The following two lines of Yoga Sutras are dedicated to mantras and power of the sound.
तस्य वाचकः प्रणवः ॥२७॥
1.27. tasya vācakaḥ praṇavaḥ
Sutras 27 and 28 tell that “the expression of that (Isvara) is OM (pranava)” and “the repetition of it (Om) in one mind’s eye allows one to experience it (the Isvara)”. Without getting into specifics of what the Om-sound means let us raise a more important question: why do mantras generally exist and how do they work?
It is only in the language like Sanskrit that the phenomenon of mantra-yoga and mantra in general could have emerged. Theoretically, all cultures have their own vibrational means of affecting human body. Shamans have their throat singing, karate-kas use the "“kiai” yell, Sufis perform zikrs while in Aikido they have a specific technique of cleaning the channels – 5 vowels sung in a certain succession, and so on. But all the said techniques fade in comparison to Sanskrit approach to mantra-yoga. And this is not without cause. The matter is that Sanskrit is a fairly unique language both historically and in terms of its structure. First of all, this is the language that has undergone practically no changes over the period of several millennia. A modern expert on Sanskrit after adjustment of several grammatical elements will be generally able to read Vedas that were written none knows when but probably more than three thousand years ago. He is already able to read classical Sanskrit, i.e. the medieval one; he can read epic Sanskrit – that is, Mahabharata, and so on.
The second thing is that for a great while Sanskrit existed in form of phonetic language, that is, unlike the rest of the world languages familiar to us it was oriented first of all to utterance and pronouncing rather than writing. The period between creation of Vedas and writing them down amounts to several thousand years. And all over this period the Vedas were being very carefully passed. Now how did they manage with such accuracy of communication in the absence of written records? For we know that even in case there are written records available there are usually errors occurred therein. Let us take, for instance, the Niconian epoch in Orthodoxy with its purposeful reformation. Why was it done? Because due to numerous scribers there came a great many of their mistakes accumulated in church books. Bear in mind – books! While here we deal with oral recitation. And in case of oral delivery the number of mistakes seems to be bigger. So in fact the main reason of why they have nevertheless managed to preserve all major records prior to the moment of their written fixation is because the bearers of this Tradition were extremely meticulous in their work with the language.
Let us start with the language title – “saṃskṛt”. The root “kṛ” means “to do/make” while together with the “saṃ” prefix it shall get the meaning of “self-made”. One can say that Sanskrit is the language that was initially “made for”, “adjusted to”, “customized for” etc. Another interesting fact is that the first grand philological and linguistic treatises in Sanskrit appeared more than 2.5 thousand years ago – at the time when our territory [modern Ukraine – translator’s note] was yet ridden around by the Scythians, and not even a germ of such science as academic grammar existed. The grammar that was in the 18th century established as academic grammar and philology in large part appeared due to the fact that Western science had got acquainted with Sanskrit treatises in which words had been for the first time classified into parts of speech, these parts of speech systemized, clustered and so on. One of the oldest tractates on Sanskrit, the Panini’s “Ashtadhyayi” (Eight Chapters) was created 2.5 thousand years ago and as of today it is still considered to be the canon of Sanskrit grammar. This is how ancient this tradition is! Moreover, because of very “energetic” Vishuddha this tradition has been continuously accomplished and refined. For instance, in the 6th century AD, the epoch of another “decline of Sanskrit scholarship”, a scholar named Bhatta decided he should fix this problem - and so he wrote Ramayana anew. Of course he wrote it in verse for in terms of grammar it was nothing but poetic form that they all by definition used. But not only that he wrote Ramayana anew – he wrote it in such a way that in every successive chapter there was only a certain verbal form or a case structure used. Thus even an undereducated person was able to learn in a rather entertaining form the basics of fairly complex Sanskrit grammar. Another example is “Amarakosha” (translated as “immortal casket”) that comes as the first ever linguistic thesaurus of the mankind (in Sanskrit, of course). This thesaurus included 10 thousand of key nouns that were classified into synonymic rows and semantic circles. It begins with telling us that first there comes the “heaven” and gives 9 denominations of it followed by 25 names of gods, 12 names of asuras, 45 names of Vishnu, 54 names of Brahma succeeded by demi-gods, people, 4 castes, subjects of their labour, household items, animals, plants, minerals and so on. Ten thousand of notions! And all of them defined in verses only. And now just think about the idea of drawing down a language thesaurus – and there it was, in the 6th century AD! Around the same time, 6th century AD, there was another interesting treatise written. Entitled as “Decinamamala” it is translated as (not verbatim) “The dictionary of Roots of Non-Sanskrit Origin”, that is, in fact, the dictionary of foreign words. Prior to it they had drawn the “Dhatupatha” list of roots in which the entire language was systematized as grounded upon 2 thousand of verbal roots from that ca. 64 million of words can be derived – all of them having their meaning and the right to exist, etc. It was in very ancient times that the language was systematized. And it was being refined and polished up by means of very capacious Vishuddha. I tell all this in order that the reader understands how grand the study of language in this culture was. So that of course Sanskrit scholars could not have passed over the very basis of the language – the sound, the understanding of which has been brought to perfection due to the aforementioned factors of necessity of using the language for pronouncing rather than writing.
