You are viewing a free preview of this lesson.
Subscribe to unlock all 10 lessons in this course and every other course on LearningBro.
The study of English Language at A-Level is built upon a systematic analytical toolkit known as the language levels framework. Rather than responding to texts impressionistically — saying that a text "feels persuasive" or "sounds angry" without being able to say how it achieves those effects — linguists use this framework to identify, categorise, and evaluate the specific features of language that create meaning. The framework is, in effect, a shared professional vocabulary: it lets you move from the vague reactions of an ordinary reader to the precise, evidence-based observations of an analyst. Understanding the language levels is the foundation for everything else you will study on this course, and it is the single most important habit of mind to develop early, because every subsequent topic — phonology, graphology, lexis, grammar, pragmatics, discourse — is simply one level examined in detail.
This lesson sits at the head of the Textual Analysis course because it underpins Paper 1: Language, the Individual and Society. Paper 1 is a written exam of 2 hours 30 minutes, worth 100 marks and 40% of the A-level. Section A — Textual Variations and Representations (70 marks) presents two texts, one contemporary and one older, and asks you to analyse text 1 (25 marks), analyse text 2 (25 marks), and compare the two (20 marks). Section B — Children's Language Development is worth 30 marks. Across the paper you are assessed against five Assessment Objectives: AO1 (apply linguistic methods, use terminology, and write with technical accuracy and coherence — 26%), AO2 (demonstrate critical understanding of concepts and theories — 26%), AO3 (analyse contextual factors and how they shape meaning — 23%), AO4 (explore connections across texts — 15%), and AO5 (demonstrate expertise in your own writing — 10%). The language levels are the practical instrument through which AO1, AO2 and AO3 are demonstrated: you name the feature (AO1), explain its function with reference to concepts (AO2), and account for it through context (AO3).
The language levels are a set of interconnected analytical categories that allow us to examine how language works at every scale — from the smallest contrastive sound or visual mark in a text right through to its overall structure and the way it functions in social contexts. They form a kind of ladder, from the micro level of individual sounds and letters up to the macro level of whole-text organisation. There are several core language levels that you need to understand and apply:
| Language Level | Focus | Key Question |
|---|---|---|
| Phonology | Sound patterns and features | How does the text sound, and what effects do sound choices create? |
| Graphology | Visual and spatial features | How does the text look on the page or screen, and what effects do layout choices create? |
| Lexis | Word choices | What words have been chosen, and what effects do these choices create? |
| Semantics | Meaning | What meanings are conveyed, including connotations, figurative language, and implied meaning? |
| Grammar | Sentence and word structure | How are words and sentences constructed, and what effects do grammatical choices create? |
| Pragmatics | Contextual and implied meaning | What is implied or understood beyond the literal meaning of the words? |
| Discourse | Text-level organisation | How is the whole text structured and organised, and how does it achieve coherence? |
Key Definition: Language levels — a set of analytical categories (phonology, graphology, lexis, semantics, grammar, pragmatics, discourse) used by linguists to systematically examine how language creates meaning.
Note that lexis and semantics are often treated together because word choices (lexis) and their meanings (semantics) are closely interrelated — you cannot really discuss why a word was chosen without discussing what it means and connotes. Similarly, grammar operates at both the word level (morphology — the internal structure of words) and the sentence level (syntax — how words combine into phrases, clauses and sentences). The boundaries between levels are not walls but membranes: an alliterative headline involves phonology and graphology and lexis at once. The framework is a way of focusing attention, not a set of sealed boxes.
A crucial misconception to avoid from the outset is that the levels are separate stages to be ticked off one at a time. In a real text they operate simultaneously and reinforce one another. Consider a tabloid headline such as COPS NAB THUG. At the level of graphology it is set in large, bold, sans-serif capitals; at the level of lexis it uses short, punchy, monosyllabic words from an informal, almost slang register (cops, nab, thug); at the level of phonology the plosive sounds in cops and the abrupt monosyllables create a percussive, urgent rhythm; at the level of grammar it is a compressed declarative with the determiners and auxiliary verbs elided, a structure typical of "block language"; and at the level of pragmatics the noun thug carries a strong evaluative implicature, positioning the accused as guilty before any trial. The most sophisticated analysis explains how these levels work together to construct a particular representation of an event and to position the reader. This integrative habit is precisely what separates a top-band response from a middling one.
