Non-Pulmonic and Secondary Articulations
Chapter Objectives
By the end of this chapter, you will be able to:
- Distinguish pulmonic, glottalic, and lingual airstream mechanisms by the location of the air-pressure source and the direction of airflow.
- Describe clicks, ejectives, and implosives in articulatory terms and locate their symbols on the official IPA chart.
- Explain why these consonant types are not exotic additions to the IPA but systematic extensions of the same place-manner-voicing framework already used for pulmonic consonants.
- Define secondary articulation and distinguish it from a primary articulation and from a segment sequence.
- Interpret labialization, palatalization, velarization, and pharyngealization notation using the IPA diacritic and superscript system.
- Analyze complex consonant transcriptions, identifying primary and secondary articulatory information separately.
- Complete the Consonant Analysis Portfolio that integrates the systems introduced in Chapters 6–13.
An Observable Problem
In the 1930s, linguists documenting languages of southern Africa encountered a challenge that existing phonetic notation could not easily handle. The languages they were transcribing — among them Zulu, Xhosa, and Khoekhoe — used a set of consonants produced by drawing air inward through a tongue-based mechanism rather than by pushing air outward from the lungs. These sounds had no equivalents in any European language the linguists knew. They were not merely unusual; they were produced by a fundamentally different mechanism.
The same linguists also encountered languages in which a single consonant appeared to carry two simultaneous constrictions — something English does not permit in a single segment. Describing these sounds required extending the existing phonetic framework, not abandoning it.
Non-pulmonic consonants and complex articulations are distributed across the world's languages in patterns that reflect their phonological functions, not their difficulty for European ears. A click used as a dental consonant in a Nguni language is no more exotic than a uvular fricative used as the "r" in French. Both require articulatory description rather than a language-based label.
- You have already learned that the IPA framework for pulmonic consonants uses three dimensions: airstream and phonation, place of articulation, and manner of articulation. Before reading on, predict how you might extend this same three-dimensional framework to describe a consonant produced with an inward-moving air stream.
Airstream Mechanisms Revisited
Purpose: consolidate the three airstream mechanisms from Chapter 6 and establish the conceptual framework for click, ejective, and implosive description.
Chapter 6 introduced the concept that consonants — and the speech stream generally — require an airstream: a movement of air through the vocal tract. The vast majority of the world's speech sounds, and all the consonants studied in Chapters 7 through 12, rely on the pulmonic egressive airstream: air pushed outward from the lungs. But the lungs are not the only possible source of air-pressure changes in speech. Two other mechanisms are attested in the world's languages as the basis for complete consonant systems.
| Mechanism | Air-pressure source | Direction | Resulting consonant types |
|---|---|---|---|
| Pulmonic | Lungs and respiratory muscles | Egressive (outward) | All consonants in Chapters 6–12; the vast majority of consonants in the world's languages |
| Glottalic | Larynx (raised or lowered glottis) | Egressive → ejectives; ingressive → implosives | Ejectives (glottalic egressive); implosives (glottalic ingressive, with modification) |
| Lingual (velaric) | Back-of-tongue closure against velum or uvula | Ingressive (inward) | Clicks |
Understanding these mechanisms requires separating two independently variable properties: the source of the pressure change and the direction of air movement. In the pulmonic mechanism, lungs provide the source and the normal direction is egressive. In the glottalic mechanism, a raised or lowered larynx modifies pressure above or below the glottis. In the lingual mechanism, the tongue body creates two separate closures, and the space between them is the source of the pressure change.
The terms egressive and ingressive describe the direction of airflow relative to the speaker's body. Egressive sounds are produced with air flowing outward: away from the lungs, through the vocal tract, and out of the mouth or nose. Ingressive sounds are produced with air flowing inward. In English, you may use ingressive sounds paralinguistically — a gasp of surprise, or the sound sometimes written "tsk" in American English to express disapproval — but English has no phonemic ingressive consonants. Do not confuse ingressive airstream with inhalation: during an ingressive consonant, the speaker is not necessarily breathing in through the lungs. The inward air movement is local, produced by the articulatory mechanism described for each sound type below.
Clicks
Purpose: explain the lingual airstream mechanism, introduce click notation and place categories, and establish that clicks function as regular phonemic consonants in languages that have them.
A click is a consonant produced by the lingual (velaric ingressive) airstream mechanism. Two closures are formed simultaneously: a back closure at the velum or uvula, and a front closure at some point forward in the mouth. The tongue body lowers between these two closures, creating a pocket of rarefied air. When the front closure is released, the inward rush of air to equalize the pressure produces the sharp click sound. The back closure is released afterward and may be accompanied by pulmonic airflow.
The result is a consonant defined entirely by what happens at the front closure: where the tongue tip or blade makes contact determines the click's place of articulation. The IPA identifies five major click types based on this front place of articulation:
| Place of front closure | IPA symbol | Description | Note |
|---|---|---|---|
| Bilabial | ʘ | Both lips form the front closure; release produces a bilabial click | Occurs in some Khoisan languages; rare cross-linguistically |
| Dental | ǀ | Tongue tip behind upper teeth forms the front closure | Found in Zulu, Xhosa, Khoekhoe, and other Khoisan and Nguni languages; corresponds to the informal English "tsk" sound, though in English this has no phonemic status |
| Alveolar | ǃ | Tongue tip at the alveolar ridge forms the front closure | Found in Zulu, Xhosa, and many Khoisan languages; sharper and more prominent than the dental click in many descriptions |
| Palatal | ǂ | Tongue blade at the hard palate forms the front closure | Found in several Khoisan languages including ǂKhomani and Khoekhoe |
| Lateral | ǁ | Tongue tip contacts the alveolar ridge; release occurs laterally on one or both sides | Found in Zulu, Xhosa, and Sotho languages; sometimes described as the sound an English speaker makes to urge a horse along |
Note: All five click symbols are recognition-level for this chapter. Productive transcription work in this chapter focuses on reading and interpreting click notation rather than generating click transcriptions independently.
Clicks interact with the pulmonic airstream through their back closure and release. The pulmonic component allows clicks to be voiced or voiceless, nasalized, aspirated, or followed by other phonation types — producing a large number of phonetically distinct click consonants in languages like ǃXóõ (also written ǃKung), which has been described as having more than eighty distinct click phonemes when pulmonic modifications are included.
