(Material Cause)
Because the Palestrina Choir School holds as its primary end the beautiful singing for the Traditional Roman Liturgy; these very texts are of utmost importance to the potential chorister. 
The texts of the Mass, Liturgy of the Hours, hymns, motets and such are the material that the primary principles of beauty of tone and purity of vowel are applied to.
Habits of singing are deeply rooted and connected to speech habit and accent. The correct singing voice has a qualitative distinction from the voice used for everyday speech. Singing with the everyday speaking voice is the cause of a multitude of problems and vowel impurities as related to choral singing. The late Dr. Edwin Gordon put it like this:
“There is a difference between a speaking voice and singing voice, regardless of chronological age. Although range relates to that difference, more important is quality of voice. Because vocal folds are thicker for speaking than singing, a speaking voice sounds somewhat heavier than a singing voice. In speech, vocal folds are stretched very little, and the larger mass vibrates more slowly.”[1]
Also:
“A singing voice is more a matter of quality than range and is characteristically lighter and more flexible than a speaking voice. There is absence of tension, strain, and tightness in the sound, because when the throat feels open, tone flows with breath.”[2]
This distinctly different voice used for singing has also been called ‘voce di finte’ or ‘feigned’ voice by the Italians. Cornelius L. Reid in his book Bel Canto Principles and Practices writes:
“In reality the ‘feigned’ voice is nothing more than an outgrowth of the falsetto, and is an adjustment of this register giving it a somewhat ‘edgy’ quality of tone.”[3]
According to Reid this quality of voice is essential to perfecting vocal technique and uniting the registers of the head and chest voices:
“Thus, the ‘feigned’ voice becomes the only means by which the register action may be satisfactorily joined together, so that the singer becomes able at all times to maintain an absolute control over tone color and intensity variations at all pitch levels.”[4]
So in order to correct wrong singing habits one needs to develop a speaking habit pattern that is congruent with the ‘feigned’ or singing voice. This quality of voice is produced by the parallel approximation of the thinned out vocal cords.
In order to conceptualize the ideal of the correct singing voice it must first be presented as an aural perception. This is developed through repetition of liturgical text in dialogue fashion, such as when the teacher speaks and the 
student responds. The teacher speaks in an elongated Sotto voce with every word and syllable seated on the profound low “oo” vowel. Anthony Rooley’s contribution to the book: A Performer’s Guide to Renaissance Music, in his essay: “Practical Matters of Vocal Performance” wrote of the necessity of this exercise and described it thus:
“This is not a soggy unfocused kind of sotto voce, but a precise and detailed delivery, rather like looking at an object under a microscope.”[5]
When this is done with the thinned out razor like edges of the vocal cords it conditions the muscle memory for precise lyricism, articulation and intonation. It is this thinning out of the cords to the tiniest microscopic edges without undue weight or pressure that make the tone quality pure and beautiful. It also promotes the habituation of free and uninhibited singing. Such speech patterns, when presented and responded to as aural stimulus, (aural/oral)[6] will lower the larynx and raise the soft palate by intuition and instinct without the pupil ever having to fixate on the volitional control of the vocal organs.
Click for: Foundational Principles
Click for Article: Palestrina and the Perfecting of the Medieval Ideal of Music as Rational
[1]Gordon, Edwin E., Music Learning Theory for Newborn and Young Children, (Chicago, IL, 2013) 131.
[2]Ibid. 173.
[3]Reid, Cornelius L. Reid, Bel Canto Principles and Practices, (New York, NY, The Joseph Patelson Music House, 1971) 103.
[4]Ibid.
[5]Rooley, Anthony: “Practical Matters of Vocal Performance,” A Performers Guide to Renaissance Music, (Bloomington, IN, Indian University Press) 103.
[6] “The most elementary level in the hierarchy of discrimination learning is aural/oral. It is the necessary readiness for every other level of discrimination learning and for all levels of inference learning.” Gordon, Edwin E., Learning Sequences in Music: Skill, Content, and Patterns, (Chicago, IL, GIA Publications, Inc. 1993) 56.