Relational Cuts: On Images — 7. What an Image Must Be

We now reach a point where the earlier refusals can no longer be treated as merely provisional.

They have accumulated into a single constraint structure.

We cannot begin with objects.

We cannot assume relations between objects.

We cannot posit a containing field.

We cannot treat differentiation as derivative of units.

We cannot introduce a subject who attends to an object.

We cannot treat stability as a property of things.

And yet:

something is given to be seen.

So the question that remains is no longer exploratory in tone.

It is structural.

What must be the case for anything to be given in this way?

We are now in a position to state the requirement more directly, though still cautiously.

What we have been circling is not a collection of components.

It is a single condition with multiple aspects:

a situation in which differentiation is maintained without collapse into uniformity, such that what later becomes describable as “object” can temporarily stabilise.

But even this formulation is already too generous.

Because it still suggests that there is a “situation” in which things occur.

We must be more precise.

There is no prior situation.

There is only the sustained possibility of distinction.

From this, everything else follows—but not as derivation from a base, and not as construction from elements.

Rather, as stabilisation of what can be held apart long enough to be encountered.

So we can now say, with as much restraint as possible:

an image is not a thing.

It is not a representation of a thing.

It is not a structure composed of parts.

It is not a field in which objects are arranged.

It is not an interface between subject and object.

All of these are later descriptions, imposed after stability has already been achieved.

What must an image be, then?

It must be:

a temporarily stabilised configuration of differentiation in which collapse into uniformity is prevented sufficiently for distinction to occur.

Nothing more than that.

Nothing less than that.

This sounds almost too minimal to be useful.

But that is because we are used to adding structure too early.

Once we resist that impulse, the simplicity becomes more revealing.

For it implies several things at once.

First:

An image does not require objects as prerequisites.

Objects are effects of stabilised differentiation within it.

Second:

An image does not require a viewer as an external condition.

Whatever is later called “viewing” is already part of the same stabilisation process that allows differentiation to persist.

Third:

An image does not contain meaning.

Meaning is not carried inside it like a substance.

Meaning is a secondary stabilisation of relations that are already active at the level of differentiation.

Fourth:

An image is not composed.

Composition is a retrospective description of how stabilised differentiations are distributed.

These are not interpretations of images.

They are consequences of the constraint we began with.

But we must be careful not to overstate what has been achieved.

We have not defined images.

We have not classified them.

We have not explained their varieties.

We have done something more basic, and more restrictive:

we have specified the minimum conditions under which anything could be given in a way that allows distinction to persist.

This is why the term “image” now feels almost secondary.

It names not a category of objects, but a threshold phenomenon:

a point at which differentiation has stabilised sufficiently to become encounterable.

And yet this stability remains fragile.

It is not guaranteed.

It is not permanent.

It is not independent.

It is continuously contingent on the maintenance of conditions that prevent collapse into uniformity.

So if there is a final formulation, it is this:

an image is what happens when differentiation is held open long enough for something to be seen without requiring that anything be presupposed as already seen.

At this point, the circle closes.

Not because we have returned to an initial definition, but because the initial assumption—“something is given to be seen”—has now been unfolded into its minimal conditions.

We began with givenness.

We end with constraint.

And between them, nothing has been added except precision in what must be true for appearance not to fail.

What remains is not a theory of images.

It is the recognition that images are not secondary to relation, object, or subject.

They are the temporary achievement of a system in which differentiation does not collapse.

And if anything is still missing, it is not another concept.

It is the question of how such a system sustains itself at all.