Phonology

• 1. Introduction

  ›

  Phonology is the branch of linguistics that asks a deceptively simple question: how do languages organize the building blocks of words? In English, the difference between bat and gnat comes down to just one sound. In American Sign Language, two signs glossed as CAR and WHICH look nearly identical except for the shape of the hand. In both cases, one small change carries a completely different meaning. That is phonology at work.

  But the field is far older and stranger than most people realize. Its roots reach back to a Sanskrit grammar from the 4th century BCE. Its modern form was shaped by a Polish scholar in a series of lectures in 1876-1877. And over the past century, competing schools of thought have argued fiercely over what a sound even is, and how the human mind handles the difference between two sounds that are physically distinct but psychologically identical.

  This documentary follows phonology through those arguments and discoveries, from Pāṇini's ancient inventory to the optimality theory developed at a summer institute in 1991.

• 2. Ancient Roots

  ›

  Evidence for systematic investigation of the sounds of a language appears in the Ashtadhyayi, a Sanskrit grammar composed by Pāṇini in the 4th century BCE. Within its auxiliary work, the Shiva Sutras, Pāṇini provided what amounts to an inventory of Sanskrit phonemes, complete with a notational scheme that runs throughout the main text. The main text itself concerns morphology, syntax, and semantics, but the phoneme inventory underpins it all.

  Ibn Jinni of Mosul brought similar rigor to Arabic in the 10th century. He was a pioneer in phonology and wrote prolifically on Arabic morphology and phonology in works including Kitāb Al-Munṣif and Kitāb Al-Muḥtasab. These scholars were not simply cataloguing curiosities. They were trying to understand the machinery behind meaning, centuries before the word phonology even existed.

  The word itself derives from Ancient Greek. Phonē meant voice or sound, and the suffix logy came from logos, meaning word, speech, or subject of discussion. Both halves of the term have roots in the ancient world, which is fitting for a discipline with such a long history.

• 3. The Phoneme Takes Shape

  ›

  Jan Baudouin de Courtenay, a 19th-century Polish scholar, is considered the starting point of modern phonology. Working together with his students Mikołaj Kruszewski and Lev Shcherba at the Kazan School, he shaped the modern usage of the term phoneme in a series of lectures in 1876-1877.

  The word phoneme itself had been coined a few years before those lectures, in 1873, by the French linguist A. Dufriche-Desgenettes. At a meeting of the Société de Linguistique de Paris on the 24th of May, Dufriche-Desgenettes proposed that phoneme serve as a one-word equivalent for the German term Sprachlaut. Baudouin de Courtenay then took that term and built a theoretical framework around it, work that was often unacknowledged at the time.

  Baudouin de Courtenay also worked on the theory of phonetic alternations, the area now called allophony and morphophonology. According to E. F. K. Koerner, he may even have had an influence on the work of Ferdinand de Saussure, whose distinction between langue and parole later became central to how phonology separated itself from phonetics.

• 4. The Prague School

  ›

  Prince Nikolai Trubetzkoy stands as one of the defining figures of interwar linguistics. A leading member of the Prague school, he defined phonology in his Grundzüge der Phonologie as the study of sound pertaining to the system of language, directly contrasting it with phonetics, which he described as the study of sound pertaining to the act of speech. The book was published posthumously in 1939 and is among the most important works in the field from that period.

  Directly influenced by Baudouin de Courtenay, Trubetzkoy is considered the founder of morphophonology, though he acknowledged that Baudouin de Courtenay had already recognized the concept. Trubetzkoy also developed the idea of the archiphoneme, a unit that represents what two phonemes have in common when a contrast between them is neutralized in certain positions.

  Roman Jakobson, another central Prague school figure, became one of the most prominent linguists of the 20th century. His work on distinctive features later fed directly into the generative revolution that would reshape the field in the 1960s. Louis Hjelmslev's glossematics offered a parallel contribution, focusing on linguistic structure in a way that was deliberately independent of phonetic realization or semantics.