This is also why the order in which the levels are usually listed — from phonology and graphology, through lexis and grammar, up to pragmatics and discourse — should be understood as a rough scale from the smallest units (sounds and letters) to the largest (whole-text organisation and social meaning), not as a compulsory running order for your essay. In practice you will move up and down the scale fluidly, often starting wherever the text's most striking features lie. A poster might demand that you begin with graphology; a political speech might demand that you begin with grammar and rhetoric; a private letter might demand that you begin with pragmatics and tenor. The framework guarantees coverage and precision; it does not dictate sequence. What it always demands is that you connect the level you are discussing to the meaning the text is making, so that terminology is in the service of interpretation rather than an end in itself.
Using a systematic framework has several important advantages over simply responding to a text based on personal impressions:
Key Definition: Linguistic analysis — the systematic examination of language using established frameworks and terminology to identify features and evaluate their effects on meaning.
One of the first things to establish when analysing any text is its mode — the channel through which communication takes place.
| Mode | Characteristics | Examples |
|---|---|---|
| Written | Planned, edited, permanent, typically no immediate audience present | Novels, newspapers, letters, academic essays |
| Spoken | Often spontaneous, temporary, audience usually present, paralinguistic features | Conversations, speeches, interviews, phone calls |
| Multimodal | Combines written, spoken, and/or visual elements | Websites, social media posts, advertisements, textbooks |
The linguist Michael Halliday argued that spoken and written language are not simply different versions of the same thing — they are systems with distinct grammatical and structural tendencies. Written language tends to be lexically dense, packing many content words (nouns, lexical verbs, adjectives) into relatively few clauses; spoken language tends to be grammatically intricate, spreading meaning across longer chains of loosely linked clauses. Compare the written sentence "The committee's rejection of the proposal caused widespread disappointment" (lexically dense, heavily nominalised) with its spoken equivalent "well the committee turned the proposal down and so a lot of people were really disappointed about it" (more clauses, fewer content words per clause, more grammatical "joining" words). Neither is more sophisticated; they are adapted to different conditions of production.
Some texts deliberately blur the boundaries between modes. Text messages and social media posts, for instance, are written but often display features more typical of speech — contractions, informal vocabulary, ellipsis, and non-standard spelling that represents pronunciation (such as gonna or wanna). The linguist David Crystal coined the term Netspeak to describe the hybrid character of online communication, which he argued is best understood not as "spoken" or "written" but as a third medium with features of both. Recognising that a text occupies a blended mode is itself an analytical observation worth making, because it usually signals something about the text's informality, immediacy, and relationship with its audience.
Key Definition: Mode — the channel or medium through which a text is communicated (written, spoken, or multimodal). Mode significantly influences the language features a text is likely to display.
Register refers to the variety of language appropriate to a particular situation. In Halliday's model it is analysed through three variables, which conveniently map onto the three dimensions of any communicative situation:
| Variable | Description | Example |
|---|---|---|
| Field | The subject matter or topic of the text | Medical terminology in a doctor's report; sporting vocabulary in a match commentary |
| Tenor | The relationship between participants — including power dynamics and social distance | Formal language between a judge and defendant; informal language between close friends |
| Mode | The channel of communication (as discussed above) | A spoken lecture vs. a written textbook on the same topic |
These three variables combine to produce the register of any given text. A legal contract, for example, has a specialised field (law), a formal and distant tenor (between parties who may not know each other, with significant consequences), and a written mode — producing a register characterised by technical jargon, passive and impersonal constructions, archaic connectives such as hereinafter, and long, complex sentences. Change any one variable and the register shifts: a solicitor explaining that same contract to a nervous first-time buyer over the phone keeps the field but softens the tenor and switches the mode, producing simpler syntax, more second-person address, and reassuring lexis. Register is therefore a powerful diagnostic tool, because a mismatch between the register you would expect and the register you actually find is almost always significant — an advertisement written in mock-legalese, or a politician adopting matey colloquialism, is making a calculated rhetorical move.
Key Definition: Register — the variety of language determined by the social context in which it is used, shaped by field (topic), tenor (relationship between participants), and mode (channel of communication).
It is also worth introducing here Halliday's broader claim, from his systemic functional grammar, that language simultaneously performs three metafunctions: the ideational (representing experience and ideas about the world), the interpersonal (enacting relationships and attitudes between participants), and the textual (organising the message into a coherent whole). You will meet these again in later lessons, but notice already how neatly they correspond to the analytical questions the levels prompt: lexis and semantics largely serve the ideational; pragmatics and tenor serve the interpersonal; discourse and cohesion serve the textual. (Be careful: Halliday also proposed a set of seven functions of language — but those belong to his work on child language development, not to register analysis, and you should not confuse the two.)