Zulu and Xhosa are Bantu languages spoken by millions of people in South Africa. Both languages use dental, alveolar, and lateral clicks as regular phonemic consonants appearing in everyday vocabulary. In Zulu, the dental click [ǀ] appears in words like icala (side), where ⟨c⟩ represents the dental click in Zulu orthography. The alveolar click [ǃ] appears in words spelled with ⟨q⟩, and the lateral click [ǁ] in words spelled with ⟨x⟩. Clicks in Zulu and Xhosa are not expressive additions borrowed from neighboring languages; they were incorporated into these Bantu languages through sustained contact with Khoisan communities and now function as fully integrated consonants with the same phonological privileges as any other consonant. The name Xhosa itself begins with a lateral click followed by a voiced aspirated stop.
The perception that clicks are exotic or rare often reflects the perspective of speakers whose languages do not have them. From a typological standpoint, clicks are systematic, stable, and productive in the languages where they occur, used in frequent everyday vocabulary, and learned as naturally by children as any other sound type. Approximately thirty languages — concentrated in southern and eastern Africa and including several languages with large speaker communities — use clicks as phonemic consonants. The challenge for the phonetician is not that clicks are inherently unusual, but that describing them accurately requires the lingual airstream mechanism to be added to the pulmonic-centered framework built in Chapters 6–12.
Ejectives
Purpose: explain the glottalic egressive mechanism, introduce ejective notation, and provide verified cross-language examples.
An ejective is a consonant produced with the glottalic egressive airstream: the glottis is closed tightly, and the larynx is raised, compressing the air in the pharynx and oral cavity above the vocal folds. When the oral closure is released, the compressed air bursts outward without any contribution from the lungs. The IPA marks ejective consonants with an apostrophe following the primary symbol: [pʼ], [tʼ], [kʼ], [sʼ], [tsʼ].
Ejectives share their place and manner of articulation with their pulmonic equivalents. A bilabial ejective [pʼ] involves the same lip closure as a bilabial plosive [p]; what distinguishes them is the airstream source. This means that the same place-manner classification system from Chapter 7 and Chapter 8 applies directly to ejectives, extended only by the glottalic egressive mechanism.
| IPA symbol | Place and manner | Verified example |
|---|---|---|
| pʼ | Bilabial ejective stop | Amharic (Semitic, Ethiopia): [pʼ] contrasts with pulmonic [p] and [b]; found in everyday vocabulary |
| tʼ | Alveolar ejective stop | Haida (isolate, British Columbia and Alaska): ejective stops are a regular feature of the consonant inventory; also found in Amharic and many Caucasian languages |
| kʼ | Velar ejective stop | Georgian (Kartvelian, Georgia): [kʼ] contrasts with [k] and [ɡ]; ejectives are a regular and frequent part of the Georgian inventory |
| sʼ | Alveolar ejective fricative | Some varieties of Amharic and other Ethiosemitic languages; ejective fricatives are less common cross-linguistically than ejective stops |
| tsʼ | Alveolar ejective affricate | Amharic, Georgian, and many languages of the Pacific Northwest coast of North America, including Tlingit |
Note: The ejective apostrophe notation and recognition of the five examples above are the productive targets for this chapter. Productive transcription focuses on correct apostrophe placement and articulatory identification.
Because ejectives require a closed glottis and a raised larynx, they cannot be voiced: a voiced ejective is articulatorily contradictory, since voicing requires air to flow through vibrating vocal folds while the ejective mechanism seals them shut. Ejectives are therefore always voiceless.
Georgian is a Kartvelian language spoken by approximately 3.7 million people, primarily in the country of Georgia. It is notable phonetically for having a three-way contrast among voiceless aspirated stops, voiced stops, and ejective stops at multiple places of articulation. Where English has a two-way contrast between [p] and [b] at the bilabial place (realized in English with aspiration on the voiceless member), Georgian distinguishes three categories: [pʰ] (aspirated), [b] (voiced), and [pʼ] (ejective). This three-way contrast is maintained at the alveolar, velar, and uvular places as well. Georgian ejectives are described by learners of the language as having a distinctive sharp, popping quality compared with the aspirated stops, which is consistent with the compressed-air burst produced by the raised larynx.
On a waveform or spectrogram, ejective stops show the same three phases as pulmonic stops — closure, hold, and release — but with differences. The release burst of an ejective is typically more intense and shorter than that of a comparable pulmonic aspirated stop, and the burst is not followed by aspiration noise, because there is no pulmonic airflow to produce turbulence at the larynx after release. The period of closure may also show no voicing, since the glottis is sealed. These acoustic properties can help confirm an ejective analysis in a language where the contrast is documented, though the definitive identification always requires language-specific context and distributional evidence rather than waveform inspection alone.
Implosives
Purpose: explain the glottalic ingressive mechanism, introduce implosive notation, clarify that implosives are typically voiced, and provide verified examples.
An implosive is produced by the glottalic ingressive airstream mechanism: the glottis is nearly closed but allows some air to pass through the vocal folds, and the larynx is lowered, creating a partial vacuum in the pharynx. The resulting negative pressure draws a small amount of air inward at the oral closure. When the closure is released, the implosive produces a sound with a characteristic "sucked" or ingressive quality. The IPA represents implosives with symbols distinct from their pulmonic counterparts: [ɓ] (bilabial), [ɗ] (alveolar), [ʄ] (palatal), [ɠ] (velar), and [ʛ] (uvular).
