A Tibetan child learning to write begins with ka, and so does every grammar, every code chart and every recitation of the alphabet. The thirty consonants are laid out in eight lines, seven of four letters and a last of two, and ka opens the first of them.
The letter writes a voiceless unaspirated velar stop and carries the inherent vowel a, so the bare sign is read ka and not k. That inherent vowel is not decoration. It is what makes the Tibetan script an abugida rather than an alphabet, and it is why the four vowel signs are described as replacing a vowel rather than as supplying one.
In the mechanics of the syllable, ka is a root letter and nothing else. It can carry a superscript above it, a subscript beneath it and a vowel sign as well, so a single stack built on ka can run to four written elements read as one syllable. What ka never does is the reverse work: it is not one of the five prefixes, not one of the ten suffixes, not one of the three superscripts and not one of the four subscripts. That restriction is worth noticing early, because roughly a third of the thirty letters do carry such duties and the difference between the two groups is the whole of Tibetan orthography.
Because it is first, ka has become the name of the thing. Ka kha ga nga stands for the alphabet in ordinary speech much as ABC does in English, and a grammarian reaching for a specimen consonant to demonstrate a rule reaches for ka by reflex.