Towards a Systemic-Functional Theory of Images 10. Towards a Hallidayan Theory of Visual Semiosis: Reclaiming the View from Above

This series began with a simple observation.

Much work on visual semiosis proceeds differently from the way Systemic Functional Linguistics approaches language.

When analysing language, Halliday begins from systems rather than structures, from meaning rather than form, and from the view from above rather than the view from below.

When analysing images, however, researchers often begin with visible features and ask what they mean.

The image becomes a collection of observable forms awaiting interpretation.

Colour is assigned meanings.

Compositions are assigned meanings.

Framing relations are assigned meanings.

The distinction between content and expression frequently becomes blurred.

The result is not necessarily wrong.

But it is no longer recognisably Hallidayan.

This series has been an attempt to recover the theoretical resources needed for a genuinely Hallidayan approach to visual semiosis.

Returning to Halliday's Architecture

The central claim of the series has been deceptively simple.

Visual semiosis should be approached through the same global dimensions that organise language:

• stratification,

• instantiation,

• metafunction.

These dimensions are not peculiar to language.

They are dimensions of semiotic organisation.

What distinguishes language is not the existence of these dimensions but the particular way in which language instantiates them.

Most importantly, language possesses a stratified content plane consisting of semantics and lexicogrammar.

Visual semiosis does not appear to.

This difference has profound consequences.

It means that visual semiosis cannot simply be treated as a language without words.

Nor should it be analysed by searching for visual equivalents of clauses, phrases, or grammatical structures.

Images are not languages.

The task is therefore not to extend linguistic categories into visual space.

The task is to understand visual semiosis on its own terms.

Content and Expression

One of the most persistent themes of the series has been the distinction between content and expression.

All semiosis requires both.

Yet visual analysis often oscillates between them without clearly distinguishing the two.

The problem becomes especially visible in discussions of colour.

Colour belongs to expression.

Meaning belongs to content.

The significance of colour lies not in what colour means but in how colour participates in the realisation of meaning.

The same principle applies more broadly.

Visual expression is not meaning.

Nor is meaning reducible to expression.

The relation between them is one of realisation.

Maintaining this distinction is essential if visual semiosis is to be theorised systematically.

Beyond Visual Grammar

A second theme has concerned the widespread search for visual grammar.

The attraction of this project is understandable.

Grammar occupies a central place within the architecture of language.

If images are semiotic, it seems natural to ask whether they too possess a grammar.

Yet this question may be misleading.

The existence of semiosis does not entail the existence of lexicogrammar.

A semiotic system may possess content and expression without reproducing the particular architecture of language.

Visual semiosis therefore invites a different line of inquiry.

Instead of searching for visual clauses, we should investigate visual meaning.

Instead of searching for visual syntax, we should investigate visual systems.

The challenge is not to demonstrate that images are languages.

It is to understand how images function as semiosis.

Metafunction Beyond Language

The series has also argued that metafunction remains indispensable.

Visual semiosis, no less than language, participates simultaneously in:

• ideational meaning,

• interpersonal meaning,

• textual meaning.

These metafunctions do not belong to language alone.

They are principles of semiotic organisation.

Consequently, visual meaning can be investigated from the same metafunctional perspectives that have proved so productive in linguistic analysis.

The result is not a transfer of grammar into visual space.

It is a recognition that semiotic organisation extends beyond language.

Context and Register

A Hallidayan theory of visual semiosis must also take context seriously.

Meaning does not arise within isolated artefacts.

It emerges through participation in social situations.

Visual semiosis therefore stands in the same relation to context as language.

Images do not merely depict contexts.

They realise contextual values.

This observation has important implications for register.

Visual registers are not collections of recurring features.

They are regions of visual meaning potential associated with situation types.

The explanatory task is therefore not to catalogue visual forms but to understand how visual meanings are organised and distributed across contexts.

Photography and the View from Below

Photography provided a particularly revealing case.

The authority of photography rests largely upon the transparency illusion: the belief that photographs simply show reality.

Yet photographs do not escape semiosis.

Their apparent naturalness often conceals the semiotic work they perform.

The transparency illusion therefore mirrors a broader methodological tendency.

It encourages analysis to begin from the visible image itself.

A Hallidayan perspective insists on a different direction of explanation.

The photograph is not explained by what it depicts.

It is explained through its participation in semiosis.

Again, explanation proceeds from above.

What Remains to Be Done

This series has not attempted to provide a completed theory of visual semiosis.

Indeed, one of its conclusions is that such a theory remains largely undeveloped.

Many questions remain open.

What systems organise visual content?

How are visual meanings differentiated?

What expressive systems organise visual form?

How are content and expression related within visual semiosis?

What kinds of registers emerge across different visual domains?

These questions require extensive theoretical and empirical investigation.

The purpose of the present series has been more modest.

It has sought to establish a coherent architecture within which such investigations might proceed.

Reclaiming the Hallidayan Project

As announced at the outset, this series has been concerned with Reclaiming a Hallidayan Approach to Visual Semiosis.

The choice of the word reclaiming was deliberate.

Halliday provided a remarkably powerful account of semiosis.

Yet much of that power derives not from particular analytical techniques but from a distinctive theoretical orientation.

The priority of system over structure.

The priority of content over expression.

The priority of explanation over description.

The priority of the view from above.

These commitments transformed the study of language.

There is no reason they cannot also transform the study of visual semiosis.

What is required is not the extension of linguistic categories into visual domains.

What is required is the application of Halliday's theoretical principles to visual phenomena.

A Final Reflection

The question posed by visual semiosis is not whether images can be analysed like language.

The question is more interesting than that.

What happens when Halliday's theory of semiosis is taken seriously beyond language?

The answer remains unfinished.

But one conclusion already seems clear.

Images are not languages.

They are not failed languages.

They are not incomplete languages.

They are not languages at all.

They are visual semiosis.

And if visual semiosis is approached through the architecture Halliday developed for understanding semiosis itself, a genuinely systemic-functional theory of images may finally begin to emerge.

Towards a Systemic-Functional Theory of Images 9. Photography and the Transparency Illusion: The Most Persuasive Image

Among all forms of visual semiosis, photography occupies a distinctive position.

Paintings are generally recognised as constructions.

Drawings are generally recognised as representations.

Diagrams are generally recognised as abstractions.

Photographs, however, often appear to be something else entirely.

They appear to be reality itself.

This appearance has given photography a remarkable cultural authority.

Photographs are routinely treated as evidence.

They are taken to document events.

They are used to establish identities.

They are invoked as proof.

