IPA number
An IPA number is a unique three-digit numerical identifier assigned to each symbol, diacritic, and other element in the International Phonetic Alphabet (IPA), designed to standardize the reference and computational representation of phonetic notation. Introduced by the International Phonetic Association at the 1989 Kiel Convention, these numbers serve as a communication-interchange standard independent of specific character encodings, enabling consistent use in linguistic databases, typesetting, and software for transcribing the sounds of the world's languages.[1] The numbering system categorizes IPA elements systematically, with the first digit indicating broad groups: 100–199 for consonants (pulmonic and non-pulmonic), 200–299 for non-IPA or retired consonants, 300–399 for vowels, 400–499 for diacritics, 500–599 for suprasegmentals, and higher ranges (600–999) for tones, extensions, and other elements. For instance, the voiceless bilabial plosive is assigned number 101, the close front unrounded vowel is 301, and the centralization diacritic (ˑ) is 425. This structure, revised periodically by the International Phonetic Association, supports unambiguous identification even as the IPA evolves, with the 2020 chart extending numbers up to 533 to accommodate updates in symbol usage.[2][1] IPA numbers have proven essential in digital linguistics, underpinning systems like X-SAMPA for ASCII-based phonetic transcription and facilitating the integration of IPA into Unicode standards. They allow linguists to reference symbols precisely without ambiguity from similar glyphs across scripts, aiding in phonetic analysis, language documentation, and computational phonology. While the core system remains stable, ongoing revisions reflect advances in phonetic theory, ensuring the IPA's adaptability to newly described speech sounds.[2]Introduction
Definition
The IPA numbers constitute a three-digit numerical coding system assigned to symbols within the International Phonetic Alphabet (IPA) and its extensions, facilitating standardized representation and computer processing of phonetic notation.[1] For instance, the code 101 corresponds to the voiceless bilabial plosive symbol /p/.[3] This system was established at the 1989 Kiel Convention to enable consistent communication of IPA symbols across linguistic and technical contexts.[1] The structure of these codes organizes symbols by category through the first digit, with subsequent digits providing finer distinctions such as place or manner of articulation. The 100 series designates pulmonic consonants, the 300 series denotes vowels, the 400 series covers non-tone diacritics, and the 500 series addresses suprasegmentals, while other ranges like 200 and 600–800 handle retired symbols, extensions, or reserved categories.[1] This hierarchical arrangement supports systematic indexing and retrieval in phonetic databases and transcription tools.[4] The scope of IPA numbers encompasses core IPA symbols for pulmonic consonants and vowels, along with diacritics, suprasegmentals, and select extensions for atypical or voice-quality features, including additions from post-1989 revisions.[1] For example, the 100 series includes codes for standard pulmonic consonants like 101 for /p/, while the 300 series assigns codes to vowel symbols such as 301 for the close front unrounded /i/.[3] This coverage ensures compatibility with foundational phonetic transcription needs.[4]History
The IPA numbering system originated at the 1989 Kiel Convention of the International Phonetic Association (IPA), where approximately 120 members, convened at the initiative of IPA President Peter Ladefoged, addressed revisions to the alphabet amid increasing demands for digital standardization of phonetic symbols. During the convention, Workgroup 9 specifically proposed assigning unique three-digit numerical codes to IPA symbols and diacritics to facilitate computer coding, serving as a communication-interchange standard for translation between phonetic character sets and software systems, particularly as digital publishing and computational linguistics grew in the late 1980s.[1] This initiative aimed to provide a stable, unambiguous reference for symbols, avoiding reliance on variable fonts or encodings, with numbers structured in ranges such as 100–199 for pulmonic consonants and 300–399 for vowels.[5] Initial assignments of these IPA numbers were detailed in the 1993 IPA chart, published in the Journal of the International Phonetic Association, where John H. Esling and Harry Gaylord outlined the linear ordering of codes corresponding to the revised symbol set from Kiel.[6] Esling, a prominent IPA council member and phonetics researcher, played a central role in refining the system for practical implementation, ensuring compatibility with emerging digital tools like X-SAMPA.[5] Peter Ladefoged further contributed to the system's evolution through his involvement in subsequent IPA revisions.[5] These early codifications covered most core symbols, with expansions in 1996 incorporating additional diacritics and suprasegmentals to reflect updated chart layouts. The system reached full codification in the 2005 IPA chart, which integrated comprehensive number assignments for all approved symbols, including recent additions like the labiodental flap (IPA 176), while maintaining the original structure from 1989.[7] Post-2005 revisions have been minor, primarily involving symbol substitutions or layout adjustments in charts from 2015 onward, with no major overhaul to the numbering scheme to preserve backward compatibility in digital applications.[8] The numbering system was further updated in the 2020 IPA chart, extending assignments up to 533 to include all current symbols while preserving backward compatibility.[2] This stability underscores the system's enduring role as a foundational reference, refined through collaborative efforts by IPA council members like Esling and Ladefoged.[5]Purpose and Applications
Original Purpose
