In ‘The Rhetoric of the Image’, Roland Barthes offers a structure for breaking down the many ways in which the image and text of an advertisement can convey meaning to us.1 He starts by identifying three types of message within the image: the linguistic message (the text), the symbolic message (connotations and connections hinted at by the image) and the literal message (in which, for instance, an image of a tomato represents literally a tomato).

What happens if this framework is applied to an instance where image and linguistic signification are combined into one small mark: to punctuation? And one mark in particular whose shape lends itself to being read as an image?

In considering what ~ might signify as an image, things are made extra-interesting by the fact that the linguistic message (whatever punctuational use the ~ has been put to) and the symbolic message (whatever its form connotes to any individual reader) are tightly rolled into one. It is inevitable that we must consider whether the form of a mark affects the meaning that is attributed to – or read from – it: what came first: the shape or the meaning? Do we have a chicken-and-egg situation here? When confronted on the page or screen with a certain (usually small, often black, probably familiar) mark, it is, as Anne Toner writes, ‘the associative nature of reading that brings punctuation’s graphic dimensions to life.’2

All waves are to do with energy. In physics, a wave is an oscillation that travels through space, accompanied by a transfer of energy. The blowing of wind over the surface of water results in the lapping of waves on a beach. The vibration of drum skin kicks out mechanical waves of pressure and displacement that reach the ear to be heard as sound.

Waviness is loose, fluid, moving and imprecise. It is not straight, and not straightforward. Not straight to the point. It curves away, or expands from a central point, bleeding out and absorbing, providing breathing space. It is shaky ground, earthquake vibrations, destabilising.

Adorno sees on the punctuation mark’s face – its ‘physiognomy’ – an inseparable link to its syntactic function.3 In this case it has personality, and therefore personhood, agency to enact its own agenda on the characters it is paired with. But to see a mark and take some sense or meaning from it, it must be recognisable; must fit a certain convention, be discernibly one thing and not another. When is a punctuation mark not a punctuation mark? What are the tolerances within which it can be read as such?

There are places where you’d expect to find a punctuation mark, and places where you wouldn’t. Context, scale, dimensionality; if enough of these variables are changed does it cease to be a piece of punctuation? A grapheme? Has it become a sculpture, a pattern, just a wiggly line?

And what of the exact form? A character can go about in the disguise of many different glyphs. What makes them all related? What makes each a tilde, and not something else? Is there a point at which a mark has moved so far away from the convention that, despite all contextual clues to the good, it cannot be recognised? What is it then?

And how will you know what it means?