Genre refers to the category or type that a text belongs to, based on shared conventions. Genres are defined by their typical content, structure, style, and purpose. Recognising genre matters because readers and listeners bring genre expectations that shape how they interpret a text before they have read a single word of the body copy: the masthead of a newspaper, the layout of a recipe, or the salutation of a letter all activate a template of expectations.
Common written genres include:
Genres are not fixed categories — they evolve over time and can be subverted or blended for effect. A novelist might borrow the conventions of a recipe or instruction manual for literary effect; an advertisement might mimic the layout and register of a news report (an "advertorial") to borrow the credibility of journalism. Part of your analysis should always consider how a text conforms to, exploits, or departs from the conventions of its genre, because conformity reassures the reader while subversion surprises and foregrounds.
It is worth distinguishing genre from mode and register, three terms students often blur. Mode is the channel (spoken, written, multimodal); register is the situational variety determined by field, tenor and mode; genre is the recognisable text-type with its accumulated conventions. A single text has all three at once: a wedding speech is spoken in mode, informal-but-ceremonial in register, and belongs to the speech / toast genre. Keeping the three apart lets you make precise, layered contextual claims rather than collapsing everything into a vague sense of "the type of text." Genre conventions are also a shared resource between producer and audience: because both parties know the template, a producer can communicate efficiently by invoking it and can create meaning by breaking it.
Every text is produced for an audience and with a purpose (or, very often, several purposes at once). Identifying these is essential because audience and purpose shape every language choice a writer or speaker makes — they are the why behind the features you will go on to analyse.
Audience can be described in terms of:
A subtle but examinable point is that texts often address more than one audience at once. A children's picture book has a primary audience of young children but a secondary audience of the adults who read it aloud and pay for it, and many of its jokes and cultural references are aimed over the children's heads. Identifying a dual or layered audience is exactly the kind of nuanced contextual observation that earns AO3 credit.
The linguist Roman Jakobson identified six functions of communication, which remain a useful framework for thinking about purpose:
| Function | Focus | Example |
|---|---|---|
| Referential | Conveying information | A news report |
| Emotive/Expressive | Expressing feelings or attitudes | A personal diary entry |
| Conative | Influencing the audience | An advertisement |
| Phatic | Maintaining social contact | Small talk ("Nice weather, isn't it?") |
| Metalingual | Discussing language itself | A grammar textbook |
| Poetic | Drawing attention to the form of the message | Poetry, advertising slogans |
Most texts serve multiple purposes simultaneously. A charity appeal letter, for instance, is referential (providing information about the cause), emotive (expressing the suffering of those affected), and conative (persuading the reader to donate) — and it may open phatically to establish rapport. The strongest analyses identify a hierarchy of purposes, distinguishing the dominant purpose from subordinate ones, rather than simply listing them.
Because AO3 (analysing how contextual factors shape language and meaning) carries 23% of the marks, context deserves dedicated attention rather than a token sentence. "Context" is not one thing but several overlapping layers, and the strongest analyses move between them:
A reliable way to earn AO3 marks rather than merely gesturing at context is to connect a specific language feature to a specific contextual factor. Not "this is from a newspaper, so it is formal," but "the agentless passive mistakes were made suits an institutional producer who wishes to acknowledge fault without naming a responsible individual." Context should explain why the language is as it is, closing the loop between feature, effect and situation.
Section A is titled Textual Variations and Representations, and representation is one of the governing concepts of the whole paper. The core idea is that language does not simply reflect a pre-existing reality; it actively constructs a particular version of people, places, events, groups and ideas. Every text offers a representation — a selective, angled, value-laden version — and many of the choices you analyse are best understood as representational choices.
Consider how the same underlying event can be represented in radically different ways through language:
Asking whose perspective a text encodes, who is foregrounded or backgrounded, who is given agency and who is acted upon, and what values are presented as natural or obvious turns language analysis into an account of how the text positions its reader to see the world. This representational lens is what unifies the language levels into a single argument, and it is the surest route to the higher bands.