The International Phonetic Alphabet (IPA) numbers were introduced at the 1989 Kiel Convention by the International Phonetic Association to provide a standardized numerical system for uniquely identifying IPA symbols and diacritics, addressing the challenges of representing phonetic notation in early digital environments.[1] This system emerged from Workgroup 9's recommendations on computer coding, aiming to create a communication-interchange standard independent of specific encoding conventions, such as those in competing font systems or early ASCII implementations.[1] A primary motivation was the need for unambiguous referencing of IPA symbols in pre-Unicode computing, where similar glyphs—such as the voiceless dental fricative /θ/ and the mid central vowel /ɵ/—could be confused or misrepresented due to limited font support and inconsistent character encodings.[9] This ambiguity hindered the storage of phonetic data in databases, the implementation of IPA in linguistic software, and the transcription of languages across platforms, particularly for research and documentation of endangered languages.[9] By assigning unique numerical equivalents, the system facilitated reliable data interchange and translation tables for phonetic applications, ensuring consistency without reliance on visual symbols alone.[1] The numbering also advanced the IPA's overarching goal of universal phonetic notation, especially for symbols derived from non-Latin scripts or involving complex diacritics that were difficult to standardize in print or digital media.[9] It served as a typesetter's guide for publishers and a foundation for cross-organizational collaboration among linguists, programmers, and typographers.[1] Design principles emphasized hierarchical organization to enhance memorability and utility: numbers were structured as three-digit codes, with the first digit denoting broad phonetic categories (e.g., 100–199 for pulmonic consonants, 300–399 for vowels, 400–499 for diacritics), followed by subcategories based on articulatory features.[9] This approach retained identifiers for retired symbols to maintain historical continuity, promoting systematic access and organization in phonetic resources.[1]Modern Usage and Alternatives
The IPA numbering system has attained legacy status since the post-2000 adoption of Unicode, which superseded it by assigning dedicated code points to phonetic symbols for digital representation and processing, exemplified by U+03B8 for the voiceless dental fricative /θ/. This shift rendered the numbers, originally devised post-1989 Kiel Convention for pre-Unicode computer encoding like X-SAMPA, largely obsolete for mainstream applications. Remaining applications persist in niche contexts, including specialized linguistic databases from SIL International that map IPA numbers to Unicode for phonetic keyboard and transcription tools.[10] The system also features in legacy software for phonological analysis and in scholarly citations referencing the 2005 IPA chart, with the International Phonetic Association maintaining an updated number chart through the 2020 revision for reference purposes. IPA numbers continue to be used in official IPA charts and software such as LaTeX.[2] Unicode serves as the dominant modern alternative, enabling seamless integration of IPA symbols across platforms without reliance on numeric codes, as detailed in resources like the Unicode Standard's phonetic extensions blocks.[11] Coverage limitations are evident in the numbering's failure to incorporate post-2005 extensions; for instance, the 2015 extIPA additions for disordered speech, including novel friction and articulation symbols like the linguolabial fricative, receive no assigned numbers.[12]Core IPA Symbols
Consonants (101–199)
The core IPA consonants, numbered 101–199 per the 2020 revision, encompass pulmonic consonants produced using lung-powered airflow and non-pulmonic consonants generated by alternative airstream mechanisms such as velaric ingressive for clicks or glottalic egressive for ejectives. These symbols are organized primarily by manner of articulation (e.g., plosives, nasals, fricatives) and secondarily by place of articulation (from bilabial to glottal or epiglottal), reflecting their phonetic properties in human languages. Descriptions include the standard name, articulatory features (place, manner, voicing), and representative usage examples from documented languages. All symbol assignments and descriptions are standardized by the International Phonetic Association.[9][13]Pulmonic Consonants
Pulmonic consonants form the majority of the IPA's core inventory and are subdivided into plosives, nasals, trills and flaps, fricatives, lateral fricatives, and approximants. They occur in nearly all languages, with voicing contrasts common in many (e.g., voiceless vs. voiced pairs).Plosives (101–113)
Plosives involve complete closure of the vocal tract, building pressure before release, and are found across languages for stops at various places.| Number | Symbol | Name | Description | Example Usage |
|---|---|---|---|---|
| 101 | p | Voiceless bilabial plosive | Voiceless stop at the lips | English "pie" [paɪ] (fruit pie)[9] |
| 102 | b | Voiced bilabial plosive | Voiced stop at the lips | English "buy" [baɪ] (purchase)[9] |
| 103 | t | Voiceless alveolar plosive | Voiceless stop at the alveolar ridge | English "tie" [taɪ] (neckwear)[9] |
| 104 | d | Voiced alveolar plosive | Voiced stop at the alveolar ridge | English "die" [daɪ] (perish)[9] |
| 105 | ʈ | Voiceless retroflex plosive | Voiceless stop with tongue curled back | Toda (Dravidian) "top" [ʈop] (sugar cane)[9] |
| 106 | ɖ | Voiced retroflex plosive | Voiced stop with tongue curled back | Hindi "ḍāl" [ɖaːl] (lentils)[9] |
| 107 | c | Voiceless palatal plosive | Voiceless stop at the hard palate | Hungarian "tyúk" [cuːk] (hen) |
| 108 | ɟ | Voiced palatal plosive | Voiced stop at the hard palate | Norwegian "kjuke" [ˈçʏːkə] approximating [ɟ] in some dialects[9] |
| 109 | k | Voiceless velar plosive | Voiceless stop at the soft palate | English "kite" [kaɪt] (bird)[9] |
| 110 | g | Voiced velar plosive | Voiced stop at the soft palate | English "guy" [gaɪ] (person)[9] |
| 111 | q | Voiceless uvular plosive | Voiceless stop at the uvula | Arabic "qalb" [qalb] (heart)[9] |
| 112 | ɢ | Voiced uvular plosive | Voiced stop at the uvula | Tlingit "gwáak" [ɢwaːk] (duck) |
| 113 | ʔ | Glottal stop | Voiceless stop at the glottis | Hawaiian "uh-oh" [ʔʌhoʊ] (exclamation)[9] |
Nasals (114–120)
Nasals feature airflow through the nose due to velum lowering, often sonorant and voiced, serving as resonances in many languages.| Number | Symbol | Name | Description | Example Usage |
|---|---|---|---|---|
| 114 | m | Bilabial nasal | Voiced nasal at the lips | English "me" [mi] (pronoun)[9] |