An important distinction between ejectives and implosives: ejectives are always voiceless because the glottis must be sealed. Implosives, however, are typically voiced, because the glottis is not fully closed — it allows vocal-fold vibration to continue even as the larynx lowers. This makes implosives unusual: they are consonants produced with an ingressive airstream mechanism but with voiced pulmonic phonation running simultaneously. The net direction of airflow can be ambiguous, and some phoneticians describe implosives as involving a mixture of mechanisms rather than a purely ingressive airstream.
| IPA symbol | Place | Verified example |
|---|---|---|
| ɓ | Bilabial implosive | Hausa (Chadic, West Africa): bilabial and alveolar implosives are phonemic; Sindhi (Indo-Aryan, South Asia): implosives occur at multiple places of articulation |
| ɗ | Alveolar implosive | Hausa, Sindhi, Fulfulde (Atlantic-Congo, West and Central Africa); alveolar is the most common implosive place cross-linguistically |
| ʄ | Palatal implosive | Sindhi; some varieties of Maasai (Nilotic, East Africa); less common than bilabial or alveolar |
| ɠ | Velar implosive | Lendu (Central Sudanic, Democratic Republic of the Congo); relatively rare cross-linguistically |
Note: All implosive symbols are recognition-level for this chapter. Productive work focuses on identifying place of articulation and mechanism from supplied transcriptions, not on generating implosive transcriptions independently.
It is easy to confuse implosives with clicks because both involve unconventional airstream mechanisms and both can produce a "pop" that English speakers associate with unusual sounds. The differences are fundamental. A click uses the lingual (velaric) mechanism: two closures trap air between the tongue and the back of the mouth, and the front closure's release draws air inward locally. An implosive uses the glottalic ingressive mechanism: the larynx lowers, pulling a small amount of air inward at the glottis, while the oral closure holds. The front closure of a click is released at the point of the click's articulation; an implosive is released like a normal stop. Clicks also have no inherent voicing from the lingual mechanism itself — their voicing is added by the pulmonic component — whereas implosives are typically voiced as part of their basic mechanism.
Hausa is a Chadic language spoken by more than 40 million people across northern Nigeria, Niger, and surrounding regions. It is one of the most studied languages with phonemic implosives. Hausa has two implosives — the bilabial [ɓ] and the alveolar [ɗ] — which contrast with the pulmonic plosives [b] and [d] in minimal pairs. The word baa (no) contrasts with a word containing the bilabial implosive [ɓ], and similar contrasts exist at the alveolar place. Because implosives and plosives share the same place and manner — the only difference is the airstream mechanism — the contrast between them is purely in the laryngeal behavior during the closure phase.
What Is Secondary Articulation
Purpose: define secondary articulation and distinguish it from a primary articulation and from a segment sequence.
Every consonant in Chapters 6–12 was described in terms of a single primary articulation: a constriction at one location in the vocal tract, with a specific manner of contact or approximation. But the vocal tract has enough independent controllers — tongue tip, tongue blade, tongue dorsum, velum, lips, and larynx — that a speaker can maintain a second, simultaneous constriction during a primary articulation. When a less prominent constriction is sustained throughout a primary articulation, it is called a secondary articulation.
Secondary articulations are distinguished from two other phenomena they might be confused with:
- A primary articulationThe main constriction that defines the consonant's manner and place. In [k], the primary articulation is the velar stop closure. Everything else during that closure is secondary if it involves an additional constriction.
- A segment sequenceTwo separate sounds occurring one after another, each with its own primary articulation. The sequence [kw] — as in quick — involves a velar stop followed by a labial-velar approximant. These are two segments in sequence, not one segment with a secondary articulation, because the labial-velar element is a separate syllabic onset position or offset following the stop.
- A secondary articulationA simultaneous additional constriction maintained throughout the primary articulation. In [kʷ] — a labialized velar stop — the velar closure and the lip rounding occur at the same time, within a single segment. The rounding is secondary because it modifies the primary consonant rather than following it.
The IPA represents secondary articulations with diacritics or superscript letters placed after the primary symbol. The choice of notation reflects the secondary articulation type:
- ʷ — labialization (simultaneous rounding)Superscript w, indicating secondary lip rounding and dorsal raising.
- ʲ — palatalization (simultaneous tongue dorsum raising toward hard palate)Superscript j.
- ˠ — velarization (simultaneous tongue dorsum raising toward velum)Superscript turned gamma, or alternatively the velarization diacritic through the symbol.
- ˤ — pharyngealization (simultaneous tongue root retraction)Superscript reversed glottal stop.
Labialization
Purpose: introduce labialization as the most common secondary articulation cross-linguistically and provide verified examples.
Labialization is a secondary articulation in which simultaneous lip rounding accompanies a primary consonant. The IPA notation uses a superscript ʷ after the primary symbol: [kʷ], [tʷ], [sʷ]. The labial gesture should be understood as simultaneous with — not following — the primary articulation.
Labialization is extremely common cross-linguistically. English itself has labialized consonants allophonically before rounded vowels: the [k] in cool is typically labialized — [kʷuːl] — because the lip rounding of the following vowel begins during the velar closure. This is a predictable allophonic effect in English, not a phonemic contrast. In many other languages, labialization is phonemic.
- [kʷ] — labialized velar stopKwak'wala (Wakashan, British Columbia): labialization contrasts phonemically with plain consonants; Twi (Akan, Ghana): labialized consonants form a regular series.
- [tʷ] — labialized alveolar stopLakhota (Siouan, North America) and other languages with labialized coronal consonants.
- [ɡʷ] — labialized velar voiced stopNumerous languages in West Africa, the Caucasus, and the Americas; contrasts with plain [ɡ] in languages where labialization is phonemic.
Whether labialization is a phonemic property (a consonant is always labialized regardless of context) or an allophonic effect (a consonant becomes labialized before rounded vowels) varies by language. In English, the labialization of [k] before [uː] is allophonic: it is predictable from the following vowel and does not distinguish words. In Twi, labialized consonants and plain consonants contrast in identical vowel environments, making labialization phonemic. The distinction between phonemic and allophonic labialization can only be established by examining distribution within the specific language — the articulatory description of the secondary gesture is the same in both cases.
Palatalization
Purpose: introduce palatalization, its phonetic basis, IPA notation, and cross-language significance, especially in Slavic languages.
Palatalization is a secondary articulation in which the tongue dorsum simultaneously raises toward the hard palate during a primary consonant. The effect is to give the consonant a "bright" or front-resonance quality. The IPA notation uses a superscript ʲ after the primary symbol: [tʲ], [dʲ], [nʲ], [lʲ], [sʲ].
Like labialization, palatalization occurs allophonically in English — the [k] in key is fronted toward the palate compared with the [k] in car — but English does not use palatalization phonemically. In Russian and other Slavic languages, however, palatalization is among the most important and pervasive phonemic contrasts in the entire consonant inventory.
