Phonetics & Phonology
Every human language is built on a finite inventory of contrastive sounds. Phonetics studies how those sounds are physically produced (articulatory), transmitted (acoustic), and perceived (auditory). Phonology studies how sounds function within a particular language -- which distinctions are meaningful, which are automatic, and how sounds interact with each other in sequence. This skill treats language-universal principles rather than any single language's system, making it applicable to any language a learner encounters.
Agent affinity: crystal (phonetic description, language diversity), chomsky-l (phonological rules, underlying representations)
Concept IDs: lang-phoneme-inventory, lang-ipa-notation, lang-ear-training, lang-suprasegmentals
The Sound Landscape at a Glance
| Domain | Focus | Key Tools |
|---|---|---|
| Articulatory phonetics | How sounds are produced | Place/manner/voicing, vocal tract diagrams |
| Acoustic phonetics | Sound waves and formants | Spectrograms, F1/F2 vowel plots, VOT |
| Auditory phonetics | How sounds are perceived | Categorical perception, ear training drills |
| Phonemics | Contrastive sound units | Minimal pairs, complementary distribution |
| Phonotactics | Legal sound sequences | Onset/coda constraints, syllable structure |
| Prosody | Suprasegmental features | Stress, tone, intonation contours |
Articulatory Phonetics: Producing Sounds
Consonant Classification
Every consonant is classified along three dimensions:
- Voicing. Whether the vocal folds vibrate (voiced: [b, d, g]) or not (voiceless: [p, t, k]).
- Place of articulation. Where in the vocal tract the constriction occurs: bilabial (both lips), labiodental (lip + teeth), dental, alveolar, postalveolar, retroflex, palatal, velar, uvular, pharyngeal, glottal.
- Manner of articulation. How the airstream is modified: plosive/stop (complete closure then release), fricative (narrow channel producing turbulence), affricate (stop + fricative), nasal (oral closure, air through nose), lateral (air around sides of tongue), approximant (narrowing without turbulence), trill, tap/flap.
Example. English [t] is a voiceless alveolar plosive. Japanese [ts] in "tsunami" is a voiceless alveolar affricate. The distinction matters: substituting one for the other changes meaning or intelligibility in context.
Vowel Classification
Vowels are classified by tongue position along two axes, plus lip rounding:
- Height (close/high to open/low): close [i, u], close-mid [e, o], open-mid [ɛ, ɔ], open [a, ɑ]
- Backness (front to back): front [i, e, ɛ], central [ə, ɐ], back [u, o, ɑ]
- Rounding (rounded or unrounded): French [y] is a close front rounded vowel -- close and front like [i] but with rounded lips
The IPA vowel trapezoid plots these dimensions, and formant analysis (F1 for height, F2 for backness) provides acoustic confirmation.
Voice Onset Time (VOT)
VOT measures the delay between the release of a stop consonant and the onset of voicing for the following vowel. It distinguishes:
- Prevoiced (negative VOT): vocal folds vibrate before release. Spanish [b], French [d].
- Short-lag (VOT near zero): voicing begins at release. English [b], Mandarin unaspirated [p].
- Long-lag (positive VOT): significant aspiration before voicing. English [pʰ], Mandarin aspirated [pʰ].
This is a major source of cross-linguistic interference: a Spanish speaker's unaspirated [p] sounds like English [b] to an English listener, and vice versa.
The International Phonetic Alphabet (IPA)
The IPA provides a universal transcription system where one symbol corresponds to one sound. This skill assumes familiarity with IPA as a meta-tool for describing any language.
Transcription Types
- Broad (phonemic) transcription uses slashes: /kæt/. It records only contrastive distinctions. Two sounds that are phonemically distinct in the language get different symbols.
- Narrow (phonetic) transcription uses brackets: [kʰæt]. It records phonetic detail including allophonic variation -- the aspiration on English [kʰ] that is not contrastive but is physically present.