| 115 | ɱ | Labiodental nasal | Voiced nasal between lip and teeth | English "emphasis" [ˈɛmfəsɪs] (stress)[9] |
| 116 | n | Alveolar nasal | Voiced nasal at the alveolar ridge | English "knee" [ni] (joint)[9] |
| 117 | ɳ | Retroflex nasal | Voiced nasal with tongue curled back | Hindi "paṇḍit" [pəɳɖɪt] (scholar)[9] |
| 118 | ɲ | Palatal nasal | Voiced nasal at the hard palate | Spanish "niño" [ˈniɲo] (child)[9] |
| 119 | ŋ | Velar nasal | Voiced nasal at the soft palate | English "sing" [sɪŋ] (vocalize)[9] |
| 120 | ɴ | Uvular nasal | Voiced nasal at the uvula | Japanese "ŋ" in some loanwords like "Washington" [waʃiŋton] approximating uvular in dialects |
Trills and Taps/Flaps (121–126)
Trills involve vibration of an articulator against another, while taps/flaps are brief single contacts; these are common in Romance and African languages for r-sounds.| Number | Symbol | Name | Description | Example Usage |
|---|---|---|---|---|
| 121 | ʙ | Bilabial trill | Voiced trill at the lips | Kele (Papuan) "bʙa" (eat) |
| 122 | r | Alveolar trill | Voiced trill at the alveolar ridge | Spanish "perro" [ˈpero] (dog)[9] |
| 123 | ʀ | Uvular trill | Voiced trill at the uvula | French "rue" [ʀy] (street, traditional)[9] |
| 124 | ⱱ | Labiodental flap | Voiced flap between lip and teeth | Mono (Ubangi) "aⱱa" (arrive) |
| 125 | ɾ | Alveolar flap | Voiced tap at the alveolar ridge | Spanish "pero" [ˈpeɾo] (but)[9] |
| 126 | ɽ | Retroflex flap | Voiced flap with tongue curled back | Norwegian "rød" [ɽøːɖ] (red)[9] |
Fricatives (127–148)
Fricatives produce turbulent airflow through a narrow constriction, with voiceless/voiced pairs in many languages for sibilants and continuants.| Number | Symbol | Name | Description | Example Usage |
|---|---|---|---|---|
| 127 | ɸ | Voiceless bilabial fricative | Voiceless friction at the lips | Japanese "fū" [ɸɯ] (wind)[9] |
| 128 | β | Voiced bilabial fricative | Voiced friction at the lips | Spanish "abuela" [aˈβwela] (grandmother)[9] |
| 129 | f | Voiceless labiodental fricative | Voiceless friction lip-to-teeth | English "fat" [fæt] (plump)[9] |
| 130 | v | Voiced labiodental fricative | Voiced friction lip-to-teeth | English "vat" [væt] (container)[9] |
| 131 | θ | Voiceless dental fricative | Voiceless friction at the teeth | English "think" [θɪŋk] (consider)[9] |
| 132 | ð | Voiced dental fricative | Voiced friction at the teeth | English "this" [ðɪs] (demonstrative)[9] |
| 133 | s | Voiceless alveolar fricative | Voiceless friction at alveolar ridge | English "see" [si] (perceive)[9] |
| 134 | z | Voiced alveolar fricative | Voiced friction at alveolar ridge | English "zoo" [zu] (animal park)[9] |
| 135 | ʃ | Voiceless postalveolar fricative | Voiceless friction behind alveolus | English "ship" [ʃɪp] (vessel)[9] |
| 136 | ʒ | Voiced postalveolar fricative | Voiced friction behind alveolus | English "vision" [ˈvɪʒən] (sight)[9] |
| 137 | ʂ | Voiceless retroflex fricative | Voiceless friction with tongue curl | Mandarin "shī" [ʂʰux] (lion)[9] |
| 138 | ʐ | Voiced retroflex fricative | Voiced friction with tongue curl | Polish "żółć" [ʐulʦ] (bile) |
| 139 | ç | Voiceless palatal fricative | Voiceless friction at hard palate | German "ich" [ɪç] (I)[9] |
| 140 | ʝ | Voiced palatal fricative | Voiced friction at hard palate | Greek "ναι" [ʝe] (yes)[9] |
| 141 | x | Voiceless velar fricative | Voiceless friction at soft palate | Scottish Gaelic "loch" [lɔx] (lake)[9] |
| 142 | ɣ | Voiced velar fricative | Voiced friction at soft palate | Dutch "goed" [ɣut] (good)[9] |
| 143 | χ | Voiceless uvular fricative | Voiceless friction at uvula | German "Bach" [bax] (stream)[9] |
| 144 | ʁ | Voiced uvular fricative | Voiced friction at uvula | French "rue" [ʁy] (street)[9] |
| 145 | ħ | Voiceless pharyngeal fricative | Voiceless friction in pharynx | Arabic "ḥaqq" [ħaqː] (truth)[9] |
| 146 | ʕ | Voiced pharyngeal fricative | Voiced friction in pharynx | Arabic "ʿayn" [ʕajn] (eye)[9] |
| 147 | h | Voiceless glottal fricative | Voiceless friction at glottis | English "hot" [hɑt] (warm)[9] |
| 148 | ɦ | Voiced glottal fricative | Voiced friction at glottis | Czech "vrah" [vrɑɦ] (murderer)[9] |
Lateral Fricatives (149–150)
These are fricatives with lateral airflow, rare but attested in Celtic and Salishan languages.| Number | Symbol | Name | Description | Example Usage |
|---|---|---|---|---|
| 149 | ɬ | Voiceless alveolar lateral fricative | Voiceless lateral friction at alveolus | Welsh "Llan" [ɬan] (church land)[9] |
| 150 | ɮ | Voiced alveolar lateral fricative | Voiced lateral friction at alveolus | Zulu "dlala" [ɮaːɮa] (play) |
Approximants (151–161)
Approximants have close but non-turbulent articulations, often functioning as glides or rhotics in Indo-European languages.| Number | Symbol | Name | Description | Example Usage |
|---|---|---|---|---|
| 151 | ɹ | Alveolar approximant | Voiced approximation at alveolar ridge | English "red" [ɹɛd] (color)[9] |
| 152 | ɻ | Retroflex approximant | Voiced approximation with tongue curl | Mandarin "rén" [ʐən] approximating [ɻ] (person)[9] |
| 153 | j | Palatal approximant | Voiced approximation at hard palate | English "yes" [jɛs] (affirmative)[9] |
| 154 | ɰ | Velar approximant | Voiced approximation at soft palate | Spanish "agua" [ˈaɰwa] (water)[9] |
| 155 | l | Alveolar lateral approximant | Voiced lateral approximation at alveolus | English "lee" [li] (shelter)[9] |
| 156 | ɭ | Retroflex lateral approximant | Voiced lateral with tongue curl | Tamil "ḷa" [ɭa] (play) |
| 157 | ʎ | Palatal lateral approximant | Voiced lateral at hard palate | Italian "fiore" [fjoˈʎe] (flower)[9] |
| 158 | ʟ | Velar lateral approximant | Voiced lateral at soft palate | Mid-Wahgi "ʟa" (name) |
| 159 | ʍ | Voiceless labial-velar approximant | Voiceless lip-to-velum approximation | Scottish English "which" [ʍɪtʃ] (interrogative)[9] |
| 160 | w | Voiced labial-velar approximant | Voiced lip-to-velum approximation | English "we" [wi] (pronoun)[9] |
| 161 | ɥ | Labial-palatal approximant | Voiced lip-to-palatal approximation | French "lui" [lɥi] (him)[9] |
Non-Pulmonic Consonants (169–178)
Non-pulmonic consonants include implosives (glottalic ingressive), clicks (velaric ingressive), and ejectives (glottalic egressive), occurring in about 13–20% of languages, primarily African and Caucasian. The ejective marker is a diacritic applied to pulmonic symbols (e.g., pʼ).Implosives (169–173)
