1100 SPEECH. [BOOK m. 



the teeth in sucn a way that a narrow channel is formed 

 between the upper incisors and the tip of the tongue curved 

 into a groove the sound is called S (soft (7) if without voice, 

 and Z if with voice. If the constriction be formed a little far- 

 ther back behind the front teeth by the approximation of the 

 tongue to the front of the hard palate, the sound uttered with- 

 out voice is Sh ; if voice be added the sound becomes the French 

 /, which we represent by z as in ' azure ' or by g as in ' badger.' 



If instead of being formed by the teeth the constriction be 

 carried farther back from the region of the teeth and hard 

 palate to that of the soft palate and fauces, a guttural aspirate 

 is formed. Without voice this is the hard Oh as in the Scotch 

 ' loch,' with voice the soft Ch. 



Y appears to be the vowel / (ee) used as a consonant, much 

 in the same way that, as stated above, W is U used as a conso- 

 nant. 



Lastly, a consonantal sound may be formed by the glottis 

 itself supplying a constriction but in such a way that the vocal 

 cords are not thrown into musical vibrations. When in utter- 

 ing a vowel we begin with the glottis not closed as in the 

 spiritus lenis but open, and send through the glottis an expira- 

 tory blast which creates irregular vibrations by friction before 

 the cords are brought into a proper position for their regular 

 vibrations, the result is the aspirate H, the spiritus asper. The 

 particularly powerful H of Arabic is produced by bringing the 

 processus vocales into contact with or near to each other but so 

 as to leave the cartilaginous glottis widely open and the mem- 

 branous glottis more or less open; at the same time the ven- 

 tricular bands are approximated, and the superior aperture of 

 the larynx is forcibly constricted. The expiratory blast driven 

 through the series of irregular passages gives rise to the irregu- 

 lar vibrations which constitute the sound, the vocal cords being 

 motionless or at least not giving rise to the regular vibrations of 

 voice. 



We have seen that in whispering no true voice is uttered, 

 no regular vibrations are generated in the vocal cords, though 

 the passage of the air through the glottis produces vibrations 

 which serve as the basis of the whisper, being modified by the 

 vowel cavity so as to form vowels. To these vibrations may 

 be added the vibrations of the consonants, so that a whispei 

 becomes complete though feeble speech. Since the irregular 

 glottic vibrations of a whisper are very weak compared with the 

 relatively powerful true vocal vibrations, the distinction between 

 consonants with voice and without voice is in a whisper largely 

 obscured; it is difficult for instance in a whisper to distinguish 

 between P and B. 



