we get from the kind of immanent engagement with the concrete world that dogs seem unable to tear themselves out of to the position of disengaged spectator naming things willy-nilly?
In another difficult essay, Benjamin reads the story of Adam and Eve’s exile from the Garden of Eden as a kind of parable about the ‘externalisation’ of meaning. ‘On Language of Such and the Language of Man’ (1916) begins by positing an absolutely general definition of meaning: ‘we cannot imagine a total absence of language in anything’. And so Benjamin understands ‘language as such’ as a very basic sense of meaning – which Genesis construes as God’s creative speech. It is a fundamental given and constituent aspect of reality, being itself. ‘Language as such’ means in the way that pictures and ‘hmm’ mean. It means itself.
Human language begins by naming this always-already meaningful reality, engaging with it by imitating it in onomatopoeia and figurative articulation. Adam’s naming of the animals is the biblical parallel. But in the Fall, this immanent and expressive meaning – which means in the same way that literally everything else means – is externalised into the human word. We abandon ‘the communication of the concrete’ and immediate, Benjamin writes, for abstract and mediate words that vainly purport to stand for things instead of just imitating aspects of them. As a result of this Fall, we exile ourselves from Eden – immediate engagement with the natural world – and from our own bodies, which we now also experience, to our great shame, as objects.
For Benjamin, designative language and the world of objects that it brings about are made possible by a kind of forgetfulness. Our language is composed of words that have undergone countless transformations and whose original mimetic connections to reality have been lost. The immediate meaning of figurative imitations and metaphors dies and detaches from the concrete contexts in which it originally communicated.
Language is dead art, still connected to the things but so withered it now appears as only an arbitrary and abstract sign. Words appear to us like a faded landscape painting appears to a half-blind man. He no longer makes out the figure depicted but, remembering what it signifies, he instead takes the shapes as tokens. Using them, he can now to refer to mountains and streams in the abstract, no longer constrained by the immediacy of the scene, and free to replace the shapes with another expression if he likes. He gains a world only by losing his ability to really see it.
Does this new kind of interchangeable expression constitute a different way of meaning from the kind of meaning that ‘tells me itself’? Wittgenstein is wary of the very concept of meaning detached from expression, not only in the latter case but also in the case of interchangeable words. He doesn’t refer to the two sentences as meaning the same thing, but only to the practice we have of replacing one with another.
Whenever we enquire after the meaning of a word, we never get the thing that is meant – a permanent definition that underlies the word – but only another way of saying it. Despite its pretensions, the dictionary is no more than a pedantic and overexacting thesaurus. It doesn’t offer meaning, only other words.
Dictionary definitions can encourage in us a sense of words as signs representing fuller meanings or content that are in some sense ‘inside’ or ‘underneath’ them. But when we analyse meaning, we are usually making only lateral moves, not ‘excavating’ anything. These are in reality interpretations that exist ‘on the same level’ as what is interpreted. ‘Every interpretation,’ Wittgenstein writes, ‘hangs in the air together with what it interprets, and cannot give it any support.’
We side with the words even when they begin to contradict the reality
When we understand – despite how the word sounds – we don’t get at something beneath what is understood, but are simply able to provide another, perhaps better, way of saying it. The way the intellectual domain has been figured by words originating in our interaction with the physical environment can mislead us into the abstruse, multilevel ontologies that plague philosophy.
Wittgenstein is not suggesting that we discard these metaphors of our intellectual life that are constituted by our language. It is unclear that we even can, or what it would mean if we did. They are part of what Wittgenstein calls our ‘form of life’. Likewise, distance from the world that language provides and reinforces is indispensable, both in everyday communication and in modelling the world scientifically, but it can lead us astray if we take it too seriously, as we often do. ‘The best I can propose,’ Wittgenstein says of one of the pictures arising from our designative language, ‘is that we yield to the temptation to use this picture, but then investigate what the application of the picture looks like.’ If we look at how these words are used, Wittgenstein thinks, the question of what they really mean or refer to will dissolve.
This solution distinguishes Wittgenstein from postmodern theorists who take the limitations of our language and the impossibility of pure objectivity as reason to reject ‘Enlightenment’ reason. Those who pretend to see through ‘the myth’ of objectivity are on no firmer ground than those who cling to it. If anything, the former pull themselves out even further than the latter, pretending to watch the watchers.
One version of this view sees us stuck in what Friedrich Nietzsche called the ‘prison-house of language’. To Nietzsche and others, we are confined within our own meagre language and its presumptuous abstractions, which fall short of the real world even while they purport to describe it truthfully. Language is deemed inadequate to the world, an implausible instrument for pursuing and expressing truth.
But this view assumes exactly the same division between language and world as the one it criticises: it’s just less sanguine about reaching across the divide. To both ways of thinking, whether we can reach it or not, there is something out there: the way things are, which language is meant to designate. But ‘the great difficulty here’, Wittgenstein writes, ‘is not to represent the matter as if there were something that one couldn’t do.’ For him, it is the divide itself, which places language on one side and the world on the other, that needs to be questioned, not whether the divide can be bridged.
This is not to say that the divide should be regarded as a fiction. It is, rather, an achievement, but one with certain limits that are easily forgotten. Wittgenstein’s later writing takes on the aspect of therapy because it tries to draw attention to the moments, in philosophy especially, where removing language from the contexts in which it has a use, lends that language a kind of magical power and leads to confusion. We begin to puzzle about what the word refers to out there in the world, instead of attending to what it actually does in particular linguistic practices – what it tells us.
These problems are not only philosophical. In all kinds of domains – science, technology, politics, religion – we are prone to taking useful interpretations and turning them into frozen and potentially dangerous ideologies. Instead of looking at the concrete application of the words, we disengage them from practice, and instil them and the pictures they generate with greater reality than reality itself. We side with the words even when they begin to contradict the reality.
There is, in the end, only one kind of meaning. As Wittgenstein puts it, if the abstractions of philosophy are to have a use, ‘it must be as humble as that of the words “table”, “lamp”, “door”.’ He might have added ‘hmm’.