Implosives involve inward airflow via glottal closure, voiced and common in South Asian and African languages.| Number | Symbol | Name | Description | Example Usage |
|---|---|---|---|---|
| 169 | ɓ | Voiced bilabial implosive | Inward stop at lips | Sindhi "ɓara" [ɓəɾə] (big)[9] |
| 170 | ɗ | Voiced alveolar implosive | Inward stop at alveolar ridge | Fula "ɗumma" [ɗʊm.mə] (hide) |
| 171 | ʄ | Voiced palatal implosive | Inward stop at hard palate | Sereer "ʄam" [ʄam] (know) |
| 172 | ɠ | Voiced velar implosive | Inward stop at soft palate | Pnar "kaɠaw" [kəɠaw] (crow)[9] |
| 173 | ʛ | Voiced uvular implosive | Inward stop at uvula | Yapese "ʛab" [ʛap] (moon) |
Clicks (174–178)
Clicks use velum lowering for ingressive airflow, basic in Khoisan languages but also ingressive in Bantu for emphasis.| Number | Symbol | Name | Description | Example Usage |
|---|---|---|---|---|
| 174 | ʘ | Bilabial click | Inward click at lips | Central !Kung "ʘa" [ʘa] (person)[9] |
| 175 | ǀ | Dental click | Inward click at teeth | Zulu "caca" [ǀaǀa] (search) |
| 176 | ǃ | Alveolar click | Inward click at alveolar ridge | Xhosa "gq" [ǃʰʷ] in compounds (deep)[9] |
| 177 | ǂ | Palatoalveolar click | Inward click behind alveolus | Nama "ǂkhā" [ǂ͡χâː] (to laugh) |
| 178 | ǁ | Lateral click | Inward lateral click | !Kung "ǁâ" [ǁɑ̀] (all)[9] |
Ejectives (401)
Ejectives are pulmonic plosives modified by glottalic pressure, unmarked in base but indicated by diacritic; prevalent in Quechuan and Northeast Caucasian languages.- 401 ʼ Modifier letter apostrophe: Ejective marker, applied post-consonant (e.g., 101p + 401 = pʼ voiceless bilabial ejective, as in Quechua "pʼuru" [pʼuɾu] 'straw').[9]
Vowels (300–399)
The core IPA vowel symbols, numbered 300–399 per the 2020 revision, encompass monophthongs that form the basis of vowel transcription in the International Phonetic Alphabet, positioned on a trapezoidal vowel chart according to tongue height (from close to open), frontness or backness, and lip rounding (rounded or unrounded). These symbols capture the primary vowel qualities used across languages, with acoustic properties primarily defined by the first two formant frequencies (F1 and F2), where F1 inversely correlates with vowel height and F2 with frontness.[3] Representative examples from various languages illustrate their usage, while typical formant values (approximate for adult male speakers) provide context for their perceptual distinctiveness.[3][14] Monophthongs are organized below in a table by approximate chart position, including central vowels for completeness. The table lists 28 core symbols, focusing on their articulatory and acoustic characteristics without exhaustive variants.| Number | Symbol | Description (Height, Frontness/Backness, Rounding) | Example Language/Word | Typical Formants (F1/F2 in Hz, approx.) |
|---|---|---|---|---|
| 301 | i | Close, front, unrounded | English "see" | 280 / 2230 |
| 302 | y | Close, front, rounded | German "über" | 300 / 2100 |
| 303 | e | Close-mid, front, unrounded | French "été" | 405 / 2080 |
| 304 | ø | Close-mid, front, rounded | Danish "ø" | 450 / 1850 |
| 305 | ɛ | Open-mid, front, unrounded | English "bet" | 600 / 1930 |
| 306 | œ | Open-mid, front, rounded | French "œuvre" | 650 / 1750 |
| 307 | a | Open, front, unrounded | Spanish "casa" | 850 / 1610 |
| 308 | ɶ | Open, front, rounded | Norwegian "mann" | 900 / 1500 |
| 309 | ɑ | Open, back, unrounded | English "father" | 750 / 700 |
| 310 | ɒ | Open, back, rounded | English "lot" (some dialects) | 700 / 800 |
| 314 | ʌ | Open-mid, back, unrounded | English "cut" | 680 / 1310 |
| 314 | ɔ | Open-mid, back, rounded | English "thought" | 560 / 820 |
| 315 | ɤ | Close-mid, back, unrounded | Chinese "ge" | 450 / 1200 |
| 316 | ɯ | Close, back, unrounded | Japanese "ku" | 300 / 1400 |
| 317 | o | Close-mid, back, rounded | Spanish "no" | 430 / 980 |
| 320 | u | Close, back, rounded | English "food" | 330 / 1260 |
| 319 | ɪ | Near-close, near-front, unrounded | English "bit" | 370 / 2090 |
| 320 | ʏ | Near-close, near-front, rounded | German "fünf" | 380 / 1900 |
| 321 | ʊ | Near-close, near-back, rounded | English "put" | 400 / 1100 |
| 310 | ɨ | Close, central, unrounded | Romanian "bucurie" | 300 / 1600 |
| 311 | ʉ | Close, central, rounded | Swedish "du" | 320 / 1350 |
| 322 | ə | Mid, central, unrounded | English "about" | 500 / 1500 |
| 323 | ɵ | Close-mid, central, rounded | Norwegian "sol" | 450 / 1250 |
| 324 | ɐ | Near-open, central, unrounded | German "besser" | 700 / 1400 |
| 325 | æ | Near-open, front, unrounded | English "cat" | 860 / 1550 |
| 326 | ɜ | Open-mid, central, unrounded | English "nurse" (non-rhotic) | 550 / 1650 |
| 395 | ɞ | Open-mid, central, rounded | Dutch "uur" | 600 / 1000 |
| 397 | ɘ | Close-mid, central, unrounded | Samoan "e" | 450 / 1600 |
Modifiers
Non-Tone Diacritics (400–499)
Non-tone diacritics in the International Phonetic Alphabet (IPA), assigned numbers 400–499, are segmental modifiers that alter the articulatory or phonatory features of individual consonants or vowels without indicating tone or suprasegmental properties. These diacritics allow for precise transcription of phonetic variations, such as changes in voicing, place of articulation, or secondary articulations, and are essential for describing sounds in languages with fine-grained distinctions. The International Phonetic Association (IPA) officially recognizes over 50 such diacritics, including subtypes (e.g., A/B variants for positioning above or below the base symbol to accommodate descenders like ŋ or ɳ), as detailed in their revised chart (2020). Stacking rules permit multiple diacritics on a single symbol, typically with below-base marks placed first, followed by above-base ones, though legibility may require adjustments for complex combinations; for instance, voicelessness can stack with dental articulation as /t̪̥/.[15]Voicing and Breathiness (400–409)
These diacritics modify the phonation type, indicating whether sounds are produced with vocal cord vibration, aspiration, or breathy/creaky qualities, applicable to both consonants and vowels. They are among the most commonly used for transcribing contrasts in languages like Hindi (aspirated stops) or Zulu (breathy vowels). Subtypes ensure compatibility with symbols having descenders.| IPA Number | Symbol | Effect | Example |
|---|---|---|---|