- [nʲ] — palatalized alveolar nasalRussian: the word нет (net, no) contains a palatalized nasal, transcribed [nʲɛt]. The palatalized and non-palatalized nasals contrast in many minimal pairs.
- [tʲ] — palatalized alveolar stopRussian and Polish: contrasts with plain [t] in identical vowel environments in Russian, forming one of the central dimensions of the consonant inventory.
- [lʲ] — palatalized lateral approximantRussian, Italian (as a distinct phoneme spelled ⟨gli⟩), and many other languages; sometimes written with its own IPA symbol [ʎ] when analyzed as a separate phoneme rather than an allophone.
Russian is one of the most frequently cited languages for phonemic palatalization. The language has approximately fifteen pairs of phonemes that contrast solely in palatalization — plain versus soft (palatalized) consonants — making it a central organizing feature of the sound system rather than a marginal phenomenon. The phonemes /t/ and /tʲ/, /n/ and /nʲ/, /l/ and /lʲ/ each contrast in minimal pairs. Learners of Russian typically find palatalization challenging to perceive and produce because no comparable phonemic contrast exists in English, making the secondary articulatory dimension initially difficult to detect auditorily. The key acoustic correlate is a rise in the second formant — the same formant that distinguishes front from back vowels — consistent with the palatal secondary gesture.
Velarization
Purpose: introduce velarization, connect it to the "dark l" from Chapter 11, and clarify the relationship between phonemic and allophonic velarization.
Velarization is a secondary articulation in which the tongue dorsum simultaneously raises toward the velum during a primary consonant. The effect is a back-resonance quality often described as "dark." The IPA notation uses a superscript ˠ after the primary symbol, or — for the specific case of a velarized lateral — the symbol [ɫ].
You encountered the most familiar case of velarization in Chapter 11: the dark or velarized lateral [ɫ] in English. Recall that in many English accents, [l] is produced with a secondary velar gesture in syllable-coda position — bell, milk, pill — while the clear [l] without velarization appears in syllable-onset position — lid, let, blue. This allophonic distribution makes velarization a positional variant of /l/ in English, not a phonemic contrast.
- [ɫ] — velarized lateral approximantEnglish (coda position in many accents): [bɛɫ] bell; Arabic: velarized consonants are phonemic and form a major dimension of the inventory.
- [tˠ], [dˠ], [sˠ] — velarized alveolar consonantsIrish English and some varieties of Irish Gaelic: velarized alveolar consonants contrast with palatalized ones, producing a system parallel to the Russian palatalization pairs but oriented toward the back of the mouth.
Arabic uses a set of "emphatic" consonants — [sˤ], [dˤ], [tˤ], [ðˤ] — that have historically been described as both velarized and pharyngealized, depending on the variety and the analysis. In Modern Standard Arabic and many spoken varieties, these consonants involve a secondary gesture in which the tongue root retracts (pharyngealization) and the tongue dorsum raises toward the velum (velarization) simultaneously. The emphatic consonants contrast phonemically with their plain counterparts — minimal pairs are well attested — and they have dramatic effects on neighboring vowels, shifting vowel formants in ways that are acoustically prominent. Whether to analyze these as primarily velarized or primarily pharyngealized is a matter of ongoing phonetic debate for different Arabic varieties; the productive notation uses ˤ for pharyngealization, while some analysts prefer ˠ or a velarization-dominant analysis.
Pharyngealization
Purpose: introduce pharyngealization as tongue-root retraction, clarify its IPA notation, and connect it to Arabic emphatic consonants.
Pharyngealization is a secondary articulation in which the tongue root retracts toward the back wall of the pharynx during a primary consonant. The IPA notation uses a superscript ˤ after the primary symbol: [sˤ], [dˤ], [tˤ]. The pharynx narrows, producing a back-cavity resonance that affects neighboring vowels.
Pharyngealization is most systematically studied in Arabic, where the emphatic consonants described above function as phonemically distinct from their plain counterparts. In Semitic languages more broadly, pharyngealized consonants are attested in languages including Tigrinya, Maltese, and some varieties of Hebrew. The pharyngealization diacritic ˤ is also used for the pharyngeal approximant or fricative in IPA notation when it appears as a primary articulation rather than a secondary one.
A practical observation for transcription: pharyngealization and velarization are both described as back-resonance secondary articulations and often co-occur. When the phonetic literature on a language uses the term emphatic or dark, examine whether the analysis supports pharyngealization, velarization, or both before selecting a notation convention. State the convention you are using.
Double Articulation and the Tie Bar
Purpose: distinguish a double articulation (two primary constrictions of equal status) from a secondary articulation, and explain tie-bar notation.
A double articulation involves two simultaneous constrictions of equal phonological status — unlike a secondary articulation, where one constriction is clearly less prominent and modifies the other. The labial-velar approximant [w], introduced in Chapter 12, is the most familiar example: its lip rounding and dorsal raising are both primary gestures with equal status, which is why the symbol is listed under labial-velar in the IPA chart rather than as a labialized velar.
When a double articulation cannot be represented by a single IPA symbol, the IPA uses a tie bar connecting two separate symbols: [k͡p] (a simultaneous bilabial and velar stop, with both closures releasing simultaneously) or [ɡ͡b] (its voiced counterpart). Labial-velar stops of this type are found in many West African languages, including Ewe, Yoruba, and Igbo, where they function as regular phonemic consonants.
- [k͡p] — simultaneous bilabial and velar voiceless stopEwe (Gbe, Ghana and Togo): [k͡pɛ] is attested in Ewe vocabulary; the two closures — at the lips and at the velum — are made and released simultaneously.
- [ɡ͡b] — simultaneous bilabial and velar voiced stopYoruba (Niger-Congo, Nigeria): [ɡ͡bà] contrasts with [k͡pà] in Yoruba; both are common consonants in everyday vocabulary.
The tie bar and the secondary-articulation superscripts look similar in function — both involve combining symbols — but they encode different analytical claims. A superscript such as ʷ in [kʷ] represents a secondary articulation: the lip rounding modifies the primary velar stop, which remains the main consonant. A tie bar in [k͡p] represents two articulations of equal status: there is no single primary closure and one modifier; both closures are primary. The distinction matters for phonological analysis. A language might have [kʷ] as a labialized velar (one phoneme, with a secondary labial gesture) in one system, and [k͡p] as a genuinely double-articulated stop (one phoneme with two equal primary closures) in another system. Read the available phonological description carefully before choosing notation.
