The Seven Parts of Speech
The familiar seven parts of speech — noun, pronoun, verb, adverb, preposition, conjunction, and interjection — have a lineage stretching back over two millennia. Their origins lie in the work of Dionysius Thrax (c. 170–90 BCE), whose Téchnē Grammatikē (The Art of Grammar) classified Greek words into eight categories. The Roman grammarian Aelius Donatus (c. 350 CE) later adapted this scheme for Latin. Since Latin has no articles (words like the, a, and some in English), he omitted the article and thus fixed the list at seven. Through Priscian (c. 500 CE) and the Latin grammars of the Middle Ages, this model passed into the English tradition, reaching its canonical form in William Lily’s Grammar of the sixteenth century.
Lily and the Standardisation of Learning
William Lily (c. 1468–1522) wrote his grammar to provide a standard Latin textbook for English schools, unifying what had become a confused patchwork of competing grammars.
By the late fifteenth century, every English grammar school used its own Latin grammar. Many were based on medieval compendia full of inconsistencies and errors. Pupils moving from one school to another often had to start again, and teachers argued endlessly about rules and terminology. Latin was still the foundation of education, law, and the Church — so the lack of standardisation was a serious problem.
Lily, educated at Oxford and in Italy, brought back the humanist methods of Renaissance scholarship. Around 1510 he composed a new Latin grammar in clear, classical style, influenced by Erasmus and other reformers. His goal was to restore Latin teaching to the purity of ancient models and to replace medieval “barbarisms” with genuine classical usage.
After Lily’s death, his grammar was revised and refined by John Colet, Thomas More, and Erasmus, whose humanist ideals had inspired its composition. In 1540, under Henry VIII, the work was published officially as The Grammar (Rudimenta Grammatices). Acting on the educational principles earlier promoted by these reformers, Henry proclaimed that all schools in England should use this single grammar book, thereby establishing the first national standard in English education.
Latin and the Discipline of the Mind
Unless they were schooled in Latin from birth — as Montaigne famously was — learning Latin was arduous work. For centuries, schoolboys laboured daily under the eye of the master, parsing, translating, and copying until grammar became reflex. It was not a language of conversation but of construction: built word by word through rule, correction, and repetition. Girls were largely excluded from this intellectual forge.
Montaigne’s case is especially revealing. His father arranged for Latin to be his native tongue: the household spoke it from his infancy so that he might think in it before he learned French. Yet in later life he confessed that he had forgotten most of it. The admission is telling. Even when acquired as a mother tongue, Latin could not survive without constant practice, for it had no living community to sustain it. It remained a learned language — exacting but brittle — a discipline that sharpened minds even as it confined them.
Henry VIII and the Politics of Language
Henry was trained from childhood in Latin composition and rhetoric, could read classical authors, and sometimes wrote letters and marginal notes in Latin. His early treatise Assertio Septem Sacramentorum (1521), though probably polished by Thomas More or Bishop Fisher, shows that Henry could dictate or draft extended Latin argumentation and understand subtle theological vocabulary. By modern standards, Henry VIII’s Latin would have been around the B2 level — fluent enough for correspondence and formal writing, but not of the scholarly depth of men like More or Erasmus.
Henry VIII came to power in a country still recovering from civil war. The Tudor dynasty’s survival depended on unity, both political and religious. To Henry, “diversity” — whether of faith, schooling, or law — meant potential rebellion. Centralisation was the price of peace. Once he broke from the Papacy (1534), he could no longer rely on Rome to define doctrine or educational orthodoxy. The Church had been the main provider of schooling; monasteries and clergy controlled much of the curriculum. With their dissolution, Henry had to rebuild authority in his own name.
A New Order
Standardising grammar teaching — along with the Bible in English and the Book of Common Prayer — was part of this new order.
- A single language of learning (Latin) ensured intellectual conformity across schools.
- A single doctrinal loyalty (to the Crown, not to Rome) secured religious and moral unity.
- A single national identity followed naturally from both: uniform education and uniform faith created a uniform polity.
Henry’s reforms were primarily religious and administrative, not linguistic. He never issued any decree about spelling, pronunciation, or dialect. His immediate aim was uniformity of authority — that everyone should read, worship, and obey under the same crown, not necessarily in the same dialect.
However, several of his key policies had unifying effects on English usage:
- The Great Bible (1539): first authorised English translation for public worship.