| 402A | ◌̥ | Voiceless (no vocal cord vibration) | /n̥/ (voiceless nasal, as in whispered )[3] |
| 402B | ◌̊ | Voiceless (variant for descenders) | /ŋ̊/ (voiceless velar nasal)[3] |
| 403 | ◌̬ | Voiced (added vibration) | /s̬/ (voiced fricative )[3] |
| 404 | ◌ʰ | Aspirated (post-release breath) | /tʰ/ (aspirated [tʰ], as in English "top")[3] |
| 405 | ◌̤ | Breathy voiced (murmur, breathy phonation) | /b̤/ (breathy [b̤], as in Hindi /bʱ/)[15] |
| 406 | ◌̰ | Creaky voiced (laryngealized, glottal creak) | /a̰/ (creaky vowel, as in some Danish realizations)[15] |
Articulation Adjustments (410–449)
This group includes diacritics for adjusting place or manner of articulation, such as tongue position, rounding, or root advancement, crucial for vowels (e.g., tongue root harmony in African languages) and consonants (e.g., dental vs. alveolar in Romance languages). They often stack with voicing marks, with below-base diacritics like retraction applied innermost.| IPA Number | Symbol | Effect | Example |
|---|---|---|---|
| 408 | ◌̪ | Dental (tip/blade at teeth) | /t̪/ (dental [t̪], as in Spanish /t/)[3] |
| 409 | ◌̺ | Apical (tip-up articulation) | /t̺/ (apical [t̺], contrasting alveolar)[3] |
| 410 | ◌̻ | Laminal (blade-flat articulation) | /t̻/ (laminal [t̻], as in some Australian languages)[3] |
| 413 | ◌̟ | Advanced (tongue forward) | /u̟/ (advanced [u̟], closer to )[3] |
| 414 | ◌̠ | Retracted (tongue back) | /e̠/ (retracted [e̠], closer to [ɛ])[3] |
| 417 | ◌̘ | Advanced tongue root (+ATR) | /e̘/ ([e̘], open pharynx, as in Igbo vowels)[3] |
| 418 | ◌̙ | Retracted tongue root (-ATR) | /e̙/ ([e̙], constricted pharynx)[3] |
| 411 | ◌̹ | More rounded (increased lip protrusion) | /ɔ̹/ (labialized [ɔ̹])[3] |
| 412 | ◌˞ | Rhoticity (r-colored, bunched/retroflexed) | /ɚ/ or /a˞/ (rhotacized vowel, as in American English "bird")[3] |
| 415 | ◌̈ | Centralized (tongue towards center) | /ë/ (centralized [ë], as in some Turkish vowels)[3] |
Other Modifiers (450–499)
This category encompasses additional segmental modifications like nasalization, releases, and co-articulations, used for features such as syllable structure or secondary articulations in languages like Arabic (pharyngealization) or Polish (palatalization). Many involve superscript letters or tilde-like marks, with stacking allowing combinations like nasal unreleased stops.| IPA Number | Symbol | Effect | Example |
|---|---|---|---|
| 424 | ◌̃ | Nasalized (air through nose) | /ã/ (nasal vowel, as in French "vin")[16] |
| 420 | ◌ʷ | Labialized (lip rounding addition) | /tʷ/ (labialized [tʷ], as in some Salish languages)[15] |
| 421 | ◌ʲ | Palatalized (palate contact addition) | /tʲ/ (palatalized [tʲ], as in Russian soft consonants)[15] |
| 422 | ◌ˠ | Velarized (velum contact addition) | /lˠ/ (velar [ɫ], as in English "feel")[15] |
| 427 | ◌ˤ | Pharyngealized (pharynx constriction) | /tˤ/ (emphatic [tˤ], as in Arabic /ṭ/)[15] |
| 490 | ◌̴ | Velarized or pharyngealized (simultaneous) | /t̴/ (co-velar-pharyngeal [t̴])[16] |
| 431 | ◌ⁿ | Nasal release (nasal after-release) | /dⁿ/ (nasal-released [dⁿ])[15] |
| 432 | ◌ˡ | Lateral release (lateral after-release) | /dˡ/ (lateral-released [dˡ], as in English "middle")[15] |
| 433 | ◌̚ | No audible release (unreleased) | /t̚/ (unreleased [t̚], as in English "cat" stop)[3] |
| 428 | ◌̩ | Syllabic (syllable nucleus) | /n̩/ (syllabic nasal, as in English "button")[15] |
| 429 | ◌̯ | Non-syllabic (glide-like) | /i̯/ (non-syllabic , as in diphthong onset)[15] |
Suprasegmentals (500–599)
Suprasegmentals in the International Phonetic Alphabet (IPA) encompass symbols that denote prosodic features extending over multiple segments, such as stress patterns, vowel or consonant lengthening, and tonal contours, which are crucial for representing rhythm, intonation, and pitch in languages.[17] These notations, assigned IPA numbers 500–599, bridge segmental phonemes with broader utterance-level phenomena, allowing precise transcription of supraphonemic elements like syllable boundaries and global pitch movements.[3] Unlike segmental diacritics, which modify individual sounds, suprasegmentals apply to syllables or groups, often placed before or after the affected unit to indicate scope.[17] Stress and accent markers, numbered 500–509, indicate prominence within words or phrases, with primary stress denoted by a high vertical stroke (501, ˈ) placed before the stressed syllable, as in English /ˈsɪləbl/ for "syllable," where the first syllable receives the strongest emphasis.[17] Secondary stress uses a low vertical stroke (502, ˌ), also preceding the syllable, exemplified in compounds like /ˌɪŋɡlɪʃ ˈtiːtʃər/ for "English teacher," highlighting subordinate prominence.[3] Additional markers include the thick vertical line (507, |) for minor phrase boundaries and double vertical line (508, ‖) for major intonation groups, as in /ðə ˈwɔːr | əv ðə ˈrəʊzəz ‖/ to separate feet in English poetry recitation.[17] A linking symbol (509, ͜) below segments indicates absence of juncture, such as in rapid speech where vowels blend without pause.[3] Length and duration symbols, 510–519, quantify the temporal extent of segments, primarily vowels but also consonants, with the triangular colon (503, ː) following the prolonged element to mark full length, as in Finnish /kɑːtʰu/ for "roof," where the vowel duration contrasts meaning.[17] Half-length is shown by a half-triangular colon (504, ˑ), and extra-shortness by a breve (505, ̆) above the segment, useful in languages like Japanese for mora-timed short vowels in /kʰăɕi/ "chopsticks."[3] The period (506, .) denotes syllable breaks within words, as in English /ˈkɒp.ə.leɪt/ "copulate," preventing misparsing of clusters.[17] Global intonation contours include an upward arrow (522, ↗) for overall rise and downward arrow (523, ↘) for fall, applied to phrases like a questioning /ðɪs ɪz ɑːt? ↗/ in English.[3] Tones and intonation symbols, 520–599, primarily describe pitch levels and contours using diacritics over vowels or Chao tone letters above syllables, essential for tonal languages.[17] Level tones employ accent marks: double acute (510, ̋) for extra high, acute (511, ´) for high, macron (512, ¯) for mid, grave (513, ̀) for low, and double grave (514, ̏) for extra low, placed above the vowel, as in African languages like Yoruba /àkʰá/ with low tone on the first syllable.[3] Contour tones use caron (517, ̌) for rising and circumflex (518, ̂) for falling, with more complex forms like macron-acute (519, ᷄) for high rising; a Mandarin example is /ma˧˥/ "scold" with rising contour on the vowel.[17] Chao tone letters provide a numerical pitch scale from 1 (low) to 5 (high), with extra-high (˥, IPA number 518 variant), high (˦, 517), mid (˧, 516), low (˨, 515), and extra-low (˩, 514) bars above the tone-bearing unit, as in Mandarin's four tones: high level /ma˥/ "mother," rising /ma˧˥/ "hemp," falling-rising /ma˨˩˦/ "horse," and falling /ma˥˩/ "scold."[3] Registers shifts include downstep (515, ꜜ) after a tone to lower the register and upstep (516, ꜛ) to raise it, common in tone languages like Igbo.[17] Contour combinations, such as rising-falling (521, ᷈), are composites but numbered for reference, facilitating detailed prosodic analysis.[3] As of the 2020 revision, no major changes have been made to these symbols.| Category | IPA Number | Symbol | Function | Placement | Example |