When you approach any text for analysis, you should begin by establishing the contextual factors before applying the language levels:
The key to effective analysis is linking features to effects to context. It is never enough simply to identify a feature ("the writer uses alliteration"). You must always explain what effect that feature has in context and how it contributes to the text's overall meaning and purpose. A reliable structure for every analytical point is the feature → effect → context → theory chain: name the feature precisely, explain the meaning or effect it creates, account for it through the audience/purpose/context, and where appropriate frame it with a relevant concept (register, synthetic personalisation, face, and so on).
Take the opening of a fundraising email: You've already changed one child's life this year. Imagine what we could do together next. The lexis is simple and warm; the second-person pronoun you and the inclusive we/together build solidarity and shared agency; the grammar moves from a completed declarative (a confident factual claim that flatters the reader) to an imperative Imagine, drawing the reader into co-constructing a hopeful future; the pragmatics rests on a presupposition (that the reader has already given) which positions them as a committed supporter; and the whole functions to fuse a referential purpose with a strongly conative one. This is what integrated analysis looks like — every level pulling towards the same interpretive claim about how the text positions its reader.
The Section A comparison task (20 marks) asks you to explore connections between the two texts, and the language levels framework is what makes a genuine comparison possible. A weak comparison analyses text 1, then separately analyses text 2, and leaves the reader to spot the links. A strong comparison is organised around shared points of comparison — usually a language level or a contextual factor — and moves between the texts within each point.
A productive way to structure comparison is to ask, for each level or factor, whether the two texts are similar, different, or similar-but-for-different-reasons:
Because Section A pairs a contemporary text with an older one, comparison almost always invites attention to language change and to shifting context of reception: differences in spelling, lexis, register or social assumptions between the two texts can frequently be explained by the gap in time and the different worlds the texts were produced for. Connective discourse markers — similarly, by contrast, whereas, in the same way — are the surface signal of genuine comparison, but they must sit on top of a real shared analytical category, not be sprinkled decoratively. Plan your comparison around three or four substantial points of contact rather than attempting to compare everything.
Task: Analyse how the opening of a charity appeal positions its reader. (Extract: "Right now, a child is going to sleep hungry. You can change that tonight.")
Mid-band response: The writer uses the word "hungry" to make the reader feel sorry for the child. They also use "you" to talk directly to the reader. This makes the appeal persuasive and makes the reader want to donate money to the charity.
Top-band response: The appeal opens with the fronted adverbial "Right now," a deictic time reference that collapses the distance between the reader's comfortable present and the child's suffering, creating a sense of immediate moral urgency. The present-tense progressive "is going to sleep hungry" foregrounds an ongoing, unresolved action, implying that the suffering continues as the reader reads. The second-person pronoun "you" then enacts what Fairclough terms synthetic personalisation, simulating a one-to-one address within a mass-distributed text and positioning the individual reader as personally responsible. The modal "can" construes donation not as an obligation but as an empowering possibility, while the temporal adverb "tonight" presupposes that meaningful action is immediately available. Across the lexical, grammatical and pragmatic levels, the text constructs the reader as an agent whose single decision can resolve the suffering it has just made vivid.
Examiner-style commentary: The Mid-band response correctly spots two features but stops at labelling and asserts a generic effect ("makes it persuasive"). The Top-band response names features precisely across multiple levels (deixis, progressive aspect, second-person address, modality, presupposition), integrates a relevant concept (synthetic personalisation), and consistently ties each feature to a specific effect on the reader's positioning. This integration of method (AO1), concept (AO2) and contextual effect (AO3) is exactly what the framework is designed to produce.
Knowing the language levels is necessary but not sufficient; you also need a reliable way to write analysis under exam conditions. A dependable paragraph structure follows the chain introduced above — feature → effect → context → (concept) — and it can be remembered as a sequence of moves:
Two further principles distinguish strong responses. First, selectivity: you cannot say everything, so choose the most significant features — the ones doing the most representational or persuasive work — rather than marching mechanically through every level. Second, an argument, not a list: the best answers are organised around a controlling idea about how the text constructs meaning or positions its reader, with each paragraph contributing to that thesis. Examiners reward a response that argues a coherent reading over one that merely inventories features, however accurately labelled.
The single most common weakness is feature-spotting — identifying a feature and either stopping there or attaching a vague, interchangeable effect ("this makes it more interesting," "this draws the reader in"). The remedy is to interrogate every observation with "so what?" until you reach a specific, text-grounded effect. The second most common weakness is narration or paraphrase — retelling what the text says rather than analysing how it says it. Keep the focus on linguistic choices and their consequences, and you will stay on the analytical side of the line.