The underlying assumption is simple:

the camera merely records what is there.

From a Hallidayan perspective, however, this assumption deserves closer scrutiny.

The Transparency Illusion

The power of photography rests upon what might be called a transparency illusion.

A photograph appears transparent to the world it depicts.

The viewer's attention passes through the image towards the scene beyond it.

The photograph seems not to mediate reality but merely to reveal it.

Consequently, the photograph often disappears as a semiotic object.

The image becomes a window.

The act of representation becomes invisible.

The photograph appears not to mean but simply to show.

This is the transparency illusion.

Why the Illusion Is Persuasive

The illusion is not arbitrary.

Unlike many other visual artefacts, photographs are causally connected to the scenes they depict.

Light reflected from objects contributes directly to the formation of the image.

This relation encourages the belief that photographs somehow bypass semiosis.

If the image is caused by the scene, then perhaps the image simply presents the scene.

Yet this conclusion does not follow.

A causal relation is not the same thing as a semiotic relation.

The fact that a photograph originates in a physical process does not eliminate the semiotic processes through which it functions.

Indeed, the very intelligibility of a photograph depends upon semiosis.

Selection Before Interpretation

Even the simplest photograph presupposes selection.

Someone chooses:

• where to position the camera,

• what to include,

• what to exclude,

• when to capture the image,

• which lens to use,

• which perspective to adopt.

These decisions occur before any subsequent editing or manipulation.

They are not accidental.

They shape what becomes visible and what remains invisible.

The resulting image is therefore not reality as such.

It is a particular visual construction of reality.

The transparency illusion arises because these selections become difficult to see once the photograph has been produced.

The Myth of Neutral Observation

The authority of photography often rests upon a deeper assumption.

The assumption is that observation itself can be neutral.

If the camera merely records what is present, then the resulting image appears free from interpretation.

Yet this is precisely where the Hallidayan perspective becomes valuable.

Semiosis is not something added after observation.

The world does not first appear in a neutral form and then acquire meaning.

Rather, meaning is intrinsic to semiotic activity.

A photograph functions as a semiotic artefact from the moment it enters social life.

Its significance depends not merely upon what it depicts but upon how it participates in contextual meanings.

The image is never simply a record.

It is always a semiotic event.

Expression Without Transparency

The transparency illusion also obscures the role of expression.

Because photographs appear natural, expressive resources are often overlooked.

Yet photographs employ many of the same expressive resources found elsewhere in visual semiosis:

• colour,

• brightness,

• contrast,

• texture,

• focus,

• framing,

• spatial organisation.

These resources do not cease to function because they appear natural.

A shallow depth of field remains a semiotic resource.

A high-contrast image remains a semiotic resource.

A particular colour palette remains a semiotic resource.

The apparent naturalness of the image does not abolish expression.

It merely conceals it.

The View from Below

Photography encourages a strongly downward perspective on semiosis.

The viewer begins with the visible image and proceeds directly to the depicted scene.

The photograph therefore seems to support the view from below.

Meaning appears to arise from what is observed.

The image appears to derive its significance from reality itself.

A Hallidayan perspective reverses this movement.

The question is not simply what the photograph depicts.

The question is what meanings are being realised through the photograph.

The image is not explained by the scene.

The image is explained by its participation in semiosis.

Again, explanation proceeds from above.

Photography and Context

This becomes particularly obvious when identical photographs appear in different contexts.

The same image may function as:

• evidence,

• advertising,

• journalism,

• memorialisation,

• artistic expression,

• political persuasion.

What changes is not necessarily the image itself.

What changes is the contextual value within which the image participates.

The photograph remains physically identical.

Its semiotic significance changes.

This fact alone should make us cautious about locating meaning in the image itself.

The photograph functions within a larger semiotic architecture.

Beyond Representation

The transparency illusion ultimately rests upon a particular conception of images.

Images are treated as representations that stand between observers and reality.

The photograph appears to succeed precisely because it minimises this distance.

Yet from a systemic-functional perspective, semiosis is not fundamentally a matter of representation.

It is a matter of meaning.

The crucial question is therefore not how accurately a photograph reproduces reality.

The crucial question is how a photograph participates in the organisation of meaning.

Once this shift occurs, photography appears in a new light.

The photograph is no longer a transparent window.

It becomes a semiotic resource.

Towards a Hallidayan Theory of Visual Semiosis

Throughout this series, we have repeatedly encountered the same principle.

Visual semiosis becomes clearer when approached from above rather than below.

Content and expression must be distinguished.

System and structure must be distinguished.

Context and semiosis must be distinguished.

Photography has provided perhaps the most compelling test of these distinctions because it constantly invites us to collapse them.

The photograph appears to be reality.

The Hallidayan perspective reveals it to be semiosis.

The final post draws these threads together.

What would a genuinely Hallidayan theory of visual semiosis look like?

And what might such a theory contribute to the future development of systemic functional linguistics?

Towards a Systemic-Functional Theory of Images 8. Colour Revisited: Meaning, Expression, and a Persistent Confusion

Throughout this series, a recurring theme has emerged.

Visual semiosis has often been analysed from below rather than from above.

The analyst begins with what is visible and then attempts to determine what it means.

From a Hallidayan perspective, however, visible forms are not the starting point of explanation. They are what require explanation.

This principle becomes particularly important when we turn to one of the most discussed features of visual semiosis: colour.

Few topics have generated more interpretive commentary.

Blue is said to signify tranquillity.

Red is said to signify danger.

Green is said to signify nature.

Black is said to signify authority.

White is said to signify purity.

Such claims are familiar enough that they often pass without comment.

Yet from a systemic-functional perspective, they raise an important theoretical question.

What exactly is being described?

Where Is Colour?

Before asking what colour means, we need to ask a simpler question.

Where does colour belong within the architecture of semiosis?

The answer is straightforward.

Colour belongs to the expression plane.

It is a visual resource through which semiosis is realised.

Whatever meanings may be associated with colour, colour itself is not meaning.

It is expression.

This distinction may seem obvious.

Yet visual analysis frequently proceeds as though colour and meaning were interchangeable.

The result is a continual slippage between content and expression.

The Problem with Colour Meanings

Consider a familiar example.

An analyst observes the colour blue and proposes that it signifies tranquillity.

At first glance, the interpretation appears reasonable.

But what has actually been established?

Certainly a colour has been identified.

And certainly a meaning has been proposed.

What remains unclear is the relation between them.

If blue literally means tranquillity, then tranquillity should be inseparable from blueness.

Yet this is obviously not the case.

