Suprasegmental phonology studies phonological properties that extend over units larger than a single segment — primarily stress, tone, and intonation. Stress languages (like English or Russian) use relative prominence to distinguish words (REcord vs. reCORD) and organize rhythmic patterns, while tone languages (like Mandarin or Yoruba) use pitch distinctions on individual syllables to differentiate lexical meaning. Intonation operates at the phrase and sentence level, conveying pragmatic information such as whether an utterance is a statement, question, or expression of surprise. Prosodic structure organizes speech into hierarchical constituents (syllables, feet, prosodic words, intonational phrases) that do not always align with syntactic boundaries, creating a semi-independent phonological hierarchy.
Learn to produce and perceive the four tones of Mandarin Chinese to internalize how pitch functions lexically. Analyze English stress patterns by comparing noun-verb pairs (PERmit/perMIT, CONtract/conTRACT) and noticing where stress shifts. Record yourself reading a paragraph and mark the intonational phrases, noting where pitch rises and falls and how these contours signal discourse structure.
From your study of phonological systems, you're familiar with how individual sounds — segments — are organized by distinctive features and phonotactic rules into the syllables and words of a language. Suprasegmental phonology studies the layer above the segment: phonological properties that stretch across multiple segments simultaneously, shaping the rhythmic and melodic texture of speech. The three major suprasegmental systems are stress, tone, and intonation, and while they all involve pitch and duration, they operate at different levels and serve different functions.
Stress is relative prominence among syllables. In English, stress distinguishes words (the noun *REcord* vs. the verb *reCORD*) and organizes the rhythmic pattern of connected speech. English is often called a stress-timed language because stressed syllables occur at roughly regular intervals — unstressed syllables in between are compressed and reduced, which is why English often sounds hurried to speakers of syllable-timed languages like French or Spanish. The key insight is that stress is relational, not absolute: a syllable is stressed relative to its neighbors, and what characterizes a stressed syllable is a combination of increased pitch, duration, and intensity, with pitch typically most salient.
Tone operates on individual syllables or morae to encode lexical or grammatical distinctions. In Mandarin, the four tones — high level, rising, dipping-then-rising, falling — distinguish completely different words that share the same consonants and vowels: *mā* (mother), *má* (hemp), *mǎ* (horse), *mà* (scold). Yoruba uses three level tones (high, mid, low) to similar effect. This is not a quirky feature of unusual languages: the majority of the world's approximately 7,000 languages use lexical tone, making tone languages the typological norm and English's non-tonal grammar the minority case. Learning a tone language requires training the phonological system to treat pitch differences as contrastive at the lexical level, which is why tone acquisition is difficult for speakers of non-tonal first languages.
Intonation operates at the level of phrases and utterances, encoding pragmatic information — whether something is a statement, question, request, or expression of surprise. English falling intonation at the end of a clause typically signals a statement; rising intonation signals a yes-no question. But intonation goes beyond sentence-type marking: it also tracks information structure, indicating which elements are new versus given, signaling whether a speaker has finished or is expecting a turn-change, and expressing attitude. Crucially, tone languages have intonation systems too — lexical tones encode word meaning while sentence-level intonation contours are overlaid on top. Separating these two pitch systems in analysis is one of the technical challenges of tonal phonology.
Underlying all three systems is prosodic structure — the hierarchical organization of speech into syllables, metrical feet, prosodic words, phonological phrases, and intonational phrases. This prosodic hierarchy is semi-independent of syntactic structure: prosodic breaks don't always align with syntactic boundaries, and the rules governing stress assignment, tone sandhi (tonal changes at boundaries), and intonational phrasing reference prosodic categories, not syntactic ones. Understanding suprasegmental phonology means understanding that the melodic and rhythmic shape of speech is itself a structured, rule-governed system — not an expressive overlay on an otherwise complete linguistic signal, but an integral part of the signal itself.