Department of Linguistics
PHONETICS AND PHONOLOGY
Complex Consonant Articulations
|Important: If you have not yet either installed the phonetic font "Charis SIL" or tested this installation to determine if the phonetic characters installed properly then click here to go to the phonetic font help pages.|
- Clark ,Yallop & Fletcher: Chapter 3
- Ladefoged, chapter 9, has some discussion of complex articulations in vowels. .
The following book was the source of many of the examples included below. It is not required reading for this course, but is an excellent reference for anyone with a continuing interest in phonetics.
- Ladefoged, P. and Maddieson, I., 1996, The Sounds of the World's Languages, Blackwell, Oxford
You should also examine the following IPA topics for details of some of the symbols used in the transcription of complex articulations:-
Definition of "Complex Articulation"
For the purposes of this course, complex articulations can be considered to belong to two categories:-
- Simultaneous articulations (double articulation or secondary articulation)
- Sequence of two (or more) gestures within the same phoneme
In reality, all speech is complex, involving the coordination of the movement of numerous articulators. The articulatory movements of the various articulators can be assumed to be attempting to reach particular articulatory targets for each phoneme in a sequence of speech sounds. The articulatory gestures belonging to one phoneme overlap with the articulatory gestures of other phonemes as each of the articulators attempts to reach its target positions. Because some articulators move slower than others these movements are not synchronised.
The key to this concept of "complex articulation" revolves around the notion of the "articulatory target" for each articulator. To determine whether a sound is a complex articulation you should ask the following questions:-
- Does the phoneme have more than one articulatory target (simultaneous or in sequence)?
- If yes, is more than one of these targets contrastive?
The second question is essential as all phonemes involve more than one articulator with its own particular articulatory target. For example, a low vowel has both a low tongue target position and a low jaw target position. The jaw target articulation does not add to the phonological contrastiveness of the vowel, however, as the low tongue position alone is sufficient to identify this vowel sound as a low vowel.
On the other hand, the contrast between /i/ and /y/ in a language such as French depends upon both the target articulation of the tongue (high, front) and the target articulation of the lips (spread or rounded). Changing either target articulation will change the phonemic identity of the sound. In other words the two targets (lingual and labial) are both contrastive.
We can now reduce the two questions, above, to a single question:-
- Does the phoneme have more than one contrastive articulatory target (simultaneous or in sequence)?
If the answer to this question is yes then we can conclude that we have a complex articulation.
Three main types of Complex Articulation
It is useful to divide Complex articulations into three main types:-
- Simultaneous Articulations
- Secondary Articulations
- Co-ordinate or Double Articulations
- Transitional or Sequential Complex Articulations
Primary and Secondary Articulations
Complex simultaneous articulations are usually described in terms of primary and secondary articulations. The primary articulation is the articulation with the greater degree or rank order of stricture. Stricture order, from highest to lowest, is stop, trill, fricative, approximant, resonant. Secondary articulations very often have approximant stricture, especially when the primary and secondary articulators are different parts of the tongue.
Lingual (of the tongue) articulations sometimes take precedence over nasal articulations of the same degree of stricture so that the lingual articulation would be primary and the nasal articulation would be secondary. This means that lingual articulations would be considered primary when the other contrastive articulation is nasal. For this reason we have nasalised vowels in numerous languages and contrastive nasalised approximants and fricatives in a small number of languages.
Co-ordinate or Double Articulation
When a sound has two simultaneous contrastive articulations and these articulations are of the same rank order of stricture (eg. both have stop stricture or both have fricative stricture) then it is not possible to distinguish between primary and secondary articulations. Both articulations in such a case are considered to be primary and the sound is said to be produced by a co-ordinate or double articulation.
The most common double articulations are stop-stop double articulations although fricative-fricative double articulations also exist. Another very common double articulation is the labial-velar approximant-approximant double articulation /w/.
Transitional (Sequential) Complex Articulations
This type of complex articulation involves "...separate and successive articulatory activities which together can be identified as a single segment" (Clark and Yallop, 1995, p63). Such segments include affricates and diphthongs as well as a number of other classes of segment which will be dealt with below.