A sense of tranquillity may be realised through:

• muted colour palettes,

• reduced visual contrast,

• open spatial arrangements,

• balanced compositions,

• soft visual boundaries,

or many other expressive resources.

The meaning remains recognisable even when the colour changes.

This suggests that tranquillity cannot be identified with blue itself.

The colour belongs to expression.

The meaning belongs to content.

The Direction of Explanation

This distinction reveals a broader methodological issue.

Many analyses move from colour to meaning.

The reasoning proceeds roughly as follows:

Here is a colour.

Therefore, what does it mean?

A Hallidayan approach reverses the direction.

The question becomes:

Here is a meaning.

How is it realised?

Colour then becomes one possible expressive resource among many.

The explanatory priority shifts from expression to content.

Again, the view from above proves decisive.

The analyst is no longer asking what colours mean.

The analyst is asking how meanings may be realised chromatically.

Colour and Context

The difficulties become even clearer when context enters the picture.

The same colour may participate in very different meanings across different situations.

Red may be associated with warning in one context, celebration in another, political affiliation in a third, and aesthetic preference in a fourth.

Blue may evoke calmness in one setting and institutional authority in another.

These variations are often treated as exceptions.

From a systemic-functional perspective, they are exactly what we should expect.

Meaning does not reside in colour itself.

Colour participates in the realisation of meanings that are intelligible within particular contexts.

The contextual value remains primary.

The colour is one resource through which that value may be realised.

Colour and Choice

At this point an important objection may arise.

Surely colour choices matter.

A designer may deliberately choose blue rather than red.

A painter may deliberately construct a particular palette.

A photographer may deliberately seek particular chromatic effects.

Is this not evidence that colour itself carries meaning?

Not necessarily.

What it demonstrates is that colour functions as a semiotic resource.

A resource can be selected precisely because it is capable of realising particular meanings.

The existence of choice does not collapse the distinction between content and expression.

On the contrary, it presupposes it.

The expressive resource is selected because it is capable of realising a content value.

The Special Case of Photography

Colour becomes particularly interesting when we turn to photography.

A painter may choose to depict a blue sky.

A graphic designer may choose to create one.

A photographer, however, appears merely to record what is already present.

This appearance has encouraged the belief that photographic colour is somehow more natural and therefore less semiotic.

Yet the situation is more complicated.

Photographers choose:

• where to point the camera,

• when to take the photograph,

• what lighting conditions to exploit,

• how to expose the image,

• how to process the result,

and increasingly, how to manipulate colour after capture.

The apparent transparency of photography conceals a substantial semiotic process.

Indeed, the assumption that photographs simply show reality may be one of the most powerful illusions in visual culture.

Beyond Colour Symbolism

The goal of a Hallidayan approach is therefore not to compile dictionaries of colour meanings.

Such projects begin from expression and attempt to work upward.

Instead, the task is to understand how colour functions within a broader semiotic architecture.

Colour is neither meaning nor reality.

It is an expressive resource.

Its significance lies in the meanings it may realise and the contexts within which those meanings become intelligible.

Once this distinction is maintained, many familiar debates about colour begin to look rather different.

The question is no longer:

What does blue mean?

The question becomes:

What meanings are being realised through this chromatic organisation, and how does colour participate in that realisation?

Looking Ahead

Colour provides a useful reminder that visual semiosis cannot be understood simply by inspecting visible forms.

Expression is not self-explanatory.

It becomes intelligible through its relation to content and context.

Nowhere is this lesson more important than in photography.

Photography has often been treated as though it bypasses semiosis altogether—as though the camera simply delivers reality directly to perception.

A Hallidayan perspective suggests otherwise.

The next post examines what might be called the transparency illusion: the belief that photographs merely show the world rather than participating in the semiotic organisation of meaning.

Towards a Systemic-Functional Theory of Images 7. Visual Systems and Visual Structures: Axis Beyond Language

The previous post argued that visual semiosis participates in context through register.

Different social situations draw upon different regions of visual meaning potential, giving rise to distinctive patterns of visual semiosis.

This observation, however, leaves an important question unanswered.

How is visual semiosis organised internally?

To answer this question, we must turn to another of the fundamental dimensions of Systemic Functional Linguistics: axis.

The dimension of axis is often discussed through grammar. Yet it is not peculiar to grammar, nor even to language itself. It is a general semiotic principle.

If visual semiosis is a semiotic system, then it is reasonable to ask whether it too exhibits paradigmatic and syntagmatic organisation.

The Dimension of Axis

Halliday distinguishes between two complementary forms of organisation:

• paradigmatic order (system)

• syntagmatic order (structure)

A system is a set of related alternatives.

A structure is a configuration.

The relation between them is one of realisation.

Structure realises system.

Importantly, this distinction is not a distinction between meaning and form.

Nor is it a distinction between potential and instance.

It is a distinction between two modes of semiotic organisation.

The dimension of axis is therefore independent of both stratification and instantiation.

System and Structure Are Not Restricted to Meaning

One reason the distinction is often misunderstood is that discussions of system and structure frequently focus on grammar.

This can create the impression that systems belong to meaning while structures belong to form.

But Halliday's architecture is more general than this.

The distinction between paradigmatic and syntagmatic organisation applies wherever a semiotic system exhibits relations of choice and configuration.

Thus, in language, we find systems and structures not only in lexicogrammar but also in phonology.

The distinction therefore cannot be reduced to the content plane.

Axis cuts across the semiotic architecture.

It is a way of organising semiotic phenomena rather than a particular kind of semiotic phenomenon.

Why This Matters for Visual Semiosis

Once this point is recognised, a common assumption about images becomes less convincing.

Many approaches to visual analysis begin by describing structures.

An image is analysed in terms of:

• composition,

• framing,

• balance,

• symmetry,

• salience,

• contrast,

and other observable configurations.

Such descriptions are often valuable.

But they leave a theoretical question unanswered.

Are these structures related to systems?

Or are they merely collections of visible features?

A Hallidayan approach cannot be satisfied with structural description alone.

If visual semiosis exhibits axis, then structures should be intelligible as realisations of systems.

The challenge is therefore not simply to identify structures.

It is to understand the systems those structures realise.

The View from Above Revisited

This brings us back to the central methodological principle of the series.

The view from above gives explanatory priority to system rather than structure.

This does not mean structures are unimportant.

On the contrary, structures are indispensable.

Without structures, systems could never be realised.

The point is that structures do not explain themselves.

A configuration becomes intelligible through its relation to alternatives.

To explain a structure is therefore to show how it participates in a system.