Sanskrit is to the full extent a language that the science of today refers to phonocentric ones, i.e. initially focused on pronouncing of sounds rather than their writing. And Sanskrit alphabet is the only alphabet in the world that is arranged not by some arbitrary or symbolic principle yet by the principle of phonetics. That is, its letters are organized by the principle of dependence upon the area of larynx that they are uttered with.
Sanskrit grammarians have in a very sophisticated manner (probably not without help of yoga) reflected the experience of a person uttering these or those sounds. They have observed the following: how the vowel sounds ([ʌ], [o], [u] and others) emerge. The vowels appear because the air passes through the vocal tract and something buzzes there. Whatever vowel we take – be they [ə, ʌ, e, i, æ, ɔ, u] – things shall always happen like this. How do the consonants occur? It happens when there is constriction for the air: [k], [ʧ] and so on. We block the glottis in some place with our tongue, there occurs an obstacle, and this obstacle is articulated as a consonant. Now here comes the question: which zones in our mouth can we in general use for emitting sounds? We can make sounds by the back of the tongue where the soft palate begins – one can press the [back of the] tongue against it and articulate the simplest sound [k]. Note here that when we utter the sound [k] the tip of the tongue can be pressed against the front teeth with the tongue thereby atrophied, but the back of the tongue must be pressed a bit up.
There is another interesting point. All sounds except for one were considered to originate from the areas of one’s mouth, throat, head etc. The only sound emitted by means of a zone located somewhat deeper is the [h]. That is why every sound that can be made with help of the tongue can come either with [h] or without it. So in fact there are two sounds that we can produce in the k-zone – [k] and [kh], the latter one – by means of bringing in additional sound from the “bottom” – the larynx. Or, if these are not sounds but letters that we name, these shall be the ‘ka’ and ‘kha’letters. Now let us once again reflect on the position of the tongue in the larynx – these are the so called guttural sounds. If we now try to make the sound [k] with our tongue a bit relaxed and moved forward, it shall then sound like [g] that is articulated by exactly the same zone yet with one’s tongue relaxed! Try to reflex upon this [k] vs [g]. Besides we can emit a similar sound from approximately the same place by making use of our nasal apparatus. In order to do this one needs to cast the tip of the tongue backwards and utter the nasal [ng]. It is a separate sound that in Sanskrit is attributed with a separate letter [nasal]. Clear, isn’t it? It is lettered there by diacritic symbol ‘n’ with upper dot. In fact, the sound like this does exists so far as we can pronounce it. Can we do it? Yes, we can. Then why not introduce it into the alphabet? That is the beauty of Sanskrit. All sounds that we can utter are available in the alphabet. K-kh, g-gh, l-ṅ. Next, we can make sounds by means of cranial vault. The simplest sound that can be emitted by the cranial vault is [ʧ]. These sounds are referred to as palatal since they are made by means of the palate. So there is [ʧ] and accordingly there is [ʧh]. If we make a fainter sound with our tongue relaxed, we shall have [dʒ] and [dʒh] accordingly. From that very zone we can utter another nasal sound [symbolized by the latter specified as] ‘ña’. Can we make it? Yes, we can. That is why it exists and I have put it down in form of ‘ña’, but in fact it is just another type of ‘n’. Note here that sounds differ in different languages. For instance, take the English ‘table’ and the same word uttered with, say, Russian pronunciation. These are two different things. And this English [t] – it sounds in another way; it is just like the French “gargle” the sound [r] while the Germans trill it. What is the beauty of Sanskrit as a language? It is that it has all sounds in it. Even the sound [z] that is as if absent in Sanskrit is in actual fact just the reduced variant of hard [dʒ]. All other sounds that can be found in other alphabets in Sanskrit are either present or can be put down in one form or another.