It will quickly become clear in the discussion below that many cases of complex articulations are not contrastive but are the result of the modification of the articulation of a particular sound as a consequence of the influence of an adjacent sound. When the pronunciation of a sound is influenced by another sound, co-articulation is said to have occurred. Such co-articulation involves the modification of a particular articulator's gesture so that it is more like the influencing sound. These modifications are usually to non-contrastive gestures as there is a linguistic need to maintain phonological contrasts so that the word can be understood. For this reason, vowels are nasalised when adjacent to nasal consonants and this nasalisation is greatest in languages which do not have an oral-nasal vowel contrast. Co-articulation occurs very strongly for non-contrastive articulators and gestures and is the main contributor to the allophonic richness of many phonemes.
When a contrastive gesture is modified this often results in a change in the phoneme. Such changes are usually referred to as assimilation. These assimilations can potentially change the meaning of a word. There are, however, many contexts in English, for example, where such phoneme-changing assimilation does not result in communicative difficulty. These are typically highly predictable contexts, contexts in which the contrast is neutralised, or contexts in which the listener can be reasonably expected to rely upon linguistic context to determine meaning.
Naming of Complex Articulations
When naming a complex simultaneous articulation, the primary articulation provides the main name of the sound and the secondary articulation usually acts as a pre-modifier of that name. As a consequence we have "labialised velar", "velarised alveolar", "nasalised labial", etc. Sometimes we split a secondary articulation into sub-categories and use those sub-category names as pre-modifiers, eg. "rounded high-front vowel" instead of "labialised high-front vowel".
To add a further complication, when the primary articulator is the tongue then the name of the articulation is most often taken from the passive articulator that the tongue articulates with (eg. alveolar, velar, etc.). Sometimes the name of the articulation is modified by the part of the tongue being used (eg. "apico-dental" versus "lamino-dental"), but only when that extra information is necessary. This means that we could potentially talk about complex articulations with names like "pharyngealised apico-dental".
Types of Complex Articulation
Simultaneous nasalisation of consonants is very rare as a contrastive feature in languages but is very common in vowels. See the topic "vowel nasalisation" for more information.
Prenasalisation most commonly occurs in prenasalised stops where the initial part of the occlusion phase of the oral stops involves nasal output.
Prenasalisation is transcribed by placing the appropriate nasal stop symbol before the oral stop symbol (generally with a ligature tying them together).
eg. [m͡b] [n͡d] [ŋ͡k]
For weaker prenasalisation it has been customary for the nasal symbol to precede the stop symbol as a superscript (eg. [mb] [nd]). Note that the latest IPA standard does not appear to support the preceding superscripted forms and so only transcriptions of the type [m͡b] and [n͡d] will be used in this course.
|Margi (Northern Nigeria) (Note also the four pre-nasalised affricates)|
|[n͡tʃà]||"point at"||[n͡dʒà]||"open wide"|
Postnasalisation is effectively the reverse of prenasalisation and most commonly occurs in nasally released oral stops. During the occlusion both the oral cavity and the velum are closed. Release occurs initially into the nasal cavity by the opening of the velum a short time before the opening of the oral cavity.
Postnasalisation can be transcribed by placing the appropriate nasal stop symbol after the oral stop symbol (generally with a ligature tying them together).
eg. [b͡m] [d͡n]
The latest IPA standard recommends the use of a superscripted "n" following the oral stop symbol to indicate nasal release of any oral stop. [bm] and [dn] have been used traditionally, but the IPA standard appears to be recommending [bn] and [dn]. In other words n is used for all nasal release regardless of the place of articulation.
eg. [bn] [dn]
|Arrernte (Northern Australia) This language has both pre-nasalised and post-nasalised stops|
Simultaneous labialisation is the protrusion or rounding of the lips at the same time as the primary articulation. This may be used phonemically when it contrasts with a neutral or spread lip posture which would be the normal or default posture for this type of primary sound. Simultaneous labialisation can be used contrastively or may be only contextual labialisation. For example, consonants adjacent to rounded vowels may be rounded through coarticulation with the lip posture of that sound. For example, in English [s] will most often be labialised [sʷ] when surrounded by rounded vowels (eg. [ʉː]) in sequences such as "two suits".