Guided Observation
The observations below use sounds already available to you as a speaker of English or any language you know. Do not attempt to produce clicks, ejectives, or implosives through unguided force. Observation 4 involves a safe approximation only.
Observation 1: Identifying a dental click in English. Make the sound sometimes written "tsk" or "tsk tsk" — the sound English speakers produce to express mild disapproval or impatience. (If this sound is unfamiliar from your variety of English, make the sound used to urge a horse to move, which is the lateral click.) Focus on what your tongue is doing. Notice that the tongue tip is in contact with something at the front of the mouth while the back of the tongue makes a simultaneous contact further back. The sound is produced by releasing the front contact while the back contact briefly remains. What did the front contact feel like — were your tongue tip, your upper front teeth, or the alveolar ridge involved? Describe this without reference to the written symbol.
Observation 2: Approximating an ejective. Produce a [p] sound, but before you release it, close your throat tightly — as if you were about to strain or lift a heavy object — and then release the lip closure with the throat still closed. You may feel a small pop produced by the compressed air between your lips and your closed larynx. This is not an accurate representation of a phonemic ejective, but it gives you a tactile impression of how compressing air above the glottis, rather than pushing from the lungs, can produce a release burst. Compare this with a normal [p] and describe any difference you notice in where the pressure seems to come from.
Observation 3: Secondary palatalization. Say the English word key slowly and then say car slowly. Focus on the position of the tongue body just before you release the velar closure. In key, the tongue body is positioned further forward and higher than in car. This anticipatory tongue-body raising in key is the English allophonic equivalent of palatalization: the dorsum moves toward the palate during the velar closure in anticipation of the following front vowel. In languages with phonemic palatalization, this front-raised position is present regardless of what vowel follows. Describe the tongue body position you feel during the closure phase of key versus car.
Observation 4: Secondary velarization. Say the English word pill and then lip. Focus on the [l] in each word. In pill, the [l] comes before the vowel, in syllable-onset position. In lip, the [l] comes after the vowel, in syllable-coda position. Many English speakers produce a darker, more back-resonant quality for the coda [l] in lip. Try to feel whether the back of the tongue is involved differently in the two words. The coda [l] in lip likely involves a secondary dorsal raising toward the velum — the velarization that produces the dark [ɫ].
- Based on Observation 1, describe the two simultaneous contacts you felt during the click sound. Which contact was released first to produce the click, and which remained briefly afterward?
- Based on Observations 3 and 4, explain in your own words why palatalization and velarization are called secondary articulations rather than separate consonants. What makes them secondary rather than primary?
Dataset Analysis Lab
Purpose: read and interpret transcriptions involving non-pulmonic consonants and secondary articulations across languages.
The transcriptions below come from four languages. Each has been verified against standard phonological descriptions. Your task is to analyze the transcriptions using only the information in this chapter and in the preceding chapters.
| Orthographic form | Gloss | Phonetic transcription | Language and variety |
|---|---|---|---|
| ⟨მე⟩ | I, me | [mɛ] | Georgian (Kartvelian) |
| ⟨კი⟩ | yes | [kʼi] | Georgian (Kartvelian) |
| ⟨ტა⟩ | forest | [tʼa] | Georgian (Kartvelian) |
| ⟨კარი⟩ | door | [kʰaɾi] | Georgian (Kartvelian) |
| ⟨нет⟩ | no | [nʲɛt] | Russian |
| ⟨да⟩ | yes | [da] | Russian |
| ⟨люди⟩ | people | [ˈlʲudʲɪ] | Russian |
| ⟨لا⟩ | no | [laː] | Modern Standard Arabic |
| ⟨صبر⟩ | patience | [sˤabr] | Modern Standard Arabic |
| ⟨طريق⟩ | road, way | [tˤaˈriːq] | Modern Standard Arabic |
Source: Georgian forms consistent with standard phonological descriptions of Georgian; Russian forms consistent with standard descriptions of contemporary spoken Russian; Arabic forms consistent with descriptions of Modern Standard Arabic phonology.
Step 1 — Identify airstream mechanism
For each consonant in the Georgian data, identify the airstream mechanism. Use the delimiters correctly. Which consonants are pulmonic egressive, and which are glottalic egressive? What is the evidence for your classification?
Step 2 — Identify secondary articulations
List every secondary articulation that appears in the Russian and Arabic data. For each, identify the primary consonant, the secondary gesture, and the IPA notation used to represent it. Then describe the secondary gesture in articulatory terms.
Step 3 — Compare primary and secondary
The Russian data includes the sequence [lʲudʲɪ]. Identify every secondary articulation in this word. Then compare the consonant [lʲ] with the consonant [l] from the Arabic data [laː]. Are these the same phone? Explain using place, manner, voicing, and secondary articulation.
Step 4 — Contrast and distribution
The Georgian data includes both [kʼ] and [kʰ]. Apply the contrast test: do these form a minimal pair? What is the phonological status of [kʼ] in Georgian based on the evidence in this dataset? What additional data would you need to make a more complete claim?
Practice — Description tasks: Write a full articulatory description for each transcription below, including primary articulation (airstream, phonation, place, and manner) and any secondary articulation. Do not use the words "click," "ejective," or "implosive" without providing the articulatory description those terms are shorthand for.
- [ǀ]
- [kʼ]
- [ɓ]
- [nʲ]
- [sˤ]
- [ɡ͡b]
Common Mistakes
Error: Describing non-pulmonic consonants as "exotic," "rare," or "difficult" without articulatory justification, and treating them as curiosities rather than systematic phonemic categories.
Correction: The analytical framework is the same as for pulmonic consonants. Non-pulmonic sounds differ in airstream mechanism, not in kind. A click has a place of articulation, a release direction, and a phonological distribution just like an alveolar stop does. An ejective has a voicing status and a place and manner just like a pulmonic fricative does. The appropriate analysis describes these properties, not the European speaker's initial reaction.