The question is not merely:

How is this image organised?

but:

What alternatives make this organisation meaningful?

A Different Question from Visual Grammar

At this point it is important to avoid a familiar misunderstanding.

To ask whether visual semiosis exhibits systems and structures is not the same as asking whether it possesses a grammar.

These are different theoretical questions.

The existence of paradigmatic and syntagmatic organisation does not entail the existence of a lexicogrammatical stratum.

Language possesses both.

Visual semiosis may possess the former without possessing the latter.

Indeed, this possibility may be one of the most important implications of a Hallidayan approach.

Images need not be languages in order to exhibit systemic organisation.

What Might Visual Systems Be?

This question remains open.

The purpose of the present post is not to provide a completed inventory of visual systems.

Rather, it is to establish the theoretical space in which such an investigation becomes possible.

If visual semiosis exhibits axis, then we should expect to find:

• paradigmatic relations among visual alternatives,

• syntagmatic configurations realising those alternatives,

• and systematic relations between the two.

The task of analysis would then be to reconstruct those relations.

Not to catalogue structures in isolation, but to understand them as moments within a larger semiotic organisation.

Beyond Structural Description

This perspective suggests a shift in emphasis.

Instead of treating composition, framing, or salience as self-explanatory categories, we begin asking what systems they may realise.

Instead of accumulating inventories of visual features, we investigate networks of semiotic alternatives.

Instead of beginning from structure alone, we seek the paradigmatic relations that make structure meaningful.

The goal is not to abandon the study of visual structures.

The goal is to place them within a systemic account of visual semiosis.

Looking Ahead

If visual semiosis possesses both content and expression planes, and if both may potentially exhibit paradigmatic and syntagmatic organisation, then a further question immediately arises.

How are the resources of visual expression related to visual content?

This question becomes particularly pressing when we consider colour.

Colour has often been treated as though it possessed meanings in its own right.

Yet from a systemic-functional perspective, colour first appears as an expressive resource.

How, then, does colour participate in semiosis?

The next post revisits one of the most familiar topics in visual analysis and asks a deceptively simple question:

What exactly are we talking about when we talk about colour?

Towards a Systemic-Functional Theory of Images 6. Context, Register and Visual Potential: From Meaning Potential to Situation Type

The previous post argued that visual content is organised metafunctionally.

Like language, visual semiosis participates simultaneously in ideational, interpersonal, and textual meaning. These metafunctions do not constitute separate systems but complementary perspectives on the same meaning potential.

Yet meaning potential never appears as an undifferentiated whole.

Not every meaning possibility is equally relevant in every social situation.

A technical drawing does not draw upon visual meaning in the same way as an advertising campaign.

A weather map does not organise visual meaning in the same way as a religious icon.

A family photograph does not operate in the same region of visual potential as a scientific diagram.

The question is therefore unavoidable:

How does visual meaning potential become organised for particular social situations?

To answer this question, we must turn to two of Halliday's most important concepts: context and register.

Context as Semiotic Potential

In everyday discussion, context is often treated as a surrounding environment.

A text exists, and context is imagined as the external situation in which it occurs.

Halliday's conception is considerably more powerful.

Context is itself semiotic.

It is a higher level of symbolic abstraction realised by semiotic systems such as language.

From the perspective of the view from above, context is not something added to meaning after the fact.

It is the source of the values that semiotic systems realise.

This point is particularly important for visual semiosis.

Images do not acquire meaning independently of context.

Nor do contexts merely influence meanings that already exist.

Rather, visual meanings are intelligible because they participate in contextual values.

A courtroom diagram, a corporate logo, a traffic sign, and a wedding photograph each realise different regions of contextual potential.

To understand visual semiosis, we must therefore understand its relation to context.

The Instantiation Cline

The relationship between context and visual instances becomes clearer when viewed through the dimension of instantiation.

Instantiation is the cline from potential to instance.

At one pole lies the full semiotic potential.

At the other lies the individual semiotic event.

Between these poles lies an intermediate region.

Viewed from the potential end of the cline, this region appears as subpotential.

Viewed from the instance end, it appears as instance type.

These are not two different phenomena.

They are two perspectives on the same position within the cline.

This midpoint is where register emerges.

Register as Visual Subpotential

Register is often misunderstood as a collection of textual characteristics.

From a Hallidayan perspective, this is too narrow.

Register is a region of meaning potential associated with a situation type.

It is a subpotential.

A scientific article does not simply happen to exhibit certain linguistic characteristics.

It draws upon a particular region of linguistic potential.

Likewise, a technical illustration does not simply possess certain visual features.

It draws upon a particular region of visual potential.

This distinction is crucial.

The visual register is not the image itself.

Nor is it a list of recurring visual forms.

It is a region of visual meaning potential associated with a particular type of social situation.

From below, this same phenomenon appears as an instance type.

From above, it appears as a subpotential.

Both perspectives are legitimate.

The view from above, however, gives explanatory priority to the latter.

Field, Tenor and Mode

Halliday characterises context through three variables:

• field

• tenor

• mode

Field concerns the nature of the social activity.

Tenor concerns the social relations among participants.

Mode concerns the role played by semiotic activity itself.

These contextual variables are realised by meaning.

Consequently, they shape the organisation of visual meaning potential.

A medical imaging system, for example, participates in a different field from a fashion magazine.

A warning sign participates in a different tenor from a commemorative portrait.

A navigation map participates in a different mode from a work of gallery art.

The point is not that field, tenor, and mode determine visual form directly.

Rather, they differentiate regions of visual meaning potential.

Different contexts call forth different visual registers.

Why Register Matters

The concept of register allows us to avoid two common errors.

The first is the assumption that all images draw upon the same visual resources.

The second is the assumption that meaning resides entirely within individual images.

Neither position is satisfactory.

Visual meaning is neither completely general nor completely particular.

Instead, it is organised through subpotentials associated with recurrent social situations.

A scientific diagram and a political cartoon may both be visual artefacts.

Yet they operate within different visual registers because they actualise different regions of visual meaning potential.

The distinction is not merely descriptive.

It is explanatory.

Register helps explain why particular meanings become available in particular situations.

Register and the View from Above

The significance of register becomes particularly clear when contrasted with feature-based approaches to visual analysis.

Suppose we observe that a set of scientific diagrams share certain visual characteristics.

A description of those characteristics may be useful.

But it does not explain why those characteristics recur.

A Hallidayan approach asks a different question.

What region of visual meaning potential is being actualised?

What situation type makes those meanings relevant?

What contextual values are being realised?

Again, explanation proceeds upward rather than downward.

The image is explained through its register.