|---|---|---|---|---|---|
| Stress | 501 | ˈ | Primary stress | Before syllable | English /ˈfɒkəs/ "focus"[17] |
| Stress | 502 | ˌ | Secondary stress | Before syllable | English /ˌɪn.təˈvjuː/ "interview"[17] |
| Length | 503 | ː | Long | After segment | Finnish /tuːli/ "wind"[17] |
| Length | 505 | ̆ | Extra-short | Above segment | Japanese /sit̆i/ "city" (short vowel)[3] |
| Tone (Level) | 510 | ̋ | Extra-high | Above vowel | Yoruba high tone example[17] |
| Tone (Contour) | 517 | ̌ | Rising | Above vowel | Vietnamese /mǎ/ rising tone[3] |
| Intonation | 522 | ↗ | Global rise | After phrase | English question /wɒt ↗/ "what?"[17] |
Extensions and Variants
Retired and Non-IPA Consonants (200–299)
The range 200–299 in the IPA numbering system encompasses consonants that were either officially retired during the 1989 Kiel Convention or employed as auxiliary symbols outside the standard IPA repertoire. These symbols reflect the evolution of the alphabet, where redundancy, legibility issues, and the preference for composable diacritics led to their obsolescence. Retired symbols (primarily 200–249) were part of pre-1989 IPA charts and used for sounds like clicks and affricates in languages such as Khoekhoe and English, but were replaced to promote uniformity. Non-IPA or auxiliary consonants (250–299) include those from regional traditions, like Bantuist notation or Slavic transcriptions, which persist in specialized literature despite not being endorsed by the International Phonetic Association. Overall, this category highlights approximately 20 symbols that underscore the IPA's refinement, prioritizing core letters (from the 100–199 range) combined with modifiers over dedicated glyphs. Key examples of retired symbols include the following, drawn from historical IPA revisions:| IPA Number | Symbol | Description | Reason for Retirement | Historical Usage | Modern Equivalent |
|---|---|---|---|---|---|
| 201 | ʇ | Turned t (dental click) | Superseded for better legibility and standardization of click notation at Kiel 1989 | Transcriptions of Khoekhoe and other click languages in early 20th-century publications | ǀ (pipe, IPA 177) with optional retraction diacritic (IPA 430) [18] |
| 202 | ʗ | Retroflex click (alveolar click in some notations) | Replaced as part of click symbol overhaul at Kiel 1989 | Early phonetic studies of African languages, e.g., Khoekhoe (1921) | ǃ (IPA 178) [18] |
| 204 | ʖ | Lateral click | Retired for standardized click letters at Kiel 1989 | Khoekhoe and Nama transcriptions pre-1989 | ǁ (IPA 180) [18] |
| 211 | ʦ | Ts ligature (voiceless alveolar affricate) | Ligatures retired at Kiel 1989 in favor of tie-bar combinations to avoid ambiguity with sequences | Common in English "ts" sounds and Germanic languages pre-1989 | t͡s (IPA 103 + 132 with tie bar, IPA 410) [19] |
| 212 | ʣ | Dz ligature (voiced alveolar affricate) | Ligatures phased out at Kiel 1989 for composable notation | Slavic and English "dz" transcriptions, 1947–1989 IPA charts | d͡z (IPA 104 + 133 with tie bar) [19] |
| 215 | ʧ | Ch ligature (voiceless postalveolar affricate) | Retired ligatures at Kiel 1989 to standardize affricate representation | Widespread in English "ch" (e.g., "church") and Romance languages pre-1989 | t͡ʃ (IPA 103 + 135 with tie bar) [19] |
| 216 | ʤ | J ligature (voiced postalveolar affricate) | Eliminated as part of ligature retirement at Kiel 1989 | English "j" (e.g., "judge") and similar sounds in global languages, 1947–1989 | d͡ʒ (IPA 104 + 136 with tie bar) [19] |
| 220 | ʞ | Turned k (velar click) | Deemed impossible post-1949, fully retired at Kiel 1989 | Early African linguistics for rear-released velar clicks | k͡q or similar non-pulmonic notation; revived informally in West African studies [18] |
| 225 | ʆ | Palatalized s (voiceless alveolo-palatal fricative) | Redundancy with diacritics led to retirement at Kiel 1989 | Slavic language transcriptions, e.g., Russian soft sibilants (Laver 1994) | sʲ or ɕ (IPA 132 + 450 palatalization diacritic) [20] |
| 226 | ʓ | Palatalized z (voiced alveolo-palatal fricative) | Retired for diacritic-based alternatives at Kiel 1989 | Slavic phonetics, e.g., soft z in Polish or Russian | zʲ or ʑ (IPA 133 + 450) [20] |
| IPA Number | Symbol | Description | Status and Reason | Historical Usage | Modern Equivalent |
|---|---|---|---|---|---|
| 255 | ƛ | Voiceless lateral affricate | Non-IPA; from older Americanist systems, not adopted due to redundancy | North American indigenous languages, e.g., Athabaskan (early 20th century) | t͡ɬ (IPA 103 + 154 with tie bar) [21] |
| 260 | ȹ | Inverted f with stroke (voiceless labiodental plosive) | Para-IPA; unofficial for Bantu sounds, excluded from standard | Bantu languages like Zulu (Doke 1954; Ladefoged & Maddieson 1996) | p̪ (IPA 102 with dental diacritic, IPA 433) [20] |
| 261 | ȸ | B with stroke and hook (voiced labiodental plosive) | Non-IPA auxiliary; not standardized | Bantu phonetics for labiodental stops | b̪ (IPA 101 with dental diacritic) [20] |
| 271 | ǂ | Variant palatoalveolar click | Non-IPA variant; overlaps with core but used in older systems | Khoisan languages, pre-Kiel extensions | ǂ (standard IPA 179) [18] |
| 276 | ɼ | R with long leg (fricative trill) | Retired at Kiel 1989 for diacritic preference | Czech and other European languages for trilled fricatives | r̝ (IPA 158 + fricative diacritic, IPA 427) [20] |
