In 2010, artist Marina Abramović performed for 700 hours at the Museum of Modern Art in New York, in a piece called The Artist is Present. It involved her sitting still in the middle of the gallery’s soaring atrium, wearing one of a selection of striking, block-colour dresses that pooled over her feet. Members of the audience could come and sit with Abramović, and face her across a table or empty space, in silence. The emotion and intensity of their responses was astonishing. Some laughed; many cried.

Arthur Danto, the late Columbia University philosopher and art critic, compared his time with Abramović to “a shamanic trance”, and described the show as “magic” in the New York Times. More than 1,500 people came and sat with Abramović, and 750,000 attended as observers. A recurring sentiment among the visitors was that the performance was a deep revelation for which words were not sufficient. If this is true, then something about it was ineffable.

We’re used to the idea that some of life’s most meaningful experiences are difficult, if not impossible, to describe. But what, precisely, does it mean to say that something is unsayable? Philosophers from Arthur Schopenhauer to Theodor Adorno and Roger Scruton have tended to see ineffability as a mere mark of the extraordinary, rather than being something extraordinary in itself. Yet I’d argue that we should take the concept of ineffability seriously – that we should ask what it is and where it comes from.

Clearly, this enquiry is full of pitfalls. If something is beyond words, then it’s hard to get a handle on what, if anything, it means. Ludwig Wittgenstein, for example, was convinced that it was nonsensical to try to speak about what lies outside the limits of language. Even so, he wrote an entire book about what cannot be said, the Tractatus Logico-Philosophicus (1921), concluding with the observation: “Whereof one cannot speak, thereof one must be silent.”

We might never be able to eff the ineffable, to paraphrase Douglas Adams’s comic detective Dirk Gently. But perhaps we can pinpoint the nature of the thing that can’t be expressed, or find a way to describe what it consists of. I believe that there are at least four possible candidates for a non-nonsensical answer: ineffable objects, ineffable truths, ineffable content, and ineffable knowledge.

Dao, God, One

First, can there be such a thing as an ineffable object – a being, a thing, an entity? The Daoists of ancient China, for example, believed in something called the “Dao”, the source of all reality that transcended characterisation. A similar idea animated the Greek philosopher Plotinus, who claimed that some sort of indescribable “One” lay behind all existing things and was the guiding principle of reality. Over the course of the European Middle Ages, Plotinus’ idea of the One was slowly absorbed into the notion of God, transferring the source of ineffability from the fundamental order of reality to the creator of that order. The medieval Jewish philosopher Maimonides argued that, since God cannot be compared to anything in the world, the only way to describe Him was by means of negative attributes, by describing what He was not.

There’s a trivial sense in which all objects are “ineffable”: we can express sentences and propositions, thoughts and emotions, but never objects themselves. I can describe a chair to you but I cannot “express” a chair – simply because chairs are not the kinds of thing that can be transported via language. However, this is surely not what ancient and medieval philosophers had in mind. Rather, they were convinced that there was at least one object (the Dao, the One, God) that could not be captured by means of ordinary human language, because no description would do it justice.

From the perspective of analytic philosophy, however, this theory is difficult to maintain. Let’s define “D” as the property that makes the Dao unique. D thus refers, first, to whatever is unique about the Dao’s nature; second, to what distinguishes the Dao from every other object in the world; and third, to whatever it is that makes the Dao ineffable. The analytic philosopher can respond to these claims with an argument put forward in 1989 by the philosopher William Alston. Attributing a property to an object implies having formed a concept of the property, which in turn implies having cognitive access to it (that is, we must have formed some idea of what having that property would entail). But if a property is cognitively accessible to one person, then in principle it must be cognitively accessible to other persons, too. It can thus become the meaning of a term in their language. That means it’s possible for an expression to signify the property that uniquely characterises the Dao, God or the One. So the somewhat romantic idea of a profound, ineffable object is not tenable.

It’s worth noting that ineffability is not an exclusively religious concern. It survived the reorientation of philosophy away from the divine and towards the “human” during the 18th-century Enlightenment, and transformed itself in response to the social and political upheavals that were to follow. Immanuel Kant, an influential Enlightenment philosopher, argued that language is an insufficient medium for capturing the “manifoldness, order, purposiveness, and beauty” of the world; the only appropriate response to these wonders, he said, is “a speechless, but nonetheless eloquent, astonishment”.

From Kant onwards, philosophers’ interest in ineffable objects gave way to the idea that ineffability is a symptom of the insufficiency of language as a tool for capturing the ultimate truths of the world. Søren Kierkegaard suggested in 1844 that humans are trapped in the “ultimate paradox of thought”, wanting to discover things “that thought itself cannot think”. The arch-skeptic of reason and the Enlightenment, Friedrich Nietzsche, said in 1873 that truth was akin to an army of metaphors on the march – a host of powerful illusions, which we humans have forgotten are illusions.

Ineffable truth

The second concept we should examine is the idea of an ineffable truth. Let’s think of a truth as an entity that correctly represents (a part of) reality, and that has, or could be, converted into a proposition. For example, a mental image of a popsicle in my freezer could count as a truth (provided it corresponds to reality) because its content can be converted into the statement: “There is a popsicle in my freezer!” By contrast, a geographical map could not count as a truth (even though it correctly represents a part of reality) because its contents cannot be rendered propositionally. An ineffable truth, then, would be an entity that accurately depicts something, which could be converted into propositional form, but which, for some reason, cannot be expressed. Can we think of an example for such an entity?