The register is explained through context.

The observable instance is not treated as theoretically primary.

Towards Visual Systems

At this point, visual semiosis begins to appear in a new light.

We have identified:

• content and expression,

• context and semiosis,

• potential and instance,

• register as visual subpotential,

• metafunction as the organisation of visual content.

What remains unclear is how visual meaning potential itself is organised.

If visual registers are regions of visual potential, what internal systems constitute that potential?

How are meaning distinctions related to one another?

How are selections made?

These questions lead directly to the next stage of the investigation.

The next post turns to the relation between system and structure. If visual semiosis possesses meaning potential organised into registers, what kinds of systems make that potential available, and how are those systems realised in visual structures?

Towards a Systemic-Functional Theory of Images 5. Metafunction Beyond Language: The Organisation of Visual Meaning

The previous post argued that visual semiosis possesses a content plane that cannot be reduced either to visual expression or to representation.

Visual content consists of meaning potential.

Particular images actualise selections from that potential and realise those selections through visual expression.

This conclusion immediately raises a new question.

How is visual meaning potential organised?

A systemic-functional answer cannot simply list meanings.

Meaning is not a collection of isolated entities.

It is organised.

The question is therefore not merely what meanings are available but how those meanings are systematically related.

At this point, Halliday's theory offers an indispensable clue.

Meaning is organised metafunctionally.

Metafunction Is Not Peculiar to Language

The metafunctional principle is often introduced through the analysis of language.

Language simultaneously organises:

• ideational meaning

• interpersonal meaning

• textual meaning

Because discussions of metafunction frequently focus on grammar, it is easy to assume that metafunction itself is fundamentally linguistic.

This would be a mistake.

Metafunction is not a property of lexicogrammar.

It is a principle of semiotic organisation.

Indeed, it is difficult to imagine how any semiotic system could function without some means of organising experience, enacting social relations, and creating coherent semiotic wholes.

If visual semiosis possesses a content plane, then that content plane must also be organised metafunctionally.

The challenge is not whether metafunction applies to visual semiosis.

The challenge is understanding how.

The Ideational Metafunction

The ideational metafunction concerns meaning as the organisation of experience.

In language, this includes both experiential and logical meanings.

Visual semiosis clearly participates in this domain.

Images distinguish:

• entities

• relations

• activities

• processes

• circumstances

• classifications

A scientific diagram, a technical illustration, a portrait, a map, and a landscape painting all organise experience in different ways.

Yet a Hallidayan account must proceed carefully.

The goal is not to identify visual equivalents of grammatical processes or participants.

The goal is to understand what kinds of experiential distinctions visual semiosis makes available as meaning potential.

The emphasis remains on systems of meaning rather than structures of expression.

The Interpersonal Metafunction

The interpersonal metafunction concerns social relations.

No image simply presents information.

Images also position viewers.

They invite, distance, confront, persuade, intimidate, seduce, reassure, celebrate, condemn, and countless other things besides.

A passport photograph and a political campaign poster may depict human beings.

Yet they establish profoundly different social relations between image and viewer.

Again, the point is not to catalogue visual devices.

The point is to recognise that visual semiosis possesses interpersonal meaning potential.

Different visual systems make available different possibilities for negotiating social relations.

The task of analysis is to understand those possibilities.

The Textual Metafunction

The textual metafunction concerns the organisation of meaning into coherent semiotic wholes.

This domain is often treated as though it belonged exclusively to language.

Yet visual semiosis also exhibits remarkable capacities for textual organisation.

Images establish centres and margins.

They create relations of prominence and background.

They organise pathways of attention.

They coordinate multiple meanings within a unified semiotic event.

Without such resources, visual communication would be impossible.

The textual metafunction therefore cannot be regarded as an optional addition to visual meaning.

It is constitutive of visual semiosis itself.

Three Perspectives on One Semiotic Event

An important feature of Halliday's metafunctional theory is that the metafunctions are not separate components.

They are simultaneous perspectives on the same semiotic phenomenon.

A visual instance does not first establish experiential meaning and then add interpersonal meaning before finally organising the result textually.

Rather, all three metafunctions are present together.

A political poster simultaneously:

• construes aspects of experience,

• enacts social relations,

• organises itself as a coherent semiotic whole.

A map simultaneously:

• construes spatial relations,

• positions users,

• organises information textually.

The metafunctions are distinct but inseparable.

They represent complementary dimensions of meaning.

Metafunction and the View from Above

The significance of metafunction for visual analysis extends beyond classification.

It provides a principled way of approaching visual content from above.

Without metafunction, the analyst is easily drawn back toward visible forms.

Attention drifts toward colours, lines, layouts, and compositions.

Metafunction redirects attention toward meaning potential.

The question is no longer:

What does this visible feature mean?

The question becomes:

What ideational, interpersonal, and textual distinctions are being actualised here?

This shift is fundamental.

It transforms visual analysis from the interpretation of expressive forms into the investigation of meaning systems.

Again, explanation proceeds from above.

A Different Kind of Question

At this point, we can begin to see why visual semiosis should not be approached through linguistic analogy.

The crucial question is not:

What is the visual equivalent of a clause?

Nor is it:

What is the visual equivalent of a sentence?

Such questions begin from linguistic structure.

A systemic-functional approach begins elsewhere.

It asks:

What ideational meanings are available?

What interpersonal meanings are available?

What textual meanings are available?

Only after those systems of meaning have been identified does it become possible to investigate how they are realised visually.

The priority remains with content.

Expression enters the picture later.

Towards Visual Registers

The metafunctional organisation of visual content provides the first glimpse of visual meaning potential as a system.

Yet meaning potential never exists as an undifferentiated whole.

Different social contexts call forth different regions of that potential.

Scientific diagrams, advertising campaigns, architectural plans, news photography, religious iconography, and technical manuals do not draw upon visual meaning in identical ways.

This observation leads directly to the next dimension of Halliday's architecture.

If visual semiosis possesses meaning potential organised metafunctionally, how is that potential distributed across different social situations?

The next post turns to this question through the concepts of context, register, and instantiation.

Towards a Systemic-Functional Theory of Images 4. The Content Plane of Visual Semiosis: Recovering Meaning from Expression

The previous post argued that images should not be understood as languages.

Visual semiosis participates in the same general semiotic architecture as language, but it does not possess a lexicogrammatical stratum. Consequently, the goal of visual analysis cannot be the discovery of visual equivalents of words, clauses, sentences, or grammatical structures.

This conclusion, however, immediately raises a more difficult question.