- The Book of Common Prayer (1549): fixed forms of worship in English reinforced uniformity.
- Centralised administration: government and legal documents increasingly used standard written English.
Together these developments meant that southern, courtly English — the dialect of London and Westminster — became the model for literacy, gradually eclipsing regional forms.
The Formation of Standard English
England’s monarchy and island geography thus encouraged the emergence of a single standard language far earlier than on the Continent. Unlike Germany, which remained a loose federation of principalities until the nineteenth century, England was governed by a single crown — notwithstanding the occasional contest between claimants. The same centrifugal pattern held in Italy and France, where regional identities and local dialects long resisted central control, and even Spain and Portugal retained strong provincial variations in speech and custom.
The decisive break in English linguistic history came after 1066, when the Norman Conquest imposed French at court and in administration for nearly three centuries, dealing the death blow to Old English. Yet once English re-established itself as the national tongue, it did so with remarkable uniformity. There was never in England the profusion of dialects found in Germany, Austria, and Switzerland, where local varieties continue to flourish to this day.
This early linguistic consolidation had far-reaching consequences. A common language fostered not only administrative coherence but also a shared sense of identity that made national policy, education, and expansion abroad far more cohesive. In that sense, linguistic unity was one of the hidden preconditions of Britain’s later world power: the empire spread a single language outward because, unusually for Europe, it had first been unified by one at home.
Because linguistic diversity was normal in Germany, accent indicated origin, not class. In England, where one form of English dominated public life, speech became a social marker, distinguishing the “educated” from the “provincial.”
Aristotle and the Origins of Grammar
Before Aristotle, philosophers such as Plato and the Sophists had debated the nature of words and names, but it was Aristotle (384–322 BCE) who first connected language to logic. In On Interpretation he distinguished between the parts of speech and analysed how meaning arises when words combine to form propositions capable of truth or falsity. From this insight grew the later discipline of grammar.
Through the Stoics and Dionysius Thrax, Aristotle’s framework passed into Latin scholarship: Donatus, Priscian, and finally Lily were all heirs to his method, even if they knew it only at second hand. Every parsing of a noun and verb, every rule of agreement or predication, rests on Aristotle’s original idea that speech mirrors the structure of reasoning.
In a deeper sense, Aristotle also articulated a problem that would not be solved until the age of computation: the relation of the sentence to the text. His logic analysed propositions, but not the larger coherence of discourse — a task only revived in the twentieth century by theories of text grammar (Harweg, Beaugrande, Dressler) and, later, by the algorithms of artificial intelligence. What modern computing formalised in code — the linkage of clauses, dependencies, and reference — is the direct descendant of Aristotle’s attempt to map thought onto language. These early theories — despite their use of logical terms — produced only descriptions of mind, not rigorous definitions. They began to trace the forms of coherence but did not uncover the mechanism that sustained them.
From Sentences to Texts
Chomsky’s work concentrated on the internal structure of the sentence — how words combine into phrases and clauses according to hidden rules. Yet the implications went further. If sentences have an underlying grammar, might not texts have one too? Chomsky himself hinted at this but left the task to others. He had, in effect, set the ball rolling. Linguists began to search for the principles that make a sequence of sentences into a coherent whole — the grammar of the text. What they envisaged was a kind of universal text grammar, governed by rules of connection and coherence rather than syntax. Two early insights emerged: first, that a “well-formed” text must be complete and coherent; and second, that these requirements mirror what every schoolchild learns in composition — that a story should have a beginning, a middle, and an end. Sometimes a re-examination of the obvious opens the way to new understanding.
When Chomsky’s theory moved beyond the sentence, it exposed the next frontier: the organisation of meaning across discourse. A text, after all, is not just a string of grammatical sentences but a living structure of reference and continuity. The search for a “text grammar” became an attempt to formalise how ideas link and develop — how one statement calls up the next, how topics are introduced, maintained, and resolved. The difficulty, as later research showed, was that such rules blur the line between linguistic structure and cognition itself.
The pioneers of text linguistics were often feeling their way forward, groping for concepts that language itself had not yet named. Their vocabulary sometimes grew grander as their clarity diminished. Petöfi, in particular, tried to combine transformational grammar, logic, and semiotics into a single “text theory,” but the result was less a model than a maze. Dressler’s work was more cautious and empirical: instead of constructing abstract systems, he observed how real texts achieve coherence through theme, reference, and cohesion, moving toward a descriptive rather than a deductive theory of textual structure.
