Topics:
The sibilants
Palato-alveolar sibilants
Physical description
The learning problem

The Sibilants [ s ], [ z ], [ ʃ ], [ ʒ ]

Among the English fricatives, [ s ] and [ ʃ ] and their 'voiced' counterparts, [ z ] and [ ʒ ] are characterized by very audible high-pitched friction. These sounds are noticeably more 'fricative' than other English fricative sounds such as [ f ] and [ v ] or [ θ ] and [ ð ], and for this reason are referred to as sibilants.

English sibilants are of two basic types. On the one hand, there are so-called alveolar sibilants, [ s ] and [ z ], which occur in words such as sea, boss, zoo, and rose. On the other hand, there are the palato-alveolar sibilants [ ʃ ] and [ z ] (as in she and measure).

Palato-alveolar sibilants

[ ʃ ]

Remember that the [ s/ʃ ] contrast is one of the major phonological contrasts of English. Acquiring authentic-sounding palato-alveolar sibilants is far more important than remembering to use liaison at word boundaries, or using native-sounding vowels or intonation patterns. Start by practicing your palato-alveolars in isolation.

If you have difficulty in producing an authentic-sounding [ ʃ ] you should check these points:
- [ ʃ ] is more retracted sound than [ s ]. If you start by saying [ s ] and then move to [ ʃ ] you should be able to feel the tip of your tongue slide back up the gum ridge, from its [ s ] – position just behind the front teeth to a more retracted position on the steeply-rising part of the ridge, where it goes up towards the hard palate. Check that when you say [ ʃ ] you can insert a finger behind your front teeth, without touching the retracted tongue tip.

- The [ ʃ ] sound is typically produced with a marker protrusion, or pouting-out of the lips. Check your degree of lip-rounding using a mirror. You may find that using more rounding helps move your [ ʃ ] sound from ‘terävä-s’ to ‘suhu-s’.

- The [ ʃ ] sound is produced with the whole ‘front’ of the tongue raised towards the hard palate, as if you were saying [ j ] at the same time. You may find it easier to produce [ ʃ ] if it is immediately followed by
[ j ]; as in these sequences:
wash your hair
Finnish year
cash your cheque

- If all else fails you can probably force ‘hushing’ sound by clamping your jaws together when say [ ʃ ] or at any rate bring them closer together than you usually do in the course of speech. If you do this, you are forced to draw in your tongue tip and sides to prevent their being bitten. With the whole area of the tongue blade drawn inside the area bounded by the teeth, the channel down the centre of the tongue must necessarily be larger and looser than it is for [ s ]; since the small, ‘tight’ channel used for [ s ] only requires a limited area of tongue surface, and leaves the edges overlapping the edges of the teeth. An extensive, wider channel results in lower air pressure and a more ‘hushing’ sibilant sound.

Back to the short version

[ ʒ ]

This sound is the voiced counterpart of [ ʃ ]. If you can say [ ʃ ], you can achieve [ ʒ ] by starting off with [ ʃ ] and “switching on” your voice during production.

Practice saying [ ʃʃʃʃʒʒʒʒʃʃʃʃʒʒʒʒʃʃʃʃʒʒʒʒ ]. Distribution is limited to word-medial position, except in the case of small number (mainly French) loan words, such as prestige, beige, rouge, barrage, collage, where the sound occurs word-finally. It occurs word-initially in the Italian loanword gigolo.

Spelling
When the sound occurs word-finally in a French loan word, the spelling is –ge, as in French. Otherwise, the spelling to look for is the following:

- ‘si’ occurring in the sequence –SION in words such as decision, occasion, conclusion, collusion, erosion, delusion, explosion, evasion, innovation; fusion and its compounds confusion, infusion, suffusion, diffusion, profusion; vision and its compounds provision, division, revision, supervision. Note however that the –SION sequence often represents [ ʃn ].

- ‘su’ or ‘zu’ in the sequenes –SURE, -ZURE, in words like closure, exposure, erasure, pleasure, measure, leisure, seizure.
- ‘su’ in the sequence –SUAL in usual, casual, visual.

Phrases with [ ʒ ]; all occurrences of [ ʒ ] are in bold.
illusions of power
leisure activities
the prestige of the President
casual revision

Back to the short version

Physical description

This is complicated by the fact that, in spite of the traditional labelling of these sounds according of place of articulation, it is probably the shape of the tongue, which is critical in producing the characteristic sibilant sound (or either type). Sibilants are produced by channelling the air stream along a groove that runs down the centre of the blade of the tongue (see The diagram); and for this reason, these sounds are sometimes termed ‘grooved fricatives’.