Minimal Pair Analysis
A minimal pair is two words that differ by exactly one sound in the same position and have different meanings: "bat" /bæt/ vs. "pat" /pæt/. Minimal pairs prove that two sounds are separate phonemes. If no minimal pair can be found, the sounds may be allophones of the same phoneme.
Worked example. In English, [l] and [ɫ] (clear and dark L) are allophones: "leaf" [liːf] vs. "feel" [fiːɫ]. No minimal pair distinguishes them because they are in complementary distribution (clear before vowels, dark before consonants or word-finally). In Russian, [l] and [ɫ] ARE separate phonemes: "лук" /luk/ (onion) vs. "лук" /ɫuk/ would be meaningful (Cyrillic spelling distinguishes palatalized from non-palatalized).
Phonological Rules and Processes
Common Cross-Linguistic Processes
| Process | Description | Example |
|---|---|---|
| Assimilation | A sound becomes more like a neighboring sound | English "input" /ɪnpʊt/ often pronounced [ɪmpʊt] -- /n/ becomes [m] before bilabial /p/ |
| Elision | A sound is deleted | French "je ne sais pas" -> [ʃɛpa] in rapid speech |
| Epenthesis | A sound is inserted | Japanese borrows "strike" as "sutoraiku" -- vowels inserted to fit CV syllable structure |
| Metathesis | Sounds swap positions | English "ask" pronounced [æks] in some dialects |
| Lenition | A sound weakens | Spanish intervocalic /b/ -> [β] (fricative): "lobo" [loβo] |
| Fortition | A sound strengthens | Korean word-initial lenis stops become tense after nasals |
| Nasalization | Vowel acquires nasal quality | French nasal vowels: "bon" [bɔ̃] |
| Palatalization | Consonant shifts toward palatal place | Russian consonants before front vowels: /t/ -> [tʲ] |
| Vowel harmony | Vowels in a word agree on a feature | Turkish: "evler" (houses) but "odalar" (rooms) -- front/back harmony |
Rule Notation
Phonological rules are written as: A -> B / C __ D
"A becomes B in the context of C preceding and D following."
Example. English aspiration rule: voiceless stops become aspirated word-initially before a stressed vowel.
/p, t, k/ -> [pʰ, tʰ, kʰ] / # __ V[+stress]
This explains why "pin" has [pʰ] but "spin" has unaspirated [p] -- the /p/ in "spin" is not word-initial.
Suprasegmental Features
Stress
Stress is relative prominence of a syllable through louder volume, higher pitch, longer duration, or fuller vowel quality. Languages differ in whether stress is:
- Fixed (predictable from word structure): French (final), Polish (penultimate), Czech (initial)
- Free (unpredictable, must be learned per word): English, Russian, Italian
In free-stress languages, stress can be contrastive: English "CONtract" (noun) vs. "conTRACT" (verb); Russian "мука" /mukÁ/ (flour) vs. /múka/ (torment).
Tone
Tone uses pitch to distinguish word meaning. Approximately 60-70% of the world's languages are tonal.
- Level tones: Mandarin has four tones (high, rising, dipping, falling) plus neutral. "ma" means "mother" (T1), "hemp" (T2), "horse" (T3), or "scold" (T4) depending on tone.
- Contour tones: Thai, Vietnamese, Cantonese use complex pitch contours (rising, falling, low-rising, high-falling, etc.).
- Register tones: Many Bantu languages use high vs. low tone with grammatical as well as lexical function.
Intonation
Intonation is the pitch pattern over an entire phrase or sentence. Unlike tone, it is not lexical but grammatical and pragmatic:
- Rising intonation for yes/no questions in many European languages
- Falling intonation for statements
- Rise-fall for emphasis or surprise
- Language-specific patterns: Japanese uses pitch accent (one mora per word carries a pitch drop), which differs from both stress and tone
Ear Training
Perceiving non-na