If visual semiosis possesses a content plane but not a grammar, what exactly constitutes visual content?

The question may appear straightforward.

In practice, it turns out to be remarkably difficult.

Indeed, many of the persistent problems in visual analysis arise precisely because the distinction between content and expression becomes blurred.

The Pull of Expression

One reason for this difficulty is obvious enough.

Visual expression is immediately available.

We can see colours, shapes, textures, spatial arrangements, contrasts, boundaries, and countless other visual features.

Content, by contrast, is not directly visible.

As a result, visual analysis often begins with expression and remains there.

The analyst identifies a visible feature and then attributes meaning to it.

Blue may signify tranquillity.

Red may signify danger.

A vertical arrangement may signify power.

A diagonal arrangement may signify movement.

The pattern is familiar.

Visible form is treated as the starting point, and meaning is inferred from it.

From a Hallidayan perspective, however, this procedure reverses the explanatory direction.

Expression is not the phenomenon that explains.

Expression is the phenomenon that requires explanation.

The task is not to determine what visible forms mean.

The task is to understand what meanings are being realised through visible forms.

Meaning Is Not Colour

Consider a simple example.

Suppose an analyst claims that blue signifies tranquillity.

The claim may be plausible.

But what exactly has been identified?

Has a meaning been identified?

Or has an expressive resource been associated with a meaning?

The distinction matters.

Blue belongs to expression.

Tranquillity belongs to content.

To say that blue means tranquillity risks collapsing the distinction between them.

The problem becomes clearer when we consider alternatives.

Tranquillity may be realised through blue.

But it may also be realised through:

• muted contrast

• soft boundaries

• spatial openness

• balanced composition

• reduced visual density

or numerous other expressive resources.

The meaning remains recognisable while the expression changes.

This suggests that the content cannot be identified with any particular expressive form.

Content must therefore be understood independently of the expression that realises it.

The Priority of Content

At this point the methodological significance of the view from above becomes apparent.

If visual semiosis possesses a content plane, then content must become the primary object of theoretical investigation.

This does not mean ignoring expression.

Expression remains indispensable.

Without expression, there is no realised meaning.

The point is rather that expression should be approached through content rather than content through expression.

Instead of asking:

What does this colour mean?

we ask:

What meaning distinctions are being realised through this colour?

Instead of asking:

What does this compositional arrangement signify?

we ask:

What content values are being realised through this arrangement?

The difference may appear modest.

In reality, it changes the entire direction of inquiry.

Content as Meaning Potential

A further complication arises once visual content is approached from the perspective of instantiation.

The content plane is not simply a collection of meanings attached to particular images.

Like every semiotic resource, visual content exists as a potential.

Particular images actualise selections from that potential.

Consequently, the content plane should not be understood as a catalogue of interpretations.

It should be understood as a system of meaning possibilities.

This distinction is crucial.

When analysts assign meanings directly to visible forms, they often move immediately from expression to instance.

A particular colour is observed and a particular interpretation is assigned.

The underlying potential remains invisible.

A systemic-functional account proceeds differently.

The task is to reconstruct the system of meaning distinctions that makes particular visual instances intelligible.

Again, explanation proceeds from above.

The Problem of Representation

At this point an objection often arises.

Surely images depict things.

Surely visual content consists simply of what is represented.

A photograph of a tree depicts a tree.

A portrait depicts a person.

A map depicts a landscape.

Is that not visual content?

The difficulty is that representation alone cannot account for meaning.

Two photographs may depict the same tree while realising very different meanings.

Two portraits may depict the same person while establishing radically different interpersonal relations.

Two maps may depict the same terrain while organising information in entirely different ways.

Representation contributes to meaning.

It does not exhaust it.

Visual content therefore cannot be reduced to what an image depicts.

The content plane concerns the semiotic organisation of meaning, not merely the identification of represented objects.

Towards a Theory of Visual Meaning

What, then, constitutes visual content?

At this stage of the discussion, a complete answer would be premature.

The purpose of this post is not to provide a finished taxonomy of visual meanings.

Rather, it is to establish the theoretical problem.

Visual semiosis possesses a content plane.

That content plane cannot be reduced to expression.

Nor can it be reduced to representation.

It consists of meaning potential realised through visual expression and actualised in visual instances.

The challenge is to determine how that potential is organised.

Fortunately, Systemic Functional Linguistics already provides a powerful clue.

Meaning is not organised randomly.

It is organised metafunctionally.

The next post therefore turns to the metafunctional organisation of visual semiosis. If visual content constitutes meaning potential, how is that potential organised ideationally, interpersonally, and textually?

Towards a Systemic-Functional Theory of Images 3. Why Images Are Not Languages: The Significance of Lexicogrammatical Absence

The previous post located visual semiosis within Halliday's broader semiotic architecture. Like language, visual semiosis possesses content and expression planes, participates in context through realisation, exhibits potential and instance through instantiation, and is organised metafunctionally.

At the same time, a crucial difference emerged.

Language possesses a content plane that is itself stratified into semantics and lexicogrammar.

Visual semiosis does not.

This distinction may appear technical. In fact, it has profound consequences for how images are understood.

For much of the history of visual analysis, images have been approached through analogy with language. Analysts have sought visual vocabularies, visual syntax, visual clauses, visual sentences, and visual grammars. The assumption behind such approaches is rarely stated explicitly, but it is easy enough to reconstruct.

Language is often taken as the paradigm of semiosis.

If images are semiotic, they must therefore possess structures analogous to those found in language.

The task becomes one of discovering the visual equivalent of grammar.

From a Hallidayan perspective, however, this assumption is far from self-evident.

Indeed, it may obscure the very phenomenon it seeks to explain.

Language Is the Exception

One of the most remarkable features of Halliday's account is that language is not treated as the model for all semiotic systems.

Quite the opposite.

Language is unusual.

Its content plane exhibits an additional level of symbolic abstraction not generally found elsewhere.

Semantics is realised through lexicogrammar, which in turn is realised through expression.

This architecture gives language extraordinary semiotic resources.

Yet precisely because language is unusual, it should not automatically become the template through which all other semiotic systems are interpreted.

The question should not be:

How do images possess grammar?

The question should be:

What follows if they do not?

The Temptation of Analogy

The attraction of linguistic analogy is understandable.

Language provides a highly developed theoretical apparatus.

When confronted with a less familiar semiotic system, it is tempting to borrow linguistic concepts and apply them to the new domain.

The result is often a search for visual equivalents:

• words become objects

• clauses become scenes

• syntax becomes composition

• grammar becomes layout

Such analogies can occasionally be useful as heuristics.