| 290 | ƞ | Turned y (Japanese nasal coda) | Retired at Kiel 1989; underspecified nasal | Japanese phonology for variable nasals (pre-1989) | ŋ̊ or context-dependent nasal [20] |
extIPA Symbols (600–699)
The Extensions to the International Phonetic Alphabet (extIPA) provide a specialized set of symbols and diacritics numbered 600–699 in the IPA system, designed primarily for transcribing disordered speech in clinical phonetics and linguistics. These extensions address atypical articulations, voice qualities, and paralinguistic features not adequately covered by the core IPA, such as those observed in speech pathology cases like lisps, stutters, or velopharyngeal dysfunctions. Developed by the International Clinical Phonetics and Linguistics Association (ICPLA), the extIPA symbols were first proposed in 1989 and officially adopted in 1991 at the ICPLA symposium, with revisions published in charts in 1997, 2005, 2015, and 2025 to incorporate feedback from clinical practice.[12][22] The consonantal extensions (600–649) introduce letters for non-standard places or manners of articulation encountered in pathological speech, such as percussives or fricatives with unusual airflow. For instance, IPA number 601 represents the bidental percussive ʭ, used to transcribe a clicking sound produced by the teeth striking each other, common in some substitution errors in children's speech disorders. Another example is 602, the velopharyngeal fricative t̪͡θ̪ (often notated with a ligature), which captures nasal airflow turbulence in cleft palate cases. These symbols facilitate precise documentation in clinical transcriptions, allowing speech therapists to analyze and target specific articulation deviations. Adoption of these consonantal symbols dates to the 1991 extIPA chart, with refinements in the 2005 version to better reflect observed disorders like apraxia of speech.[9][23]| IPA Number | Symbol | Description | Clinical Use Example |
|---|---|---|---|
| 601 | ʭ | Bidental percussive | Transcription of dental clicking in lisping substitutions, e.g., /ʭ/ for /s/ in "sun" as [ʭʌn].[9] |
| 602 | t̪͡θ̪ (ligature) | Velopharyngeal fricative | Noting turbulent nasal emission in post-velar fricatives for velopharyngeal insufficiency.[12] |
| 603 | s͡ɬ (ligature) | Lateralized | Describing lateral lisps where air escapes sideways, e.g., /s͡ɬ/ in "see" as [si͡ɬ].[9] |
| 604 | z͡ɮ (ligature) | Lateralized | Capturing voiced lateral airflow in stuttering or dysarthric speech.[23] |
| IPA Number | Symbol | Description | Clinical Use Example |
|---|---|---|---|
| 651 | ◌͆ | Dentolabial | Marking lip-dental contact in fricatives, e.g., /f͆/ for interdental lisp in "fish" as [f͆ɪʃ].[12] |
| 652 | ◌͇ | Labial spreading | Indicating spread lips in approximants for tense disorders.[9] |
| 656 | ◌͗ | Velopharyngeal friction | Noting friction at the velum in hypernasal speech, e.g., /s͗/ in cleft-related sibilants.[23] |
| 670 | ◌̼ | Laminal dental | Denoting blade-of-tongue dental contact in plosives for articulation therapy.[9] |
| 661 | ↓ | Ingressive airflow | Transcribing implosive-like ingressives in atypical pulmonic airflow disorders. |
Capital Letters (700–799)
The capital letters in the IPA numbering system, assigned to the range 700–799, consist of uppercase variants of core phonetic symbols derived from the Latin alphabet. These are primarily the uppercase forms A through Z, numbered 701 to 726, which mirror the lowercase symbols used in standard IPA transcriptions for consonants and vowels. For example, 701 designates A as the uppercase equivalent of the open front unrounded vowel a (IPA number 324), 711 designates P as the uppercase of the voiceless bilabial plosive p (IPA number 101), and 726 designates Z as the uppercase of the voiced alveolar fricative z (IPA number 139). This numbering facilitates computational encoding and legacy systems like X-SAMPA, though the system is largely superseded by Unicode.[9][25] In practice, these uppercase symbols are infrequently employed, as the IPA convention favors lowercase letters throughout transcriptions to maintain clarity and avoid conflation with orthographic conventions. They appear sporadically in phonemic representations to denote abstract units, such as archiphonemes (e.g., /P/ for an underspecified bilabial plosive) or cover symbols for sound classes in analyses of language acquisition. For instance, in transcribing proper names, a form like "London [ˈLʌndən]" might use capitals to align with written capitalization, though this is discouraged in formal IPA usage to prevent ambiguity with specialized notations like small capitals for approximants. Some uppercase letters, such as V (722) for voice and W (723) for whisper, overlap with voice quality symbols in extensions, but their primary role here is as capital variants.[26][27] The higher portion of the range, 760–799, covers limited uppercase forms of diacritics and non-Latin-based symbols, with incomplete coverage compared to lowercase counterparts. A notable example is 771 for Ə, the uppercase schwa corresponding to the mid central vowel ə (IPA number 322), occasionally used for emphasis in phonemic charts or titles involving unstressed syllables. These special uppercase forms are even rarer, appearing mainly in pedagogical materials or historical texts rather than standard transcriptions, due to the predominance of lowercase in IPA practice. Overall, the 700–799 range includes approximately 30 symbols, focusing on essential Latin-derived forms while omitting dedicated uppercase for most modifiers.[9]| IPA Number | Uppercase Symbol | Corresponding Lowercase IPA Symbol | Usage Notes |
|---|---|---|---|
| 701 | A | a (324, open front unrounded vowel) | Rare; for phonemic emphasis or proper name alignment, e.g., [ˈɑkt] as [ˈAkt]. [25] |
| 702 | B | b (102, voiced bilabial plosive) | Limited to archiphoneme notation; avoids confusion with orthography. |
| 703 | C | (no direct core; overlaps with creaky voice in extensions) | Redundant with extIPA 624; uncommon in core transcriptions. |
| 704 | D | d (104, voiced alveolar plosive) | Used in cover symbols for alveolar obstruents in acquisition studies. |
| 705 | E | e (302, close-mid front unrounded vowel) | Occasional in vowel phoneme sets; rarity emphasizes lowercase standard. |
| 706 | F | f (127, voiceless labiodental fricative) | Overlaps with falsetto in voice quality; not standard for core f. |