Simultaneous labialisation is transcribed by placing a superscript "w" symbol ʷ to the right of the sound being labialised.
eg. [sʷ], [kʷ], [lʷ]
Some phoneticians have used a "w" placed under the symbol of the sound being labialised. This does not appear to be supported by the latest IPA standard and will not be used in this course.
In transitional labialisation the lip rounding or protrusion only becomes evident in the later part of the sound being labialised. This phenomenon most commonly occurs when a sound is followed by a lip rounded or protruded sound (such as [w] or [u]) and occurs because of coarticulation with that sound. Such examples are generally not phonemic, but rather are allophones of the normally unlabialised phoneme.
Transitional labialisation is transcribed by placing a superscript "w" symbol ʷ to the right of the sound being labialised.
eg. [sʷ], [kʷ], [lʷ]
The following examples are probably of simultaneous labialisation.
|Kwakw'ala Labialised and non-labialised Velar and Uvular stops
(nb. ˈ indicates that the following syllable is stressed, whilst ʼ indicates that the preceding stop is an ejective stop)
|Voiceless velar and uvular oral stops|
|Voiced velar and uvular oral stops|
|Voiceless velar and uvular fricatives|
|Voiceless velar and uvular ejective stops|
|[ˈkʼata]||"writing"||[kʷ ʼeˈsa]||"light (weight)"|
|[ˈqʼasa]||"sea otter"||[ˈqʷ ʼasa]||"crying"|
In simultaneous palatalisation the front of the tongue is raised to approximate the tongue position for the vowel [i] below the hard palate. This is a secondary articulation for a sound which has a primary articulation anterior to the hard palate, such as alveolar, dental and labial articulations. Simultaneous palatalisation very often occurs as a result of co-articulation when a sound with an anterior primary articulation is uttered adjacent to palatal consonants such as [ j] or high front vowels such as [i]. For example, in English, the [p] in the words "peel" (/piːl/) and "pure" (/pjʊə/) can be pronounced with either simultaneous or transitional palatalisation. In English the consonant [l] is usually pronounced with palatalisation when it precedes a vowel at the beginning of a syllable (this is generally referred to as the "clear l").
Clark and Yallop show simultaneous palatalisation as being transcribed with a subscript "j" following the symbol of the primary articulation, as in [lj]. This does not appear to be supported by the recent IPA standards and will not be used in this course. Instead, the IPA recommended superscript "j" will be used to indicate both simultaneous and transitional palatalisation, as follows:-
eg. [lʲ] [pʲ]
Transitional palatalisation is most often seen as a palatal feature in the transition to the next segment. If the next feature is a palatal sound then this feature may be merely a non-contrastive co-articulatory anticipation. Transitional palatalisation can also be used contrastively and is produced in such cases independently of the nature of the following sound.
The IPA standard recommends a superscript "j" following the symbol of the sound being palatalised. This is identical to the transcription of simultaneous palatalisation.
eg. [lʲ] [pʲ] [kʲ]
The following examples may involve either simultaneous or transitional palatalisation. No distinction is made in these examples between simultaneous and transitional palatalisation.
|Russian Contrasting plain and palatalised bilabial stops (and also a [pj] sequence) as well as giving two examples of palatalised [rʲ]|
It is possible to imagine co-articulatory influences resulting in a transitionally velarised labial sound into a following velar consonant (such as [...apxa...]). Only simultaneous velarisation is possible for sounds with lingual primary articulations as velarisation involves an adjustment of the entire tongue body. Contrastive velarisations are always simultaneous.
The IPA standard recommends a superscript "ɣ" following the symbol of the sound being velarised.
eg. [sˠ] [tˠ] [dˠ]
Also, a tilde character "~" is sometimes placed through another symbol to indicate velarisation or pharyngealisation. This is traditionally done most often to indicate a "dark" or velarised [l].
|Marshallese (North Pacific)
Normal and velarised bilabial nasal stop
|Normal and velarised lateral approximant|
Secondary pharyngealisation appears to be always simultaneous with lingual primary articulations and involves an adjustment of the tongue body. Only simultaneous secondary pharyngealisation appears to be used contrastively in human language. It may be possible, however, for co-articulatory forces to result in transitional pharyngealisation of a bilabial such as [b] when followed (probably across a word or syllable boundary) by the pharyngeal fricative [ʕ] in a language such as Arabic.