The whole enterprise soon became vast, and many lost sight of the larger question. Scholars were bogged down in technicalities, as if the details of cohesion could explain consciousness itself. My own master’s dissertation (1979) — on the use of paraphrase as an empirical method for establishing text typologies — was based on Egon Werlich’s Eine Typologie der Texte (1975) at the University of Dortmund. Mine was a narrow study comparing newspaper prose with poetry. We were all trying to bring deep processes to the surface, to formalise what language does instinctively. Yet language operates at a far deeper level, where learning and intuition work below awareness. We were, as I later realised, sitting in one train watching another move — and getting nowhere. It was not the linguists who discovered artificial intelligence, but the computer scientists and programmers working in other laboratories, who approached mind not as theory but as construction.
Chomsky and the Logic of Language
Aristotle gave us the form of reasoning — the framework and vocabulary through which reasoning itself could be examined and tested. Chomsky, two millennia later, sought to explain why language naturally follows those same patterns, exploring the origin of that framework in the mind.
His early transformational-generative grammar proposed that a finite set of rules could generate an infinite set of sentences, moving from an abstract deep structure to a spoken surface structure by orderly transformations. In that vision, grammar is not ornament but mechanism — the hidden machine of meaning.
Later, Chomsky reframed the idea as Universal Grammar: not a cookbook of rules but an innate endowment of the human mind — principles and constraints that make rapid language acquisition possible. Whether expressed as Principles and Parameters or, later, as the Minimalist Program (in which a single operation, Merge, recursively builds structure), the point was the same: human language is rule-governed, hierarchical, and potentially unbounded in principle.
Recursion and the Challenge of the Pirahã
Daniel Everett’s work with the Pirahã, a small indigenous community living along the Maici River in the Brazilian Amazon, challenged not a footnote of Chomsky’s theory but one of its central pillars — the claim that recursion is the defining universal property of human language.
After years of fieldwork, Everett reported that Pirahã grammar lacks recursive clause embedding: speakers express complex ideas through sequences of simple statements rather than nesting one within another. He glossed this with a simple idea often dressed in jargon: they talk about what they can see or know firsthand. The claim shook generative linguistics, which had treated recursion as the hallmark of all human speech. If Everett was right, language could no longer be seen purely as a biological constant, but also as a cultural adaptation shaped by lived experience.
The response from the Chomskyan establishment was intense and often personal. Everett’s findings were dismissed, his interpretations questioned, and he was largely ostracised from mainstream generative linguistics. The dispute became a clash between two paradigms: language as mental computation versus language as cultural behaviour. Everett’s academic isolation was real. His work was criticised in major journals, funding became scarce, and Chomsky himself publicly called his claims “absurd.” Yet his data have never been decisively refuted — only reinterpreted by those unwilling to abandon the universality of recursion. The controversy left a lasting mark: even among Chomsky’s supporters, few now insist that recursion alone defines human language. The field has moved toward a more pluralist view, where recursion is one manifestation of a broader human capacity for hierarchical thought, but not its sole expression.
From Aristotle to Artificial Intelligence
This was more than linguistics; it was a bridge back to logic and forward to computation. Chomsky’s formalisation linked naturally to formal language theory — the Chomsky hierarchy of grammars (regular, context-free, context-sensitive, recursively enumerable) and their companion automata (finite automata, pushdown automata, Turing machines). Where Aristotle analysed valid inference, Chomsky mapped well-formed expression; both revealed that thought is not merely expressed in language — it is structured by it.
When Chomsky’s model was applied beyond English, its elegant machinery began to strain. Many of the world’s languages organise meaning in very different ways: some allow free word order, where position matters little because grammatical endings carry the sense; others have rich morphology, packing into a single word what English spreads across several; still others use ergative alignment, where the subject of a transitive verb is marked differently from the subject of an intransitive one; or non-configurational syntax, where relationships depend on context rather than fixed sequence. In such systems, the tidy hierarchy of subject, verb, and object — so natural to English — breaks down, and the idea of a universal grammar built on those relations becomes harder to sustain.
Even so, Chomsky’s central legacy remains widely accepted: that language is not a flat string of words but a structured system in which ideas combine hierarchically. Beneath the surface variety of speech lies an architecture of meaning that reflects how human thought itself is organised.