| 707 | G | ɡ (110, voiced velar plosive) | Phonemic use only; lowercase preferred to distinguish from orthographic G. |
| 708 | H | h (173, voiceless glottal fricative) | Rare; sometimes for breathy voice extensions. |
| 709 | I | i (300, close front unrounded vowel) | For high vowel phonemes in abstract notation. |
| 710 | J | (no direct core; overlaps with jaw protrusion) | Limited; extIPA overlap (626). |
| 711 | P | p (101, voiceless bilabial plosive) | Common in phonology for /p/-like sounds; e.g., /Pin/ for pin phoneme. |
| 712 | L | l (148, alveolar lateral approximant) | Overlaps with extIPA 625; rare for core lateral. |
| 713 | M | m (106, bilabial nasal) | Nasal class cover symbol. |
| 714 | N | n (114, alveolar nasal) | Used in underspecified nasal representations. |
| 715 | O | o (304, close-mid back rounded vowel) | Vowel emphasis in phonemic tiers. |
| 716 | (no direct; vertical stroke in suprasegmentals) | Prosodic overlap; not core uppercase. | |
| 717 | Q | (no core equivalent) | Rarely used; reserved for extensions. |
| 718 | R | ɹ (122, alveolar approximant) | For rhotic phonemes in broad transcription. |
| 719 | S | s (135, voiceless alveolar fricative) | Sibilant class notation. |
| 720 | T | t (105, voiceless alveolar plosive) | Common archiphoneme for /t/- /θ/ merger. |
| 721 | U | u (306, close back rounded vowel) | High back vowel in phoneme lists. |
| 722 | V | v (128, voiced labiodental fricative) | Overlaps with modal voice (extIPA 621); dual use. |
| 723 | W | w (136, labio-velar approximant) | Overlaps with whisper (extIPA 623); rare for core. |
| 724 | X | x (177, voiceless velar fricative) | For velar fricatives in phonology. |
| 725 | Y | y (301, close front rounded vowel) | Front rounded vowel emphasis. |
| 726 | Z | z (139, voiced alveolar fricative) | For voiced sibilants; e.g., /Z/ in buzz phoneme. |
| 771 | Ə | ə (322, mid central vowel) | Uppercase schwa for titles or emphasis; highly limited. |
Transcription Delimiters (900–999)
Transcription delimiters in the International Phonetic Alphabet (IPA) serve to enclose entire transcriptions and mark internal boundaries, distinguishing between phonetic representations of actual speech sounds and phonemic representations of abstract sound units. These symbols facilitate clear structure in linguistic analysis, adhering to conventions outlined in the IPA Handbook, which emphasizes their role in indicating transcription levels and prosodic divisions.[9] The delimiters in the 900–999 range primarily include brackets for enclosing transcriptions and bars or hashes for boundaries, with fewer than 20 symbols total, prioritizing distinctions between narrow (detailed phonetic) and broad (approximate or phonemic) transcriptions.[25] Square brackets [ ] (IPA numbers 901 for the opening bracket and 902 for the closing bracket) enclose phonetic transcriptions, capturing realized speech sounds including allophonic variations and articulatory details.[25] For instance, a narrow phonetic transcription might use [kʰæt] for the English word "cat" to denote aspiration on the initial consonant, while a broader phonetic version could omit diacritics as [kat]. The IPA Handbook specifies that square brackets denote actual pronunciations, suitable for both narrow and broad phonetic levels, with narrow transcriptions incorporating suprasegmental diacritics for precision.[9] Slashes / / (IPA number 903 for the slash, used in pairs) delimit phonemic transcriptions, representing underlying phonemes without allophonic details.[25] An example is /kæt/ for "cat," focusing on contrastive units rather than surface forms. This convention aligns with broad transcription practices, as per IPA guidelines, where slashes indicate abstract phonological structures, often used in rule-based analyses or language descriptions.[9] For internal structuring, the single vertical bar | (IPA number 507, categorized under transcription delimiters in extended conventions) marks minor boundaries such as word or syllable divisions within a phonetic transcription.[9] In [naɪ.treɪt], it separates syllables in "nitrate," aiding segmentation analysis. The double vertical bar || (IPA number 508) denotes major phrase or intonation group boundaries, signaling stronger pauses, as in transcriptions of connected speech where prosodic units are divided.[9] The hash mark # (conventionally numbered around 921 in legacy IPA extensions) indicates utterance or morpheme boundaries, separating complete linguistic units.[9] For example, [kæt # dɒg] distinguishes two utterances, useful in phonological studies of juncture effects. Parentheses ( ) (IPA numbers 906 for opening and 907 for closing) enclose silent or inaudible articulations, such as mouthed but unvoiced segments in disordered speech.[9] Double parentheses (( )) extend this for obscured speech or noise annotations, like ((cough)) interrupting a transcription.[9] Curly braces { } delimit prosodic notations, briefly referencing suprasegmental features within bounded units, such as {stress} markings inside [ ].[9] Pauses are indicated by bracketed periods [. . .] (IPA numbers 631–633 for short, medium, long durations), placed within phonetic transcriptions to show hesitations without altering intonation registers.[9] These delimiters collectively ensure transcriptions reflect both segmental accuracy and suprasegmental context, with the IPA Handbook advising consistent use to avoid ambiguity in phonetic versus phonemic interpretations.[9]| IPA Number | Symbol | Usage Level | Convention and Example |
|---|---|---|---|
| 901 | [ | Phonetic (open) | Begins narrow or broad phonetic transcription; e.g., [kæt].[25] |
| 902 | ] | Phonetic (close) | Ends phonetic transcription; per IPA guidelines for realized sounds.[25][9] |
| 903 | / | Phonemic | Pairs for underlying forms; e.g., /kæt/, broad level without details.[25][9] |
| 507 | | | Boundary (minor) | Word/syllable division; e.g., [hɑʊ.naʊ], for internal structure.[9] |
| 508 | || | Boundary (major) | Intonation group pause; e.g., [go: | ho:m], phrase separation.[9] |
| ~921 | # | Utterance | Separates utterances; e.g., [kæt # dɒg], morpheme or full boundary.[9] |
| 906 | ( | Uncertain/silent (open) | Inaudible parts; e.g., [kæ(t)], mouthed but silent.[9] |
| 907 | ) | Uncertain/silent (close) | Ends silent notation; for phonetic detail in analysis.[9] |
| 631–633 | [. . .] | Pause | Duration-marked hesitations; short [.], medium [..], long [...].[9] |