Error: Writing an ejective as a stop followed by a glottal stop: [pʔ] instead of [pʼ].
Correction: The ejective notation [pʼ] uses an apostrophe, not the glottal stop symbol [ʔ]. The two symbols are visually similar but analytically distinct. A glottal stop [ʔ] is a separate pulmonic segment — a stop articulated at the glottis. An ejective apostrophe ʼ is a diacritic modifying the primary symbol to indicate the glottalic egressive mechanism. Writing [pʔ] implies a sequence of two separate segments, not a single ejective consonant. Use [pʼ] for ejectives.
Error: Treating [kʷ] and [k] followed by [w] as the same thing.
Correction: [kʷ] is a single segment with a simultaneous secondary labialization: the rounding occurs during the entire velar closure. The sequence [k.w] or [kw] represents two separate segments — a velar stop followed by a labial-velar approximant — in consecutive syllable positions. The distinction is phonological: a language might have one but not the other, or both as separate phonemes. In transcription, use the superscript when the labial gesture is simultaneous with and part of the primary consonant; use a sequence when the labial approximant occupies a separate syllabic position.
Error: Assuming that implosives are ingressive pulmonic sounds — sounds produced by breathing in.
Correction: Implosives use the glottalic ingressive mechanism, not the pulmonic ingressive mechanism. The speaker does not simply inhale. The larynx lowers, creating a small partial vacuum above the glottis, while the vocal folds continue to vibrate (producing voicing). The inward movement of air is very slight and is generated by the lowering larynx, not by the lungs. An English speaker who inhales a pulmonic ingressive sound is producing something qualitatively different from an implosive — an ingressive pulmonic sound has different acoustic properties and a different articulatory source.
Error: Using the same transcription for a secondary articulation and a phoneme sequence without checking the language's phonological analysis.
Correction: Whether [tʲ] (palatalized t) or [tj] (t followed by the palatal glide) is the appropriate notation for a given language depends on the phonological analysis of that language. In Russian, palatalization is a feature of the consonant phoneme, not a following glide segment, so [tʲ] is correct. In English, the sequence tune in British English may be realized as [tjuːn] (with a separate glide) or as [tʃuːn] (with yod coalescence), depending on the variety — not as [tʲuːn] with phonemic palatalization. Consult the language's phonological description and state the convention you are using.
Chapter Review
Part A — Definitions and Mechanisms
1. What distinguishes the lingual airstream mechanism from the glottalic airstream mechanism? Identify the source of the pressure change in each.
2. Ejectives are always voiceless, but implosives are typically voiced. Explain why this is the case in articulatory terms.
3. Define secondary articulation. What two phenomena must secondary articulation be distinguished from, and what is the key criterion for each distinction?
4. Explain when a tie bar is the appropriate notation and when a secondary-articulation superscript is appropriate. Use an example for each.
Part B — Symbol Identification
For each IPA symbol or notation below, supply: (a) a full articulatory description identifying the airstream mechanism, voicing, place, and manner as applicable; (b) the type of non-pulmonic mechanism or secondary articulation, if any; and (c) one language in which the sound functions as a phonemic consonant, with the language family if known.
- ǃ
- kʼ
- ɗ
- tʲ
- ɫ
- sˤ
- k͡p
Part C — Notation and Analysis
1. A student transcribes a Georgian word as [kʔali]. Identify the error in this transcription and rewrite it correctly, explaining what the original notation implied and what the corrected notation claims.
2. Russian has pairs of consonants that contrast in palatalization: [n] versus [nʲ], [t] versus [tʲ], and so on. In English, the allophone of /k/ before front vowels is fronted toward the palate. Explain why the Russian contrast is described as phonemic palatalization while the English fronting is not, even though both involve the same secondary palatal gesture.
3. The word for "road" in Modern Standard Arabic is transcribed [tˤaˈriːq]. Identify every secondary articulation present in this transcription. Then describe the articulatory difference between the first consonant [tˤ] and a plain alveolar stop [t].
Cumulative Review
1. From Chapter 6: Non-pulmonic consonants differ from pulmonic consonants in airstream mechanism. Ejectives and implosives use the glottalic mechanism. Using what you know about voiced versus voiceless phonation from Chapter 6, explain why ejectives are necessarily voiceless while implosives are typically voiced.
2. From Chapter 9: Ejective stops share their place and manner of articulation with pulmonic stops. Describe the three phases of a pulmonic stop (closure, hold, release) and identify how each phase differs in an ejective stop.
3. From Chapter 12: The chapter introduced the labial-velar approximant [w] as a double articulation involving simultaneous lip rounding and dorsal raising. Using the distinction between double articulation and secondary articulation introduced in this chapter, explain why [w] is analyzed with a tie bar (or single double-articulation symbol) rather than as a plain velar with labialization superscript [ɰʷ].
Consonant Analysis Portfolio
Unit II Portfolio — Chapters 6–13
Demonstrating Consonant Competence
This portfolio closes Part II of the course. It asks you to apply, in an integrated way, the consonant analysis systems introduced across the eight chapters of this unit. The portfolio is not an additional quiz; it is an opportunity to demonstrate that you can use the complete framework coherently and that you understand the analytical principles behind the notation.
Complete all four tasks. Where you are asked to make analytical decisions, explain your reasoning. Correct notation alone is not sufficient; the explanation demonstrates that the notation is not mechanical substitution.
Task 1: Framework Description (Chapter 6–8)
In 150–250 words, explain the three-dimensional framework used to classify pulmonic consonants. Your explanation should address: the role of the airstream and phonation dimension; the role of place of articulation; and the role of manner of articulation. Then identify which of the three dimensions must be extended or supplemented to account for (a) non-pulmonic consonants and (b) secondary articulations, and explain how each extension works.
Task 2: Consonant Classification Set (Chapters 6–13)
For each of the following IPA transcriptions, provide a complete phonetic classification. Include airstream mechanism, phonation, place of articulation, manner of articulation, and any secondary articulation or special notation features. Identify the class (stop, fricative, approximant, etc.) and any non-pulmonic mechanism. Use correct delimiters throughout.
