Ranjana La

La is the twenty eighth consonant of the Ranjana varnamala and the third of the four semivowels, the lateral made by letting air pass along the sides of the tongue while the tip touches the ridge behind the teeth. The Indian analysis pairs it with the vocalic l that stands among the vowels, in the same way it pairs ya with i, ra with vocalic r and va with u, and the pairing is stated as a rule about how sounds alternate rather than as a similarity noticed after the fact. Sanskrit uses the consonant widely, unlike its vocalic partner, which is one of the rarest letters in the whole chart.

In Ranjana it is written as any consonant sign, carrying the inherent short a, hanging from the head stroke and taking its vowel marks around it. It joins with a following consonant into a stacked form where no vowel intervenes, on the same terms as the letters of the grid. The letter is also one of the three whose reduced forms appear most often beneath other consonants in the northern scripts, so a reader working through a dense passage will meet its joined shape long before meeting an isolated instance of it. Learning the reduced form is therefore part of learning the letter rather than an advanced refinement, and the same is true of ra and va.

The four pairings that link a semivowel to a vowel are among the most elegant things in the Indian analysis, and they explain why the vocalic liquids were classed as vowels in the first place. If ya is the consonantal counterpart of i and va of u, then a system that recognises those pairs must ask what the counterparts of ra and la are, and the answer is the vocalic r and vocalic l. The rare letters at the head of the chart exist because the pattern demands them. Once that is seen, the whole arrangement of the varnamala reads as a single argument rather than as two lists joined together, and the rare vowels stop looking like curiosities.

For Ranjana the point is not decorative. The script inherited the argument along with the letters, and its charts, its recitation and its teaching carry all of it, which is why a Ranjana alphabet has fourteen independent vowels rather than the six a European reader would predict and thirty three consonants rather than a number fitted to any particular language. It also shows why the chart cannot be shortened without damage. Removing the rare vowels would leave the semivowel pairings with nothing to pair against, and the argument that holds the two halves of the order together would no longer be visible on the page.