The problem arises when they become explanations.

At that point, the image is no longer being understood on its own terms.

Instead, it is being redescribed in linguistic language.

The analysis begins to tell us less about visual semiosis than about the analyst's familiarity with grammar.

Absence Is Not Deficiency

Perhaps the most persistent misunderstanding concerns the notion of absence.

To say that visual semiosis lacks lexicogrammar is often heard as a claim of inadequacy.

The image appears deficient because it lacks something language possesses.

This conclusion does not follow.

A fish lacks wings.

That does not make it an unsuccessful bird.

Likewise, visual semiosis need not be interpreted as an incomplete language simply because it lacks lexicogrammatical organisation.

The absence of lexicogrammar is not a deficiency.

It is a theoretical clue.

It suggests that visual meaning operates differently.

The task is therefore not to compensate for what images lack.

The task is to understand what visual semiosis is.

A Different Kind of Semiotic Organisation

Once the search for visual grammar is abandoned, a different question emerges.

If visual semiosis does not organise meaning through lexicogrammatical systems, how does it organise meaning?

This question is considerably more difficult.

It also happens to be considerably more interesting.

The answer cannot simply be imported from linguistics.

Nor can it be derived by cataloguing visible forms.

Instead, it requires a systematic investigation of visual meaning potential itself.

This is precisely where a Hallidayan approach becomes valuable.

The view from above directs attention away from the image as an object and toward the meaning potential that the image actualises.

The issue is no longer whether a particular arrangement resembles a clause or a sentence.

The issue is what systems of visual meaning make the arrangement meaningful in the first place.

Why the Difference Matters

The distinction between language and visual semiosis affects more than terminology.

It affects the entire direction of inquiry.

If images are treated as languages, analysis tends to focus on formal structures. The task becomes identifying visual equivalents of linguistic units.

If images are treated as semiotic systems in their own right, analysis begins elsewhere.

The task becomes understanding the organisation of visual meaning potential.

This shift may appear subtle.

In reality, it transforms the theoretical project.

Instead of asking:

What is the grammar of images?

we ask:

What systems of visual meaning are available?

Instead of asking:

What visual structure corresponds to a clause?

we ask:

What meaning distinctions are being actualised here?

Instead of beginning with visible forms, we begin with semiotic potential.

Again, the view from above proves decisive.

Beyond the Linguistic Shadow

The challenge facing a systemic-functional account of visual semiosis is therefore not to reproduce linguistic theory in another medium.

Nor is it to discover hidden grammatical structures lurking within images.

The challenge is to understand a semiotic system whose content plane is not organised lexicogrammatically.

This requires a certain intellectual discipline.

Language is the most elaborated semiotic system available to us. Consequently, it is very easy to interpret all semiosis through linguistic categories.

Yet a genuinely systemic-functional theory must resist that temptation.

The goal is not to place images within the shadow of language.

The goal is to understand visual semiosis as a semiotic system in its own right.

Only then can we begin asking what kinds of meaning potentials visual systems make available and how those potentials are realised in visual expression.

The next post turns directly to that question. If visual semiosis possesses a content plane but not a lexicogrammatical stratum, what exactly constitutes visual content?

Towards a Systemic-Functional Theory of Images 2. Visual Semiosis in Halliday's Architecture: Locating Images Within a Systemic-Functional Theory of Meaning

The previous post argued that any genuinely systemic-functional account of visual semiosis must begin from what Halliday calls the view from above. Rather than starting with visible forms and attempting to infer meanings from them, we begin with meaning potential and ask how that potential is realised in particular instances.

Before we can investigate visual meaning itself, however, we need to answer a more fundamental question.

Where does visual semiosis sit within the architecture of Systemic Functional Linguistics?

This question is important because visual semiosis is frequently discussed either as though it were simply another language or as though it were an entirely different kind of phenomenon requiring a separate theoretical framework. From a Hallidayan perspective, neither position is satisfactory.

Visual semiosis belongs within the same general theory of semiosis as language. At the same time, it is not merely a variant of language, nor should it be analysed by importing linguistic categories wholesale into the visual domain.

To understand why, we need to begin with Halliday's account of semiotic systems.

Content and Expression

Following Hjelmslev, Halliday treats a semiotic system as involving two planes of symbolic abstraction:

• a content plane

• an expression plane

The content plane concerns meaning.

The expression plane concerns the forms through which meaning is realised.

Neither plane exists independently of the other. A semiotic system is constituted by their relation.

This immediately places visual semiosis within the same general semiotic architecture as language.

Images are not merely physical artefacts.

Nor are they collections of visual stimuli.

They are semiotic phenomena because they involve the relation of content and expression.

Visual semiosis therefore possesses both a content plane and an expression plane.

This much it shares with language.

Where Language Differs

The crucial difference emerges when we look more closely at the content plane.

Language possesses a highly unusual architecture.

Its content plane is itself stratified into two further levels of symbolic abstraction:

• semantics

• lexicogrammar

This additional stratification allows language to organise meaning through a complex system of grammatical resources.

Visual semiosis does not appear to possess this additional level.

Its content plane is not further stratified into semantics and grammar.

This distinction is fundamental.

Much work in visual analysis has been driven by the search for visual equivalents of linguistic categories. Analysts have sought visual vocabularies, visual grammars, visual clauses, visual syntax, and visual sentences.

Such analogies may occasionally be suggestive.

But they risk obscuring the very feature that makes visual semiosis theoretically interesting.

Visual semiosis is not language.

The challenge is not to discover a hidden visual grammar.

The challenge is to understand how a semiotic system operates when its content plane is not organised through lexicogrammar.

Context Above Semiosis

The architecture becomes clearer when we look upward rather than downward.

In Halliday's account, semiotic systems do not exist in isolation.

They participate in a higher level of symbolic abstraction: context.

Context is not simply an environment surrounding semiotic activity.

It is itself semiotic.

Language realises context.

Visual semiosis also realises context.

Indeed, this is one of the reasons different semiotic systems can participate in the same social situations. They are related not because they share the same expressive resources, but because they realise the same contextual values.

From the perspective of the view from above, context remains primary.

Language becomes intelligible through the contextual values it realises.

Visual semiosis becomes intelligible through the contextual values it realises.

The explanatory movement remains the same.

Instantiation and Visual Potential

Visual semiosis is also subject to the dimension of instantiation.

Just as there is no language apart from its potential and its instances, there is no visual semiosis apart from its potential and its instances.

Particular images are instances.

The semiotic possibilities available to a visual system constitute its potential.