Geoffrey Hinton’s work in neural networks marked the essential jump: from describing the mind to producing one in function. Yet, curiously, even now we have no formal explanation of how that invention works. The network behaves intelligently without our fully understanding the grounds of its intelligence — a mirror, perhaps, of our own condition.
From Aristotle’s early analysis of language as an instrument of logic to the refinement of English as a medium of thought and feeling, this long evolution traces the effort to bring order to expression.
English as the Natural Instrument of Logic
While logic was being formalised by philosophers and mechanised by computers, one natural language — English — evolved almost accidentally into a parallel instrument of reason.
By shedding the inflections of its ancestors, English made order itself carry meaning.
It became linear, procedural, and explicit — a language of sequence and relation rather than case and gender.
Where Greek and Latin allowed ideas to weave and nest, English insisted on flow: subject, verb, object.
Each clause executes its thought in turn, like the steps of a program.
It is the closest thing to a spoken algorithm humanity has produced.
English syntax mirrors the logical structure of thought: conditional, progressive, and hierarchical.
Its reliance on primary auxiliaries (tense) — do, be, have, will — makes meaning explicit rather than implicit, whereas the use of secondary auxiliaries (mode) — may, can, shall, might, could, should — introduces shades of uncertainty, intention, or doubt that render meaning implicit.
Its fixed word order reveals relations that other languages conceal in endings.
Even its connective words — if, then, else — are the atoms of procedural reasoning.
By historical accident, English evolved into a language that externalises logic itself.
Over years of teaching and analysis, I came to see that every self-contained passage of great writing forms a kind of living ecosystem — what I now call a semantic network. Beneath the surface structure of words lies a pattern of recurring images and associations that bind the text into coherence. It is a modern extension of the old text-immanente Methode of explication de texte, which treated even a stanza or paragraph as a world complete in itself. In Keats’s La Belle Dame sans Merci, for example, the alternation of light and darkness, life and desolation, through the poet’s choice of words, builds an invisible web of meaning. Such networks reveal that language operates not only by grammar but by resonance — that thought itself is patterned long before it is spoken.
The Music Within the Machine
And yet — within that tight syntax, within those logical bones — English carries the opposite potential: music.
The same language that frames the world’s codebases and legal contracts also gave birth to Shakespeare’s cadences, to Hopkins’s rhythmic ecstasies, and to Carroll’s nonsense.
The paradox of English is that its logic never stifled its lyricism; it intensified it.
Hopkins, in his “sprung rhythm,” worked with an almost mathematical precision of stress, yet his verses ignite with spiritual fire:
Glory be to God for dappled things —
For skies of couple-colour as a brindled cow…
He takes the procedural clarity of English and fills it with counterpoint, as though the language itself were straining against its logical scaffolding toward pure sound.
Logic becomes pulse; precision becomes prayer.
Lewis Carroll, in Jabberwocky, achieves the same miracle in play.
He keeps the syntax flawless but dissolves the lexicon:
’Twas brillig, and the slithy toves
Did gyre and gimble in the wabe.
The sentence structure is immaculate; the meaning vanishes — yet we still understand.
The grammar alone carries the ghost of sense.
Carroll’s nonsense depends on the same thing that makes computers work: syntax without semantics.
It proves how deeply logical English is — and how, within that logic, imagination can still breathe.
Even at the level of sound, English demonstrates its dual nature.
Its minimal pairs — ship/sheep, pin/pan, light/right — give poets and speakers the subtlest levers of contrast.
A single vowel or consonant shift can change everything.
That precision of sound mirrors the precision of syntax: both rest on the smallest possible distinctions — the digital heart of the language.
Coda: The Depth Beneath Speech
It seems counterintuitive to think that thought can exist before consciousness, yet the mind repeatedly proves it so. Images and insights appear unbidden, like lightning across an unseen landscape. In dreams we witness the mind’s power to create whole worlds from nothing — complete, vivid, and instantaneous.
What we call consciousness is only the surface where these deeper currents break into sound and symbol. God touching the finger of Adam in Michelangelo’s fresco is not merely the birth of man, but the meeting of the unconscious with awareness — the moment the depths of being find a form.
Language, for all its brilliance, records only the ripple, not the sea. The grammar of thought is older than speech; the music of meaning was playing long before the first word was heard.