Further, the ‘correctness’ of a sibilant of either type is primarily a matter of how it sounds; and different speakers may produce the right sounds by slightly different articulatory means. Some speakers, for example, use the tongue tip to produce the alveolar sibilants; while others use the blade . To produce a palato-alveolar, the blade is always used; but while some speakers may tuck their tongue tips down behind their lower front teeth, other hold the tongue tip up near the alveolar ridge .

Back to the short version

Alveolars
Alveolar sibilants are produced in the following manner: with the sides of the tongue in firm contact with the side teeth, the airstream is channelled along a narrow groove in the tongue blade. Friction occurs between the tongue blade (or with some speakers, the tip and blade) and a very small area of the alveolar ridge located right behind the speaker’s two front teeth.

The tongue blade is hollowed for [ s ], while it is domed for the contrasting [ ʃ ] sound. With [ s ], there is usually no free space between the tongue tip and the front teeth; although this may not be true for those speakers who use their tongue tips rather than blades to make this sound.

The articulation involves considerable muscular tension in order to maintain the narrow but deep groove in the tongue blade. This is why drunkards may sometimes be heard to substitute the [ ʃ ] for [ s ] in their attempts at speech, since alcohol serves to relax the muscles generally. The alveolar sibilant sound is much higher-pitched (that is, it includes more high-frequency sound) than the palato-alveolar sound.

Back to the short version

Palato-alveolars
In the production of palato-alveolar sibilants, the tongue tip is retracted, so that the speaker can easily insert a fingertip between his/her front teeth and the tongue tip. The tongue blade is raised and approaches the hard palate approximately at the point where the alveolar ridge merges into the main body of the hard palate. Thus the tongue blade follows the upward rise of the alveolar ridge towards the hard palate (instead of being hollowed, and curving away from it as with [ s ]).

The channel in the tongue blade is much wider, and also shallower, than is the case with [ s ]. Air pressure is lower, and the escape of air is more diffuse, with the friction occurring over a wider area of the tongue and the alveolar ridge. The friction, while still very audible, is considerably lower in pitch than in the case of the [ s ] sound.

The front of the tongue is also raised towards the hard palate; thus the articulation may be described as ‘palatal’. It is for this reason that native English speakers usually substitute [ ʃ ] for [ s ] when /s/ is immediately followed by the palatal continuent [ j ], as an effect of assimilation; thus expressions such as I’ll miss you or this year are typically pronounced with palato-alveolar rather than alveolar sibilant sounds.

The palato-alveolar sibilants of some speakers may be accompanied by slight lip-rounding in all positions, while for others this only occurs when the sound is followed by a rounded vowel.

Back to the short version

The learning problem

In the native Finnish system there is only one sibilant, Finnish /s/, to correspond to the English /s/, /z/, /ʃ/, and /ʒ/. This means that the Finnish learner of English has to go from his/her own less- differentiated native system to a more highly-differentiated one, and is faced with problems of perception and production.

Since there is only one sibilant in Finnish, speakers have certain amount of freedom in the precise quality they give it; and so Finnish /s/ may sometimes have a more palato-alveolar quality than English /s/ does. Nonetheless, the Finnish sound is broadly acceptable as a token for English /s/; because the empirical fact are that except in rare individual cases, Finns have very little trouble with the English alveolar sound. Finnish learners’ problems begin when they try to produce the English palato-alveolars [ ʃ ] and [ ʒ ].

Finns often have great difficulty in maintaining a consistent ‘palato-alveolar’ quality, and preventing an ‘alveolar’ sound from creeping in. But it is much rarer for a Finnish learner’s alveolars to be spoiled by being too palato-alveolar; and when this does happen, it is usually the result of interference from neighbouring palato-alveolar sound. For example, the learner may pronounce a word such as position with the alveolar and palato-alveolar sounds reversed; or produce a sound intermediate in quality between the alveolar and palato-alveolar values, to represent both sibilant sounds.

The sibilants, in fact are the only English consonants that present some Finnish learners of English, at University level, with a problem at the level of primary acquisition. That is to say that there are some cases in which learners find it hard to produce acceptable English palato-alveolars even in isolated words (for example, if they are asked to read from a word list). Many more find it hard to produce good palato-alveolar sibilants, of any type, in the context of sentences read aloud in laboratory drills; even if the sentences are grouped together in lists which focus exclusively on one particular sound (such as [ ʃ ]).

Back to the short version

Theory adapted from: English Pronunciation (opetusmonisteita) 3rd edition by Michael Peacock.