Between these poles lies the familiar midpoint of the instantiation cline: the region where subpotentials and instance types emerge.

This observation may seem obvious, but it has important consequences.

A photograph, diagram, painting, map, infographic, advertisement, or technical drawing should not be treated simply as an object.

Each is an instance of a meaning potential.

Consequently, the theoretical task is not merely to describe the image itself.

The task is to understand the potential of which it is an actualisation.

Again, the view from above remains decisive.

Metafunction Is Global

There is one further point that will become increasingly important as the series develops.

The metafunctional organisation of meaning is not peculiar to language.

Halliday's metafunctions are principles of semiotic organisation.

Visual semiosis therefore cannot be understood apart from:

• ideational meaning

• interpersonal meaning

• textual meaning

The precise manner in which these meanings are organised visually remains an open question.

But the question itself is unavoidable.

If visual semiosis is a semiotic system, it must participate in the metafunctional organisation of meaning.

This will become a central concern in later posts.

A Semiotic System in Its Own Right

At this point, we can locate visual semiosis within Halliday's architecture.

Like language, it possesses:

• content and expression planes

• participation in context through realisation

• potential and instance through instantiation

• metafunctional organisation

Unlike language, however, its content plane is not further stratified into semantics and lexicogrammar.

This difference is not a deficiency.

Nor does it make visual semiosis a primitive form of language.

It simply means that visual semiosis must be understood on its own terms.

The goal of a systemic-functional theory of visual semiosis is therefore not to discover visual equivalents of linguistic structures.

It is to understand how visual meaning operates within the same general semiotic architecture while exhibiting a different internal organisation.

The next post explores the implications of this distinction directly. If images are not languages, what exactly follows from the absence of lexicogrammar?

Towards a Systemic-Functional Theory of Images 1. The View from Above: Reclaiming a Hallidayan Approach to Visual Semiosis

Anyone familiar with Systemic Functional Linguistics will have encountered the phrase the view from above.

It is one of Halliday's most important methodological insights, yet it is often easier to recognise in principle than to maintain in practice. This becomes particularly evident when attention shifts from language to visual semiosis.

The moment analysts begin discussing images, there is a strong temptation to start with what can be seen.

The analysis begins with colours, lines, shapes, layouts, compositions, salience, framing, vectors, and countless other visible features. Having identified these expressive forms, the analyst then attempts to determine what they mean.

The procedure appears natural enough. Images are visible artefacts. Why not begin with what is visible?

From a Hallidayan perspective, however, this move reverses the direction of explanation.

The central insight of systemic-functional theory is that explanation proceeds not from the observable toward the abstract, but from the abstract toward the observable. Halliday  (IFG4: 49) describes this as giving priority to the view from above:

Giving priority to the view ‘from above’ means that the organising principle adopted is that of system: the grammar is seen as a network of interrelated meaningful choices. In other words, the dominant axis is the paradigmatic one: the fundamental components of the grammar are sets of mutually defining contrastive features. Explaining something consists not in stating how it is structured but in showing how it is related to other things: its pattern of systemic relationships, or agnateness.

This principle is often understood as a preference for paradigmatic analysis over syntagmatic analysis. While that is certainly true, the implications are much broader.

The view from above permeates the architecture of systemic-functional theory itself.

In the dimension of axis, structure is explained through system. A structural configuration becomes intelligible through its relation to a network of systemic alternatives.

In the dimension of stratification, expression is explained through content. An expressive form becomes intelligible through the meanings it realises.

In the dimension of instantiation, an instance is explained through potential. A particular semiotic event becomes intelligible through the potential of which it is an actualisation.

The nature of these relations differs. In axis and stratification, explanation proceeds through relations of realisation. In instantiation, explanation proceeds through a relation of potential and instance. What remains constant is not the grammatical form of the relation but the explanatory orientation. The observable term is not treated as theoretically primary. Explanation proceeds from above.

This is not merely a preference. It is built into the architecture of the theory.

The distinction becomes particularly important when visual semiosis enters the discussion.

Visual analysis frequently begins from the opposite direction. Analysts identify visible features and then attempt to infer meanings from them. Colour is perhaps the most familiar example. A particular colour is identified and then assigned an interpretation: blue may signify tranquillity, red may signify danger, green may signify nature, and so forth.

Whatever the merits of particular interpretations, the explanatory movement is already problematic. The analysis begins with an expressive form and attempts to work upward toward meaning.

The difficulty is not that colours cannot realise meanings. They clearly can.

The difficulty is that the explanatory order has been reversed.

From a Hallidayan perspective, the question is not:

What does blue mean?

The question is:

What meaning potential is being realised through this chromatic expression?

The difference may appear subtle, but it changes the entire character of the analysis.

The first question treats the visible feature as the starting point. The second treats the visible feature as the phenomenon to be explained.

This distinction becomes even more important when we recall that visual semiosis is not language.

Halliday's account of language is unusual among semiotic theories because language possesses a content plane that is itself stratified into semantics and lexicogrammar. Visual semiosis does not appear to possess this additional level of symbolic abstraction. It operates through content and expression, but not through a further lexicogrammatical stratum.

This difference has often encouraged analysts to search for visual equivalents of linguistic categories. Images are described as though they possess vocabularies, grammars, clauses, or syntactic structures.

Yet the absence of lexicogrammar does not imply the absence of system.

Nor does it justify abandoning the view from above.

On the contrary, it makes that methodological commitment even more important.

Without the guidance provided by lexicogrammatical analysis, it becomes very easy to drift toward a theory organised around visible forms. The analyst begins with the image and attempts to discover meanings hidden within it.

A systemic-functional approach proceeds differently.

It begins with meaning potential.

It asks what systems of meaning are available.

It asks how those systems are actualised in particular visual instances.

It asks how visual content is realised through visual expression.

And only then does it turn to the expressive forms themselves.

This series begins from the conviction that any genuinely systemic-functional account of visual semiosis must preserve this commitment.

The challenge is not to discover a visual grammar that mirrors language.

Nor is it to catalogue the meanings of colours, shapes, or layouts.

The challenge is to understand visual semiosis from the same theoretical perspective that Halliday brought to language: a perspective in which systems take precedence over structures, content takes precedence over expression, potential takes precedence over instances, and context takes precedence over the semiotic systems through which it is realised.

In short, the challenge is to learn once again how to look from above.

The next post turns to the place of visual semiosis within Halliday's broader semiotic architecture. If visual semiosis is neither language nor a lesser version of language, where exactly does it sit within the dimensions of stratification, instantiation, and metafunction that organise systemic-functional theory?