The IPA standard recommends a superscript "ʕ" following the symbol of the sound being pharyngealised.
eg. [sˁ] [tˁ] [dˁ]
|Arabic has a number of normal and pharyngealised ("emphatic") consonant pairs (nb. for some Arabic dialects the secondary articulation is velar).
The consonant pairs are [s, sˁ], [z, zˁ], [t, tˁ], [d, dˁ].
Below are examples for [s, sˁ] only.
|[sad]||"to prevail"||[sˁ]||name of the letter|
|Even (Siberia) has pharyngealised vowels|
Co-ordinate or Double Articulation
Co-ordinate or double articulations occur when two simultaneous articulations of the same stricture occur simultaneously. Double articulations are most commonly double stops (both double oral stops and double nasal stops), typically involving a simultaneous lingual and a non-lingual stop articulation. There are also some (very rare) examples of double fricative articulations. One could also call the very common labial-velar approximant [w] a double articulation, in this case a double approximant, as the labial and the velar articulations both have the same stricture, approximant stricture.
Naming Double Articulations
Since double articulations consist of two primary articulations it is not appropriate to talk of "labialised velar stops" or "velarised labial stops" when talking of a double labial and velar stop. The normal name for such a double stop would be a "labial-velar stop". You will sometimes see a term like "labio-velar stop". Such a name appears to be too similar to names like "apico-dental" where the first part of the name refers to the active articulator and the second part of the name refers to the passive articulator of a single primary articulation. The term "labio-velar" might, therefore, be seen to imply an impossible combination of active and passive articulators. The published IPA standard refers to "labial-velar" approximants and fricatives and so it seems desirable to use a similar terminology with the double stop articulations.
nb. It is customary to place the name of the non-lingual articulation before the lingual articulation.
Since the two articulations are of the same rank order of stricture, that is they are both primary articulations, neither symbol is subscripted or superscripted, but both are tied together by a ligature. Usually the lingual articulation symbol is placed before the non-lingual symbol.
eg. [k͡p] [ɡ͡b] [ŋ͡m] [t͡ʔ]
|Idoma (West Africa) has three bilabial-velar double articulations|
|Eggon (West Africa?) has single stops, double stops and pairs of single and double stops in sequence, all contrasting|
|SePedi (Northern Sotho, Africa) has three double fricatives (but acoustic analysis suggests that they may actually be fricative sequences)|
An affricate is a stop with an extended and controlled fricative phase following the occlusion. The stop and aspiration or fricative phase are homorganic (same place of articulation). Affricates form one extreme of the following continuum:-
- Unaspirated stop
- Aspirated stop
- Affricated stop
In this continuum the stops have increasing strength and duration of aspiration, with an affricate effectively being a single phoneme with clear stop and fricative stages.
The affricate [tʃ] is very common, being found in about 45% of the world's languages. Dental and alveolar affricates are also very common.
Affricates are very often written with a ligature tying together the symbol for the stop and fricative, but sometimes it is customary to write the two symbols without the ligature as is often the case for [tʃ] and [dʒ] in English.
eg. [t͡ʃ], [d͡ʒ], [t͡s] (or [tʃ], [dʒ], [ts])
Affricated stops have been traditionally written in the form [ts] but this does not appear to be supported by the current IPA standard.
(Tones are omitted in the examples, below)
([ɕ] is a voiceless alveolo-palatal fricative)
|[tsa]||"to smack lips"||[ʈʂa]||"to pierce"||[tɕa]||"to add"|
|[tsʰa]||"to wipe"||[ʈʂʰa]||"to stick in"||[tɕʰa]||"to pinch tightly"|
- A context where a contrast is neutralised is one where a normal contrast on a particular language does not apply. For example, in English /p/ and /b/ contrast with respect to voicing (well, actually VOT). In the context /sp.../ at the beginning of a syllable the contrast is neutralised. That is the sequences /sp.../ and /sb.../ never contrast at the beginning of a syllable in English. The same is also true for the alveolar and velar oral stops. It can be said that in English the voicing contrast for oral stops is neutralised in this context. Click here to return to the text above.