All these we have uttered by means of the palate – [ʧ] - [ʧh], [dʒ] – [dʒh] and [ɲʌ] [the respective letters of Sanskrit alphabet been c, ch, j, jh, ñ – transl.note]. Now we go on. In the place where we set our tongue against the palate (when we do Bhastrika, for instance, or another exercise of the same type) we can make the tongue take a shape of circle – here the symbol of the sound coincides with the shape that it stands for – and articulate the sound [t ] that shall differ from the dental [t] of the Russian language. It is close to that in English, yet even softer. Respectively, it can sound like [th]. Soft [d] is [t]. ‘Ṭ’ and ‘ḍ’. And ḍh and ṇ - another nasal that very mush resembles the standard English –ing. Remember this [ŋ] in English transcriptions? This is how this ṇ sounds like. Next. We can utter five sounds in the same way, by means of teeth. How? We simply set out tongue against the teeth and say [t]. And [th], respectively. Soft [d] and respectively the [dh]. We can also pronounce another [n]. And now, which organ of speech is left? The lips, right you are! Now you see, we proceed from the depth and outwards, from the guttural, palatal, cerebral – i.e., cranial – and dental sounds to labial with our lips as “external” organ. Which sounds can we articulate by means of lips? We can utter [p] and [ph] respectively. If we pronounce [p] in soft manner what shall we have? Yes, it shall be [b] and, respectively, [bh]. We can use lips to utter one mooing sound ‘mmm’ – that is, [m]. And here it is - a nice table in a 5 to 5 format.
As a matter of fact this table is the sequence of consonants of the Sanskrit alphabet. It contains all possible sounds that a man can utter at expiration. Of course, for instance, in Africa there exist Khoisan languages with ‘clicking’ consonants emitted at inspiration. Russian oral speech also contains a number of archaic elements like ts-ts-ts [a ‘tut-tut’ saying – transl.note], lipsmacking and so on…
Sanskrit obviously has this phonetic regularity – the involvement of absolutely all areas of speech delivery, that is, articulation zones. But this is not all. The thing is that except for consonants Sanskrit also contains the so-called semivowels. What are semivowels? They are sounds that are similar to both consonants and vowels. They in a way sound like vowels because one can sing them. Their similarity to consonants lies in their association with closure of the vocal tract. For instance, what sound can definitely be sung? We can sing [rrrrr], [vvvv], [llll]. While [b] cannot be sung, just like it is also impossible to do with [p]. And now pay attention to the following amazing fact. Each of the vowels and its respective semivowels is uttered by means of placing the tongue at some zone of the vocal apparatus. The same very used for a row (varga) of consonants. While the sounds [h], [y], [r], [l], [v] are bija-mantras of chakras. That is, on all chakras there are bija-mantras represented by semivowels uttered by means of different zones. And the petals of every chakra are attributed with those very consonants that correspond approximately to the zone of its bija semivowel. There is a slight deviance in Vushuddha because of the Anahata petals that are big in number, but it is anyhow centered there. That is, the bija-manta of the entire chakra, the principal chakra mantra, correlates with a certain zone. The petals are juxtaposed to consonants present in this very zone, that is, uttered from this zone. And in a point of fact we understand at once that the logic of chakra system – at least, the psychosomatic logic – has been initially embedded within the structure of the Sanskrit alphabet. It is the logic of human vocal apparatus being the projection of the brain and psyche. That is why mantra-yoga is possible only in Sanskrit – the language in which almost all features of this apparatus are used.
Unlike prayer or singing it is not its sense and meaning that a mantra affects a person by. For instance, in case one prays that “give us this day our daily bread…” it shall be emotional and semantic addressing involved. The influence exerted by mantra is different. Its effect is stipulated by one’s using these or those zones of vocal apparatus that in their turn are related to some zones of the brain or, as it has been observed by Sanskrit grammarians, to some chakras. What happens when we utter a correspondent consonant? We activate a respective zone in our brain. It has been observed that when someone reads a text to himself: a) he makes micro-motions of the tongue similar to those occurring during reading aloud; b) the respective zones of his brain become active, as if he reads the text aloud. Moreover, the example of studies of fistulated people – the cases when the air no longer passes through the larynx – shows that when a person is not able to utter sounds his vocal apparatus nevertheless arranges itself as if for articulating them. The vocal apparatus can be characterized as closely related to late cerebral zones. There is a feeling of reverse association as well, though there’s been very few scientific research done in this sphere. So that we, just like ancient Indian scholars, rather ground ourselves upon empirical knowledge and reflection.