The most promising arguments for ineffable truths were invented by philosophers of mind, and proceed from a premise of inaccessibility to a conclusion of ineffability. In his seminal paper What Is It Like to Be a Bat? (1974), Thomas Nagel argues that, since bats are conscious beings who make conscious experiences, there must be something it is like to be a bat. A bat, he elaborates, must have access to facts that embody the bat’s particular form of subjectivity, which is determined by its body, its sensory apparatus, its mind or consciousness. But given that these facts depend on the bat’s particular point of view, they are inaccessible to beings occupying non-bat perspectives. Consequently, truths about what it is like to be a bat are “beyond the reach of human concepts” – ineffable.

Such vivid thought-experiments are intuitively compelling, but they turn out to be incoherent. If it’s right that our subjective point of view is not only a feature of our representations of reality, but also of reality itself, then the sum of all facts about the world would include incompatible facts – that is, facts corresponding to incompatible perspectives (a bat’s, a human’s, an antelope’s, and so on). Yet, if two facts are incompatible, then they cannot both be the case. Put differently, the sum of all facts about the world would not add up to a uniform, seamless picture of reality, an outcome that is unacceptable for most analytic philosophers. So, since the notion of ineffable truths relies on the notion of subjective facts, we should also give up the notion of ineffable truths.

Ineffable content

Next there’s the possibility of ineffable content as the source of the unsayable. Colour is a good way to think about this: does each of the many shades of colour we can perceive correspond to its own concept, with its own content? Imagine a colour spectrum for varieties of red: we might have a concept for “vermillion” and “scarlet”, maybe also for “carmine” and “crimson”, but definitely not for every single one of the infinitely many kinds of red.

For some time, philosophers wondered whether the mental contents generated by our perception of nameless shades of red were non-conceptual. However, an argument by John McDowell did away with this idea in the 1980s. If we’re able to discriminate a piece of perceptual content, he claimed, such as a particular shade of colour, then we’re also able to form a concept for it. All we need to do is pick out and gesture at the content (for example, by pointing our finger at a blob of the relevant colour) and employ a higher-order term to give it a name: “I will call this [point with finger] shade of colour [higher-order term] tongue-red [new concept].”

McDowell’s argument refutes the existence of ineffable content that we can perceive, but it can be applied to the idea of ineffable contents more generally. As soon as we can discern the existence of a piece of content, in principle we’re able to conceptualise it. So, if we take concepts to be the elementary constituents of language, a piece of content that can be captured in concepts cannot be ineffable. (Note the similarity of this argument, by the way, to the one that was made against ineffable objects.)

Ineffable knowledge

Let’s consider the final possibility for what gives rise to ineffability: the notion of ineffable knowledge. Ineffable experiences are often described as “meaningful” or “revelatory”, but the alleged meanings or revelations are claimed to be ineffable. If this is correct, then perhaps what ineffable experiences give us are bits of ineffable knowledge.

We can think of states of knowledge as enabling states, “disposition[s] on the part of the subject to act on his, her, or its desires in a certain way”, as the philosopher AW Moore put it in Points of View (1997). For example, my knowledge that it will rain tonight (which I can express by saying: “It will rain tonight!”) enables me to act on my desire not to get wet (for example, by taking an umbrella or staying at home). The question then becomes whether all knowledge can be expressed verbally, or whether some kinds of knowledge are ineffable. If there are ineffable forms of knowledge, then we might have finally found the right way to describe the basis of ineffability.

It turns out that we know many things without being able to express them, and there is nothing mysterious about that at all. I know how to play the violin, but my explanations about how to hold the instrument and move the bow would not be enough to impart this knowledge to you. You would have to acquire “knowledge-how” to play a violin yourself, through practice. Similarly, I could try to explain the colour red to you, in terms of the latest scientific theories about wavelengths, retina receptors and human colour-perception. But no matter how comprehensive my description, phenomenal knowledge can’t be passed on through language.

Chasing the origins of the unsayable doesn’t explain how inexpressible experiences can transform us.

Similarly, an amnesiac might know all the facts about, and properties of, a man called Rudolf (where he was born, what he looks like, whom he is married to, and what his favourite ice-cream flavour is) without knowing that he himself is Rudolf. The piece of knowledge that would enable Rudolf to ascribe those properties to himself, and to recognise the facts about Rudolf as facts about himself, cannot be rendered in language – at least, not in language that is free of expressions that index facts to a particular person or spatiotemporal location, such as “I”, “you”, or “here”. No description of himself, no matter how detailed or complete, would convey to him the indexical knowledge that he is Rudolf.

There are forms of knowledge, then, that cannot be expressed in language. Unlike ineffable objects, truths and contents, ineffable knowledge is neither incoherent nor untenable for any other reason. In fact, it’s vital for day-to-day life. Knowledge-how enables us to act, indexical knowledge to recognise ourselves, and phenomenal knowledge to understand the world with our senses.

But what about the ineffable knowledge that we might gain during a performance by Abramović, or in a moment of religious ecstasy, or when we are moved by a piece of music? Chasing the origins of the unsayable doesn’t account for how inexpressible experiences can affect, enable and transform us. Perhaps there is no universal answer, but that shouldn’t stop us from reflecting on the question. “If philosophy can be defined at all,” Adorno said, “it is an effort to express things one cannot speak about.”

This article first appeared on Aeon.