• 5. The Generative Revolution

  ›

  In 1968, Noam Chomsky and Morris Halle published The Sound Pattern of English, known in the field as SPE. That book became the basis for generative phonology, a framework in which phonological representations are sequences of segments built from distinctive features.

  Those features were an expansion of earlier work by Roman Jakobson, Gunnar Fant, and Morris Halle. In the generative view, every feature describes some aspect of articulation or perception, draws from a universally fixed set, and carries a binary value of either plus or minus. The grammar then operates through ordered rules that transform an underlying representation into the actual surface pronunciation.

  One significant consequence of SPE's influence was the downplaying of the syllable as an analytical unit, with much greater emphasis placed on segments instead. The generativists also folded morphophonology into phonology proper, a move that resolved certain problems but created new ones. A year later, in 1969, David Stampe began publishing the work that would become natural phonology, a competing theory arguing that phonological processes are universal and that languages differ mainly in which of those processes they suppress.

• 6. New Architectures

  ›

  John Goldsmith introduced autosegmental phonology in 1976, and the field has not looked the same since. Rather than treating phonological phenomena as operating on a single linear sequence of segments, autosegmental phonology proposes parallel sequences of features residing on multiple tiers at the same time.

  That framework later evolved into feature geometry, which became a standard theory of representation across approaches as different as lexical phonology and optimality theory. Government phonology arrived in the early 1980s, attempting to unify theoretical ideas from syntax and phonology. Its proponents, including Jonathan Kaye, Jean Lowenstamm, Jean-Roger Vergnaud, Monik Charette, and John Harris, argued that all languages follow the same small set of inviolable principles but differ in how they select certain binary parameters.

  At an LSA summer institute in 1991, Alan Prince and Paul Smolensky developed optimality theory, which reframed the whole enterprise. In this approach, languages choose a pronunciation that best satisfies a ranked list of constraints. A lower-ranked constraint can be violated when doing so is required to obey a higher-ranked one. John McCarthy and Alan Prince soon extended the approach to morphology, and it has become a dominant trend in the field. Critics, notably Mark Hale and Charles Reiss, have challenged what they call the phonetic grounding of constraints, advocating instead for substance-free phonology.

• 7. What Counts as the Same Sound

  ›

  In English, the letter p produces two physically different sounds. In the word pot, the p is aspirated, written as ph in phonetic notation. In the word spot, the same letter produces an unaspirated p. Native English speakers intuitively hear both as the same sound, making them allophones of the single phoneme /p/.

  In Thai, Bengali, and Quechua, the situation is different. Those languages have minimal pairs, two words with different meanings where aspiration is the only contrasting feature. The same physical difference that English speakers ignore is, in those languages, the entire point.

  The question of what counts as one phoneme versus two has not stayed static within a single language either. In older forms of English, f and v were allophones of the same phoneme. They share the same place and manner of articulation and differ only in voicing. Over time they separated into distinct phonemes, a shift that illustrates what historical linguistics calls one of the main mechanisms of language change. Research in speech perception has further complicated the picture. Interchanged allophones of the same phoneme can produce words that are genuinely unrecognizable, and real speech is so heavily co-articulated that cleanly splicing it into segments is often impossible without distorting perception.

• 8. Beyond the Phoneme

  ›

  The phoneme is not the end of the story. Phonology also covers how sounds alternate across different forms of the same morpheme, what linguists call allomorphs. It studies syllable structure, stress, tone, intonation, and feature geometry.

  Phonological constraints on which sounds can appear in which positions are called phonotactics. The rules governing how a sound changes in context, sometimes in a fixed order where one rule feeds into or bleeds another, make up phonological alternation. Above the level of the segment, prosody examines stress and intonation as suprasegmental features, meaning features that extend across more than one sound.

  Since the early 1960s, many theoretical linguists have moved away from the traditional phoneme concept toward a more abstract unit tied to morphemes, called the morphophoneme. Phonological principles also extend beyond spoken language entirely. The same analytical tools have been applied to sign languages, where sublexical units are not speech sounds at all but components of gesture. In American Sign Language, the contrast between APPLE and ONION turns on location alone, the same logical structure as a minimal pair in spoken language, just in a different medium.