- [ŋ]
- [ʒ]
- [ɾ]
- [kʼ]
- [ɓ]
- [ǁ]
- [lʲ]
- [sˤ]
- [ɡ͡b]
- [ʔ]
Task 3: Distribution Analysis (Chapters 5, 6–13)
The table below presents a small dataset from an unnamed language. Study the distribution of [p] and [pʼ] in the data.
| Phonetic transcription | Gloss |
|---|---|
| [pata] | leaf |
| [pʼata] | fire |
| [apo] | water |
| [apʼo] | stone |
| [mipa] | bird |
| [mipʼa] | tree |
Answer the following: (a) Do [p] and [pʼ] appear in the same environments? Identify the shared environments. (b) Do they form a minimal pair? Identify it if present. (c) Based on the evidence, are [p] and [pʼ] separate phonemes or allophones in this language? Justify your conclusion. (d) Using correct delimiters, write the phonemic symbols for these phones according to your analysis.
Task 4: Cross-Language Comparison (Chapters 6–13)
Choose any two consonant sounds from different languages that share an IPA place label but differ in at least one of the following: airstream mechanism, manner, secondary articulation, or phonological status. Write a comparison of 100–200 words that uses correct phonetic notation throughout and explains what the shared place label does and does not tell you about the similarity of the two sounds.
Chapter Performance Task
Performance Task
Reading and Interpreting Complex Transcriptions
The transcriptions below represent words and short forms from languages that use non-pulmonic consonants or secondary articulations. Each transcription is provided along with its gloss and language. Your task is to read and interpret each transcription — not to produce the sounds, but to demonstrate that you can extract all the phonetically relevant information from the notation.
| Phonetic transcription | Gloss | Language and variety |
|---|---|---|
| [kʼiʒi] | stone (rough approximation of a Georgian form) | Georgian |
| [ɓolo] | end, tail | Hausa |
| [ˈnʲɛbo] | sky | Russian |
| [sˤalaː] | prayer (approximate citation form) | Modern Standard Arabic |
| [ǁàbà] | lion (approximate form illustrating lateral click) | Zulu-type inventory illustration |
| [ɡ͡bɛ] | name (Ewe syllable illustration) | Ewe (illustrative) |
Note: Some transcriptions above are broad phonetic representations used for instructional purposes. The Zulu and Ewe entries are illustrative of consonant types in those languages and should be treated as exercises in reading notation rather than precise lexical citation forms requiring independent verification.
For each transcription:
- Identify every consonant present. For each consonant, state its IPA symbol and provide a complete articulatory description including airstream mechanism, phonation, place, manner, and any secondary articulation.
- Identify any diacritics, superscripts, or special notation beyond the primary symbols and explain what each encodes.
- State whether any consonants in the transcription involve a non-pulmonic mechanism and, if so, identify which mechanism.
- Write one sentence for each transcription stating what you could not determine from the transcription alone without consulting additional language-specific evidence.
Answer Key (click to expand)
Dataset Analysis Lab — Step 1
Georgian consonants and their mechanisms: [m] in [mɛ] — pulmonic egressive, voiced bilabial nasal; [kʼ] in [kʼi] — glottalic egressive (ejective), voiceless velar stop; [tʼ] in [tʼa] — glottalic egressive (ejective), voiceless alveolar stop; [kʰ] in [kʰaɾi] — pulmonic egressive, voiceless aspirated velar stop. Evidence for ejective status: the ejective apostrophe ʼ explicitly encodes glottalic egressive mechanism; pulmonic consonants carry no mechanism diacritic. Additional evidence would include acoustic features such as a compressed release burst without aspiration noise.
Dataset Analysis Lab — Step 2
Secondary articulations in Russian: [nʲ] in [nʲɛt] — palatalization superscript ʲ; the tongue dorsum simultaneously raises toward the hard palate during the alveolar nasal closure. In [ˈlʲudʲɪ]: [lʲ] — palatalized alveolar lateral approximant (superscript ʲ); [dʲ] — palatalized alveolar stop (superscript ʲ). Secondary articulations in Arabic: [sˤ] — pharyngealization (superscript ˤ); tongue root retracts toward back wall of pharynx during alveolar fricative. [tˤ] — pharyngealization of alveolar stop. For each: primary consonant identified by base symbol; secondary gesture identified by superscript; articulatory description states what gesture accompanies the primary constriction.
Dataset Analysis Lab — Step 3
Secondary articulations in [ˈlʲudʲɪ]: palatalization on [lʲ] (initial position) and on [dʲ] (medial position). Comparison of [lʲ] and [l]: both are voiced alveolar lateral approximants with pulmonic egressive airstream. They differ in secondary articulation: [lʲ] includes a simultaneous raising of the tongue dorsum toward the hard palate; [l] in the Arabic data has no secondary articulation noted at the broad phonetic level. They are not the same phone: the secondary palatal gesture is an additional articulatory feature that changes the resonance properties of the sound, produces a different auditory quality, and in Russian differentiates phonemes.
Dataset Analysis Lab — Step 4
The Georgian dataset contains [kʼi] (stone) and [kʰaɾi] (door). These are not a minimal pair because the surrounding environments differ: [kʼi] has a front high vowel following, while [kʰaɾi] has a low vowel followed by additional consonants. A minimal pair requires identical environments with only the target sound differing. The data is consistent with [kʼ] being a phoneme in Georgian (as established by standard descriptions), but this dataset alone is insufficient to demonstrate the contrast — a pair such as Georgian words differing only in [kʼ] versus [kʰ] in the same environment would be needed.
Practice — Description Tasks
1. [ǀ]: Dental click; lingual (velaric ingressive) airstream; front closure formed by tongue tip behind upper teeth; back closure formed by the tongue dorsum against the velum or uvula; released at the front (dental) contact; voicing and pulmonic phonation type depend on accompanying notation not present in this bare symbol, which represents the click type only. 2. [kʼ]: Voiced — no, voiceless; glottalic egressive (ejective) airstream; velar stop; the larynx is raised with the glottis closed, compressing air above the vocal folds; the velar oral closure is released, producing a compressed burst without aspiration. 3. [ɓ]: Typically voiced; glottalic ingressive (implosive); bilabial stop; the larynx lowers, creating a partial vacuum above the glottis; vocal folds continue to vibrate, producing voicing; the bilabial oral closure holds and then releases. 4. [nʲ]: Voiced; pulmonic egressive; primary articulation: alveolar nasal (velopharyngeal port open, tongue tip at alveolar ridge, complete oral closure); secondary articulation: palatalization, tongue dorsum simultaneously raised toward hard palate. 5. [sˤ]: Voiceless; pulmonic egressive; primary articulation: alveolar fricative (turbulent airflow at the alveolar constriction); secondary articulation: pharyngealization, tongue root simultaneously retracted toward back wall of pharynx. 6. [ɡ͡b]: Voiced; pulmonic egressive; double articulation — simultaneous bilabial stop closure and velar stop closure, both primary with equal status; both closures released simultaneously; tie bar notation indicates double articulation rather than sequence.
Chapter Review — Part A (model answers)
1. The lingual mechanism uses the tongue dorsum to form a back closure against the velum or uvula, and the tongue tip to form a front closure; the space between them is modified to create ingressive air movement when the front closure releases. The glottalic mechanism uses the larynx as the pressure source: raised for ejectives (compressing air above), lowered for implosives (rarefying air above). 2. Ejectives require the glottis to be tightly closed so the larynx can compress air above it without any air escaping through the vocal folds. Voicing requires air to pass through vibrating vocal folds. These two conditions are incompatible, so ejectives are necessarily voiceless. Implosives lower the larynx but do not seal the glottis — the vocal folds continue to vibrate for voicing while the larynx lowers. 3. A secondary articulation is a simultaneous additional constriction that is less prominent than and modifies the primary articulation. It must be distinguished from a primary articulation (the main constriction that defines place and manner) by its lesser prominence and from a segment sequence (two separate sounds in time) by being simultaneous. The criterion for the first distinction is relative prominence; the criterion for the second is temporal simultaneity versus sequencing. 4. Use a superscript when the secondary gesture is less prominent and modifies the primary consonant — for example, [kʷ] where lip rounding modifies a primary velar stop. Use a tie bar when two articulations are simultaneous and of equal phonological status — for example, [k͡p] where bilabial and velar closures are both primary.
Chapter Review — Part C
1. The student wrote [kʔali], which implies a velar pulmonic stop followed by a separate glottal stop segment, then the vowel sequence. The correct transcription for a Georgian ejective velar stop would be [kʼali]. The difference: the glottal stop symbol [ʔ] represents an independent pulmonic segment at the glottis; the ejective apostrophe ʼ is a diacritic indicating the glottalic egressive mechanism during the primary consonant. These are analytically different claims about the segment count and the airstream mechanism. 2. In Russian, palatalized and plain consonants appear in identical vowel environments and form minimal pairs; the palatalization is a property of the consonant phoneme regardless of context. In English, the fronting of [k] before front vowels is predictable from the following vowel — it does not occur independently of that phonological environment. The Russian contrast is phonemic because it is not predictable from context; the English fronting is allophonic because it is entirely predictable. The phonetic gesture may be similar, but the phonological status differs because of the distributional facts. 3. Secondary articulation in [tˤaˈriːq]: pharyngealization ˤ on the initial [tˤ]. The plain alveolar stop [t] involves only the primary alveolar closure with no modification of the tongue root. The pharyngealized alveolar stop [tˤ] involves the same primary alveolar closure plus a simultaneous retraction of the tongue root toward the back wall of the pharynx, narrowing the pharyngeal space and producing a back-resonance quality that affects neighboring vowels.
Portfolio Task 3 (model answer)
(a) [p] and [pʼ] appear in word-initial position (before [a]) and in intervocalic position (between two vowels of the same quality). The environments are identical. (b) The forms [pata] (leaf) and [pʼata] (fire) constitute a minimal pair: the only difference is [p] versus [pʼ] in word-initial position before the same vowel sequence. (c) Because the two phones contrast in identical environments and form at least one minimal pair, they are separate phonemes in this language. The contrast is phonemic. (d) Using correct delimiters: the phonemic symbols are /p/ (pulmonic egressive bilabial stop) and /pʼ/ (glottalic egressive bilabial ejective stop).
Cumulative Review
1. Voicing requires the vocal folds to vibrate, which requires a pulmonic egressive airstream to flow through a partially adducted glottis. Ejectives seal the glottis completely to build pressure — no air can pass through the vocal folds, so voicing is impossible. Implosives lower the larynx rather than sealing the glottis; the vocal folds can remain adducted and vibrate as a small amount of air is drawn inward, so voicing is possible and typical. 2. Pulmonic stop phases: (1) closure — the active articulator forms a complete seal against the passive articulator; (2) hold — the seal is maintained while pulmonic air pressure builds behind it; (3) release — the closure is broken, releasing the compressed air as a burst. In an ejective stop: (1) closure — same oral closure as the pulmonic equivalent; (2) hold — the glottis is simultaneously sealed and the larynx raises, compressing the air above; no pulmonic pressure increases because the glottis is closed; (3) release — the oral closure releases, releasing the compressed glottalic air as a burst without aspiration. 3. In a double articulation, both gestures — here bilabial rounding and dorsal raising — are of equal phonological status: neither modifies the other. The symbol [w] was listed on the IPA chart under labial-velar precisely because both gestures are primary. The hypothetical notation [ɰʷ] would imply a primary velar approximant with a secondary labialization, implying that the velar element is primary and the labial element secondary. This misrepresents the analysis: the double-articulation notation or the labial-velar symbol treats both gestures as equal in status, which matches the phonological and phonetic evidence better.
Looking Ahead
What You Can Now Do
With this chapter, the IPA consonant chart has been introduced in full. You can now describe pulmonic consonants by place, manner, voicing, and relevant diacritics; identify non-pulmonic consonants by their airstream mechanism; read and interpret click, ejective, and implosive notation; analyze secondary articulations using superscript and diacritic notation; and distinguish secondary articulations from primary articulations and segment sequences. The Consonant Analysis Portfolio demonstrates that these systems work together as a unified analytical framework, not as an accumulation of isolated facts. You have also reinforced the foundational principle — established in Chapter 5 and applied throughout Part II — that a phone's IPA symbol describes its phonetics, while its phonological status (phoneme or allophone) requires distributional evidence from the specific language.